Forest Mensuration: An Outline of

An Outline of
Forest Mensuration
with Special Reference to Australia
L. T. Carron
L. T. Carron commenced his career
with the Queensland Forestry Depart
ment in 1938, and after war service
graduated in forestry at Queensland
University in 1948, taking his
M. Sc.For. at that University in 1955.
He lectured at the Australian Forestry
School, and is now Senior Lecturer in
Forestry at the Australian National
University. In 1958 he spent a year at
Oxford University and in 1967 was
Leverhulme Visiting Fellow at the
University of Tokyo. Forest mensura
tion and statistics have long been his
main-interest and he has- published a
number of papers onkhö'«ubjeet.'*'•** ",v
*r *» •* • »
; : ;? Wi j;
Price $A6.00
This book was published by ANU Press between 1965–1991.
This republication is part of the digitisation project being carried
out by Scholarly Information Services/Library and ANU Press.
This project aims to make past scholarly works published
by The Australian National University available to
a global audience under its open-access policy.
An Outline of
Forest Mensuration
with Special Reference to Australia
L. T. Carron
Australian National University Press

ä ä Canberra 1968
F i r s t published 1968
T his book is co p y rig h t in a ll c o u n trie s s u b sc rib in g to th e B ern e

Convention; re p ro d u c tio n in w hole o r in p a r t w ithout w ritte n p e r m iis s s io n
of th e p u b lish e r is fo rb id d en
L ib ra ry of C o n g re ss C atalo g C ard no. 68-54900

N ational L ib r a r y of A u s tra lia re g . no. AUS 68-2022
O. D. C. 5
F o r my parents and my family
Preface
Text books on fo re s t m ensuration are available fo r many co u n tries

such as the United States of A m erica, W est G erm any, F ran ce and
Italy. Many m ethods and techniques of fo re st m en su ratio n used in
A u stra lia have been adapted from those of w estern Europe and N orth
A m e ric a, but they have been so modified to su it A u stralian conditions
that u n derg rad u ates in fo re s try in this country need a text of
re fe re n c e to local p ra c tice and its background. No such text is
available in A u stra lia at p re se n t and this book attem pts to fill the need,
at le a s t in p a rt.
It is difficult to decide to what extent and depth the subject should
be covered. F o re s t m ensuration is an applied su b ject which includes
a sp e c ts of pure and applied m athem atics, s ta tis tic a l m ethods, land
surveying, the use of a e ria l photographs, and data p ro c e ssin g and
com puting. A uthors have v aried considerably in the extent to which
they have included these aspects and the m anner in which they have
p re sen te d them . It has been assum ed in this text th at th ese v ario u s
a sp ects w ill be tre a te d sep arately but c o n cu rren tly in an undergraduate
c o u rse in fo re s try as th eir application is not lim ited to fo re s t
m en su ratio n . To have given them adequate tre a tm e n t even for the
p u rposes of fo re s t m ensuration would have m eant enlarging the
p re se n t text to unreasonable siz e . Experience in le ctu rin g on the
subject over many y e a rs at the A ustralian F o re s try School and the
D epartm ent of F o re s try of the A ustralian N ational U niversity, and
continual contact with pro fessio n al p ra c tic e , suggest that the scope
and depth of this text a re adequate for undergraduate study. It should
be s tre s s e d that fo re st m ensuration is a p ra c tic a l m a tte r; th is tex t is
intended to serv e as background to p ra c tica l in stru ctio n in the fo re s t,
not to re p lace it. It has been assum ed that the student would have
a c c e ss to a lib ra ry so that e x tra c ts of lite ra tu re have been given only
when they se rv e d a p a rtic u la r purpose.
vii
v iii P refac e
To the m any colleagues and students, in A u stra lia and e lse w h e re,
who have contributed so much to this book I offer grateful
acknow ledgm ent. I am p a rtic u la rly indebted to P ro fe s s o r J . D. Ovinggton
(D epartm ent of F o re s try , A u stralian N ational U niversity), M e s srs
T . E . Edw ardson (D epartm ent of F o re s try , Oxford U niversity),
N. B. Lewis and A. Keeves (Woods and F o re s ts D epartm ent, South
A u stra lia ), and P . R . Law rence (F o re sts C om m ission of T asm an ia),
who m ade valuable suggestions fo r im proving the tex t. They a r e , of
c o u rse , in no way responsible for its sh ortcom ings.
I am also grateful to the staff of the A u stralian N ational U n iv e rs ity /
P r e s s fo r th e ir e d ito rial a s s is ta n c e , to M rs B. D riv er, who typed
the m a n u sc rip t, M rs L. H arm ey fo r s e c re ta ria l a s s is ta n c e , and to Mvfr
K. Lyon who drew the fig u res.
L. T. CJ.
Contents
P re fa c e vii
1 In tro d u ctio n 1
2 U nits of M e a su re m e n t of M ajo r F o r e s t P ro d u c ts 5
3 Logs 9
4 T re e G irth , D ia m e te r, and S ectional A re a 17
5 T re e H eight 26
6 F o rm and T a p e r 36
7 B a rk and Crow n 45
8 Stem Volum e 51
9 G row th and In c re m e n t of Individual T r e e s 63
10 Stand V a ria b le s 69
11 Stand T a b le s and Stand B asal A re a 71
12 Stand H eight 86
13 Stand V olum e 93
14 B ole A re a 123
15 Site Q uality 127
16 Stand D ensity 135
17 In c re m e n t and Y ield 146
18 F o r e s t In v ento ry 171
R e fe re n c e s 207
Index 223
ix
Figures
C h ap ter 5
5:1 T re e heigh t m e a s u re m e n t 29
5:2 L eaning tr e e 32
C h ap ter 8
8:1 Individual tr e e v o lu m e, se c tio n a l m ethod 54
8:2 Individual tr e e v o lu m e, p e rc e n tile m ethod 57
8:3 Individual tr e e v o lu m e, ta p e r lin e m ethod 60
C h a p te r 9
9:1 C h a r a c te r is tic s of grow th of individual tr e e s 67
C h ap ter 11
11:1 D .b . h .o . b . d is trib u tio n , u n ev en -ag ed stan d of P . ra d ia ta 73
11:2 D .b . h .o . b . d is trib u tio n , u n ev en -a g ed eu caly p t f o r e s t 75
11:3 D .b . h .o . b . d is trib u tio n , u n ev en -ag ed eu ca ly p t f o r e s t 77
11:4 G .b . h .o . b . d is trib u tio n , u n ev en -a g ed v irg in r a in - f o r e s t 77
11:5 P rin c ip le of angle count m ethod 79
C h a p te r 12
12:1 Stand heigh t c u rv e , ev e n -a g e d P . ra d ia ta 88
12:2 Stand heigh t c u rv e , u n ev en -ag ed eu ca ly p t f o r e s t 92
C h a p te r 13
13:1 V o lu m e -b a sa l a r e a lin e s , P . ra d ia ta 98
C h a p te r 17
17:1 In c re m e n t in d . b . h . o . b . , P . r a d ia ta 158
17:2 In c re m e n t in d. b .h . o. b . , re d and w hite m ahogany, tallo w
wood, and blue gum 159
xi
Tables
C h a p te r 4
4:1 E r r o r s in c a lc u la tio n of tr e e s e c tio n a l a r e a 21
C h a p te r 5
5:1 E r r o r s a s p e r cen t of tr u e h eig h t 32
C h a p te r 7
7:1 B a rk th ic k n e s s a t b r e a s t h eig h t, P . ra d ia ta 47
7:2 B a rk v o lu m e, P . r a d ia ta 49
C h a p te r 11
11:1 D . b . h . o . b . freq u en cy d is trib u tio n of an unthinned,
e v e n -a g e d stan d of P . ra d ia ta 73
11:2 Stand ta b le s fo r an u n ev en -a g ed stan d of m ix ed
e u ca ly p ts 75
11:3 Stand ta b le fo r an u n ev en -ag ed stan d of v irg in
r a in - f o r e s t 78
C h a p te r 13
13:1 T r e e volum e ta b le , P . ra d ia ta 107
13:2 V olum e u n d e r - b a r k to 4 in d . u . b . a s p e rc e n ta g e
of to ta l volum e 120
C h a p te r 14
14:1 T re e bole a r e a ta b le 125
C h a p te r 17
17:1 A nnual ra in fa ll, C a n b e rra 147
17:2 S tan d ard m ethod of d e riv in g d . b . h . o . b . in c re m e n t
fro m stan d ta b le s 150
17:3 S m ith e rs ’s m o d ificatio n of the m ethod of d e riv in g
d . b . h . o . b . in c re m e n t fro m sta n d ta b le s 152
17:4 P ro d a n 's m ethod of d e riv in g d . b . h . o . b . in c re m e n t
fro m stan d ta b le s 153
17:5 D . b . h . o . b . in c re m e n t fro m stan d ta b le s of s u c c e s s iv e
x iii
xiv T ables
m e asu re m e n ts, P ro d a n 's method com pared with
actual values 155
17:6 D .b .h .o .b . in crem en t from stand tab les of
su ccessiv e m e a su re m e n ts, actual in crem en t
com pared with stan d ard and P ro d a n 's m ethods 156
17:7 Stand tables of su cc e ssiv e m easu rem en t of
c y p re ss pine 157
C hapter 18
18:1 S trip s, R oses T ie r; e r r o r s as p ercen tag e of tru e
volume 190
18:2 S trip s, R oses T ie r; d istrib u tio n of e r r o r s as
p ercentage of tru e volum e 190
18:3 S trip s, R oses T ie r; e r r o r s as num ber of tr e e s
and as volume of whole a re a 191
Introduction
A f o r e s t e r is c o n c e rn e d with m anaging f o r e s t s . I r r e s p e c t i v e of the

kind of f o r e s t and p u rp o s e of m a n a g e m e n t - pro d u ctio n , p ro te c tio n ,
o r r e c r e a t i o n - he need s quantitative in fo rm a tio n about w hat he is
m anaging and how it changes with tim e . It is the function of f o r e s t
m e n s u r a tio n to p ro v id e th is.
M en su ratio n is th a t b ra n c h of m a th e m a tic s which d e a ls with the
m e a s u r e m e n t of lengths of lin e s , a r e a s of s u r f a c e s , and volum es of
so lid s; f o r e s t m e n s u ra tio n is th is kind of m e a s u r e m e n t fo r individual
t r e e s and f o r e s t s . This is the m o s t a p p r o p r ia te kind of m a th e m a tic s
b e c a u s e a f o r e s t m ight be r e g a r d e d fro m the q uantitative viewpoint as
a n u m b e r of solid bodies u p rig h t on a plane s u r f a c e . If th e se solids
w e r e few in n u m b e r, of s im p le , id e n tic a l, and unchanging shape and
s iz e , and on a r e a d ily a c c e s s ib le plane s u r f a c e , then the application
of the a p p r o p r ia te p r in c ip le s of m e n s u r a tio n would be s tr a ig h tf o r w a r d .
T r e e s in f o r e s t fo rm a tio n , h o w ev er, a r e u su ally n u m e ro u s , of
co m p le x sh ap e, differin g from each o th e r and continuously changing
in s iz e , and often in a r e a s w h e re a c c e s s is difficult. To apply the
p r in c ip le s of m e n s u r a tio n , the f o r e s t e r has to d raw on o th e r d is c ip lin e s .
Since quantitative inform ation is of little value u n le s s r e la te d to
a r e a o r s p a c e , he b e c o m e s involved w ith the definition of f o r e s t
b o u n d a rie s on land and plan by s u rv e y in g and w ith the d e te rm in a tio n
of a r e a . A e r ia l photographs a r e v e r y useful fo r m apping f o r e s t s ,
in t e r p r e ti n g qualitativ e in fo rm atio n , and getting quantitative
in fo rm atio n by p h o to g r a m m e t r y , so th a t the u se of a e r i a l photographs
is also an e s s e n tia l p a r t of f o r e s t m e n s u r a tio n . S ta tis tic a l Methods
a r e of p a r t i c u l a r im p o rta n c e . Since f o r e s t s a r e ex te n siv e and
individual t r e e s a r e often la r g e the f o r e s t e r m u s t, fo r econom ic
r e a s o n s , d e r iv e his quantitative in fo rm a tio n fro m a s a m p le . He is
a lso p a r t i c u l a r l y c o n c e rn e d with d e te r m in in g re la tio n s h ip s am o n g st
1
2 Outline of F o re s t M ensuration
tre e v a ria b les and stand v a ria b le s in o rd e r to u nderstand the behaviour
of tre e s and to estim ate a v a ria b le which is difficult o r expensive to
m easu re from another which is sim ple o r cheap to m e a su re . These
a re aspects of s ta tis tic a l m ethods. Many m en su ratio n p ro je c ts involve
la rg e am ounts of data and so re q u ire m odern m ethods of data
p ro cessin g and high speed com puting.
Depending on w hether the inform ation is to m eet a biological o r
b usin ess re q u ire m e n t, on the location, extent, and kind of fo re s t, and
on the money and labour available, so the m ost a p p ro p riate m ethod of
m easu rem en t and handling the data w ill d iffer and w ill draw v ario u sly
from the m athem atics of m en su ratio n , the p rin c ip le s and techniques
of surveying, the use of a e ria l photographs, the concepts and
m athem atics of s ta tistic a l m ethods, and p ro c e d u res of data p ro c e ssin g
and computing. To d eterm in e the m ost a p p ro p ria te m ethod, the
fo re s te r m ust know what m ethods a re av ailab le, how to apply them ,
and th e ir re la tiv e advantages and lim itatio n s in v ario u s c irc u m sta n c e s.
The m ore fa m ilia r he is with them and with the techniques of applying
them , the m ore effective w ill be h is contribution tow ard the solution
of a m anagem ent problem .
The m a tte r of e r r o r is of p a rtic u la r im p o rtan ce. If m e asu rem en t is
thought of as an exact determ in atio n of the size of som e p ro p e rty in
te rm s of a unit which re la te s to a stan d ard , then the 'm e a su re m e n t'
of fo re st m ensuration is m ost often ra th e r an e stim ate - an inexact
determ ination. The difference betw een a tru e value and an inexact
determ ination is an e r r o r . It is e sse n tia l in any m en su ratio n
procedure to reco g n ise the likely so u rc es of e r r o r , to c o n sid er th e ir
effects, and to evaluate the kind and extent of e r r o r a sso c iate d with
an e stim ate .
In fo re s t m ensuration th e re a re se v e ra l kinds of e r r o r in the
general sen se. One kind of e r r o r is the m istak e such as when a
figure is read in c o rre c tly from an in stru m e n t, called in c o rre c tly ,
h eard or reco rd ed in c o rre c tly , o r a figure is w rongly tra n s fe rre d or
in te rp re te d from re c o rd s. Such e r r o r s a re cau sed by c a re le s s n e s s
o r casu aln ess and can usually be avoided. C ertain conditions, both
in field and office, in c re a se the likelihood of th is kind of e r r o r ,
how ever, and should be anticipated in o rg an isatio n . M istakes may
be made through ignorance of re q u ire m e n ts o r lack of a p p ro p riate
training. This so u rce of e r r o r is the re sp o n sib ility of the su p erv iso r.
In c o rre c t arith m etic is a m a jo r so u rce of m ista k es but th e re is no
excuse fo r slipshod calcu latio n s, w hether due to ineptitude for
Introduction 3
arith m e tic or to the tedious n ature of som e calcu latio n s. M istakes can
be reduced to negligible prop o rtio n s by c a re and checking, the m ost
im p o rtan t p re lim in a ry check on any re s u lt being to te s t w hether it
seem s sen sib le. E very e ffo rt should be made to design calculation
p ro c e d u res with b u ilt-in checks.
A nother kind of e r r o r is accidental e r r o r . Small differen ces from
the tru e value that occur unavoidably, despite ev ery p recau tio n , due
to lim itatio n s of in stru m e n ts, inconstant environm ental conditions o r
lack of perfection in sight o r touch, com e in th is categ o ry . U sually
they a re unim portant because they tend to v ary in sign and size and
thus to balance out. The danger is in w rongly a sc rib in g to th is
categ ory e r r o r s which a re m ista k es or a re due to b ia s.
B ias im plies a tendency to d ep art in a p a rtic u la r d irectio n from the
tru e value sought, for exam ple an in stru m en t with graduations of
in c o rre c t size o r with an in c o rre c t origin would give a b iased reading.
In sam pling, ’an estim ate is said to be biased if rep etitio n of the
p ro c e d u res by which it was derived. . .would give re s u lts which, when
averaged over an extended s e r ie s of re p e titio n s, d iffer from the tru e
value of the quantity e stim a te d ' (Finney, 1947).
If the size of a p a rtic u la r v a ria b le of a heterogeneous population is
estim ated by sam pling, the e r r o r that m ay be a sso c iate d with the
e stim a te , p u rely due to the sam pling, is re f e r r e d to as sam pling
e r r o r . Under c e rta in conditions of sam pling, the sam pling e r r o r
a sso ciated with the e stim ate of the arith m e tic m ean of the population
is the standard e r r o r ,the calculation of which involves the concept
(though not the p ra c tice ) that the sam pling of the a rith m e tic m ean is
c a rrie d out a la rg e num ber of tim es and th ese e stim a te s of the mean
v ary am ongst th em selv es. The te rm p re c isio n r e f e r s to this
variation; ’by p re c isio n is m eant . . . the re p ro d u c ib ility of an estim ate
in rep eated sam pling conceiving an infinite rep etitio n of the sam pling
p ro c e d u re' (Finney, 1947). A p re c ise m e asu re m e n t is one which
re p e a ts a value co n sistently. It is im portant to a p p reciate that a
p re c is e m e asu re m e n t m ay not give the tru e value - no m a tte r how
rep ro ducible the value a bias in the p ro ced u re m ay p rev en t the
e stim ate from e v er being the tru e value. One can thus have e stim a te s
which a re p re c is e and unbiased, p re c ise and b iased , not p re c ise and
unbiased, and not p re c ise and biased.
In som e c a s e s in fo re s try the m a te ria l is so irre g u la r or the
environm ent such that an exact determ inatio n of size cannot be
obtained. In th ese c a s e s an e stim ate m ust be accepted. In som e
4 O utline of F o re s t M ensuration
c a s e s e stim a te s have to be accepted because of the lim itatio n s of tim e
and money. In many c a se s an approxim ation w ill se rv e the purpose
p erfectly w ell. The im portant thing is to know when one is estim atin g
ra th e r than m easu rin g , when one is m aking an e stim ate d e lib e rately ,
when one is accepting an e stim a te because one has to, and in a ll c a s e s
what the kind and size of the e r r o r is likely to be. The re c o rd should
not be made in such a way that a fa lse stan d ard of m e asu rem en t is
im plied nor should one re a d into a re c o rd a stan d a rd which is not
im plied.
F o re s t m ensuration is to a la rg e extent the handling of fig u res and
it dem ands sy stem atic re c o rd in g , c le a r notation, adequate labelling
of inform ation, and re s p e c t fo r the perm anence of the re c o rd .
2 Units of Measurement of Major
Forest Products
T re e s and th e ir products a re m e asu re d in v ario u s units depending on

country and locality, and on the p a rtic u la r dim ension concerned.
T h e re is an enorm ous range of units throughout the w orld and v arious
g lo s s a rie s of these; Jennings (1965), fo r exam ple, has com piled one
fo r the log and tim b e r tra d e of G reat B ritain . The units with which an
A u stra lian fo re s te r should be fa m ilia r a re as follow s.
L ength. In co u n tries using the English sy stem of u n its, such as

A u stra lia , the units for m e asu re m e n t of length a re the foot and the
inch. F o r m ost p u rp o ses, such as length of log and height of tr e e ,
the value is rounded (usually down) to the n e a r e s t foot.
In c o u n trie s using the m e tric sy ste m , the units a re the m e tre and
c e n tim e tre .
G irth, d iam eter and sectional a r e a . In A u stra lia , as in m ost

c o u n trie s using the English system of u nits, g irth and d iam eter of
tr e e s o r logs a re m easu red in feet, inches, and d ecim als of an inch.
The equivalent sectional a re a is e x p re ssed in sq u are feet and d ecim als
of a sq u are fo o t.
In G re at B ritain , for m ost standing tre e m e asu re m e n t, the units for
g irth o r d iam eter a re feet, inches, and frac tio n s o r d ecim als of an
inch, the equivalent sectional a re a being e x p re ssed in sq u are feet and
d ecim als of a square foot. F or lo g s, the c irc u m fe re n ce is m easu red
but often e x p re ssed as q u a rte r g irth (a q u a rte r of the circu m feren ce)
in inches and frac tio n s of an inch according to the Hoppus system (see
ch. 3). The size of standing tre e s is also so m etim es e x p re ssed by
q u a rte r g irth in which case the equivalent sectio n al a re a m ay be
e x p re ss e d in q u a rte r girth (Hoppus) square feet.
In co u n trie s using the m e tric sy ste m , g irth and d iam eter a re
5
e x p re ssed in m e tre s and c e n tim e tre s , and sectional a re a e x p re ssed
in square m e tre s and decim als o r in sq u are c e n tim e tre s.
T ree vo lu m e. In A u stra lia the volume of the standing tre e is m ost

com m only e x p re ssed in cubic feet and d ecim als of a cubic foot. Where
the volume of the tre e is a s s e s s e d as that of the log which can be cut
from it, and it is conventional to e x p re ss the volum e of a log in
su p erficial feet or Hoppus su p erfic ial feet (see ch. 3), then the tre e
volume is so e x p re ssed . The su p erfic ial foot is re p re se n te d by a
solid one foot by one foot by one inch, th e re being 12 su p erfic ial feet
to the cubic foot. T h ere a re 12 Hoppus su p erficial feet to the Hoppus
cubic foot (see ch. 3).
In New Zealand, the general p ra c tic e is to use the cubic foot.
In G reat B rita in , the Hoppus cubic foot and the cubic foot a re used
v ariously.
In the U. S. A. , the cubic foot is now fa irly common for re s e a rc h
and m anagem ent p u rp o se s. W here the volum e of the tr e e is a ss e ss e d
on a log b a sis the unit used is usually the board foot which is the sam e
as the su p erficial foot; but the tre e contents may in th is c a se be
a ss e ss e d as the volum e of potential sawn tim b e r by re fe re n c e to a
log ru le (see ch. 3).
Where the m e tric system is used, tre e volume is u sually e x p re ssed
in cubic m e tre s and decim als.
To avoid confusion betw een the stan d ard cubic foot and the Hoppus
cubic foot, and between the stan d ard su p erfic ial foot and the Hoppus
su p erficial foot, p a rtic u la rly since people who a re used to Hoppus
m easu re a re apt to om it the w ord 'H oppus', the stan d ard cubic and
su p erficial foot a re often qualified by the w ord 'tr u e ', thus 'tru e
2 ubic foot' o r 'cubic foot (tru e)' and 'tru e su p erficial foot' o r
'su p e rficia l foot (tru e )'. The te rm su p erfic ial is com m only abbreviated
to 's u p e r'.
Log volum e. Hanson and W ilson (1960) have outlined how m ethods
and units of log m easu rem en t differ within A u stra lia and Papua-N ew
Guinea. The position may be su m m a rised thus:
Qld, N. S. W. , Vic. - Hoppus su p e r foot.
A . C . T . , S . A. , P a p . - N . G . - su p e r foot.
T as. - Hoppus su p er foot fo r native b ro ad -leav ed sp ecies and
c o n ife rs, su p er foot fo r plantation c o n ife rs.
W. A. - cubic foot (or load, equivalent to 50 cubic feet) fo r native
Units of M easurem ent of M ajor F o re s t P ro d u cts 7
b ro a d -le a v ed sp e c ie s, su p er foot fo r plantation c o n ife rs.
The tra n s p o rt in d u stry in som e States of A u stra lia u ses the ton as
a unit fo r logs, being v ario u sly 50 cu ft for 'lig h t' tim b e rs (density
about 45 lbs p e r cu ft) and 30 cu ft for 'heavy' tim b e rs (density about
75 lbs p e r cu ft).
In New Zealand, the volume of logs is quoted v ario u sly in cubic feet
o r in board feet by the F o re s t Service and industry; the la tte r so m etim es
u ses su p erfic ial feet.
In G re at B rita in , the common unit fo r round logs is the Hoppus cubic
foot with subdivision in decim als o r duodecim als.
In the U. S. A. the conventional unit is the board foot, the contents
being a s s e s s e d as potential sawn tim b e r by re fe re n c e to a log ru le .
W here the m e tric system is used, log volum e is usually quoted in
cubic m e tre s and decim als but special te rm s a re given in som e
co u n tries to a cubic m e tre equivalent.
Sawn tim b e r (lu m b er). In A u stra lia , the common unit is the
su p erfic ial foot though quantities a re usually re fe rre d to on a 100
su p erfic ial feet b a sis . The m e a su re is so m etim es re fe rre d to as
'su p e r solid m e a s u re ' to c o n tra st with 'face m e a s u re ' w here the
volume of d re s se d (planed) m a te ria l is calcu lated on the dim ensions
before d re ssin g . F or sawn tim b e r in stan d ard s iz e s such as
flooring, battens, w eath erb o ard s, and m ouldings, the lineal foot is
often used as the unit.
In New Z ealand, the board foot is the common unit.
Round and hewn tim b e r. P ro d u cts such as p o les, p ile s, g ird e rs and
house blocks, a re usually re f e r r e d to in te rm s of lineal feet o r running
fe e t, p ric e being adjusted according to d ia m e ter.
Hewn poles and g ird e rs a re re fe rre d to sim ila rly by lineal feet, o r
in te rm s of cubic feet. R ailw ay sle e p e rs (hewn o r m illed) a re usually
re fe rre d to on a piece o r 100-piece b a s is , p ric e being adjusted
according to dim ensions.
Stacked m a te r ia l. The common unit fo r stacked m a te ria l in

A u stra lia , as in m ost co u n tries using English u n its, is the cord
re p re se n tin g 128 cu ft (8' x 4' x 4') of stack sp ace, the solid wood
content of this space varying according to such fa c to rs as the shape
and size of the m a te ria l and the way it is stacked. Some A u stralian
pulp and paper com panies use the cunit which re p re se n ts 100 cu ft of
wood with the stack space varying.
W here the m e tric system is used, stack volume is quoted in cubic
m e tre s of space but special te rm s a re given to 1 cubic m e tre
equivalent of space in som e c o u n trie s.
P ro c e s se d m a te ria l. The common unit for plywood, v e n ee r,

h ard b o ard , softboard, and the like in A u stra lia is the sq u are fo o t,
p ric e being quoted usually on 100 squ are feet b a sis o r p e r sh eet in the
c ase of standard s iz e s . In quoting s ta tis tic s it is e sse n tia l to r e f e r to
sheet th ickness.
3 Logs
A log is a portion of the stem o r of a branch of a tre e p re p a re d for

c o m m ercial u tilisatio n by felling, c ro s s-c u ttin g , and trim m in g . Its
value is usually determ ined from its volum e. Though the volum e of
such a solid could be derived by x y lo m etric o r h y d ro static m ethods,
it is usually re g a rd ed as m ost p ra c tic a l to d eriv e the volume of a log
by equating it with a g eo m etrical solid for which m ath em atical ru le s
have been estab lish ed to d erive the volume - the sim p le r the solid,
and the ru le s , the b e tte r.
The Shape of Logs

The solid which a log m ost clo se ly re se m b le s depends la rg e ly on
the position of the log in the tr e e , the sp ec ie s, and the environm ent.
In g en eral the c ro s s -s e c tio n dim ension of a stem d e c re a se s from
ground upw ards - it ta p e rs; but th e re a re d ifferen ces in the ways in
which stem s ta p e r and in the r a te s at which they ta p e r. Stem shape
and ta p e r a re d iscu ssed in som e detail la te r. F or the p re s e n t it can
be stated that con sid erab le investigation has suggested th a t, for
p u rp oses of p ra c tic a l approxim ation, th e re a re a lim ited num ber of
sim ple solids which p a rts of tr e e s re s e m b le , of c irc u la r
c ro s s -s e c tio n throughout, produced by the revolution around a v e rtic a l
ci b
axis of a curve of the general fo rm u la y = k .x w here = ra d iu s of
sectio n , x = distance from apex, the pow ers of and x d eterm in e the
way the solid ta p e rs (the shape of the solid), and the value of k
d e term in e s the ra te of ta p e r w ithin the specified shape. These solids
2 3 2
a re the neiloid (y = kx ), the paraboloid of second degree (y = kx),
3
the cubic paraboloid (y = kx) and the conoid (y = kx). F o r m ost
9
p u rp o ses, a log is re g a rd ed as a frustum of one o r another of th ese
so lids.
The shape and ra te of ta p e r of the section of a stem c lo se st to the
ground, g en erally called the b u tt, is usually influenced by b u tt-sw ell
in a way and to an extent that depends on the sp ecies and the environm ent.
In the case of a tr e e with v e ry little b u tt-sw e ll, such as a young co n ifer
in a dense plantation, little e r r o r m ay be involved in reg ard in g the butt
as a frustum of a second degree paraboloid. In m ost c a s e s, how ever,
this section is m ost likely to be neiloidal in shape. In the case of
som e ra in -fo re s t sp ec ie s and v e teran eucalypts, the low est sectio n is
v e ry irr e g u la r due to flanges, fluting, o r b u ttre s s e s , but the tre e is
usually felled above th ese irr e g u la ritie s .
M ost coniferous tre e s and young tre e s of som e b ro ad -leav ed sp ecies
have a single stem which is continuous from ground to tip. The p a rt
above bu tt-sw ell and for som e d istance into the living crown can
usually be re g a rd e d as approxim ately paraboloidal in shape; the p a rt
from the tip downwards for a sh o rt distance into the crown u sually has
a conoid shape. W here the bole m erg es into crow n by rep eated
branching, as in m ost b ro a d -le a v ed and som e coniferous sp e c ie s, the
p a rt above b u tt-sw ell is approxim ately paraboloidal up to the ra th e r
a rb itra rily defined crow n b re a k . Beyond th is is a s e r ie s of conoid o r
paraboloid fru sta . A log from anyw here within the main bole of any
tre e is thus usually re g a rd e d as a frustum of a paraboloid, usually a
second d eg ree paraboloid fo r convenience in m easu rem en t and
calculation. A log from the v e ry top of a tre e m ight b e st be re g a rd ed
as the frustum of a cone.
T re e b ran ch es tend to have a shape sim ila r to that of the bole
b earin g them so that a log cut from a branch m ight be re g a rd ed as a
frustum of a second degree paraboloid in m ost c a s e s. Short p ie ce s of
branch m a te ria l.o fte n approxim ate c y lin d ers.
F o rm u lae to Give T rue Solid Volume

Depending on the shape assum ed for a log so its volum e can be
calculated using the a p p ro p ria te form ula. The form ulae fo r the so lid s
re fe rre d to previously a re :
Logs 11
Full F ru stu m
S (S + 4 si + s)
N eiloid V L V =
4 ' 6
S (S + 4 si + s)
Conoid V .L V =
3 6
S (S + 4 si + s)
P araboloid V . L o r s i2 . L V =
(second degree) 2 6
_ (S + s)
or
2
w h ere V = volume of log, L = length of log, S = sectional a re a at
la rg e r (basal) end of log, s = sectional a re a at s m a lle r (top) end of log,
and s i = sectional a re a midway betw een base and top of log.
(S + 4 s i + s)
The form ula V = -------- ^ ------ . L (usually called N ew ton's form ula)
recom m ends itse lf since it applies to a frustum of each shape. In
p ra c tic e , irre s p e c tiv e of w here the log com es from in a tre e and the
kind of tr e e , a log is alm ost in v ariab ly reg ard ed as a frustum of a
second degree paraboloid and its volum e calculated by one o r o th er of
the two form ulae:
(S+ s)
(a) V = — - — .L (S m alian’s form ula)
(b) V = s l . L (H uber's form ula)

F o r S m alian's fo rm u la, the av erag e of the two end sectio n al a re a s
should be used and not the sectional a re a equivalent to the m ean of the
two d ia m e te rs. V ery little e r r o r is involved in using the la tte r if the
two d ia m e te rs a re n e arly the sam e; the g re a te r the difference between
th em , the g re a te r the e r r o r , fo r exam ple fo r d ia m e te rs re sp ectiv ely
6 in and 12 in, the e r r o r is 10 p e r cent. The d irectio n and extent of
the e r r o r in volum e using the H uber and Sm alian form ulae fo r logs of
a shape other than paraboloidal has been investigated th e o re tic a lly by
C haturvedi (1926) and many o th e rs. T h eo retical argum ents favour
H u b e r's form ula but it is often m uch e a s ie r to get to the ends ra th e r
than the m iddle of a log and the form ula used in p ra c tic e depends on
the p a rtic u la r c irc u m sta n c e s.
The re p re se n ta tiv e n e ss of the position of m e asu re m e n t is im p o rtan t
since it affects the extent of the e r r o r of e stim a te of volum e. Logs a re
usually in som e d eg ree irr e g u la r in outline; if the actual position of
m easu rem en t is , because of a local effect, u n re p re se n ta tiv e of the
shape assum ed for the log then the volum e e stim a te w ill be in e r r o r .
M ore co nsideration should be given to th is so u rc e of e r r o r in p ra c tice ;
it is doubtful if any other m a te ria l of equivalent value is m e asu re d in
as casual a fashion as a re m ost logs.
F or co m m ercial p u rp o ses w here logs of s im ila r size a re handled
rep eated ly , th e re a re obvious advantages in having tab les available
from which volum e can be re a d for v a rio u s re fe re n c e v a ria b le s . A
common table has length of log and c e n tre g irth o r c e n tre d ia m e te r as
the re fe re n c e v a ria b le s . A nother h as top (sm all end) d ia m e ter as the
re fe re n c e v a riab le because it d ic ta te s the saw ing dim ension in the
m ill, is s a fe r to m e asu re than c e n tre d ia m e te r in log p ile s, is e a s ie r
to m e asu re than c en tre d ia m e te r of a log on the ground, and can be
m easured d ire c tly under b a rk . The com pilation and use of tab les of
this kind in New Zealand a re d iscu ssed by Duff (1959, 1960) and in
A u stra lia by Benson (1964).
F orm u lae to Give A pproxim ations to T ru e Solid Volume

At vario u s tim e s people have sought sim ple approxim ations to tru e
sectional a re a which would give good ap proxim ations to tru e volum e,
called by sev e ra l w rite rs the 'esc ap e fromTr'. Many form ulae for
these approxim ations a re quoted in the lite r a tu r e , the com m onest
being the so -c alle d fifth g irth m eth o d ,th e fo rm u la for which is:
It
V = -lT-2 . 2L w here V ~ volum e (sup er ft), G
12
c en tre g irth (in) and
L_ = length (ft).
^ . 2 su b stitu tes for —- the re s u lt being an o v e r-e stim a te of only

(5) 4 7T
O’ 5 p e r cent. T h ere a re a num ber of v a ria tio n s of th is fo rm u la, such
G2 .f 2D .f
as V = ----------- , V = ------------ w here D = c e n tre d ia m e te r (in).
100 10 —
Sim ilar approxim ations w ere favoured by B u rt (1888) whose tab les a re
Logs 13
s till used for som e purposes in the tim b e r im p o rt tra d e of G reat
B rita in . Such approxim ations a re of little value u n less the form ulae
a re sufficiently sim ple for 'bush a rith m e tic ', and few of them a re .
Hoppus M easure: Q u a rte r G irth Method

The form ula for the q u a rte r g irth m ethod as used in G reat B ritain
<G)2
(4)
is V = ^ . L w here V = volum e (Hoppus cu ft), G = cen tre g irth
(in) and jL = length of log (ft). The form ula used in A u stra lia , and
(G)2
(4)
re f e r r e d to as Hoppus m e a su re , is V = x / . L w here V = volume
(Hoppus super ft). E. Hoppus was a Surveyor to the C orporation of
London Insurance. About 1730 he published a tex t 'P ra c tic a l
m easu rin g made easy to the m eanest capacity by a new set of ta b le s'
which w as followed in 1736 by his 'M e a su re r' based on the form ulae
above. In those days 'the logs w ere lightly hewn w here they lay felled
in the fo re s t. T h e ir average c ro s s -c a lip e r m e asu re m e n ts w ere used
as a b a sis upon which to com pute volum e thus tre a tin g the logs as if
they w ere actually sq u are in sectio n ' (L av er, 1951). It becam e the
p ra c tic e in G reat B ritain to apply the form ula to round logs. This
p ra c tic e seem s to have becom e fa irly well estab lish ed in c e rta in p a rts
of A u stra lia and is at p re s e n t used extensively in the e a s te rn States.
(G)2 o 2
Since in the Hoppus form ula, su b stitu tes for —_— in the form ula
fo r tru e solid volum e, the volum e derived is about 78 p er cent of tru e
volum e. To convert tru e volum e to equivalent Hoppus volum e, tru e
volum e is m ultiplied by 0-78; to convert Hoppus volume to tru e
volum e, Hoppus volum e is m ultiplied by 1-27. J u st as th e re a re 12
su p e r ft in 1 cu ft so th e re a re 12 Hoppus su p er ft in 1 Hoppus cu ft.
T h e re a re 9- 42 Hoppus su p er ft in 1 cu ft; 10 is often used as a
convenient conversion fa c to r. T ables fo r d ire c t reading of volum es
from log length and tru e g irth have been com piled by many authors
since H oppus's tim e . One by D. M atheson is used in some States of
A u stra lia and has led to the te rm s M atheson volume and M atheson's
sc a le . T here has been co n sid erab le co n tro v ersy on the advantages and
disadvantages of the Hoppus sy stem . On balance it would seem b e st to
re p la ce it with tru e volum e but no p ra c tic a l step s have been taken to do
th is. C are should be taken with the term inology: the te rm ’Hoppus
foot' usually m eans a Hoppus su p e r ft in A u stra lia but a Hoppus cu ft
in G reat B rita in , and the prefix Hoppus is often om itted by fo r e s te rs
in those States of A u stra lia which use the m e a su re , and th is is liable
to confuse o th e rs into thinking tru e volum e is intended.
E stim ates of Log Volume in T erm s of F o rec a ste d P roduct

The estim ation of the volume of sawn tim b e r likely to be produced
from a log of given dim ensions using log ru le s h as a long h isto ry in
N orth A m erica, a log ru le being 'a table which shows the estim ated
amount of lum ber which can be sawn from logs of given length and
d ia m e te r’ (S. A. F. , 1950) allowing for lo ss in conversion due to
saw dust, edgings, defect, and so on. G raves (1914) d iscu ssed the
subject com prehensively but m odern N orth A m erican tex ts should be
consulted for the log ru le s in p re s e n t u se. T h ere a re no such log
ru le s in A u stra lia and because of th e ir co n sid erab le lim itatio n s (which
m ost N orth A m erican tex ts s tre s s ) it would be difficult to su stain a
c ase for th e ir introduction to A u stralian p ra c tic e .
Allowance fo r Defect
In A u stra lia, as in m ost c o u n trie s, logs a re graded according to
defects by the application of log grading r u le s , the p rin c ip le s and
p ra c tic e s of which v a ry from State to State. They a re d iscu ssed in
textbooks on utilisatio n . In m any c a s e s , a d ire c t deduction of volum e
is made to allow fo r p ip e , a c e n tra l portion which is hollow o r not
u tilisab le owing to the effects of rin g sh ak es, fungi, o r te rm ite s .
U sually the volume of the pipe is com puted assu m in g its c ro s s -s e c tio n
is sq u are, of a side equal to the d ia m e te r of the pipe as a s s e s s e d by
the m e a s u re r, b ecau se, though the defect a re a is u sually c irc u la r in
section, saw ing rem oves it as a re c ta n g u la r section.
Weight M easu rem en t

If a log is to be converted into a product such as lu m b er w here end
use is re la te d to volum e, then quantity of log is m ost conveniently
e x p re ssed as volum e. F or m a te ria l such as firew ood, it does not
m a tte r much from the viewpoint of end use w hether quantity of log is
e x p re ssed as volum e o r w eight, though w eight may be a m o re p ra c tic a l
Logs 15
m e a su re if the p ieces a re sm all in siz e , ir r e g u la r in shape, and of
little value. If, how ever, the end product is stro n g ly re la te d to weight
of log it m ay be much m ore suitable to e x p re ss the log in te rm s of
w eight; and th e re m ay be v e ry p ra c tic a l re a so n s fo r doing so if th e re
a re many logs, if they a re reaso n ab ly s im ila r in size and w eight, if
e stim atin g the dim ensions of each would slow up bulk handling, and if
the b asic value of the m a te ria l is fa irly low. This is g en erally the
c a se fo r wood fibre products.
Wood w eight may be determ ined in se v e ra l w ays. Each log can be
w eighed individually on a scale situated som ew here along the
production line from fo re s t to facto ry and th is is so m etim es done.
Since the m a te ria l has usually to be loaded on to a vehicle for tra n s p o rt
a t som e stage it can be weighed in bulk w hile on the vehicle; th is is
probably the com m onest m ethod. To find the w eight of wood for a
p a rtic u la r load from the total weight of the loaded v eh icle, o ther
w eights a re needed, for exam ple the w eight of the em pty vehicle, of
b a rk , and of foreign m a tte r like mud and ice on the vehicle and the
load. A verage values of th ese a re norm ally used to save d eterm ining
them for each load. A nother p o ssib ility is to develop tre e -d im e n sio n
and tre e -w eig h t re la tio n sh ip s s im ila r to tre e -d im e n sio n and tr e e -
volum e relatio n sh ip s so that weight can be deriv ed in d irec tly . T re e -
w eight tab les of this kind have been com piled and a re p a rtic u la rly
useful fo r standing tr e e s . B ecause it m ay be m o re convenient to use
volum e instead of weight in c e rta in of the o p eratio n s between fo re s t
and fa c to ry , volum e-w eight conversion fa c to rs m ay also be n e c e ssa ry .
Irre sp e c tiv e of how the weight is d eriv ed , if d ry m a tte r equivalent
is re q u ire d then conversion fa c to rs a re needed according to the amount
of w ater in the wood.
C onsiderable in te re s t in m easu rem en t by w eight instead of volume
h a s been shown in c e rta in p a rts of N orth A m erica in re c e n t y e a rs
e sp ecially by the pulp and paper ind u stry . Much of a conference on
wood m e asu rem en t held in Toronto (Canada) in 1966 was devoted to
w eight m easu rem en t and the proceedings (U niversity of Toronto, 1967)
outline the h isto ry of 'w eight scalin g ' in N orth A m erica, m ethods of
com piling log^ and tree-w eig h t ta b le s, and investigations into the
d eterm ination of conversion fa c to rs needed to d eriv e d ry weight of
wood m a tte r from log volum e, loaded vehicle w eights and so on.
C oulter (1959) has d iscu ssed the determ in atio n of such conversion
fa c to rs fo r ra d ia ta pine in New Zealand and the effects on the fa c to rs
of region, age, season , and v ariatio n in am ount of vehicle equipm ent
and in foreign m a tte r.
In A u stra lia, weight m easu rem en t of th is kind h as been c a r rie d out
only on a lim ited experim ental b a sis . Suggestions a re being m ade for
its introduction w here conditions a re suitable and it is likely th a t in the
next decade much investigation w ill be c a rrie d out, on s im ila r lin es
to that in N orth A m erica, into its applicability to m a te ria l from
coniferous plantations.
Stacked M aterial
Two kinds of m a te ria l a re m e asu re d on a stack b a sis in A u stra lia .
One co m p rises sm all and fa irly re g u la r logs of s im ila r d ia m e te r cut
to a standard length. This m a te ria l can be conveniently m e a su re d in
stack s on ground or on tru ck as c o rd s o r cunits and converted by
a p p ro p riate conversion fa c to rs to w eight o r volum e. In the n o rth e rn
h em isp h ere, v ery la rg e quantities of th is m a te ria l a re used by pulp
and paper m ills and for in d u stria l and dom estic fuel and th e re is
extensive lite ra tu re on m ethods and techniques of m easu rin g such
stack s and estim atin g the solid wood content. M ost pulpwood and
cordwood fo r pulp and paper m ills and case saw m ills in A u stra lia is
m easu red in this way. The o th e r kind c o m p rises m a te ria l of so low
a value and so irr e g u la r in shape that m easu rem en t of individual
p ieces is uneconomic so that it is cut into stan d ard lengths and
m easu red on a stack b a sis - such as logs sp lit into b ille ts fo r
pulpwood, stem and branch wood fo r in d u stria l and dom estic firew ood.
The amount and w eight of this m a te ria l v a rie s m o re than the re g u la r
m a te ria l re fe rre d to above and conversion fa c to rs from c o rd s to
weight o r volume a re likely to v a ry co n sid erab ly depending on the
irre g u la rity of the m a te ria l, the m ethod of stacking, and so on.
M easurem ent of such m a te ria l from a poor quality eucalypt fo re s t in
the A. C. T. is d iscu ssed by C arro n (1953).
4 Tree Girth, Diameter, and
Sectional Area
In dealing with a population co m p risin g a num ber of individuals of

d ifferen t s iz e , it is often convenient for both quantitative and
qualitative purp o ses to m e a su re the sam e dim ension on each individual.
This m easu rem en t can be used to identify the individual and it also
allow s ready so rtin g of the individuals into groups of s im ila r size.
F o r standing tr e e s , the m ost convenient dim ension by g en eral consent
see m s to be a stem c ro s s -s e c tio n taken som ew here within the re a ch
of the av erag e p erso n from the ground. Ideally the m easu rem en t
should be taken at a com parable place on each tre e ; since tre e s differ
in height, the m easu rem en t should be made at the sam e re la tiv e position,
for exam ple a t som e percen tag e of total height. T his involves f ir s t
m easu rin g tre e height, which is not easy to do re lia b ly , and fu rth erm o re
a p ra c tic a b le position for m e asu re m e n t on a la rg e tre e m ay not be
p ra c tica b le on a sm all tre e and vice v e rs a . B ecause of these
d ifficu lties, standing tr e e s a re alm o st u n iv e rsally m easu red at a fixed
height above ground lev el, th is position being re fe rre d to a s b re a s t
h e ig h t.
In A u stra lia , the fixed height is 4 ft 3 in, following the use of a
height of 1- 3 m e tre s in E urope, ground level being defined as the
ground at tr e e base if the ground is level and on the h ig h er side if the
ground is sloping. In many c o u n tries of the w orld such as the U. S. A. ,
New Z ealand, B urm a, India, M alaya, and South A frica, the height
adopted is 4 ft 6 in, ground level being defined as the ground at tre e
b ase if the ground is level and m id-w ay betw een hig h est and low est
lev el if the ground is sloping. In A u stra lia , it is usually p re s c rib e d
that w here th e re is a mound of soil o r litte r around the b ase of the
tr e e , the mound should be displaced if it is v ery loose otherw ise its
upper su rface is taken as ground level; and th at any m a te ria l at b re a s t
height which is not an actual p a rt of the tr e e such as m oss and loose
17
o r hanging b a rk , should be rem oved. G enerally, if a tre e forks
between ground and b re a s t height it is conventional to re g a rd it a s two
stem s: and if it forks above b re a s t height to re g a rd it a s one stem .
In som e c a s e s, for exam ple in T asm ania (L aw rence, p riv ate
com m unication), a forked tre e is re g a rd ed as two stem s if it is at all
p ossible to m e asu re them as such from the ground. The position of
b re a s t height is often estim ated by eye o r located roughly by re fe re n c e
to the o p e ra to r’s height. F o r m ost p u rp o ses it is m o re sa tisfa c to ry
to use a tape, a sc a le , or a m arked stick to fix the position of b re a s t
height, even if one feels doubtful about being able to define ground
level s tric tly . If a tre e is to be m e asu re d p erio d ically , it is e sse n tia l
that the position be identified perm anently by som e m eans such as a
paint m ark, sc rib e m ark, o r soft m etal nail at b re a s t height o r at a
p re s c rib e d d istance from it. O therw ise d eterm in atio n s of in crem en t
at b re a s t height over tim e m ay be confounded with d ifferen ces due to
m easu rin g at a different position on each occasion, p a rtic u la rly as
tre e s often tend to build up a mound a t the b ase and as ground level
r is e s so does the position of b re a s t height.
The size of a tre e at b re a s t height can be m e asu re d and re c o rd ed
as g irth (circu m feren ce), d ia m e te r, ra d iu s, o r sectional a re a .
T here is no p a rtic u la r convention about th is except that ra d iu s is
ra r e ly used (though ra d ia l in c re m en t is a common ex p ressio n ) and
tre e s a re ra re ly re c o rd e d as individuals or ta llie d in c la s s e s
according to th e ir sectional a re a s . G irth o r d ia m e te r is m ost
com m only used, the choice depending m ore on local habit than on any
p a rtic u la r advantage of one over the o th er, though m ost people can
im agine size and change in size m ore easily in te rm s of d ia m e te r
than of girth. In stru m en ts fo r m easu rin g g irth a re com m only
graduated to read equivalent d ia m e te r as w ell; the r e v e rs e is r a r e .
V ery occasionally in stru m e n ts to m e asu re g irth o r d ia m e ter m ay be
graduated to read sectional a re a a lso .
M easurem ent of G irth at B re a s t Height

G irth is usually m e asu re d in A u stra lia with a tape o r band, the kind
varying with size of tre e and purpose of m easu rem en t. F o r rough
e stim a te s o r sm all ste m s, cloth ta p es m ay be used but b ecau se of
w ear and the effects of w eather they becom e u n reliab le and have a
sh o rt life. F o r m e d iu m -sized tr e e s up to about 12 ft in g irth , ste e l
tap es a re com m only u sed, 3 /1 6 -3 /8 in wide, 6-12 ft long, housed
T re e G irth, D iam eter, and Sectional A rea 19
in a ste e l o r le a th e r case with a re tu rn sp rin g o r w inder m echanism .
The stee l is stro n g in tension but b re a k s e asily if kinked.
F ib re g la ss tapes in m etal, le a th e r, o r p la stic c a se s a re also fa irly
popular and in many c irc u m sta n c e s a re b e tte r than steel ones. F o r
tr e e s la r g e r than about 12 ft in g irth , ste e l, c lo th -m e ta llic (cloth with
m etal th re a d s), o r fib re g la ss ta p e s, usually 66 ft (one chain) o r 100
ft in length, in le a th e r o r p lastic c a s e s a re used. The sh o rte r tap es
a re usually graduated in feet, inches, and tenths on one side and in
7Tunits on the other side to give equivalent d ia m e ter which has led to
the nam e 'd ia m e te r ta p e ’. It should alw ays be kept in mind that it is
g irth that is m e asu re d by such a tape and any value of d iam eter
obtained by a g irth tape is only as re lia b le as the m easu rem en t of g irth .
F o r m easu rin g tre e s over 2 ft in d ia m e te r, it is convenient to have a
hook at the leading end of the tape which can be pushed into the tre e
to hold that end w hile the r e s t is taken around the tre e by the o p e ra to r.
M easurem ent of D iam eter at B re a st Height

D irect m e asu re m e n t of d ia m e te r at b re a s t height (d .b .h .) is usually
m ade by som e form of c a lip e r (callip er). The m ost common form of
tre e c a lip e r c o m p rise s a sc a le , graduated in a p p ro p ria te u n its, to
one end of which an arm is fixed a t rig h t angles and on which another
a rm can slide and take up a rig h t angle position when at r e s t. The
scale and fixed arm a re placed flush ag ain st the tr e e , the movable
arm is then brought up against the tr e e , the in te rc e p t it m akes with
the graduated scale giving the tre e d ia m e ter. The m ain re q u ire m e n ts
a re that the whole in stru m en t be durable but light in weight; the scale
m ust be easy to re a d under field conditions (the kind and size of
num bers may have im portant effects on e r r o r s in reading and calling);
and the m ovable a rm m ust move easily but com e to r e s t a t rig h t angles
to the sca le when brought up against the tr e e , o th erw ise th e re is an
e r r o r in tre e d ia m e ter.
The e a r lie s t, sim p lest c a lip e rs w ere m ade of wood and a re still
used in many p a rts of the w orld. T h eir m ain fault is that .the sliding
a rm is difficult to move o r w e ars quickly and the provision of bushings
to overcom e th is has not been v ery su ccessfu l. The la te r m etal ty p es,
usually of light alloy, a re m ore su cc e ssfu l, p a rtic u la rly those with
devices to allow c o rre c t alignm ent of the arm if it gets out of
adjustm ent. T h ere is a wide v a rie ty of c a lip e rs som e of which a re
d iscu ssed by Hummel (1951). One kind h as both a rm s fixed, such as
the tre e fork d iscu ssed by E ssed (1957), and the Finnish type with
one stra ig h t a rm and one curved. F or en u m eratio n s of stands of
sm all tre e s with a lim ited range of s iz e s , a piece of plywood into which
h as been cut a s e r ie s of gaps which d e c re a se in width by step s from the
o u ter edges inw ards, is handy for enum erating tr e e s in size c la s s e s .
In another kind, the a rm s a re bowshaped and pivoted on a pin with a
sca le suspended betw een them . T here a re v a rio u s p a tte rn s which allow
autom atic r e g is te r of size on a re c o rd in g tape such as the L ignom eter
(FAO, 1958).
D iam eter can be d eterm ined in d irec tly through the u se of a
g eo m etrical p rin c ip le as in the B iltm o re stick which c o n sists of a
stra ig h t length of wood graduated in such a way th at the d ia m e ter at
any point on a standing tre e m ay be re a d from the stick when it is held
against the stem , tangential to the point, p e rp e n d icu la r to the v e rtic a l
axis of the tre e , with the o b s e r v e r’s eye at the sam e h o rizo n tal level
and a d istance from the stick fo r which the stic k is g raduated. The
stic k is placed so that the zero end is in line of sight from one eye to
one side of the tre e ; the graduation on the stic k which is in lin e of
sight with the other side of the tr e e gives the d ia m e te r. The
g eo m etrical p rin c ip le involved, and the ap p ro p ria te fo rm u lae for
graduating the stic k , a re d iscu ssed in te x ts such as Chapman and
M eyer (1949). The B iltm ore stic k is a sim ple and convenient
in stru m en t for placing medium sized tre e s in 2 in o r 4 in d ia m e ter
c la s s e s in inventory.
Sectional A rea
By convention, sectional a re a at b re a s t height is given the p a rtic u la r
nam e b a sa l a r e a . It is norm ally deriv ed in d irec tly by re fe re n c e to
p re p a re d tab les from a m e asu re m e n t of g irth through the fo rm u la
2 2
g , o r from a m e asu rem en t of d ia m e ter through ^d . In A u stra lia ,
4 ft 4
b asal a re a is e x p re ssed in sq u are feet and d e cim als. The fo rm u lae
assu m e a c irc u la r section. A p e rfe ctly c irc u la r stem c ro s s -s e c tio n
is ra re ; m ost stem sections a re irr e g u la r in outline. If a tr e e section
is assum ed c irc u la r when it is not th e re is an e r r o r involved in the
sectional a re a calcu lated from the form u lae. F o r exam ple, if a tre e
section is ellip tical with a m a jo r axis of 22 and a m inor axis of 20,
then the e r r o r s involved in calculating the a re a of the section as a
c irc le bv v a rio u s wavs a re a s shown in Table 4:1.
T able 4:1
Errors in c a lc u la tio n of tree se c tio n a l area
M ethod Area Error %
As an ellip se 3 4 5 -6 0
As a c ir c le on m ajor d iam eter 380- 1 +9-9
As a c ir c le on m inor d iam eter 3 1 4 -2 -9 - 1
As a c ir c le on m ean o f m ajor
and m inor diam eters 3 4 6 -4 +0-2
As a c ir c le on girth 3 4 6 -7 +0- 3
T hat is , calculatin g the c ro s s -s e c tio n as c irc u la r on the b a sis of the

m ean of the m a jo r and m inor d ia m e ters involves le a s t e r r o r but the
e r r o r is only slightly le s s than that involved in calculating a re a from
g irth . Both e r r o r s a re positive and re la tiv e ly v e ry sm all even for
what is an e c c e n tric ity quite obvious to the eye. In fact, a tru ly
ellip tic a l section is as r a r e a s a tru ly c irc u la r sectio n . F o r p ra c tic a l
convenience, the a re a of the section is alm o st in v ariab ly calculated
as if the section w ere c irc u la r without re g a rd fo r its actual shape
e ith e r through g irth m e asu re d by a tape o r through d ia m e ter m easu red
by a c a lip e r. If th e re is obvious e cc e n tric ity the m a jo r and m inor
d ia m e te rs m ay be m e asu re d with a c a lip e r and the sectional a re a
com puted as a c irc le from the m ean of the two d ia m e te rs.
C om parison of G irth Tape and C alip er

If the c ro s s -s e c tio n of the stem is not c irc u la r , the sectio n al a re a
will be o v e re stim ate d from a m e asu rem en t of g irth by tape sin ce, for
a p a rtic u la r p e rim e te r, the c irc le has a g re a te r a re a than any o ther
fig u re. The size and sign of the e r r o r due to one m easu rem en t by
c a lip e r w ill v ary depending on w here the c a lip e r is placed.
M easurem ent by tape is th e re fo re likely to be p re c is e for the sam e
o p e ra to r and among o p e ra to rs but m e asu re m e n t by c a lip e r is not
u n less a m ark is made on the tre e ag ain st which the c a lip e r b a r should
alw ays be placed. If the section is e llip tic a l, the c a lip e r allow s a
b e tte r e stim ate of sectional a re a than the tape through the mean of the
m a jo r and m inor d ia m e te rs, but, as shown in Table 4:1, the e stim ate
is only slightly b e tte r and unless the actual m a jo r and m inor d ia m e ters
a re m easu red the e r r o r m ay be la rg e r than th at from g irth and may
be positive or n e g ativ e .
P rovided a tape is c o rre c tly graduated th e re is no in stru m en t

e r r o r . If the sliding arm of a c a lip e r is not s tric tly p erp en d icu lar
to the sc a le when the reading is taken th e re w ill be an e r r o r which in
p ra c tic e is u sually negative.
The m ain so u rc e of o p e ra to r e r r o r re s u lts from not placing the
in stru m e n t in a plane at rig h t angles to the v e rtic a l axis of the tre e .
F o r an angle of d isplacem ent of A0 from the c o rre c t position, in the
c a se of a c irc u la r section of d ia m e ter D, a c a lip e r m e a su re s a
distance D sec A°; a tape m e a su re s the p e rim e te r of an ellip se of
m ajo r axis D sec A0 and m inor axis D.
A g irth tape is sm all, easy to use and to c a r ry . The c a lip e r is
com p arativ ely bulky and may over a long p erio d be m o re tirin g to use
and c a r r y than a tape. A tape can be c a r rie d up and used read ily in
any position on the standing tr e e . The c a lip e r cannot be c a rrie d up a
tre e conveniently. One m e asu rem en t by tape is usually fa s te r than
two by c a lip e r; one m e asu re m e n t by c a lip e r is usually fa s te r than one
by tape. It is often difficult to get the tape around felled tre e s and logs
and, if one m e asu re m e n t is acceptab le, the c a lip e r may be much m ore
p ra c tic a b le . The use of one in stru m en t ra th e r than the other probably
re fle c ts local habit m ore than objective te s t. The tape is much m ore
common in A u stra lia . One re a so n may be the in h eritan ce of the
Hoppus sy stem of m easu rin g volum e. A nother may be th at, before
coniferous plantations w ere estab lish ed , m e asu rem en t w as confined
to native sp ec ie s which w ere too la rg e fo r c a lip e rs and use of the
tape for them has p e rs is te d for all s iz e s . C alip ers a re used in some
States fo r m e asu rin g end or c e n tre d ia m e te rs of logs from coniferous
plantation thinnings. They a re likely to be used m ore in the future in
enum erations of coniferous plantation s.
M easurem ent of D iam eter at O ther P o sitio n s on Standing T re e s

G irth o r d ia m e te r is often needed at v ario u s positions on a standing
tre e beyond convenient re a ch from ground lev el, fo r exam ple to
d erive stem volum e o r an ex p ressio n of tr e e ta p e r. T h ere a re th re e
m ain ways of getting it: d ire c t m e asu re m e n t at the point by an
o p e ra to r; ra isin g a tape o r c a lip e r to the point; o r using an optical
device, on the ground and at som e distance from the tr e e , d irected at
the point of m e asu re m e n t.
D irect m e asu re m e n t is likely to be the m o st re lia b le . This is
p ra c tic a b le fo r co n ifers in plantations in A u stra lia. They can be
clim bed with the aid of the b ra n c h es, though a lad d er is n o rm ally
used fo r the low er section which is g en erally pruned o r on which the
b ra n c h es a re dead and unsafe fo r clim bing. A ltern ativ ely , light m etal
la d d e rs in 10 ft sections a re strap p ed to the whole length up to the
position of m easurem ent; the b ra n c h es usually have to be pruned along
the path of the la d d e rs. Climbing iro n s of the s tirru p type, and
clim bing spikes, a re not suitable fo r m e asu re m e n t w ork. Climbing
devices such as the Baum velo (Hummel, 1951) a re suitable in sp ecial
c irc u m sta n c e s. Sectional la d d ers a re probably the m ost suitable
equipm ent for young b ro ad -leav ed tr e e s . In A u stra lia , d ire c t
m e asu rem en t up the stem is ra r e ty c a r rie d out on old, la rg e tre e s of
b ro ad -leav ed sp ecies except for a re fe re n c e m e asu re m e n t in place of
b re a s t height in the c ase of tr e e s with extensive b u tt-sw e ll, flanges,
o r b u ttre s s e s . F or th is the o p e ra to r can sc ra m b le up the b u ttre s s e s
o r e re c t a te m p o ra ry scaffolding from litte r o r use a light lad d er, but
may s till have difficulty in passin g the tape around the stem .
The use of pole c a lip e rs o r pole tap es is re s tric te d to the height to
which a pole can conveniently be held and the m easu rin g device at its
end can conveniently be seen or operated. The m echanics of these
in stru m e n ts lim it th e ir application to sm all tr e e s and a height about
15 ft above ground level. Hummel (1951) d isc u sse s v a rio u s types.
They have not been used in A u stra lia to any extent.
Many in stru m en ts of considerable v a rie ty have been designed on
optical p rin cip les to d eterm in e stem d ia m e te rs at heights above
convenient re a ch . Hummel (1951) has an extensive d iscu ssio n which
is of p a rtic u la r h is to ric a l value. G rosenbaugh (1963) has a m odern
d iscu ssio n of in stru m en ts and th eo ry . At one end of the range of co st
and sophistication is the B a rr and Stroud D endrom eter d escrib ed by
J e ffe rs (1955). At the other end of the range a re sim ple in stru m en ts
b ased on p rin cip les s im ila r to th at of the B iltm o re stick such as
d iscu ssed by Finlaj'son and A rc h e r (1964) and N ousianen (1965).
Within th is range a re a num ber of in stru m e n ts such as the ’wide scale
re la sk o p ’ d escrib ed by Sutter (1965), the tr a n s it theodolite as outlined
by L e a ry and B urns (1963), and the P e n ta p rism c a lip e r (W heeler,
1962) a te s t of which is d e sc rib e d by Benson and Slinn (1964). In
A u stra lia the B a rr and Stroud D endrom eter has been used in re s e a rc h
and routine w ork such as d escrib ed by van Schie (1967) and sim ple
in stru m e n ts based on p rin cip les s im ila r to that of the B iltm o re stick
have been used to e stim ate d ia m e te rs of la rg e b ro a d -le a v ed stem s
above b u ttre s s e s .
The R ep resen tativ en ess of P o sitio n s fo r G irth o r D iam eter
M easurem ent
The section at b re a s t height is often u n re p re se n ta tiv e due to the

b a sa l sw elling of b ra n c h es, callu s grow th re su ltin g from w ounds, o r
o th er fa c to rs . U sually in such a c a se the o p e ra to r m e a su re s d ia m e ter
(girth) at points equidistant above and below b re a s t height outside the
affected zone, and takes the m ean of th e se two d ia m e te rs (or b e tte r
s till the m ean of the two sectional a re a s ). T his a ssu m e s th at d ia m e ter
(sectional area) above and below the affected zone changes w ith height
in a re g u la r way. U nless th is is so, and frequently it is not, the m ean
of two such points may not be s a tisfa c to ry and since the o p e ra to r has
in effect to pick subjectively two points which w ill give a s a tisfa c to ry
m ean he m ight do as w ell o r b e tte r by subjectively picking only one
position of re p re se n ta tiv e siz e . When b re a s t height has to be
p erm an en tly identified, the need fo r two positions to be m ark ed and
m e asu re d often pro v id es an irr ita tin g so u rce of e r r o r in p ra c tic e -
only one of the points is m e asu re d , only one has the m ark renew ed,
m ista k e s a re made in taking the m ean, and so on. F o r th ese re a so n s,
it is som etim es p re s c rib e d that the o p e ra to r choose subjectively only
the one point for m e asu re m e n t when b re a s t height is u n re p re se n ta tiv e .
T e sts show th at the effects of bias a re not likely to be se rio u s.
It is difficult to decide w hether th is p ra c tic e should be extended to
m e asu re m e n t of positions elsew h ere on the stem . The difference
betw een the d ia m e te rs under b ark at 5 ft and 15 ft above ground level
is com m only used in som e p a rts of A u stra lia as a m e asu re of ta p e r of
P inus ra d ia ta for re fe re n c e as an independent v a ria b le in a tre e
volum e table (see ch. 13). The chance location of th ese p o sitio n s of
m e asu re m e n t re la tiv e to branch sw ellings and gen eral su rface
irre g u la rity may give a quite u n re p re se n ta tiv e ex p ressio n of ta p e r
and so a b iased value of tre e volum e. Again, positions nom inated by
the p re s c rip tio n s of a m ethod to d eriv e tre e volum e (see ch. 8) may be
s im ila rly u n re p re se n ta tiv e . The usual p re s c rip tio n in th ese c a se s is
to take a m ean of m e asu re m e n ts equid istan t above and below the
nom inated point. One point chosen subjectively may be m ore
sa tisfa c to ry in th is c a se also . H ow ever, if too much subjectivity in
selectio n of such points of m e asu re m e n t is introduced, the effects of
b ias m ay becom e im portant. How fa r the argum ent for one
m e asu re m e n t over the m ean of two m e asu re m e n ts should be extended
m u st be judged by common sen se.
O cular E stim ate
It is difficult to justify o cu lar e stim ate of d ia m e ter o r g irth in any
se rio u s m easu rem en t p ro je c t, it is fa r too much subject to e r r o r .
E stim atin g upper d ia m e te rs by eye to calcu late volum e is so liab le
to e r r o r th at an inventory based on th is m ethod is only w orthw hile
c a rry in g out if v ery rough e stim a te s a re acceptable. The fo re s te r
should, how ever, be able to e stim ate stem d ia m e te r sufficiently well
by eye to act a s a rough check on m e asu re m e n t by in stru m en t.
5 Tree Height
The total height of a tre e is usually defined as the d istan ce from

ground level to the ’top' (highest point) of the tre e - th is a ssu m e s th at
the tre e is stra ig h t and v e rtic a l. F o r m ost co n ifers the top of the tre e
can be easily nom inated as the growing tip o r re s tin g bud and, b ecau se
of the n a tu re of the crow n, an u n in terru p ted line of sight to it from the
ground can usually be obtained. F or b ro ad -leav ed tr e e s with healthy
conical crow ns such as many young eucalypts, the top may also be
read ily nom inated and sighted as the highest point of the crow n. But
fo r tre e s with um brageous irr e g u la r cro w n s, such as m ost m atu re
eucalypts, the top m ay be difficult to nom inate, p a rtic u la rly if the
upper p a rt of the crow n is dead, and it is frequently defined as that
point w here the crown is in te rse c te d by the im ag in ary upward p ro jectio n
of the bole. In T asm ania (L aw rence, p riv ate com m unication), the
top is defined as the highest green leaf. A part from the difficulty of
nom inating the top of such a tr e e , sighting it may also be difficult
since the line of sight usually goes through the crow n. F o r such tr e e s ,
how ever, som e m e asu re of bole height is often of m ore in te re s t than
to tal height, for exam ple from som e defined stum p height to a position
on the upper bole fixed by an a rb itra ry d ia m e te r lim it o r by a quality
lim it set by branching or defect. U nfortunately, th is position has
usually to be judged by eye and d ifferen ces in e stim a te s of height m ay
re s u lt from difference in definition of what is being m easu red .
Even if height can be unequivocally defined it is not easy to get
co n sisten tly re lia b le m e asu re m e n ts from the sam e o p e ra to r o r among
o p e ra to rs and, since tre e height is a v a ria b le of fundam ental
im portance for quantitative d e sc rip tio n s of a tre e and a stand,
fo r e s te rs have devised a co n sid erab le num ber of in stru m e n ts and
m ethods to m easu re the height of a standing tr e e . The ideal
in stru m en t should give p re c ise and unbiased m e a su re m e n ts, be e asy
26
T ree Height 27
to c a r r y and use in fo re st conditions, and be re la tiv e ly inexpensive.
Height can be m easu red d ire c tly , o r deriv ed in d irectly by
in stru m e n ts of two groups based re sp e c tiv e ly on g eo m etrical p rin cip les
and trig o n o m etrica l p rin c ip le s.
D irect M ethods of M easurem ent

The m ost relia b le way to m e a su re the height of a standing tre e is
to take a tape to the top of the tr e e by clim bing o r by la d d er and to
m e asu re the height d ire c tly with the tape. If the top section is
in a cc e ssib le , a light, rig id , graduated pole may be c a rrie d o r hauled
up the tre e and used to m e a su re that section s e p a ra te ly . F o r obvious
p ra c tic a l and econom ic re a so n s d ire c t m e asu re m e n t by clim bing can
only be c a rrie d out under sp ecial c irc u m sta n c e s, and the m o st common
method of m easu rin g height d ire c tly is to use height rods o r height
s tic k s . V arious types of sin g le, folding, and telesco p ic height stick s
have been invented to suit p a rtic u la r conditions of m e asu rem en t. The
chief re q u ire m e n ts a re lig h tn ess and rig id ity , p o rtab ility , e a se , speed
and safety of operation, and cheapness of co n stru ctio n . F o r sh o rt
tre e s up to about 20 ft high a single graduated pole which folds into
two sectio n s, or a telescopic pole, is com m only used. F or ta lle r
tre e s the ro d s a re usually in sections about 5 ft long, of s im ila r
d ia m e te r, each fitting into the next by a fe rru le , with p erh ap s one to
th re e tap erin g sections at the top. The upper lim it to the use of height
ro d s of th is type under b e st conditions seem s to be about 100 ft but the
av erag e lim it over a range of w eath er and tre e conditions is probably
between 60 and 80 ft. They a re b est used on tr e e s with b ran ch es which
w ill let the rods p ass between them re ad ily yet give them la te ra l
support, such as plantation c o n ifers in A u stra lia . The m ost effective
team for an extensive program of m e asu rem en t using height ro d s is
probably th re e m en, one using the ro d s, the second looking a fte r
supply of ro d s to the f ir s t and reco rd in g re s u lts , and the th ird
d eterm ining when top of rod and tre e a re coincident (for exam ple from
a vantage point in another tr e e ) , the th re e exchanging ta sk s frequently
to avoid fatigue. The ro d s a re usually made of wooden dowelling o r
hollow m e ta l-a llo y tubing. They som etim es com e a p a rt in the
re trie v in g p ro c e ss so that w earing safety h elm ets to avoid head
injury should be m andatory. V arious c a r r ie r s fo r the ro d s have been
invented fo r easy p o rtability; a c a r r i e r like a golf bag which can be
slung over the shoulder is handy.
In d irect M ethods of M easurem ent

Instrum ents B ased on G eom etrical P rin c ip le s
These in stru m e n ts depend on the p ro p e rtie s of s im ila r tria n g le s,
and th e ir application re q u ire s e ith e r m e asu re m e n t of d istan ce of the
o p e ra to r from the tre e o r m e asu re m e n t of the length of p a rt of the
stem . Many in stru m en ts have been devised but th e re a re only a few
p rin c ip le s involved.
The sim p le st p rin cip le is illu s tra te d in the so -c alle d staff o r pole
hypsom eter which in p ra c tic e need only be a stra ig h t stic k , for
exam ple a piece of branch. The o b s e rv e r holds it p a ra lle l to the tr e e
so that the length of the stic k from its top to the point a t which it is
held is the sam e as the d istance from the eye to w here it is held, and
in such a position that the top of the tr e e is sighted p a st the top of the
stick and the bottom of the tre e is sighted p a st the point at which the
stick is held. The height of the tre e is given by the distance from the
eye of the o b se rv e r to the base of the tre e . F o r all p ra c tic a l p u rp o ses
th is distance is d eterm ined by pacing, since the n atu re of the method
does not w a rra n t m easu rin g it by tap e.
A nother sim ple p rin c ip le , in which a length on the stem is
determ ined as equivalent to som e p re d e term in ed fractio n of the total
height, is illu s tra te d in what Hummel (1951) r e f e r s to as the
h ypsom eter of V orkam pff-L aue. On a stic k of a specific length a m a rk
is made at a d istance upw ards from the bottom of the stick which is
an a rith m e tic ally convenient fractio n of the total length of the stick ,
for exam ple a t 1 in from the bottom of a 10 in stick . The o b se rv e r
holds the stick p a ra lle l to the tre e at such a position re la tiv e to his
eye and at such a d istance from the tr e e that he can sight the top of
the tre e past the top of the stick and the bottom of the tre e p ast the
bottom of the stick . The in te rc ep t on the tre e m ade by the bottom
section of the stick is then m e asu re d o r estim ate d and the height of
the tre e is given by m ultiplying the length of the in te rc ep t by the ra tio
of the length of the whole stick o v er the length of the bottom section of
the stick , by 10 in th is c ase . If the frac tio n is m ade_l_ , tr e e height
12
in feet is given by the length of the in te rc ep t in inches. T hese two
hy p so m eters a re useful fo r rough e stim a te s.
B e tte r e stim a te s a re given by in stru m e n ts such as the C hristen
hypsom eter and W eise hypsom eter (J e rra m , 1939), the M e rritt
T ree Height 29
h y p so m eter (Husch, 1963) and the JAL hy p so m eter (FAO, 1954) but
none of th ese have been used to any extent in A u stra lia w here
in stru m e n ts based on trig o n o m e tric a l p rin c ip le s a re favoured.
In stru m ents Based on T rig o n o m etrical P rin c ip le s

The basic p rin cip le of th ese in stru m en ts is sim ple and is illu s tra te d
in Figure 5:1. A A
F ig . 5:1 T re e h e ig h t m e a s u re m e n t
(a) o b serv er's eye above tre e base (b) o b se rv e r's ey e b e lo w tre e base
Sighting from O, the o b s e rv e r re a d s the angle AOC to the top of the

tre e and the angle BOC to the bottom of the tr e e . If O is above the
base of the tre e (as in Fig. 5:l(a) ), he m e a su re s the h o rizontal
d istan ce OC, com putes AC as OC tan AOC and BC as OC tan BOC, and
adds AC to BC to get the height of the tre e . If O is below the base of
the tre e (as in Fig. 5:l(b) ) he m e a su re s the slope d istan ce OB and
d e riv e s the equivalent h o rizontal OC from it; he com putes AC as OC
tan AOC and BC as OC tan BOC, and su b tra c ts BC from AC to get the
height of the tre e . The angles a re read by som e form of c lin o m e te r,
an in stru m en t for m easu rin g angles in the v e rtic a l plane. T here a re
two general c a s e s.
In one c a se , the o b se rv e r stands from the tre e a distance which
is convenient for observ atio n , m e a su re s that d istan ce and the angles
AOC and BOC, and d e riv e s the tre e height as outlined previously.
The angles can be re a d by any kind of clin o m e ter. In p a rtic u la r c a se s
a theodolite is used; for m ost p ra c tic a l p u rp o ses the sim p le r Abney
level s e rv e s v e ry well - an in stru m e n t com m only used in fo re st
surveying and usually re a d ily available. The Abney level provides a
line of sight through a tube which c a r r ie s an a rc graduated to 90°
each side of a z e ro , the a rc supporting a m ovable index arm to which
is attached a sm all s p irit level. If the bubble of the level is brought
to cen tre of run for any line of sight, the graduation on the a rc
indicated by the index gives the angle of inclination re la tiv e to
h o rizo n tal. The in stru m en t m ust be checked to en su re th at the line of
sight and the bubble tube axis a re p a ra lle l; th is collim ation ad ju stm en t
is d escrib ed in textbooks on surveying. D isadvantages of the Abney
level a re that the bubble ap p ears m o re and m o re d isto rte d with
in c re asin g angle of elevation and becom es difficult to c e n tre , and having
to move the index a rm to get the a p p ro p ria te angle of inclination m akes
it difficult to keep the in stru m e n t steady. The field of view is also so
re s tric te d that m ista k es a re often m ade in sighting the top of the tr e e ,
p a rtic u la rly in conditions of poor v isib ility .
If the o b s e rv e r can so position h im self above the level of the b ase
of the tre e that the angle of elevation to the top of the tr e e is 45°,
then AC = OC (ref. Fig 5: (a) ) and he can d eterm in e the height of the
tre e by m easuring the h orizontal distance OC and adding to it the
distance CB m easu red d ire c tly on the tr e e . The m ain disadvantage
of this method is that in fo re s t conditions the o b se rv e r m ay not be
able to see the top of the tr e e at an angle of o b serv atio n of 45°. A gain,
since it is advisable to m e a su re tr e e height from two independent
positions as a check on o p e ra to r e r r o r , the o b se rv e r m u st be able to
see the top of the tre e at an angle of 45° at two d ifferen t p o sitio n s,
p re fe ra b ly on quite d ifferent sid es of the tre e . This method is often
re fe rre d to as the 'Abney 45° m ethod’. V arious o th er clin o m e ters
u tilising a different working prin cip le a re also used fo r m easu rin g
tre e height, for exam ple Benson (1961a) d is c u s s e s the Suunto
c lin o m e ter in which the angle of inclination of a line of sight is given
by the ap p ro p riate rotation of a weighted c y lin d rical drum about its
a x is. T ables from which the values of AC and BC can be read for
various values of OC o r OB and the angles AOC and BOC a re com piled
fo r use with clin o m e ters (for exam ple, A. F . R . I . , 1964).
In the other g e n eral c a se , the clin o m e ter is graduated in te rm s of
the tangents of su c c e ssiv e angles of inclination so th at fo r a sp ecific
distance from the tr e e the values of AC and BC can be read d ire c tly
from the in stru m e n t o r derived by sim ple arith m e tic without re c o u rse
to ta b le s. In the topographic Abney level the sc a le is graduated to allow
readings of ris e (elevation) o r fall (depression) in feet p e r one chain
h orizontal distance. In the engineering Abney level the scale is
graduated in p e rc en ta g e s. The Haga a ltim e te r h as five sc a le s on a
T re e Height 31
ro ta ta b le pentagonal rod - 15. 20. 25. 30 (for d istan ces of 15, 20. 25
and 30 units) and p e r cent. The in stru m en t is about the size and
shape of a pistol: the inclination of the line of sight is indicated by a
w eighted p o in ter, the m ovem ent of which is co ntrolled by a button
re le a s e m echanism and trig g e r clam p. The B lu m e -L e iss height
m e a s u re r is s im ila r to the Haga in size and g en eral p rin cip le except
th at the p o in ter is d ire c ted downwards from its pivot in stead of
upw ards as in the Haga and the five s c a le s , 15, 20, 30. 40 and p e r
cent o r d e g re es a re in the form of five con cen tric bow s. As an
attach m ent for the Haga, and in co rp o rated into the B lu m e-L eiss
in stru m e n t, th e re is a 'ran g e fin d e r’ w hereby the o b se rv e r can set
h im se lf at a specific d istance from the tre e by hanging on the low er
p a rt of the tre e an a p p ro p ria te ly graduated sca le which he sights by
the 'ra n g e fin d e r'. These two in stru m e n ts a re in common use in
A u stra lia being in se v e ra l w ays su p e rio r to the Abney lev el. W henever
light conditions a re poor o r the top of the tre e is likely to be confused
with th a t of another, the ex tern al line of sight is su p e rio r to the
in te rn a l, re s tric te d field of view of the Abney; th e re is no bubble to
c e n tre by moving p a rt of the in stru m e n t, and so the in stru m en t gives
a g re a t flexibility for selection of d istance in dense stan d s. F u rth e r,
under suitable stand conditions and for v e rtic a l tr e e s , by using the
d ista n c e -m e a su rin g attach m en ts, the o b se rv e r can disp en se with
m e asu rin g distance to the tre e . The moving p a rts of both in stru m en ts
a re su b ject to w ear and they should be checked frequently. V arious
co m p arative studies of the in stru m e n ts have been c a rrie d out, such as
by W arren (1958) .
The Leaning T ree

In the discu ssio n so fa r it has been assum ed th at the tre e to be
m e a su re d is v e rtic a l. In fact many tr e e s in p lantations and native
fo re s t in A u stra lia a re not v e rtic a l, and the assum ption that a tre e is
v e rtic a l when it is not is a m a jo r so u rce of e r r o r in tre e height
m e asu re m e n t.
E r r o r s w ill occur in in d irec t m ethods u n less su itab le p recau tio n s
a re taken. If the o b s e rv e r, standing in the plane of tre e lean,
m e a s u re s with the lean tow ards him he w ill o v e re stim a te tre e height;
if he m e a s u re s with the lean away from him , he w ill u n d erestim ate
tre e height. The extent of the e r r o r for v ario u s angles of lean of tre e
and angles of observation under these c irc u m sta n c e s is shown in Table
5:1.
T able 5:1
Errors, as per c e n t of true h eigh t, for various angles of lea n and angles of observation
(Chaturvedi 1926)
A ngle of lean A ngle of observation to tot3 of tree

1
60° 4 5° 30°
Towards A way Towards A way Towards Away
5° +15 -1 5 + 8 - 9 + 5 - 5
10° +29 -3 2 +16 -1 9 + 8 -1 2
15° +41 -4 8 +22 -2 9 +11 -1 9
20° +53 -6 5 +28 -4 0 +13 . -2 6
The table shows that the s m a lle r the angle of o b serv atio n , the le ss
the e r r o r in m easu rin g the leaning tre e as if it w ere v e rtic a l, that for
a p a rtic u la r angle of observ atio n , the g re a te r the lean the g re a te r the
e r r o r , and that for any angle of lean and angle of o b serv atio n , the
negative e r r o r (in te rm s of tru e height) which o c cu rs when the tre e
lean s away from the o b se rv e r is g re a te r than the positive e r r o r
in c u rre d when the tr e e leans tow ards the o b s e rv e r, as can be seen
from Fig. 5:2 and the following explanation.
D
Fig. 5:2 Leanirtg tree

T re e Height 33
A ssum e AB is a leaning tr e e and P is the plumb position on the
ground of the tip A, AP being the ’v e rtic a l com ponent'. L et the angle
of lean from the v e rtic a l be d°. L et the angle of observation from
both M and N be a °, that is the sam e on both sid es of the tr e e . A ssum e
fo r convenience that the o p e ra to r’s eye is a t ground level; in p ra c tic e ,
MPN would be above ground level. Then the height m e asu re d from M
with the tre e leaning tow ards the o b s e rv e r is BD = BE + ED = BE +
AE tan a 0 . The height m easu red from N with the tre e leaning away
from the o b se rv e r is BF = BE - E F = BE - AE tan a0 . BA' is made
equal to BA (BAA' then being an is o sc e le s trian g le ). The tru e height
is then BA = BA^ = BE + EA^ = BE + AE tan d ° /2 (since it can be
shown th at the angle EAA^ = d °/2 ).
The e r r o r in BD (lean tow ards) = BD - BAi
= (BE + AE tan a0) - (BE + AE tan d °/2 )
= AE (tan a ° - tan d °/2 )
The e r r o r in BF (lean away) = BF - BA^
= (BE - AE tan a0) - (BE + AE tan d °/2 )
= - AE (tan a0 + tan d °/2 )
The conventional recom m endation is to m e asu re the leaning tre e at
rig h t angles to the plane of lean . If the o b se rv e r m oves from the base
of the tr e e at rig h t angles to the plane of lean of the tre e (at rig h t
angles to a line joining the b ase of the tre e and the plum b point of the
top) and m e asu re s the angles of inclination to the top and bottom of the
tr e e , the c o rre c t tre e height can be obtained. From any o th er
position an e r r o r of varying am ount w ill re s u lt.
If the o b se rv e r m oves from the plumb point of the top, at rig h t angles
to the plane of the lean, until he se e s the top at an angle of 45°, the
height of the tre e w ill be given by the distance from him to the base of
the tre e (in fact, since the angle is re a d from eye lev el, the distance
w ill be from the o p e ra to r's eye to eye level on the tre e plus the
d istan ce on the tre e from eye level to ground). If two m e asu rem en ts
a re taken from independent positions as a check on o p e ra to r e r r o r ,
one m ust be taken on each side of the tr e e , along the lin e at rig h t
angles to the plane of lean through the plumb point of the top, at an
angle of 45° from the top. It m ay o r m ay not be p o ssib le to see the
tre e top from this position.
Fig. 5:2 shows th at if the sam e angle of o b serv atio n (not the sam e
distance) is used on each side in the plane of lean of the tre e the mean
of the two observations is the 'v e rtic a l com ponent' (the distance from
the top of the tre e to its plumb position on the ground). The o bservations
re sp ec tiv e ly with the lean tow ard and away from the o b se rv e r a re
BE + AE tan a0 and BE - AE tan a 0 , the m ean being BE which equals
PA (defined as the v e rtic a l com ponent). The v e rtic a l com ponent is
fo r m ost p ra c tic a l purposes a clo se approxim ation to the tru e height
of the tre e ; fo r an 8° lean, it is 99 p e r cent of tru e height, for an 11°
lean 98 p e r cent, and even 95 p e r cent for an 18° lean which is a
sev e re lean indeed. T rue height can then be d eriv ed from the v e rtic a l
component by m easu rin g from the b ase of the tr e e to the plum b point
of the top and calculating the tru e height trig o n o m etrica lly o r by
re fe re n c e to p re p a red tab les such as those of F alco n er (1931). The
disadvantages of the m ethod a re th at the m e asu re m e n ts on each side
of the tre e differ from each o th er and provide no check on o b se rv e r
e r r o r in one o r both; that the top m u st be v isib le from positions along
the plane of lean which subtend the sam e angle of inclination to the top;
and that the in stru m e n t used m ust be such as can be set a t the sam e
angle of observation fo r both s id e s .
F or a ll the m ethods d iscu ssed above the top of the tre e m ust be
plumbed to the ground so as to e sta b lish the plane of lean. This being-
done, it is then m ore convenient to m e asu re the v e rtic a l com ponent
d ire c tly by any s o rt of in stru m en t and from any position. The d istan ce
from plumb point to tre e base can be m easu red and tru e height can be
derived if re q u ire d . T h ere is a check on o p e ra to r e r r o r b ecause
m easu rem en ts of the v e rtic a l com ponent from any position should be
the sam e; and th e re is com plete flex ib ility in position of o b serv atio n .
The top m ust be plumbed in from at le a s t two d ire c tio n s a t about rig h t
angles. A convenient way is for one o b se rv e r to sight the top behind
a plum b-bob strin g held a t a r m 's length in front of his eye and to
d ire c t another o b s e rv e r into that lin e, then re p e a t the p ro c e ss from
another directio n at som ething like rig h t an g les, the in te rsec tio n of
the two lines giving the plumb point.
E rro rs
T here is alw ays a likelihood of o b s e rv e r e r r o r in height
m easu rem en t - wrong reading or re c o rd in g , w rong settin g out of
d istance o r angles, m easu rin g the w rong top in a dense stand,
m easu rin g a side branch instead of the top, d ifference of opinion
am ongst o b s e rv e rs in nom inating the top of a tre e with an um brageous
crow n. P e rso n a l e r r o r is b e st checked by two independent
m e a su re m e n ts, e ith e r by the sam e o p e ra to r from d ifferen t p o sitio n s
T re e Height 35
o r by two o p e ra to rs from differen t p o sition s. The m easu rem en t of
d istan ce from o b se rv e r to tr e e m ust be done to the cen tre of the tre e ,
not to the side of the tre e facing the o p e ra to r. A sighting angle of
about 45° (less ra th e r than m ore) has th e o re tic a l and p ra c tic a l
advantages. Instrum ents a re obvious so u rc es of e r r o r and should be
checked before use to en su re they a re in adjustm ent. A m a jo r source
of e r r o r is m easuring a leaning tre e as if it w ere v e rtic a l. W hatever
method of m easu rin g the height of a leaning tre e is used it is e sse n tial
to d eterm in e the plane of the lean to avoid e r r o r s , and the only
sa tisfa c to ry way to do this is by plum bing in the top with an aid such
a s a plum b-bob; the eye cannot determ in e the plane of lean without
som e objective aid.
O cular E stim ate

F o r m anagem ent p u rp o se s, heights of tre e s in coniferous plantations
a re usually determ ined by in stru m e n t, but e stim a te s of tre e heights
a re often determ ined o cu larly in native fo re s ts b ecau se conditions of
topography and undergrow th would m ake the use of an in stru m en t
difficult. In fact, the m ore difficult it is to use an in stru m e n t the
m ore likely a re o c u lar e stim a te s subject to b ia s. If a tre e height is
re q u ire d fo r any but a casual purpose it is w orth the tro u b le and
expense to d eterm in e it by in stru m e n t.
6 Form and Taper
T h ere a re good re a so n s fo r w anting to know the shape (form) of a tre e

stem . In C hapter 3 it was pointed oilt th at the m ost p ra c tic a l way of
determ ining the volum e of a solid such as a stem is to equate it with
a geom etrical solid and d eterm in e its volume from the form ula for
deriving the volum e of that solid. We need to know the shape of the
stem so as to decide what g eo m etrical solid to equate it with. A lso,
if we can d eterm in e why a stem develops to a p a rtic u la r shape we
m ight be able, fo r exam ple by breed in g , to in c re a se the volume of
u tilisa b le wood - cy lin d rical s te m s, for exam ple, would be le s s
w asteful of wood in sawing than tap erin g stem s. Consequently, the
kind and the ra te of stem ta p e r have alw ays been of p a rtic u la r in te re s t
to fo re s te rs .
O bservation shows that th e re is a la rg e v a rie ty of stem shapes -
the bulbous bole of an A u stralian k u rrajo n g , the conoid appearance of
an open-grow n redw ood, the la rg e b u ttre s s e s of som e ra in -fo re s t
tr e e s , the slow re g u la r ta p e r of a plantation-grow n conifer.
N e v erth ele ss, tre e stem s can be c la ssifie d into two broad groups. In
one group, apical dom inance is stro n g , at le a s t to a fa irly m atu re age,
so that the stem is continuous and undivided from ground to tip and the
b ran ch es extend la te ra lly over the whole o r the upper p a rt of the stem ,
depending on fa c to rs such as the growth habits of the sp ecies and the
environm ent. In the o th er group, apical dom inance is weak at le a s t
a fte r an e arly age and the stem is not continuous from ground to the
tip of the tre e but m e rg e s into a crow n e ith e r gradually o r abruptly,
the crow n being in g en eral of an um brageous shape on top of the stem .
The te rm s conifer and b ro ad -leav ed a re conventionally used to
identify these two groups. These te rm s a re u n satisfa c to ry as the
c h a ra c te ris tic of the f ir s t group is not typical of all co n ifers under
all conditions and is typical of som e b ro a d -le a v ed tr e e s under c e rta in
36
Form and T ap er 37
conditions; and the c h a ra c te ris tic of the second group is not typical of
all b ro ad -leav ed sp ecies under all conditions and is typical of som e
co n ifers under c e rta in conditions. H ow ever, no o th er suitable te rm s
a re in general use; 'e x c u rre n t' and 'd eliq u escen t' fo r the two groups
as suggested by B aker (1950) have not been widely accepted. The
conventional te rm s conifer and b ro ad -leav ed w ill th e re fo re be used
h e re with th e ir lim itatio n s im plied.
F o r both groups, the stem at any p a rtic u la r tim e a p p ea rs as a solid
g en erated by the revolution of a tap erin g rad ial p ro file around a v e rtic a l
ax is. In the case of the f ir s t group, the p ro file has at le a s t two m ain
points of inflection - one in the butt region and one som ew here in o r
clo se to the crow n. In the case of the o th er group, if we re g a rd the
stem as stopping at som e a rb itra ry point and crown beginning th e re ,
th e re is perhaps only one main point of inflection - in the butt region.
T here a r e , how ever, many m inor points of inflection and the p ro file
of any stem is re la tiv e ly com plex; even within a group of tre e s of the
sam e sp e c ie s, d .b .h . , and height growing clo se to g eth er th e re may
well be differences between d ia m e te rs at s im ila r positions along the
stem s. It is p o ssible th e re a re never two stem s with identical p ro file s.
It is difficult to specify the whole p rofile of a single stem and even
m ore difficult to specify a pro file to re p re s e n t all stem s.
F o re s te rs have, how ever, attem pted to specify stem p ro files in two
broad w ays. One is by investigating growth p ro c e ss e s and deducing
what shape a stem m ight take as a re s u lt of them . The o ther is by
d eterm ining stem shape from em p iric a l evidence and ra tio n alisin g
th is as a re s u lt of growth p ro c e ss e s .
Growth P ro c e s s e s and T ree Shape

Even under continuously favourable environm ental conditions, a
tre e ra re ly grow s continuously but ra th e r in an episodic way. The
stem of a tre e is a solid of m a terial which has accum ulated as a s e r ie s
of la y e rs but a la y e r m ay o r may not be of the sam e thickness along
the v e rtic a l axis of the stem nor around the stem at any p a rtic u la r
height. F o r c e rta in sp ecies in a c e rta in environm ent, the developm ent
of the la y e rs is c o rre la te d with season and the la y e rs a re m ore o r
le ss re a d ily identifiable - the w ell-know n annual rin g s, each
d ifferentiated into e a rly and late wood. F o r o th er sp ecies under o th er
environm ental conditions, the la y e rs m a y b e identifiable as se p a ra te
la y e rs but a re not c o rre la te d with tim e. F o r o th e r sp ecies and o th er
conditions again, th e re m ay be no differen tiatio n into la y e rs a t all and
it is difficult to ap p reciate that grow th has not been continuous in tim e
at all points on the t r e e .
The m anner in which the m a te ria l is accum ulated depends on the
individual - its sp ecies and genetical m ake-up; its age; on the p re sen t
and past conditions of its environm ent - the so il, m eteo ro lo g ical, and
stand conditions; and on the length of tim e it has spent in those various
environm ental conditions. An enorm ous num ber and range of
investigations over a long period have shown th is. ’Growth of a tre e
is a com plex phenomenon compounded of re sp o n se s of its p rim a ry
apical m e ris te m s and its secondary cam bial m e ris te m s to both in trin sic
and e x trin sic fo rc e s which a re not uniform e ith e r in tim e o r sp ace'
(F orw ard and N olan, 1961). L arso n (1963) has an ex cellen t sum m ary
of the v a ria b le s affecting stem fo rm , a d iscu ssio n of the c h a ra c te ris tic s
of open-grow n and stand-grow n tr e e s , the effects of in h e rita n c e, site ,
thinning and pruning, and so on. The v aria tio n s in tre e form and
fa c to rs with which th ese v a ria tio n s ap p ear to be c o rre la te d a re also
d iscu ssed in many tex ts on tre e growth such as those of K ra m e r and
Kozlowski (1960) and Kozlowski (1963). To explain the v a ria tio n s in
the way tre e s accum ulate woody m a te ria l with tim e , v ario u s stem
form th e o rie s have been proposed. The b e tte r known of th ese th e o rie s
w ere proposed during the period 1880-1920.
The original p ro positions of the n u tritio n al theory by H artig , and
la te r developm ents by other w o rk e rs, and the hypothesis of J a c c a rd 's
w ater conduction theory a re su m m arised by L a rso n (1963). Though
v aria tio n s in the distrib u tio n of in c re m en t on the stem a re p resu m ed
to be explained by th e se th e o rie s , n e ith e r lead s to assu m p tio n s of
specific shape of the whole o r the m ain stem .
The m echanistic th e o ry , on the other hand, does propose a stem
shape. A ssociated m ost prom inently with M etzger, it a ssu m e s 'the
p rin cip le of wood form ation in the stem is governed by the re q u ire m e n t
of the tre e for m echanical stre n g th ’ (Busgen et a l . , 1929) and is often
re fe rre d to as 'M e tz g e r’s beam th e o ry ’. In a s e r ie s of p a p e rs,
M etzger suggested th a t the m ain stem should be p rim a rily a v e rtic a l
stru c tu ra l m em ber able to w ithstand horizontal wind fo rce s on the
crow n and stem and the v e rtic a l fo rc e s co m p risin g the weight of the
stem and the w eight of w a te r, snow, and ice on the crow n; and should
be made of minimum m a te ria l c o n sisten t with the idea that the m axim um
proportion of the food supply is d ire c ted to crow n and rep ro d u ctiv e
organ developm ent. That is, it should be a beam of uniform re s is ta n c e ,
Form and T ap er 39
having the highest possible bending stren g th in all p a rts with the
sm a lle st expenditure of m a te ria l. This im plied, according to M etzger,
th at the m ain stem should be of the shape of a th ird d eg ree paraboloid
and he dem onstrated that many tre e ste m s, p a rtic u la rly co n ifers and
to a le s s e r extent b ro ad -leav ed sp e c ie s, conform ed with th is p rin cip le
by the lin e a r relatio n of the cubes of d ia m e te rs with the resp ec tiv e
heights from ground level at which they w ere m e asu re d , o v er m ost of
the bole (excluding a section n e ar the top and a section n e a r the ground
under the influence of bu tt-sw ell). Since th is th eo ry p roposes to
account for the in tern al and ex tern al fo rce s acting on a stem , the
re su ltin g s tr e s s e s in the stem , and the way the stem develops, many
investigations have been planned to prove o r d isp ro v e th ese asp ects
of the theory. Some of th ese a re su m m a rised by L arso n (1963).
R ecently, the ro le of growth re g u la to rs (growth substances) has
been studied intensively in re la tio n to the in itiatio n and control of
cam bial activity and the developm ent of cam bial p ro d u cts,
com prehensively su m m arised by R o m b erg er (1963). L arso n (1962)
outlined how gradients of auxin tra n slo ca tio n originating in the crow n
m ight reg u late the distrib u tio n of ra d ia l growth on the stem . W hilst
th is horm onal theory provides a physiological explanation of how a
tre e grows and why tre e s differ in the way they ta p e r, it does not
specify the p a rtic u la r shapes which tre e s m ight have under varying
c irc u m s ta n c e s .
The quality of the wood is ju s t as im portan t as the quantity of the
la y e r of wood at any position in the tre e and is c o rre la te d fa irly clo sely ,
from the viewpoint of density and thus stre n g th , with the proportion of
la te wood in the growth rin g s. Many observ atio n s have been made of
the p a tte rn s of v ariatio n of e a rly and late wood but none is read ily
re la ted to stem form .
L arson (1963) su m m a rised the position of th ese th e o rie s thus:
the n utritional th eo ry is convincing in manj'- re s p e c ts but re s e a rc h
has shown that n u tritional g rad ien ts cannot possibly provide the
re q u ire d re g u lato ry control. The w ater conduction and m echanistic
th e o rie s , being functional in n a tu re , cannot in th em selv es provide
an adequate physiological explanation fo r the observed facts. From
the standpoint of strength re q u ire m e n ts and support function, the
m echanistic th eo ry provides a re a lis tic in te rp re ta tio n that has
survived experim ental study reasonably w ell. This is p a rtic u la rly
tru e when the v aria tio n s in latewood percen tag e and wood density
a re co n sid ered in conjunction with stem ta p e r. The horm onal
theory, envisioning sp atial and tem p o ral v aria tio n s in auxin g rad ien ts
throughout the stem , provides the m ost p ro m isin g approach to the
stem form problem . In te rm s of th is th eo ry , auxins e x e rt the
re g u lato ry control governing the d istrib u tio n of ra d ia l grow th on the
stem and thereby provide a physiological b a sis for the n u tritio n al
as well as the functional th e o rie s of stem form .
H eger (1965) found differen ces in the 'fo rm ' of e a rly and late wood
la y e rs in Douglas f ir which he felt could not be explained by the
m echanistic, n u tritio n al, conductive, o r horm onal th e o rie s of stem
form ation and proposed th at 'ra d ia l growth m ay p roceed a t varying
ra te s fo r unequal perio d s of tim e within d ifferent p a rts of the cam bium
cy linder depending la rg e ly on the lev els of su b co rtical te m p e ra tu re s '.
V arious w o rk ers have studied the s tr e s s e s and s tra in s in tre e stem s
as they grow: Jacobs (1965) and Boyd (1950a, b. c) have been prom inent
in work on A u stralian sp e c ie s. From th is w ork, how ever, no
suggestions have been made as to the shapes tr e e stem s m ight take in
developing and accom m odating to th ese s tr e s s e s and s tra in s .
Stem P ro file s
At the sam e tim e as stem form th e o rie s w ere being proposed to
explain how a tre e grows and the cause of v a ria tio n s in tre e shape,
v ario u s w o rk ers w ere investigating the em p iric a l evidence of tre e
shape. About the tu rn of the century in w e ste rn Europe a tte m p ts, by
Jonson in p a rtic u la r, to d e sc rib e the p ro files of co n ifers by form ulae
derived as the b e st fit to d ia m e te rs at v ario u s heights on the stem
stim ulated a la rg e amount of w ork of th is kind in many p a rts of the
w orld. D escription of a tre e p ro file by form ula is conventionally
called a stem p rofile equation. F o r som e tim e p rio r to th is , thinking
in Europe about tre e volum e was p a rtic u la rly concerned with finding
c o rre la tio n s between form fa c to r and som e index which could be
m easu red conveniently on a stem , form facto r being a fa c to r by which
the product of the height and som e c ro s s -s e c tio n a l a re a of the stem
has to be m ultiplied to give a p a rtic u la r ex p ressio n of volum e of the
stem . Depending on the sp ecies of the tr e e , its genetic background
and environm ent, on the e x p re ssio n of height used and w here the
c ro s s -s e c tio n is m e asu re d , and on the ex p ressio n of volum e re q u ire d ,
so the kind and quantitative value of a form fa c to r v a rie s . It may o r
m ay not infer by its value som ething about the form (shape) of a stem .
Such a c o rre la tio n was found between form facto r and form quotient,
F orm and T ap er 41
g e n erally defined now as the ra tio of the d ia m e te rs re sp ec tiv e ly at
half the height above b re a s t height and at b re a s t height, the ra tio being
conventionally e x p re ssed as a decim al, for exam ple -70.
Conventionally, c la s s e s of form quotient values known as form c la s s e s
w ere estab lish ed for classify in g tre e s in frequency d istrib u tio n s, for
exam ple form c la ss 70 (tree s with form quotients from -675 to -725).
F rom num erous m e asu rem en ts of d ia m e ter at p e rc en tile heights
(that is at tenth heights above b re a s t height) of c e rta in w est European
c o n ife rs , Jonson (1910-12) proposed that the p ro file of such tre e s
from b re a s t height upw ards could be d escrib ed by the form ula
L ____ (c + L - 2-5)
" dT - c l o g ------- 5-------
w here DL = the d ia m e ter at p e rc en tile d istan ce _L from the tip , _D =
d ia m e te r at b re a s t height and and c a re constants varying with form
c la s s . This form ula was a m odification of one which H ojer had
pro p o sed som e y e a rs previously. From th is equation, Jonson com piled
ta b le s giving tre e d ia m e ters at p e rc en tile h eig h ts, as a percentage of
d ia m e te r at b re a s t height, fo r v ario u s form c la s s e s . The p ro file of
a stem being defined according to its form quotient, so the form
fa c to r of the stem was autom atically defined. T his led re ad ily to
d eriv atio n of stem volume as the product of b asal a re a , height, and
form fa c to r. B ecause of the obvious p ra c tic a l difficulty in m easu rin g
d ia m e te r at half tre e height above b re a s t height, Jonson looked for an
index with which form quotient m ight be c o rre la te d . A rguing from
M etz g e r's beam theory th at the c e n tre of g rav ity of the crow n should
be such an index, he estab lish ed a c o rre la tio n between the re la tiv e
position of this form point on the tr e e and the form quotient so that
form quotient could be determ ined for a tre e from an estim ate of the
re la tiv e position of the cen tre of g ravity of the crow n. From the
n atu re of the exp ressio n of the stem p ro file, tr e e s of the sam e total
height and d iam eter at b re a s t height but of different form quotient a re
re g a rd e d as being of d ifferent shape.
Following Jon so n 's w ork, W right (1927) c a rrie d out a detailed study
of ta p e r cu rv es (stem p rofiles) of c e rta in co n ifers in Canada. In the
United S tates, B ehre (1927) proposed an equation to d e sc rib e the stem
p ro file s of se v e ra l N orth A m erican co n ifers which, in te rm s of the
Dl l
sym bols used in Jo n so n 's equation above, is - j j - = ^ '
the constants C and c having, of c o u rs e , d ifferen t arith m e tic values

from those in Jo n so n 's equation. The d ia m e te rs at p e rc en tile heights
for vario u s form c la s s e s a re reaso n ab ly s im ila r to those of Jonson
fo r the m ost p a rt. These and o th er form ulae proposed by w o rk e rs in
Europe w ere subjected to co n siderable investigation with confirm ation
o r denial of the s a tisfa c to ry natu re of stem p ro file equation depending
on the species studied and its environm ent, and subjective opinions of
what was ’s a tis fa c to ry '. One m a jo r problem was th at the form ulae
only applied above b u tt-sw ell and if bu tt-sw ell w ere p re se n t above
b re a s t height application of the form ulae was difficult.
In A u stra lia, G ray (1943, 1944, 1956, 1966) hypothesised 'th at
within the usual lim its of a cc u ra cy of tre e m e asu re m e n t, the c e n tra l
length of the bole u n d e r-b a rk approxim ates as c lo sely to a solid body
g enerated by a second degree parab o la as to any o th er g eo m etrical
body even though v ariatio n of form quotient m ay be c o n sid e ra b le '. He
suggested that th is shape m ay have differen t ta p e r (m ore s tric tly ,
d ifferent ra te s of tap er) according to fa c to rs such as the size and
shape of the crow n, the fo rc e s of wind on the crown and the w eight of
crow n and stem . T hese fa c to rs would also influence the extent and
shape of top and butt sections which a re su p erim p o sed on the m ain
bole. G ray based h is hypothesis on em p iric a l evidence from
m easu rem en t of a la rg e num ber of tre e s over a wide range of sp ec ie s.
He subsequently attem pted to show that the hypothesis was co n sisten t
with the physiological and stru c tu ra l re q u ire m e n ts of tr e e s . The
p ra c tic a l im plications of the hypothesis a re d iscu ssed in C hapter 8.
Many w o rk ers such as P e trin i (1921) have em p h asised th at the
p ro file is of a compound natu re and can only be d e sc rib e d adequately
by se p a ra te ex p re ssio n s for the top sectio n , the m ain bole, and the
b utt-sw ell region.
The Main P ro p o sitio n s of T re e Shape C om pared

If the u n d e r-b a rk d ia m e te r of a m atu re fo rest-g ro w n co n ifer,
undivided from ground to tip , is m e asu re d at v e ry clo se in te rv a ls
along the stem , th ese values plotted on re c ta n g u la r c o -o rd in a te p ap er
against the heights of the m easu rem en ts from ground lev el, and these
points a re joined, a lm o st in v ariab ly they form an ir r e g u la r line which,
how ever, if sm oothed out, ap p ears as a compound curve w ith two
m ain points of inflection - one in the crown region and one n e a r the
ground.
Some people c o n sid e r th re e se p a ra te cu rv es a re involved - one fo r
Form and T ap er 43
m o st of the stem bearing live b ran ch es; one fo r the m ain stem ,
p a rtic u la rly the p a rt without live b ran ch es; and one for the b u tt-sw ell
region; and that the sections of the stem d e sc rib e d by th ese se p a ra te
c u rv e s v a ry with sp ec ie s, siz e , age, stand s tru c tu re , environm ent,
and individual vigour so that th e re is no re a so n to expect th at even two
tr e e s of the sam e sp ecies and the sam e d .b .h .o .b . and height growing
alongside each o th er would have the sam e kind and extent of th ese
c u rv e s.
Proponents of the stem p rofile equation would say th at the ta p e r
curve could be d escrib ed along its whole length by one g en eral equation,
shapes indicated by the curve differing only as the ra tio of the d iam eter
at b re a s t height and half-height above b re a s t height d iffers for different
s te m s, the shapes of stem s being the sam e irre s p e c tiv e of o v erall size
if th e ir ra tio s a re the sam e.
T hose following the p ro p o sals of M etzger o r G ray would say that all
tre e s have a main stem of common shape - cubical paraboloid
acco rding to M etzger, second degree paraboloid according to G ray -
with different ra te s of ta p e r (s tric tly , d ifferen t ra te s of change of
tap er) according to sp ec ie s, siz e , age, environm ent, and so on; and
with varying bu tt-sw ell and top.
T re e Shape in A u stralian F o re s ts
The effects of butt-sw ell and b ark th ickness m ake it difficult to
d eterm in e u n d e r-b a rk form quotient consisten tly but if the influence of
b u tt-sw ell can be rem oved the range of u n d e r-b a rk form quotient for
plantation co n ifers in A u stra lia is about 0- 60 to 0-80. O ver th is range
of form quotient, Jo n so n 's equation gives a reaso n ab le fit fo r varying
lengths of m ain stem depending on a num ber of fa c to rs - stand density,
site quality, the extent of pruning. Stem s with a form quotient about
0- 80 also give a reaso n ab le fit to M etz g e r’s cubical paraboloid since
the form quotient of a cubical paraboloid is about 0-80. Stem s with a
form quotient about 0- 70 give a reaso n ab le fit to G ra y 's second degree
paraboloid since the form quotient fo r a second deg ree paraboloid is
about 0-70. It so happens that a la rg e percen tag e of the stem s of
plantation conifers in A u stra lia have a ’b a s ic ’ form quotient (that is ,
form quotient m easu red so that b u tt-sw ell does not influence it) of
about 0-70 so that G ray has found co n sid erab le evidence fo r the
'g en e ra lity ' of his hypothesis. Under c e rta in stand conditions which
influence the num ber, siz e , and disposition of the b ran ch es (which
sim ila rly affect the fit of Jo n so n 's equation), som e tre e s show an
evident point of inflection in th e ir p ro files within the live crow n. If
th is inflection is slight, one ta p er line may serv e as a reaso n ab le
approxim ation fo r the whole of the m ain stem ; so m etim es two se p a ra te
ta p e r lines may be p re fe ra b le ; som etim es th e section of the m ain stem
within the live crow n cannot be fitted at all s a tisfa c to rily with a lin e a r
ta p e r line and for such tre e s only a p ro file of a compound c u rv ilin e a r
n atu re would be re a lly sa tisfa c to ry . Data from the Q ueensland
D epartm ent of F o re s try (H enry, p riv ate com m unication) suggest that
hoop pine in plantations approxim ates as clo sely to a second degree
paraboloid as to anything else; but fo r a fa irly la rg e sam ple of
plantation slash pine, the Jonson equation provided a v ery good fit
over m ost of the stem over a 10-16 in range of d . b . h . o . b . and a 60-70
range of form quotient.
G ray has suggested that the ste m s of b ro ad -leav ed sp ec ie s up to
crow n b reak a re truncated second degree parab o lo id s with b u tt-sw ell
v ario u sly superim posed. As fo r c o n ife rs, this seem s to hold for m ost
native b ro ad -leav ed species much of the tim e but not for a sp ecies all
the tim e. It holds wrell fo r young second-grow th eucalypts. F o r o ld er
eucalypts, a la rg e proportion of the stem is usually influenced by
b u tt-sw ell.
7 Bark and Crown
BARK
The ra d ia l thickness of the sheath of b a rk on a tre e d e c re a se s from
ground to tip , usually in a fa irly re g u la r way though not n e c e s sa rily
with the sam e ra te of ta p e r as the stem . B ark th ick n ess v a rie s
am ongst sp ecies and also within a sp ecies according to age, s iz e , and
environm ent. Even in an even-aged stand of one sp e c ie s, b a rk
th ick n ess m ay vary am ongst individuals of the sam e d .b .h . o .b . and
height. On an individual tr e e , b a rk thickness may v a ry around the
stem at a p a rtic u la r height above ground. N e v e rth e le ss, c o rre la tio n s
can usually be estab lish ed betw een ex p re ssio n s of b a rk th ick n ess and
tre e o r stand v a ria b le s . From th ese re la tio n sh ip s e stim a te s of the
quantity of b a rk as a useful or w aste product can be m ade, and
e stim a te s of u n d e r-b a rk dim ensions obtained from o v e r-b a rk
dim ensions. U nder-bark dim ensions can also be d eterm in ed by d ire c t
m e asu rem en t on felled tr e e s by m easu rin g at the ends of logs o r
rem oving rings of b a rk , and on standing tre e s through m easu rem en t
of b a rk thickness. The te rm s d ia m e ter o v e r-b a rk (d .o .b .) and
u n d e r-b a rk (d .u .b .) a re the ones com m only used in A u stra lia . In
som e o ther c o u n trie s, the te rm s d ia m e ter outside b ark ( d .o .b .) and
inside b a rk ( d . i . b . ) a re conventional.
Since tre e volume is e x p re ssed on an u n d e r-b a rk b a sis alm o st
everyw here in A u stra lia , the m a tte r of b ark th ick n ess is v ery
im p o rtant.
M easurem ent of B ark T hickness
On felled tre e s or logs, b a rk thickness is so m etim es m easu red on a
chip. If so, c a re should be taken that no b ark is lo st, o r th at the b ark
is not c o m p ressed , in cutting the chip. F o r standing tr e e s , and as an
alte rn ativ e to m easu rin g a chip in the c a se of felled tr e e s , b ark
45
thickness is m easu red by som e form of b ark gauge. The m o st common
type is the Swedish b a rk gauge, e sse n tia lly a s e m i-c irc u la r gouge-chisel
with a wooden o r m etal handle. By p re s s u re on the handle, the ch isel
end is forced through the b a rk , the depth of p en etratio n of the chisel
being determ ined on a suitably graduated sc a le , while the gauge is in
the tre e , by the upper end of a sleeve the bottom end of which has a
flange which is p re s se d against the outside of the tr e e . T here a re
sim p le r hom e-m ade types which take v ario u s fo rm s depending on the
in sp iratio n of the m ak er. Lewis (1953) d e sc rib e s two designs which
have been used in South A u stra lia . A v e ry sim ple type can be made
by driving a graduated nail through a block of wood which s e rv e s as a
handle. U sually with the sim ple types, the level of the b a rk on the
gauge is m arked by thum b-nail while the gauge is w ithdraw n and the
graduation re a d .
Any such b a rk gauge depends fo r its function on the assum ption that
the b a rk w ill be p en etrated , and th a t the sapwood w ill not be p en etrated ,
by the gauge; o r th a t the b ark is m o re e a sily p e n etrate d than the wood
and the o p e ra to r can detect when the point of the gauge re a ch e s the
junction between the two. U nfortunately, th ese assu m p tio n s a re not
alw ays tenable, m any tre e b a rk s being co n sid erab ly h a rd e r than the
sapwood. Even when the b a rk is s o fte r than the wood, no p racticab le
way has yet been found of judging when the gauge has ju st passed
through the b ark and is about to p e n e tra te the wood. A fter co n sid erab le
ex p erien ce, p a rtic u la rly with one sp e c ie s, o p e ra to rs develop a ’feelin g ’
in using a b ark gauge and th e ir m e asu re m e n ts can be re lie d on. The
effect of e r r o r in b a rk thickness m e asu re m e n ts is not always
appreciated: an e r r o r of O’ 1 in in the ra d ia l b ark th ick n ess of a 10 in
d. o .b . log re s u lts in an e r r o r of n e a rly 5 p e r cent in the u n d e r-b a rk
volum e; an e r r o r of 0- 2 in re s u lts in a volum e e r r o r of about 9 p er
cent.
V arious m odifications to the Swedish b ark gauge have been suggested
to im prove re lia b ility without much s u c c e ss, but no alte rn ativ e way
of m easuring b ark th ick n ess, such as rem oving c o re s of b ark for
d ire c t m easu rem en t of b a rk th ic k n e ss, have proved as g en erally
sa tisfa c to ry as by the b a rk gauge. Any of the sim ple types seem
s a tisfa c to ry in sm ooth b a rk up to about 0-5 in: beyond that thickness
the Swedish gauge is usually m ore sa tisfa c to ry . When m easu rin g the
thickness of rough and deeply fis s u re d b a rk , it is im p o rtan t to keep in
mind the purpose of the m e asu re m e n t and how the o v e r-b a rk
m easu rem en t was taken, for exam ple w hether by c a lip e r o r tape.
B ark and Crown 47
B ecause of the likely v a ria tio n of b ark thick n ess around a tr e e , se v e ra l
e stim a te s of it should be m ade - two at rig h t angles (on the long and
sh o rt axes if the tr e e section is eccen tric) o r four on a c ro s s . C are
should be taken with a b a rk gauge w hile clim bing a standing tre e ; in
good o rd e r, a b ark gauge is a sh arp gouge ch isel and a guard should
be placed on the sh arp end to avoid p o ssible injury to the o p e ra to r.
The re la tio n sh ip s between o v e r-b a rk and u n d e r-b a rk dim ensions on
a stem a re :
rad iu s u n d e r-b a rk = radius o v e r-b a rk - ra d ia l b a rk th ick n ess
( r .u .b . = r . o .b . - b . t . );
d ia m e te r u n d e r-b a rk = d ia m e te r o v e r-b a rk - ra d ia l b a rk th ick n ess x 2
(d .u .b . = d .o .b . - 2b. t . );
g irth u n d e r-b a rk = g irth o v e r-b a rk - ra d ia l b a rk th ick n ess x 2 tt
(g .u .b . = g . o . b . - 2 tt b. t . ).
B ark R elationships
The relatio n sh ip betw een b a rk thick n ess at b re a s t height and d ia m e ter
o v er b a rk a t b re a s t height is of p ra c tic a l in te re s t. In estab lish in g
th is re la tio n sh ip fo r a sp e c ie s, the re la tiv e contribution of lo cality ,
age, tre e height and site to v ariatio n in b a rk th ick n ess at b re a s t
height should be investigated. Studies of ra d ia ta pine in A u stra lia
suggest th at differences occur due to lo cality . F o r exam ple, values
in South A u stra lia a re considerably le s s than in the A. C. T. Some
in vestigations have shown th a t, w ithin lo c a litie s, v aria tio n is re la te d
to both tr e e height and d . b . h . o . b . ; o th er investigations have suggested
th at a sim ple relationship betw een b a rk thickness and o v e r-b a rk d iam eter
was sufficiently p re c is e fo r all ages and s ite s . This is som etim es
lin e a r. F o r data of 2,800 tr e e s of Pinus rad iata in Kowen F o re s t,
A. C. T. , with a d . b . h . o . b . range of 4 in to 18 in. the relatio n sh ip was
c u rv ilin e a r but v e ry flat with values as shown in Table 7:1.
T a b l e 7:1
Bark thickness a t br e a s t h e i g h t P, r ad i at a , A . C . T .
(after Ca r r o n and Jacobs, 1964)
d. b. h. o . b . (in) 2b. t. (in) d. b . h. o. b. (in) 2b. t. (in) d . b . h . o . b . (in) 2b. t. (in)
5-0 0-65 1 0- 0 1-58 15-0 2-20

6-0 0-88 11-0 1-73 16-0 2- 3 3
7 -0 1-08 12-0 1-85 1 7- 0 2- 4 2
8-0 1-25 13-0 1-98 1 8- 0 2-52
9-0 1-43 14-0 2- 10
F o r some sp ecies under c e rta in conditions, b ark th ick n ess b e a rs
approxim ately the sam e ra tio to o v e r-b a rk d ia m e te r at any position on
the tre e . F o r o th er sp ec ie s o r different conditions th is is not the c ase.
The volume of a log of a broad-leav ed sp ec ie s o r a native co n ifer is
g enerally derived in A u stra lia by re fe re n ce to the cen tre girth. This
is so m etim es m easu red o v er the b ark and a g en eralised allow ance for
b a rk thickness su b tra cte d to give u n d e r-b a rk cen tre g irth .
A ltern ativ ely , a rin g of b ark is rem oved at the point of m easu rem en t
and the u n d e r-b a rk g irth m e asu re d d ire c tly . The tim e spent in
rem oving the rin g of b a rk is d isprop o rtio n ate to the tim e spent in
falling, topping, and butting under m odern conditions and th e re is an
in c re asin g tendency to m e a su re o v e r-b a rk g irth and use c o rre ctio n s
for bark th ick n ess. When the re la tiv e b ark th ick n ess is the sam e
everyw here on the tr e e the Simple relatio n sh ip d .u .b . = k d .o .b . w ill
suffice, otherw ise fa c to rs such as the position of the log in the tre e
and the log length have to be introduced into the relatio n sh ip as fu rth e r
independent v a ria b le s.
F or some sp ecies under c e rta in conditions, b a rk volume is also the
sam e re la tiv e to stem volum e, over a wide range of tre e size, for
exam ple for ra d ia ta pine in the A. C. T. the u n d e r-b a rk volume of stem s
o r logs can for m ost purposes be reg ard ed a s 85 p e r cent of the
o v e r-b a rk volum e. F o r m ore re lia b le e stim a te s , c o rre ctio n s need to
be based on fa c to rs such as lo cality , size , and site . When b ark is a
m inor product th e re is considerable advantage in com piling a b ark
volum e tab le. U sually the sam e m ethods a re used as for tre e volum e
table com pilation (see ch. 13). A b a rk volum e table fo r rad iata pine,
U r ia r r a F o re s t, A. C. T. , is shown as Table 7:2. F o r tre e s with
furrow ed b a rk , the volume of b a rk computed from conventional
m easu rem en ts of b a rk thickness by b a rk gauge w ill include the a ir
space of the fu rro w s.
CROWN
The d ia m e ters of the crow ns of tr e e s in a stand a re frequently

c o rre la te d with the d ia m e te rs of the stem s at b re a s t height. The
relatio n sh ip is com m only a sim ple ra tio and so is usually called
crow n r a tio . In many c a se s it is not a sim ple ra tio and may even be
c u rv ilin e a r. The relatio n sh ip has been used for a long tim e fo r
v a rio u s silv ic u ltu ra l p u rp o ses. Jacobs (1955) d isc u sse s its use in the
B ark and Crown 49
Table 7:2
Bark v o l ume (cu ft) to 4 in d . u . b . ; P. radiata. Uriarra Forest, A . C . T . ,
170 sampl e trees
d . b. h. o . b . Total height (ft)

(in)
40 50 60 70 80
6 0-52 0-63 0- 72 0-82

7 0-70 0-84 0-90 1- 15
8 0-90 1- 10 1-30 1-50
9 1-13 1 - 37 1-68 1-95
10 1-38 1-70 2 - 05 2-42 2-70
11 2-00 2-50 2- 9 5 3-30
12 2-40 3-00 3-50 4-00
13 3-50 4- 10 4-70
14 4-80 5-50
15 6-40
silv ic u ltu re of the native eucalypt fo re st; Dawkins (1963) d isc u sse s
its behaviour in som e tro p ic a l fo re s ts . The re la tio n sh ip is also
valuable in m ensuration on a e r ia l photographs since stem d iam eter
can be derived in d irectly from crow n d ia m e ter m e asu re d on the
photographs; and crow n d ia m e te r can be used as an independent
v a ria b le in a e ria l tre e volum e ta b le s as d iscu ssed by Spurr (1960).
To esta b lish the relatio n sh ip actual crown d ia m e ter has to be
m e asu re d . F or som e tr e e s th is can be done by clim bing the tre e and
extending a c a lib ra te d pole ho rizo n tally . In m ost c a se s the tre e s or
crow ns a re too la rg e for th is to be done and crow n d ia m e ter is
m e asu re d by pro jectin g points of the crow n p e rim e te r on to the ground
in som e way and m easu rin g on the ground the d istan ce betw een those
p ro je c tio n s. U sually the m ean is taken of two d ia m e te rs at rig h t
an g les. V arious m ethods a re used o r have been suggested to p ro ject
the edge of the crow n to the ground. F or som e p u rp o se s, visual
subjective judgm ent may be sufficient. Commonly, one o p erato r
aligning a plum b-bob strin g on to the edge of the crow n d ire c ts an
a s s is ta n t on to the p ro jected point. V arious optical devices for
d irectin g the a s s is ta n t to a position v e rtic a lly below the crown edge o r
for an o p e ra to r to so position h im se lf a re d iscu ssed by N ash (1948),
S arvas (1956), and W alters and Soos (1962).
The sectional a re a of the crow n of an individual tre e can be
calcu lated from its d ia m e te r as if the crow n w ere c irc u la r . If the
crow n is ir r e g u la r and a m ore exact e stim a te of its sectio n al a re a is
re q u ire d , the outline of the crow n can be draw n on plan by plumbing
the crow n edge to the ground at an ap p ro p ria te num ber of positions,
plotting th ese points on plan by re c ta n g u la r c o -o rd in a te s, joining the
points and calculating the enclosed a re a by p la n im e ter o r dot grid.
Mapping the outlines of crow ns on a plan in th is way is one method of
determ ining crown c lo s u r e , the ra tio of the a re a of the v e rtic a l
p ro jectio n s of crow ns to the equivalent a re a of ground occupied by the
stand. T his is a m e asu re of stand density which is of p a rtic u la r
in te re s t in silv ic u ltu re and also a useful index fo r stratify in g stands
on a e ria l photographs. Except in special c a s e s , actu al mapping of
crown projections is usually too laborious and v ario u s devices which
give a m easu re of crown c lo su re at sam pling points have been suggested
such as the ’m oose h o rn ' (Robinson, 1947; G a rriso n , 1949). The
te rm crown c lo su re b e tte r d e sc rib e s the sectional a re a of the canopy
than the te rm crow n density which is so m etim es used fo r this but b e tte r
im plies the total bulk of the crow ns which affects quality, d irectio n ,
inten sity and duration of light and so can be m easu red by a light m e te r.
Crown length is usually determ ined from ap p ro p riate heights
m easu red with som e height m easu rin g in stru m e n t. The ra tio of length
of crown to total length of tr e e is so m etim es re fe rre d to as crow n p e r
cent.
Crown su rface and crow n volum e a re so m etim es computed for
special p u rp o ses. A shape a p p ro p riate to the growth h ab its of the
p a rtic u la r species is assum ed, fo r exam ple conoid, paraboloid, o r
h e m isp h erica l, and dim ensions obtained fo r calculation of su rface and
volume by ap p ro p riate form ulae. Dale (1962) gives approxim ations
fo r easy calculation of crown su rfa ce a re a .
Crown weight is im p o rtan t in as a ss e ss m e n t of to tal production and
the weight of slash left a fte r fellings. Brown (1963) illu s tra te s a
relatio n sh ip betw een tre e crow n weight and tre e d .b .h .o .b . , site
quality and stand density, fo r stands of red pine in the U. S. A. Wile
(1964) illu s tra te s a relationship of crown d ia m e te r, length, and weight
with tre e d . b . h . o . b . for red spruce and balsam f ir in Canada. T h ere
is v e ry little inform ation of th is s o rt for e ith e r the native fo re st o r
plantations in A u stra lia .
8 Stern Volume
If a f o r e s t is m anaged fo r wood p ro d u ctio n , the volum e of wood in the

f o r e s t is a p a r tic u la r ly im p o rta n t c r ite r io n of m a n ag em e n t and the
e s tim a tio n of th a t volum e is an im p o rta n t a s p e c t of m e a s u re m e n t.
W h atev er the m ethod of e s tim a tio n u se d , it is b a se d d ir e c tly o r
in d ire c tly on e s tim a te s of th e stem volum e of in d iv id u al t r e e s . The
re lia b le e s tim a tio n of in dividual ste m volum e is th e re f o r e one of the
m o st im p o rta n t m e n su ra tio n re q u ir e m e n ts .
The D efinition of Stem V olum e

V ery few s te m s e x ac tly r e p r e s e n t s im p le , id e a l, g e o m e tric s o lid s and
m o st have ir r e g u l a r s u rfa c e s due to th e ep iso d ic grow th p a tte rn of th e
stem and sw ellin g s a t th e b a s e of b ra n c h e s . T h is ir r e g u la r ity being
n a tu ra l to th e tr e e as a r e s u lt of its grow th h a b its , we m ig h t think of
th e volum e of a stem in b io lo g ic a l te r m s ; th a t is a s th e volum e of a
stem w hose b ra n c h e s have been trim m e d off a t th e ir ju n ctio n w ith the
s te m , excluding any ir r e g u l a r it ie s th a t a r e not p a r t of th e n a tu ra l
grow th h a b it of th e stem such a s th o se due to the e ffe c ts of in s e c t and
fungal a tta c k and f ir e . T his is an o b jectiv e d efin itio n of v o lu m e. It h a s ,
h o w ev er, obvious d isa d v a n ta g e s. If o u r in te r e s t is in u tilis a tio n , the
b io lo g ical volum e of the stem m ay be quite u n re a l. F u r th e r , we would
expect s te m s w ith com m on v a lu e s of d . b . h . o . b . , h eig h t, and ta p e r to
have th e sa m e volum e and th is e x p ec tatio n should not be u p set by
i r r e g u l a r it ie s unique to in dividual t r e e s . F ro m th e se v iew points it is
lo g ical to exclude the n o rm al ir r e g u l a r it ie s of ep iso d ic grow th and
b ra n c h ju n c tio n s in d e te rm in in g stem volum e - but to w hat e x te n t? F o r
a tr e e w ith a sm ooth stem such a s E u caly p tu s g r a n d is , p o te n tia lly
u tilis a b le volum e and b io lo g ical volum e a r e the s a m e . F o r a tr e e w ith
la rg e b ra n c h e s in c lo se w h o rls su ch as P in u s r a d i a t a , b io lo g ical volum e
51
m ay have som e re a lity for a pulpwood operation but not for sawn tim b e r
production, and perhaps it is m o re re a lis tic in such a c ase to define
volum e by the minumum d ia m e te rs in internodal a re a s . T h ere a re
m any g ra d e s of irre g u la rity betw een these two c a s e s . Obviously any
volum e o ther than biological volum e involves som e subjective judgm ent
and the volum e of a stem can no longer be rig id ly defined; though th is
is not to say th a t biological volum e can be re ad ily m e asu re d .
The volum e of a stem is defined la rg e ly by the m ethod used to
d eterm in e it and so com parison of methods by re fe re n c e to som e
stan d ard volum e is difficult. A p a rtic u la r method might b e st be judged
according to how w ell it m eets the ob jectiv es. B efore applying a
m ethod one m ust decide what volum e one w ants.
F a c to rs A ffecting M ethods of D irect E stim ation of Stem Volume

If in a method of estim atin g volum e a stem o r a p a rt of it is assu m ed
to approxim ate a c e rta in solid, the re lia b ility of the e stim ate of volum e
w ill depend on the validity of that assum ption for the p a rtic u la r case
and how c lo sely the dim ensions which a re d eterm in ed conform to those
a p p ro p ria te to deriving volum e fo r that solid.
The type of tr e e m ay influence the choice of m ethod. F o r a c o n ife r,
volum e can be estim ate d fo r the whole stem o r for any p a rt betw een
c e rta in a r b itr a r y lim its. F or a b ro ad -leav ed sp e c ie s, an upper lim it
is im posed by crown b reak . A method applicable to the one may not
be applicable to the other.
W hether a tre e is felled or standing is of p a rtic u la r im p o rtan ce. A
tre e on the ground is readily a cc e ssib le fo r m e asu rem en t and th e re is
h ard ly any lim it to the num ber of m easu rem en ts th at could be taken.
H ow ever, p ra c tic a b ility and econom ics set lim its to m e asu rem en t of
a standing tr e e .
R equirem ents of the Ideal Method

The m ethod of e stim atin g stem volume should su it the purpose for
which the inform ation is re q u ire d . It should be sim ple and p ra c tic a b le ,
p a rtic u la rly if it is to be applied to standing tr e e s , and it should be
econom ical in tim e and m oney, p a rtic u la rly if a la rg e num ber of tre e s
is involved. The method should give a re s u lt which is unbiased, and
which is p re c is e for the sam e o p e ra to r and among o p e ra to rs . The
m ethod should be such as to p e rm it m istak es to be re a d ily d etected if
Stem Volume 53
they o c c u r. No method w ill m eet all th e se re q u ire m e n ts p e rfe ctly and
som e com prom ise has to be accepted.
Stem volume D erived by Volume of Sections
If a stem is considered to c o n sist of a num ber of sec tio n s, its volume

can be deriv ed as the sum of the volum es of those sec tio n s. It has
p rev io u sly been pointed out that in p ra c tic e the volum e of a log cut from
the m ain stem is calculated alm o st invariably by e ith e r the Huber or
Sm alian form ula as if it w ere a frustum of a second deg ree paraboloid.
Following th is , the volum es of the sectio n s co m p risin g the stem a re
u su ally calculated by e ith e r the Huber o r Smalian form ula.
Sectional Method
T his is a v ery old m ethod (C haturvedi, 1926) which goes by v ario u s
n a m es. It is com m only known as the sectio n al m ethod in A u stra lia .
The stem is considered to c o n sist of sectio n s of som e constant
convenient length, girth (diam eter) is m easu red at the c e n tre of each
sec tio n , and volume is calculated by H u b e r's form ula. In A u stra lia ,
the length used is 10 ft because th is is convenient in laying out the
p o sitio n s fo r m easu rem en t of g irth , and the volum e of a sectio n is
quickly calculated in cubic feet by m e re ly shifting the decim al point of
the sectional a re a in square feet one place to the rig h t. The method
can be applied to conifers o r b ro a d -le a v ed tr e e s , felled o r standing,
fo r to tal volum e, o r volume to any p a rtic u la r lim it, o v e r-b a rk or
u n d e r-b a rk . It can be applied to tr e e s with m ultiple ste m s. The
sec tio n s m ay s ta r t at ground level o r at an a r b itr a r y stum p height.
F o r c o n ife rs , if total volume is the m ain re q u ire m e n t, the la s t section
to the tip is usually re g a rd ed as a cone. If the volume re q u ire d is to
som e a r b itr a r y lim it such as 4 in d . u . b . , the la s t sectio n , norm ally
le s s than 10 ft in length and often called 'the odd lo g ', is m e asu re d at
its m id -point and its volume calculated as the product of the
m id -se c tio n a l a re a and the length. The m ethod is illu s tra te d in Fig. 8:1
on a pro fo rm a found to be convenient fo r reco rd in g the m e asu rem en ts
and calculating volum e. T here a re m any v aria tio n s of th is p ro fo rm a .
The m ethod is sim ple and p ra c tica b le fo r standing tre e s as w ell as
felled tr e e s . To the extent that each section is equivalent to a second
d e g re e paraboloid frustum and that the nom inated point of m e asu re m e n t
is re p re se n ta tiv e of it, the m ethod gives an unbiased e stim a te . The
54 Outline of Forest Mensuration
Fig. 8:1
I n d iv id u a l tree v o lu m e , s e c t io n a l m e t h o d
Location: M t S tro m lo , A . C . T . D .b .h .o .b . 1 1 .4 in Meas. by: Black
Species: P. ra d i a ta T o t. ht: 68 ft R eco rd ed by: White
Identity: 32 H t to 4 in: 57 ft D ate: 4 /5 /6 7
M ain stem to odd log
Ht O ver-bark U n d e r-b a rk
above
ground D iam . S ec. a re a 2BT D iam . S ec. area
5 ft 10-6 •613 1-2 9-4 •482
15 ft. 9 -4 •482 0-6 8-8 •4 22
25 ft 9-0 •442 0-5 8 -5 • 394
35 ft 7-6 • 315 0-4 7- 2 •283
45 ft 6 -0 • 196 0- 3 5-7 • 177
Sum 1-758
Vol. 17-58
Odd Log
D iam . O v e r-b a rk U n d e r-b a rk

m eas. Length D iam . Sec. A 2BT D iam . Sec. A
at
53 ft 6 in 7 ft 4-4 •106 0 -3 4- 1 •092
Vol. 0- 64
M erch . v o l. u . b . = 17-58 F 0 - 6 4
= 1 8 -22 c u ft
ABOVE ODD LOG

L ength (L) = 11 ft
V o lu m e - -0 87 x 11 f t = -32
' 3
T o ta l v o l . u . b . = 1 7 - 58 + O ' 64 + -32
= 1 8 -5 4 c u ft
Stern Volume 55
bottom section of m ost tre e s is affected by b u tt-sw ell and the
assum ption that the stem th e re is a frustum of a paraboloid may lead
to a b iased e stim ate . This b ias may be reduced by taking s h o rte r
sectio ns in that region, for exam ple two 5 ft sectio n s in stead of one
10 ft section, since the s h o rte r the section s the le ss the o v erall e r r o r
in assum ing neiloidal sections a re paraboloidal. The positions of
m easu rem en t a re fixed and any o b se rv e r should m e asu re a t those
points on every occasion; to th is extent the method is p re c ise fo r the
sam e o p e ra to r o r am ongst o p e ra to rs . The p recisio n am ongst o p e ra to rs
may be reduced, but usually only slightly, when positions of
m easu rem en t a re considered u n re p re se n ta tiv e and o p e ra to rs use
d ifferent techniques to overcom e th is , fo r exam ple by m easu rin g at
one position elsew here o r taking the averag e of m e asu re m e n ts at
positions equidistant above and below.
Since th e re is no visu al check on the m e a su re m e n ts, as th e re is in
graphical m ethods d iscu ssed la te r, the re c o rd e r m u st be a le r t to pick
up m istak es made in m easu rin g , and to avoid m istak es in reading
section a re a values from ta b le s, in reco rd in g them and sum m ing them .
He should see that m easu rem en ts of g irth and b a rk th ick n ess up the
stem follow a sensible tre n d . This m ethod is used in sev e ra l States
of A u stralia for felled and standing plantation c o n ife rs. U sually
m erchantable volume is d erived, th at is volume u n d e r-b a rk to a top
d . u . b . lim it of 3 in o r 4 in with or without allow ance for stum p. F or
a few tr e e s , the calculations a re c a r rie d out m anually. If many tre e s
a re involved the method is e asily program m ed for calculation by
com puter.
P e rc e n tile Method
This is also an old method which goes by se v e ra l nam es but is

com m only known as the p e rcen tile m ethod in A u stra lia . By its n atu re
it is re s tric te d to whole stem s undivided from ground to tip and is
usually asso c iate d with the com pilation of stem p ro file equations for
such tr e e s . The stem is co n sid ered in two p a rts:
(a) The p a rt below b re a s t height which is reg ard ed e ith e r as a
tru n cated paraboloid, in which case the volum e is deriv ed as the
product of a sectional a re a 2 ft l j in from ground and a length of
4 ft 3 in, o r as a cy lin d er, in which case the volum e is derived
as the product of sectional a re a at 4 ft 3 in from ground and a
length of 4 ft 3 in.
(b) The p a rt above b re a s t height to the tip which is re g a rd ed as 10
tru n cated paraboloids of equal length, the volume of each section
being derived by S m alian's form ula from the sectio n al a re a at
each end of the section. The d ia m e te rs at th ese p ercen tile points
of m easu rem en t a re re fe rre d to re sp ec tiv e ly as Dq (at b re a st
height), D]_, D2 , D3 , D4 , D5 , Dß> D7 , Dg, and D9 , Dio being
the tip and having a value of z ero . The co rresp o n d in g sectional
a re a s a re Sq , Si ......................S9 . The volum e of th is p a rt is
given by
f (So + s1) + (S-J + s2) + .........(Sfi + So) + (S9 + 0) 'I

2 2 2 2 iL
r E ü + Sl + s 2 + ............. + S 8 + S9 ,
v 2 J
w here L is the common length of section given by

total height - 4 ft 3 in
10
The volume of the whole stem is obtained by adding the volum es of the
two p a rts . The m ethod is illu s tra te d in Fig. 8:2 on a p ro form a found
to be convenient.
The m ethod was d e lib e rately designed so th at d ia m e te rs a re obtained
a t the sam e re la tiv e positions on stem s of d ifferen t size and stem
profile equations can bd developed. The m ethod can only be applied
p rac tica b ly to felled tr e e s . F o r standing tr e e s , the m o st effective
p ro ced u re is to clim b and determ in e the c o rre c t height, then m easu re
d ia m e te rs during the re tu rn to the ground. As with the sectional
m ethod, the e stim ate of the volume above b re a s t height w ill be unbiased
to the extent that each section is equivalent to a second d eg ree paraboloid
frustum and the nom inated points of m easu rem en t a re re p re se n ta tiv e .
A ssum ing the p a rt below b re a s t height is a cy lin d er of sectional a re a
equal to the basal a re a will alm ost in v ariab ly lead to a negative b ia s.
A ssum ing it is a paraboloid fru stu m m ay lead to a p ositive o r negative
bias depending on its actual shape. Since the positions of m easu rem en t
a re fixed, the m ethod should provide an e stim ate fo r a p a rtic u la r stem
which is p re c ise fo r the sam e o p e ra to r and among o p e ra to rs . There
is the sam e lia b ility to m ista k es as in the sectional m ethod, perhaps
m ore so if the o p e ra to rs a re unskilled.
Stem Volume 57
Fig. 8:2
In d iv id u al tree v o lu m e , p e rc e n t il e m e t h o d
L ocation: M t Strom lo, A . C . T . D .b .h .o .b : 11*4 in M eas. by: Black

Species: P. ra d i a ta T o t. ht: 68 f t R e c o rd e d by: White
Identity: 32 D ate: 4 /5 /6 7
Below Breast H eig h t
D iam . O v e r-b a rk U n d e r-b a rk

M eas. Length D ia m . Sec. 2 BT D iam . Sec.
at area- area
2 f t l | in 4 ft 3 in 1 1-8 •760 1-6 10-2 •568
Vol. b elo w b . h t = S ec. are a x 4 ft 3in

" " " o .b . x 4 ft 3in =
" " " u .b . - • 568 x 4 ft 3in = 2• 41
A bove Breast H e ig h t
P ercen tile l e n g t h = L = 6 8 '0 " - 4 '3 " = 6ft 4 y in

10
D ia m . O ver- b ark U n d e r-b a rk

M eas.
D iam . S ec. are a 2BT D iam . S e c . area
at
dg 4 ft 3 in 11*4 • 70 9 1-4 10-0 s0 •545

s0 / 2 •272
dj_ 10 ft 7 j in 9-8 •5 23 0-7 9- 1 si •452
d2 17 ft 9-5 •492 0-6 8-9 s2 •432
d3 23 ft 4 | in 8 -7 •4 13 0 -5 8*2 S3 • 367
d4 29 ft 9 in 8* 1 • 358 0 -5 7-6 s4 • 315
d5 36 ft l j in 7 -5 • 307 0-4 7- 1 s5 •275
dg 42 ft 6 in 6- 1 •2 03 0- 3 5-8 s6 • 184
d7 48 ft 10 in 5 -5 • 165 0-3 5-2 s7 • 147
dg 55 ft 3 in 4 -4 • 106 0-3 4- 1 s8 •092
d9 61 ft 7 \ in 3* 1 •052 0-2 2- 9 s9 •046
Sum 2-5 8 2
V ol. above b . h t = Sum sec. areas x L =

" " " o .b . = x
" " " u . b . = 2 -5 8 2 x 6 ft 4 | in = 16-46
Whole T re e
T o ta l v o l. o . b . = + c u ft
= 16-46 + 2 -41 18- 87 c u ft
O ther Methods Based on Volume of Sections

V arious m ethods have been suggested b ased on sectio n s of unequal
length, the sections usually being sm a lle r n e a r the butt w here the
effect of e r r o r s is g re a te r. T hese a re not in common u se. A method
outlined by J e ffe rs (1955) based on d ia m e ter m e asu rem en ts by the B a rr
and Stroud D endrom eter involves sections of unequal length because of
the scaling of the d en d ro m eter.
Stem Volume D erived through a G raphical P ro ce d u re

If values of d ia m e te r m easu red at vario u s positions on the stem a re
plotted against the d istance of those positions from the ground, as
re c tan g u la r c o -o rd in a te s on graph p ap er, and then a curve is fitted to
the points, a d ia m e tra l outline of the stem , usually-called a ta p e r cu rv e
o r stem p ro file , is obtained. The m easu rem en ts can be taken close
to g eth er o r fa r a p a rt and irr e g u la ritie s can be included o r excluded at
w ill in m e asu rem en t o r in fitting the ta p e r cu rv e. From the ta p e r cu rv e,
values of d ia m e ter at any position can be re a d and stem volume
calculated by sections using the Smalian o r H uber form ula. The
advantages of th is method over the sectional and p e rc en tile m ethods
a re that plotting the inform ation grap h ically provides a visual check
of the m easu rem en ts and that the graph provides a p erm anent and
convenient re c o rd of the stem which m ay find many u ses such as
calculation of volume of v ario u s a sso rtm e n ts of the stem .
If sectional a re a s ra th e r than d ia m e ter a re plotted g rap h ically ,
volume is re p re se n te d d ire c tly by the a re a under the ta p e r cu rv e.
The advantages of th is w ere outlined by R eineke (1926) who
recom m enced determ ining the a re a by p la n im e ter. The m ethod has
been applied extensively in the U. S. A. with the use of re c tan g u la r
co -o rd in ate p ap er, or p ap er with specially graduated ax es, and the
use of a p lan im eter or sq u are-c o u n t method to d eterm in e a re a . It
has also been used extensively in A u stra lia , p a rtic u la rly for felled
b ro ad -leav ed stem s; Unwin and Bowling (1951) and L aw rence (1960)
have d iscu ssed its use in T asm ania. The method has been used also
for plantation c o n ife rs, standing and felled , to d eterm in e biological
volume as a base against which volum e by o th er m ethods could be
com pared in investigations such as that of C arro n and M cIntyre (1959).
This graphical m ethod can be applied to tr e e s undivided o r divided
from ground to tip , fo r to tal volume o r volume to any p a rtic u la r
d ia m e ter lim it o r betw een any d ia m e ter lim its, o v e r-b a rk o r
u n d e r-b a rk . If m any m e asu re m e n ts a re taken, the method tak es
longer than the m ethods in which volume is calculated by fo rm u lae,
Stern Volume 59
but m any m easu rem en ts should be taken to obtain the full advantages
of the graphical m ethod. If the tre e is felled, the e x tra tim e involved
can re a d ily be ju stified in view of the re la tiv e am ount of overhead tim e
involved in applying any m ethod. With standing tr e e s , the e x tra tim e
involved is proportionately g re a te r but may still be ju stifie d on the
grounds of the amount of tim e involved in getting any m easu rem en ts
at all. The m ore m easu rem en ts taken, the c lo se r is the e stim ate to
the biological volume and the m ore p re c ise the e stim ate for the sam e
o p e ra to r and among o p e ra to rs , but th is depends to som e extent on the
p re s c rip tio n s for selecting the points of m e asu re m e n t. P lotting each
point g raphically provides a visu al check on both the m easu rem en t and
the re c o rd in g of it, since any point which v a rie s from the g en eral ta p er
curve as shown by the other plotted points prom pts a check. The
m ethod can read ily be p rogram m ed for calculation by com puter.
G ray ’s T a p e r Line Method

About 1940 at the A u stralian F o re s try School, G ray was investigating
the use of a stan d ard ised s e r ie s of ta p e r c u rv e s, p a rtic u la rly for
P. r a d ia ta . so th at having determ ined the a p p ro p ria te ta p e r curve from
a lim ited num ber of m e asu re m e n ts in the low er p a rt of a tre e the re s t
of the curve could be extrapolated without actual m e asu rem en t to reduce
the am ount of clim bing of standing tre e s ; and so th at th e re would be an
objective check on m e asu re m e n ts re c o rd ed in the field, experience
having shown this to be a continual so u rce of e r r o r s . Plotting sectional
a re a ag ain st height showed that 'excluding a v a ria b le d istan ce from the
ground and a v ariab le d istance from the tip , sectio n al a re a under b ark
d e c re a se d lin e a rly with height, at le a s t within the lim its to be expected
with biological data and usual stan d a rd s of accu racy of m e a su re m e n t'.
This condition appeared to apply to a wide range of sp ec ie s,
p a rtic u la rly conifers and young eucalypts. G ray (1943, 1944, 1956,
1966) then proposed the hypothesis a lread y outlined in C hapter 6, and
on the b a sis of this hypothesis proposed a m ethod of deriving the volume
of a tre e called the ta p e r line m ethod. The m ethod has m ost advantage
when applied to a standing tre e and this will be assum ed in the
d e sc rip tio n which follow s, illu s tra te d fo r a co n ifer in Fig. 8:3.
A m e a s u re r subjectively se le c ts se v e ra l 're p re s e n ta tiv e ' points fo r
m e asu re m e n t, usually two in the b u tt-sw ell region, and a m inim um of
th re e and a m axim um of about five along the m ain stem up to between
a half and tw o -th ird s the height of the tre e . He m e a su re s o v e r-b a rk
60 Outline of Forest Mensuration
Ind iv id u al tree v o lu m e , t a p e r line m e t h o d
L ocation: M t S trom lo, A . C . T . D .b .h .o .b : ll-4 in M eas. by: Black

Species: P. ra d i a ta T o t. ht: 68 ft R e c o rd e d by: White
Identity: 32 D ate: 4 /5 /6 7
Ht (ft) d . o . b . (i n ) 2BT (in) d . u . b . (in) s. a . u . b. (sq ft)
1 f t 11 in 11-8 1-6 10-2 •568

4 ft 1 in 11-4 1-4 10-0 •545
11 ft 2 in 9- 8 0-7 9- 1 •452
22 ft 11 in 8 -6 0-5 8 -1 •358
31 f t 3 in 7 -8 0-4 7 -4 •299
39 ft 0 in 6 -8 0-4 6 -4 • 225
•540 x 68
b u tt - s w e l l = • 09 x 7
non-m erch. )
•087 x 11
p a r a b . v o l. )
V ol. to 4 in d. u. b. 18- 36 + 0 - 3 2 + 0-48
18 -2 0 c u ft
(N .B . hp happens to be the sa m e
as t o t a l h e ig h t in this ca se ;,
n o r m a l l y i t is less for conifers. )
S 40
S.A.'J.B. (Sq ft)

Stern Volume 61
d ia m e te r at th ese points by g irth tape and b a rk th ick n ess by b ark
gauge. A re c o rd e r on the ground plots the equivalent u n d e r-b a rk
sectio n al a re a s against the re sp e c tiv e heights above ground level on
re c ta n g u la r co -o rd in ate paper as m easu rem en t p ro ceed s. As soon as
the re c o rd e r is confident of fitting a stra ig h t line to the points
re p re se n tin g the paraboloid of the m ain stem the m e a s u re r re tu rn s to
the ground.
The stra ig h t ta p e r line fitted to the points of the m ain stem is
extended to cut the sectional a re a axis at p arab o lic b ase Sp and to cut
the height axis at parabolic height h p . The a re a under th is stra ig h t
line re p re s e n ts parabolic volum e Vp and being that of a tria n g le is
com puted as s p x hp t A cu rv e is fitted to the points in the b u tt-sw ell
2
reg io n , the a re a between th is c u rv e , the ta p e r line, and the sectional
a re a axis being obtained by sq u are count o r from the dim ensions of the
tria n g le form ed by a stra ig h t line which ap p ro x im ates the b u tt-sw ell
cu rv e (as in Fig. 8:3). T his b u tt-sw ell volume is added to the
p arab o lic volum e. If the top lim it to which volume is re q u ire d is le ss
than p arabolic height, the volum e re p re se n te d by the a p p ro p riate
tria n g le is su b tra cte d from p arabolic volum e. F o r a co n ifer, to tal
height is usu ally g re a te r than p arab o lic height, and if to tal volum e is
re q u ire d a curve is run down from total height to join the ta p e r line
at an a p p ro p ria te position and the volume com puted in the sam e way
as b u tt-sw e ll volume; this is then added to p arabolic volum e and
b u tt-sw e ll volum e.
The ta p e r line method is thus a graphical method c h a ra c te ris e d by
the assum ption that the m ain stem approx im ates to a second degree
paraboloid with a ’s k ir t' of b u tt-sw e ll of varying size at the b ase and,
in the c ase of c o n ife rs, surm ounted by a co ne-like top. This
assum ption leads to the p ra c tic e of m easu rin g the m inim um num ber
of points re q u ire d to fit a stra ig h t line with confidence, with a saving
in tim e , energy, and ris k which m ay be co n sid erab le in a la rg e
m easu rin g p ro g ram , when the tr e e s a re v e ry ta ll, o r when clim bing
is difficult. In o th er re s p e c ts , the m ethod is like the o th er g raphical
m ethods - the p ro ced u re provides a v isual check of m easu rem en t
and reco rd in g , the graph form s a perm an en t and convenient re c o rd
of the stem from which the volum e of a ss o rtm e n ts of the stem can
re a d ily be determ ined at any tim e. B ecause the points of m easu rem en t
a r e few and chosen subjectively, how ever, the e stim a te fo r a p a rtic u la r
tre e is likely to be d ifferent fo r differen t o p e ra to rs and d ifferen t for
the sam e o p e ra to r on different o ccasio n s so th at the m ethod is not
p re c ise ; and any re s u lt is likely to be biased . C arro n and M cIntyre
(1959) d iscu ss a te s t of the re la tiv e p re c isio n and b ias of the ta p e r line
and sectional m ethods applied to P . ra d ia ta .
9 Growth and Increment of
Individual Trees
Though th e te r m s grow th and in c re m e n t a r e u se d in te rc h a n g e a b ly by

f o r e s t e r s , in th e s t r i c t s e n s e tr e e grow th m e a n s th e b io lo g ical
phenom enon of in c re a s e in s iz e w ith tim e and in c re m e n t m e an s th e
q u an titativ e in c re a s e in s iz e o v e r a sp ecified tim e in te rv a l re s u ltin g
fro m the phenom enon of grow th. It is th e phenom enon of grow th which
m a k es it p o s sib le to m an ag e a f o r e s t a s a re n ew ab le c ro p . The
e s tim a tio n of th e e ffe c ts of grow th is th e re f o r e an im p o rta n t p ro c e s s
in m e n su ra tio n b e c a u se m an ag em en t m u st be guided la rg e ly by
f o r e c a s ts of fu tu re y ie ld fro m in fo rm a tio n of p r e s e n t r a te s of in c re m e n t
F o r m an ag em en t p u rp o s e s , th e m ain in te r e s t is n a tu ra lly in th e
in c re m e n t of a stan d . Since th is , h o w ev er, is c o m p ris e d of th e
in c re m e n ts of in d iv id u al t r e e s we need to c o n s id e r how in c re m e n t in
an individu al tr e e can be d e te rm in e d .
In g e n e ra l, each tr e e s p e c ie s h as its own p a tte rn and r a te of grow th
but th e s e can be in flu en ced c o n s id e ra b ly in an individual by in te rn a l
fa c to rs - g enetic and p h y sio lo g ic a l f a c to r s unique to th e individual;
and by en v iro n m en ta l f a c to r s - c lim a te , edaphic and b io tic . The
e ffe c ts of p a r tic u la r f a c to rs a r e v e ry d ifficu lt to is o la te . The ep iso d ic
grow th of the m ain s te m s of tr e e s h a s been co m m en ted on e lse w h e re .
F o r p ra c tic a l m a n ag em e n t p u rp o s e s and to re d u c e the c o n s id e ra b le
v a ria tio n s in in c re m e n t, w ith in -y e a r v a ria tio n s a r e u su a lly ig n o red by
m aking the m in im u m p e rio d betw een m e a s u re m e n ts one y e a r , and
b e tw e e n -y e a r v a ria tio n s a r e sm oothed by c o n sid e rin g 'c u r r e n t' annual
in c re m e n t a s an a v e ra g e of s e v e ra l p re v io u s y e a r s . B ecau se of the
c o r r e la tio n betw een in c re m e n t and w e a th e r o v e r a s e a s o n , it is
e s s e n tia l th a t any m e a s u re of in c re m e n t be re la te d to o r q u alified by
the w e a th e r of the p e rio d . A gain, b e c a u se of th e a s y m m e tric a l n a tu re
of t r e e grow th o v e r a s te m , cau tio n m u st be e x e r c is e d in acc ep tin g
grow th in one p a r t of a ste m as ev id en ce of s im ila r gro w th e lse w h e re
on the s te m .
63
The m ain tre e v a ria b le s whose changes with tim e a re of in te re s t a re
d ia m e te r, height, and ta p e r. From a knowledge of in crem en t in th ese
v a ria b le s , increm ent in other tre e v a ria b le s such as volum e and bole
a re a can be determ ined. To fo re c a s t in crem en t from the p a st growth
p a tte rn , in crem en t over a p a st perio d can be re la te d to size a t the
beginning or end of the period w hether the age of the tre e is known o r
not. If age is known, size at a p a rtic u la r tim e can be re la te d to age
at that tim e or in crem en t over a perio d can be re la te d to an age which
re p re se n ts that period.
D eterm ining the Age of an Individual T re e

The age of a tre e can only be d eterm in ed with c e rta in ty when the y e a r
of its germ ination is known. This is usually so fo r tr e e s in
a rtific ia lly estab lish ed fo re s ts . F o r n a tu ra l fo re s ts , the age of
individual tre e s can som etim es be d eterm in ed from the date of a
catastro p h ic event such as a fire or cyclone, o r of a logging o p eratio n ,
as a re s u lt of which they developed. The age of young tr e e s of som e
sp ecies can be determ ined from a count of annual developm ents of
shoot. F o r som e sp e c ie s, age can be d eterm in ed by counting the
growth rin g s of the stem if they a re annual.
Annual growth rin g developm ent is d iscu ssed in d etail in te x ts on
th at subject. B riefly , tru e annual rin g s occur in tr e e s c h a ra c te ris e d
by the form ation of d ifferen t e a rly and late wood such as te m p e rate
c o n ife rs, som e su b -tro p ic a l c o n ife rs, m ost te m p e rate deciduous
b ro ad -leav ed s p e c ie s, and those sp ec ie s c h a ra c te ris e d by a rin g porous
condition. Annual rin g s a re g en erally not found in tro p ic a l co n ifers o r
diffuse porous non-deciduous b ro a d -le a v ed sp ecies u n less they a re
growing in sub-alpine o r alpine conditions. The tr e e s o v er m uch of
the n o rth e rn h e m isp h ere form arrfiual rin g s so that m ost m en su ratio n
te x ts , because they a re w ritten in that p a rt of the w orld, im ply that
counting rings is a stan d ard technique fQr d eterm in in g age. T h ere a re
lim ited conditions in A u stra lia w here tre e s develop tru ly annual rin g s.
The ease with which annual rin g s m ay be counted depends on the
sp ecies and the environm ental conditions of the p a rtic u la r tre e .
D ifficulties a ris e when th e re is little colour d ifference betw een e a rly
and late wood - th e re does not seem to be any staining technique
which gets over th is problem : su p p re ssio n , old age, o r ex trem e
w eather conditions m ay cause v ery n a rro w rings o r cessatio n of wood
Growth and Increm ent of Individual T re e s 65
form ation in c e rta in p a rts of the stem ; false rin g s m ay be form ed due
to ex trem e w eather conditions such as e a rly o r late fro s t o r a long
drought followed by ra in ; rin g s may be d isto rte d by b ran ch w horls o r
defects and the occu rren ce of co m p ressio n wood o r tension wood.
Bowling (1954) has d iscu ssed counting the rin g s of sm all scru b species
in stead of la rg e eucalypts of s im ila r age for determ in in g the age of
eucalypts in southern T asm ania.
D eterm ining the In crem en t of Individual T re e s

Increm ent in height o r d ia m e te r over a period can be determ in ed by
m easu rin g the v a ria b le at the beginning and end of the period. This
is done in investigations of a silv ic u ltu ra l n a tu re , and is the b a sis of
p erio d ic inventory, to be d isc u sse d la te r.
If the height is m easu red with height stic k s, the e stim a te s of height
in crem en t w ill usually be sa tisfa c to ry . If m e asu re m e n t is done with
in stru m e n ts based on g eo m etrical o r trig o n o m etrica l p rin c ip le s, the
e stim a te s m ay not be sa tisfa c to ry ; in fact, try in g to get e stim a te s of
height in crem en t by rep e ate d m e asu re m e n t using such in stru m e n ts
over a period of y e a rs is often v e ry fru s tra tin g d espite the apparent
sim p licity of the operation. E r r o r s can so m etim es be reduced by
having perm anent positions from which in stru m en t o b serv atio n s a re
m ade. R epeated m easu rem en t of d ia m e ter o r g irth g en erally gives
much m ore sa tisfa c to ry re s u lts than rep eated m e asu re m e n t of height,
provided the position of m e asu re m e n t on the stem is perm anently
m ark ed. This is com m only done by the use of cray o n , paint, o r a nail
of a soft m etal such as alum inium o r copper which w ill not cause
problem s in subsequent saw ing. If the paint o r nail is likely to affect
grow th in any way at the position it should be placed som e set distance
away. If a c a lip e r is used to m e a su re d ia m e ter the b a rk is som etim es
sc rib e d at the position, the b a r of the c a lip e r being se t in the scrib e
m ark .
The determ ination of in crem en t of individual tr e e s is often c a rrie d
out in perm anent sam ple plots for the m e asu re m e n t of which codes of
p ro ced u re a re generally p re p a re d . The F o re s t S erv ices of A u stra lia
have published these in v ario u s fo rm s for dom estic u se. Exam ples
fo r o ther co u n tries a re the S ilvicultural R e se a rc h Code (G riffith and
P ra s a d , 1949) for India and the Code of Sample P lo t P ro ce d u re
(Hummel e t a l . , 1959) for G reat B rita in .
Stem A nalysis
The p ast growth h isto ry of a tre e stem with tru ly annual rin g s can be
re a d at any tim e by determ ining the position of each rin g . This is
known as stem a n a ly s is . If the inform ation is obtained at only one
position on a stem , fo r exam ple at b re a st height, the p ro ced u re is
called p a rtia l stem a n a ly s is . If the inform ation is obtained at a num ber
of positions along the stem with a view to assem bling a h isto ry of the
whole stem from which d ia m e te r, height, and volume in crem en t can
be d eterm ined, the p ro c e ss is called com plete stem a n a ly s is . If the
tre e m ust be left standing, the analysis is done on c o re s of the stem
taken horizontally from b a rk to pith with an in crem en t b o r e r . Small
c o re s com prising the rings of the la st few' y e a rs may be taken with an
in crem en t h a m m e r. T here is an extensive lite ra tu re on the taking,
handling, sto rin g , and baeasuring of in crem en t c o re s - Spurr (1952)
gives a good bibliography. If the tre e can be felled, the coimting of
rin g s and m easu rem en t of d ia m e ters is usually done on com plete
sections one or two inches thick. This allows com prehensive visual
inspection of the section w hereas the in crem en t b o re r only sam p les it.
Depending on how many tre e s a re to be analysed and the value of the
tim b e r, so sections m ay be cut to the b est advantage of the analysis
o r to suit the re q u ire m e n ts of utilisatio n . The intention of stem
an alysis is to re p ro d u c e, from the evidence of the sectio n s, the
u n d e r-b a rk outline of the stem at each of se v e ra l previous ag es, from
which the d ia m e te r, height and volume at those ages, and th e ir
in c re m e n t. can be e stim ated . T here a re v ario u s ways of tabulating
and graphing the data; that outlined by J e rra m (1939) is sim p le,
p rac tica b le and effective.
The C h a ra c te ris tic s of Growth of Individual T ree s

The g en eral tren d of size with age for biological o rg an ism s is
re p re se n te d by a sigm oidal curve. Many attem pts have been made to
d e sc rib e such tre n d s m athem atically, but for fo re s try p u rposes fitting-
cu rv es freehand is m ore p ra c tica b le and ap p ears sa tisfa c to ry for
m ost p u rposes.
The survival and subsequent h isto ry of a tre e depend p rim a rily on
its growth in height re la tiv e to that of its c o m p etito rs. The tre n d of
height with age is likely to show the inherent vigour of the tre e and the
environm ental conditions under which it grew . F o r dom inant tre e s of
Growth and In crem en t of Individual T re e s 67
an in to le ra n t sp ec ie s, the relationship is usually a sigm oidal curve
which shows the fa irly slow initial grow th of the seedling stag e, an
ap p ro x im ately lin e a r section which re p re se n ts the m ain life of the
tr e e , then a flattening out at m atu rity . F o r su p p re ssed tre e s o r a
to le ra n t sp e c ie s, the tre n d will be ap p ro p ria te ly d ifferen t, for exam ple
the lin e a r m ain section w ill be le ss steep and the curve will flatten out
much m o re quickly.
The relatio n sh ip of d . b . h . u . b . and age is g en erally sigm oidal, the
m ain section being lin e a r o r c u rv ilin e a r depending on sp e c ie s,
environm ent, effects of thinning and pruning, and so on. If the m ain
sectio n of the d . b . h . u . b . -age relatio n sh ip is lin e a r, the b asal a re a -a g e
relatio n sh ip w ill be a risin g curve as illu s tra te d in Fig. 9:1 for a
Eucalyptus regnans re c o rd ed by Helms (1945), an old tre e by A u stralian
stan d a rd s with p a rtic u la rly p e rs is te n t vigour. M ore often, the
d . b . h . u . b . -age relationship is a falling cu rv e, and the b asal a re a -a g e
re la tio n sh ip is lin e a r o r n e arly so.
As d iscu ssed previously, d ia m e ter in c re m en t may be s im ila r o r
d ifferen t along the length of the stem depending on many fa c to rs , so
140 —
120 —
80 —
60 —
40 —
20 “
1560 1600 1640 1680 1720 1760 1800 1840 1880 1920
YEAR A.D.
F ig . 9:1 C h a ra c te ris tic s of gro w th of in d iv id u a l tre e s; E. reg n a n s, T a sm a n ia , 1 5 4 8-1942;
g . b . h . o . b . 65 ft; h e ig h t to to p of b ro k en crow n 257 ft; v o lu m e 64, 000 sq ft
(a fte r H elm s, 1945)
that it is im possible to g e n eralise about the relatio n sh ip of ta p e r with
age. B ecause of the effects of changing shape and ra te of ta p e r, the
change of volume with age is likely to be som ew hat e r r a tic . These
v aria tio n s in volum e a re sm oothed over to a la rg e extent because
changes in shape and ra te of ta p e r a re not taken into detailed account
in estim ating volum e, and the relatio n sh ip of volume and age is usually
sigm oidal.
Increm ent and the Increm ent-A ge R elationship

T here a re two conventional e x p ressio n s of in crem en t. One is known
as c u rre n t annual increm ent abbreviated to C. A. I. , CAI o r c . a. i. ,
which by its nam e im plies the increm en t of the c u rre n t y e a r. Annual
growth is often sm all and difficult to m e asu re re lia b ly , and is also
subject to seasonal fluctuations. T h erefo re CAI is often d erived as
the m ean over the previous se v e ra l y e a rs in which case it is known as
periodic m ean annual in crem en t, abbreviated to P .M .A .I . , PMAI o r
p .m .a . i. In quoting CAI o r PMAI it is e sse n tia l to define the tim es
o r dates fo r which the in crem en t is c u rre n t and the age o r size of the
tre e at the tim e. The o ther ex p ressio n , known as m ean annual
in crem en t and abbreviated to M .A .I. , MAI o r m . a . i . , re p re se n ts the
average in crem en t from origin up to a specific age and is derived by
dividing the size by the num ber of y e a rs to produce th at s iz e . In
quoting MAI it is e sse n tia l to quote the age of the tre e . To illu s tra te
the tre n d of in crem en t with age, values of CAI and MAI m ay be plotted
graphically against tim e o r age and in crem en t cu rv es draw n. As would
be expected from th e ir a rith m e tic deriv atio n , th e re is an invariable
relatio n sh ip between these c u rv e s - they m ust c ro s s when the MAI is
at a m axim um . As long as the CAI curve is above the MAI cu rv e, the
la tte r m ust be risin g since each added y e arly in crem en t im proves the
av erag e. When the CAI re a c h e s its m axim um , the im provem ent in the
MAI w ill be le s s m arked. When the CAI and MAI values a re equal,
the MAI curve will tend to rem ain horizontal o r fall a little . As the
CAI fa lls, the MAI w ill be reduced and the cu rv es w ill c r o s s , the MAI
curve falling but rem aining above the CAI cu rv e. This d escrip tio n is
an idealised one; th e re a re likely to be co n sid erab le irre g u la ritie s in
the curves due to the v a riatio n s in growth re fe rre d to p rev io u sly . The
in te rre la tio n sh ip of CAI and MAI cu rv es of a stand, th e ir re la tiv e shape
and the position of th e ir point of in te rse c tio n , a re of sp ecial in te re s t
to m anagem ent.
10 Stand Variables
The F o re s t and Stand

So fa r the m easu rem ent of the v a ria b le s of an individual tr e e has
been considered. This is usually only a m eans to the end of estim atin g
the v a ria b le s of a stand o r fo re s t of tr e e s . Though the objectives of
m anagem ent m ust obviously be re a lis e d through the m anipulation of
individual tr e e s , the only p ra c tica b le unit for m anagem ent planning
and control is a stand o r a fo re s t. A stand is an aggregation of tr e e s ,
sufficiently uniform in com position and age, o r defined by n a tu ral o r
a rtific ia l boundaries, to be re g a rd e d as a unit fo r silv ic u ltu ra l or
m anagem ent purposes (B. F. A. , 1953). A fo re s t co m p rises a num ber
of such stands aggregated fo r som e purpose of m anagem ent. It is
convenient to talk of e x p re ssio n s of the com bined d ia m e te rs, heights,
and volum es of the individual tr e e s as stand v a ria b le s .
F a c to rs Influencing th e M easurem ent of Stand V ariab les

The fa c to rs which u ltim ately determ in e what v a ria b le s a re m easu red
and how they a re m easu red a re tim e , m oney, and labour. However,
a num ber of p re lim in a ry fa c to rs have to be co n sid ered in planning the
m easu rem en t.
An im portant facto r is the purpose of the m e asu rem en t. T here a re
v ario u s re a so n s for wanting a quantitative d escrip tio n of a stand. The
m aintenance and developm ent of its fo re s ts a re im p o rtan t to the economy
of a nation. To develop and im plem ent s a tisfa c to rily a policy for that
m aintenance and developm ent, the location and extent of the fo re st
e sta te and its value for protection and productive p u rp o ses m ust be
known. This inform ation is provided by a re s o u rc e in v en to ry . To
m anage an individual fo re st efficiently m ore d etailed inform ation is
re q u ire d than is usually provided by a fo re s t re s o u rc e inventory; it is
69
usually provided by a m anagem ent inventory or w orking plan in v en to ry .
E xperim ental investigations re q u ire even m ore detailed m easu rem en t.
The stand descrip tio n re q u ire d is differen t for all th ese c a se s and
affects what is m easu red and how it is m easu red .
The general s tru c tu re of a stand also influences its m easu rem en t.
T his can be illu stra te d by co n trastin g the conditions in A u stra lia of a
plantation of a coniferous sp ecies with those of a native eucalypt o r
ra in -fo re s t stand. The fo rm e r is usually of one tre e sp ecies; the
individual tre e s a re in a re g u la r p a tte rn which allow s convenient a c c e ss
for m easurem ent; th e re is little o r no o th er vegetation; the tre e s a re
of a size and habit such that m e asu re m e n ts anyw here on the standing
stem a re p ra c tica b le . The stands a re even-aged and the age is known;
th e re is a lim ited range of size c la s s e s within any p a rtic u la r even-aged
stand; and, with few exceptions, all the tr e e s a re u tilisa b le . The
native fo re s t, on the other hand, usually c o m p rises a num ber of sp ec ie s;
the tre e s a re irre g u la rly d istrib u te d and th e re is often a le s s e r
vegetation which im pedes a c c e ss and m akes m easu rem en t difficult;
the tre e s a re of a size and habit such th at m e asu rem en ts up the stem
a re im p racticab le; the stand is uneven-aged and determ ining the age
of any tre e may o r m ay not be possible; th e re is a wide range of siz e s;
not all the tre e s a re of co m m ercial sp ecies and many tre e s of
co m m ercial sp ecies a re not u tilisa b le because of defects which m ay
not be read ily d isce rn ib le. Such d ifferen ces in s tru c tu re and
com position w ill obviously affect what v a ria b le s can be m easu red and
how they can be m easu red .
U sually the population is too extensive for the m easu rem en t of even
one v ariab le on ev ery tr e e . T his depends, of c o u rse , on the v a ria b le
and the s tru c tu re of the stand. The g irth of ev ery tre e in a
com partm ent of a coniferous plantation could be m easu red without much
tro u b le. M easuring the volume of ev ery tre e would, how ever, be a
different proposition and so would m easu rin g the g irth of every tre e
on an equivalent a re a of a dense uneven-aged native eucalypt stand.
As a ru le , w hatever the stand v a ria b le , it is m easu red on a sam ple
of the stand. Some populations can be conveniently defined for
sam pling purposes by the num ber of individuals but it is usually m ore
convenient to define the population of a stand by its a re a , which can
be determ ined by surveying. Sampling is then effected through
sam pling units, the shape and size of the units frequently affecting
what stand v a ria b le s a re m e asu re d and how they a re m easu red .
11 Stand Tables and Stand
Basal Area
Stand T a b le s
An effective way to d e s c r ib e a stand qu an titativ ely is to c l a s s i f y the
t r e e s a c c o rd in g to freq u en cy of s p e c ie s , d . b . h . o . b . , height, and the
lik e. A frequency table of c l a s s e s of d . b . h . o . b . is c a lle d a stand t a b l e .
It is v e r y useful fo r s ilv ic u ltu r a l and m a n a g e m e n t p u rp o s e s beca u se
d iffe re n t s t r u c t u r e s of sta n d s a r e c h a r a c t e r i s e d by p a r t i c u l a r freq u en cy
d is tr ib u tio n s of d . b . h . o . b . A stand table is u su ally co m p iled on a p e r
a c r e b a s is but m ay be co m piled fo r the whole of a s m a ll stand. The
p u r p o s e , r e q u ir e m e n t s , and c h a r a c t e r i s t i c s of stand ta b le s a r e the
s a m e a s th o s e of s i m i l a r kinds of fre q u e n c y ta b le fo r o th e r biological
d ata. The p u rp o s e is one of condensation o r s u m m a r is a ti o n and for
e a s i e r m anipulation of the data. U sually equal s iz e d c l a s s e s a r e
fo r m e d , the width and c e n t r a l v alu es of which should be c l e a r ly
sp ecified . In g e n e r a l, for p r e s e n ta tio n and m a th e m a tic a l p u r p o s e s ,
the n u m b e r of c l a s s e s should be fro m about ten to twenty. Once the
ta b le is c o m p iled , the individual lo s e s its identity and ta k e s the c e n tr a l
value of the c l a s s . If the c l a s s in te r v a l is wide and th e r e a r e few
individuals in the c l a s s th is m a y give anom alous r e s u l t s . The m a n n e r
in which the c l a s s fre q u e n c ie s a r e d is tr ib u te d in r e la tio n to the c l a s s e s
is the freq u en cy d is tr i b u tio n . If the valu es of the v a r ia b le a r e plotted
on a h o riz o n ta l ax is and freq u en cy of o c c u r r e n c e on the v e r tic a l axis
of r e c t a n g u la r c o - o r d in a te s , a g ra p h ic a l r e p r e s e n t a tio n of the fre q u e n c y
ta b le in the fo rm of a freq u en cy d ia g ra m can be c o m p iled , such a s a
h i s t o g r a m . freq u en cy polygon, and freq u en cy c u r v e .
The m e a s u r e m e n t of d ia m e te r and co m pilation of a stand tab le is
u su a lly r e f e r r e d to a s an e n u m e r a ti o n , the t e r m im plying a counting
of t r e e s of v a r io u s s iz e s . The width of the c l a s s u sed v a r i e s acc o rd in g
to the conventional p r a c t ic e of the f o r e s t ow ner in r e la tio n to the value
of the stand and the p u rp o s e of the e n u m e ra tio n . It is im p o rta n t to
71
define the c la s s lim its in such a way that the c la s s has an a rith m e tic ally
convenient cen tre and to m e asu re in a way a p p ro p ria te to the c la s s
lim its. F or coniferous plantations in A u stra lia , 1 in c la s s e s of d .b .h .
o .b . o r g .b .h .o .b . a re com m on. F o r native sp e c ie s, c la s s e s from
4 in to 12 in d. b .h . o. b. or g .b .h .o .b . a re used. In growth stu d ies,
w here the tr e e s a re identified individually, d ia m e te rs o r g irth s a re
usually m e asu re d to O' 1 in, w hatever the width of c la s s e s of the stand
ta b le , so th a t in c re m en t can be determ in ed to th at lev el. F o r re s o u rc e s
inventory, m anagem ent inventory, o r growth studies w here the tre e s
a re not identified individually, the tr e e s m aj’ be c la ssifie d d ire c tly in
the field. T his p ro c e ss is called tallying and the re s u lt is called a
ta lly . If the data is to be p ro c e sse d m anually, the tre e s a re usually
tallie d by the gate system w here 1 11 111 1111 11 I F re p re s e n ts 1 to 5
re sp e c tiv e ly . If com pilation o r com putation is to be done by
m echanical data p ro c e ssin g , special m ethods of re c o rd in g a re used.
D .b .h . o .b . D istribution in an Even-aged Stand

An even-aged stand is one in which all of the tr e e s a re of identical
age o r, if of different age, a re sufficiently clo se in age to be taken for
all p ra c tic a l purposes as of identical age. The frequency cu rv e of
d .b .h . o .b . size of tr e e s of an unthinned even-aged stand approxim ates
a s ta tis tic a lly norm al curve fo r much of the life of the stand. The
n o rm al cu rv e is b e ll-sh ap e d , unim odal, and sy m m e tric al about a
c e n tra l value w here a rith m e tic m ean, m edian, and mode coincide; the
c e n tra l c la s s has the g re a te s t num ber of tr e e s , th e re being
p ro g re ssiv e ly few er tr e e s in c la s s e s s m a lle r than and g re a te r than the
c e n tra l c la s s . The n o rm al d istrib u tio n is v ery common in many kinds
of n a tu ra l phenom ena and much of the th eo ry of s ta tis tic s has been
developed around it. How clo sely the stand table of an even-aged stand
approxim ates a norm al d istrib u tio n depends, how ever, on se v e ra l
fa c to rs . The d istrib u tio n of tr e e d ia m e te rs is of th is form because of
the d ifferen ces in growth of individual tre e s which fo r m ost stands
re s u lt from a compounding of the differen ces in the g enetical m ake-up,
and the m a c ro - and m ic ro -e n v iro n m en t, of the individual tr e e s .
Depending on the sp ecies and general environm ent of the stand, the
so rtin g of tre e s according to th e ir growth ra te s m ay take p lace slowly
or re la tiv e ly quickly. If it takes place slow ly, the frequency curve may
in the e a rly stag es be r e v e r s e - J shaped, the g re a te s t num ber of tre e s
being in the sm a lle st size c la s s e s with rap id ly d e crea sin g nu m b ers in
Stand T ables and Stand B asal A rea 73
the la rg e r size c la s s e s . As the stand ag es, the curve shifts to the
rig h t (along the a b sc issa of the graph) and a peak ap p ears which is at
f ir s t left of cen tre (positive skew ness) but gets c lo se r to the c en tre as
the stand ages and m ay even move to the rig h t of the c en tre (negative
skew ness). If the sorting out is re la tiv e ly quick an approxim ately
n o rm al d istrib u tio n may be estab lish ed fa irly quickly; it may rem ain
c lo se to norm alit}Tor with tim e becom e skewed positively o r
negatively. The frequency curve m ay thus v a ry within a stand and
betw een stands from sym m etry to a varying d eg ree of asy m m etry ,
depending on the sp ecies and its growth h a b its, on the quality of the
s ite , and on changing density due to n a tu ra l m o rtality . The frequency
d istrib u tio n of an unthinned stand of Pinus ra d ia ta is shown in Table
11:1 and illu stra te d in Fig. 11:1. W here two age c la s s e s a re
T able 11:1
D .b .h .o .b . frequency distribution (stand ta b le) o f an unthinned, e v e n -a g e d stand of

P. radiata ( 0- 2 acre), A. C . T . , planted 1943
D.b.h.o.b. Frequency
(in) 1956 1960 1963 1965
3 2 1 1
4 20 10 9 8
5 41 24 19 19
6 53 44 28 26
7 21 43 43 41
8 2 14 30 28
9 3 8 13
10 1 4
\ w
9 10 11 12
D.b.h.o.b. (in)
Fig. 11:1 D . b . h . o . b . distribution, unthinned, e v e n -a g e d stand of P. radiata,
Kowen. A . C . T (see T able 11:1)
re p re se n te d se p a ra te ly in a stand th e re may be two se p a ra te norm al
c u rv e s. This m ay also be so for a su p p re ssed even-aged stand.
F o re c a sts of the future condition of a stand should, if they a re to be
effective, include a fo re c a s t of the d . b . h . o . b . d istrib u tio n . If the
distribution is likely to be n o rm a l, fo recastin g the num bers of tre e s in
the vario u s c la s s e s is not difficult if the total num ber in the stand, and
the m ean and stan d ard deviation of the d istrib u tio n , can be fo re c a s t.
F o re s te rs have th e re fo re sought c o rre la tio n s between c h a ra c te ris tic s
of d ia m e ter d istrib u tio n s and re a d ily m easu red stand v a ria b le s , using
both graphical and m athem atical an aly sis of the d istrib u tio n s. A
convenient ex p ressio n of a norm al distrib u tio n can be made g rap h ically
by plotting cum ulative frequency p e r cent ag ain st size on
a rith m e tic -p ro b a b ility pap er; the points lie on a stra ig h t line if the
d istribution is n o rm al. The m ethods of m ath em atical an aly sis a re
outlined in stan d ard tex ts on s ta tis tic a l m ethods. M eyer (1930), for
exam ple, investigated the relatio n sh ip between the stan d ard deviation,
the skew ness and k u rto sis (relativ e peakedness o r flatness) of d ia m e ter
d istrib u tio n s, and stand v a ria b le s such as age, site , and stand m ean
d ia m e te r, for se v e ra l N orth A m erican co n ifers in n atu ral stan d s. He
showed that stand m ean d ia m e ter was v ery s a tisfa c to ry as an
independent v ariab le and th at, although the absolute values w ere
d ifferent for each sp ec ie s, the tre n d of the re la tio n sh ip s betw een the
c h a ra c te ris tic s of the d istrib u tio n and stand m ean d ia m e ter was s im ila r.
O thers have dem onstrated v ariatio n in the stem d ia m e ter d istrib u tio n s
of stands with the sam e average d ia m e ter but of d ifferen t age and on
different s ite s. In m any yield ta b le s, how ever, the stand ta b le s a re
h arm onised through stand m ean d ia m e ter ra th e r than through age and
site index.
If a stand is thinned a rtific ia lly , the n o rm al n atu re of the d istrib u tio n
is likely to be affected, depending on the kind and frequency of the
thinning, p a rtic u la rly the size c la s s e s on which the thinning is
concentrated. F or a num ber of thinnings of P . ra d ia ta d iscu ssed by
Jacobs (1962), C arron (1964) has shown that the d . b . h . o . b .
d istrib u tio n s of the stands before thinning, the thinnings, and the
stands a fte r thinning w ere all sufficiently n e a r to norm al d istrib u tio n s
for p ra c tic a l p u rp o ses. In other c a s e s , p a rtic u la rly with fa st grow ing
sp ecies which need frequent thinning, the tendency of a thinned stand
to approach n o rm ality again a fte r thinning is continually u p set by the
frequency of fu rth e r thinning, and at no stage is the stand d istrib u tio n
norm al once thinning s ta r ts . Lew is (1963a) com m ents on the
difficulty of fo recastin g future d ia m e ter d istrib u tio n s in A u stralian

coniferous plantations.
D .b .h .o .b . D istribution in an U neven-aged Stand

An uneven-aged stand is one in which the individual tre e s cover a
co n siderable range of age. Each tre e may d iffer in age from its
neighbours as in much single tre e selectio n silv ic u ltu re . When
reg en eratio n takes place in groups, as in group selection silv ic u ltu re ,
the tre e s within a group a re of the sam e age and. groups being
re g e n era te d at in te rv a ls , each group m ight be re g a rd ed as an even-aged
stand; but fo r m anagem ent p u rp o se s, the stand c o m p rises a num ber of
such groups and in com bination they re p re s e n t a range of age with a
co rresponding range of size from seedlings to v e te ra n s. The d .b .h .
o .b . d istrib u tio n of an uneven-aged stand is typically r e v e r s e - J shaped,
th e re being a la rg e num ber of sm all tre e s and d e crea sin g frequency as
d iam eter size in c re a s e s . An exam ple for an uneven-aged stand of
mixed eucalypt sp ec ie s (m ainly blackbutt) at Pine C reek State F o re st
(N .S .W .) is given in Table 11:2 and Fig. 11:2; the values a re num bers
T a b le 11:2
Stand tab les for an u n e v e n - a g e d stand of m i x e d eu caly p ts,

Pine C re e k S tate Forest, N . S . W .
D . b . h. o. b. class (in) 4-8 8 -1 2 12-16 16-20 2 0 -2 4 2 4 -2 8 28-3 2

No. per acre (a c tu a l) 48- 3 22*4 14 ’ 2 9‘ 1 4*4 2-4 1-8
No. per acre (for q - 1 - 7 3 ) 48- 3 27’ 9 16- 1 9- 3 5 ’4 3’ 1 1* 8
No. per acre (from 43-6 2 5 -0 1 3-0 8* 3 4-7 2 -7 1-6
regression)
D.b.h.o.b. (in)
Fig. 11:2 D . b . h . o . b . d istrib u tio n , u n e v e n - a g e d e u c a l y p t forest, Pine C re e k
S tate Forest, N . S . W . (see T a b le 11:2)
of tre e s p er a c re of all sp ecies deriv ed from sixty-nine h a lf-a c re
sam ple plots. The d ia m e te r distrib u tio n on sm all a re a s of uneven-aged
fo re s t may be much m ore irr e g u la r.
De L iocourt (1898) observed th a t, in c e rta in uneven-aged selection
fo re s ts in e a ste rn F ran c e , the ra tio between the num ber of tr e e s in
su cc e ssiv e d ia m e ter c la s s e s was roughly constant for a p a rtic u la r
fo re s t, the value of the ra tio differing from one fo re st to another.
This has since been confirm ed extensively in many kinds of uneven-aged
fo re s ts in various p a rts of the w orld. If the ra tio between the num ber
of tr e e s in su cc e ssiv e c la s s e s is re p re se n te d by q, the num ber of tre e s
in su ccessiv e d ia m e te r c la s s e s is re p re se n te d by a g eom etric s e r ie s
of the form
a, aq, a q 2, aq 3........ aqn_1
w here a re p re se n ts the num ber of tre e s in the la rg e s t size c la s s of
in te re s t and th e re a re n c la s s e s , n-1 being the sm a lle st size c la s s of
in te re s t. F o r the actual stand data of Table 11:2, a = 1 ’ 8, aq6 = 48-3
and so it follows that q = 1- 73. A stand table on th is b a sis would be as
in Table 11:2. If the lo g arith m s of the num ber of tre e s in su ccessiv e
c la s s e s a re plotted on re c ta n g u la r co -o rd in ate p ap er against the size s
of the c la s s e s , the b e st fit for such a geom etric s e r ie s is a stra ig h t
line as shown in Fig. 11:3. A ltern ativ ely , the plotting may be done on
sem i-lo g a rith m ic pap er. The re g re ss io n for the data of Table 11:2 is
log Y = 2* 000 - O’ 06014 X w here Y^ = num ber of tre e s p er size c la ss
and X = the m id-point of the size c la s s (4 in interval) in inches.
V alues of num ber of tr e e s p er size c la s s from this re g re ss io n a re also
shown in Table 11:2. The value of the ra tio c[ from these values is
1*73. C urtin (1962a) has proposed a distribution fo r m axim um
production in an uneven-aged blackbutt fo re s t in N. S. W. in which the
£ value is 1*47. Jacobs (1955) has used a c[ value of 2 in deriving
sim ple stocking guides in uneven-aged eucalypt fo re s ts such as those
at Pine C reek.
The distrib u tio n can also be re p re se n te d as an exponential cu rv e.
M eyer (M eyer and Stevenson. 1943; M eyer. 1953) has done considerable
w ork on the d ia m e te r d istrib u tio n s of an uneven-aged stand,
c h a ra c te ris in g them by the exponential function Y = k e _a^ w here Y =
num ber of tre e s p e r size c la ss X. e_ = base of n a tu ral lo g arith m s 2* 718,
and k a n d ji a re constants fo r a given d istrib u tio n .
A stand table fo r an uneven-aged stand of virgin rain fo re s t in North
Q ueensland is illu s tra te d in Fig. 11:4 (data by co u rtesy of the
Q ueensland D epartm ent of F o re stry ).
Stand T a b le s and Stand B asal A re a 77
of f r e q u e n c y
Log
8-12 12-16 16-20 20-24 24~28 28+

D.b. h. o.b. c l a s s ( i n )
Fig. 11:3 D . b . h . o . b . distribution, u n e v e n - a g e d e u c a l y p t forest, Pine C re e k
S tate Forest, N . S . W . (see T a b le 11:2)
0-12 12-24 24-36 36-48 48-60 60-72 72-84 84-96 96 +

Girth class (in)
Fig. 11:4 G . b . h . o . b . d istribution, u n e v e n - a g e d v ir g in ra in -fo r e st, North
Q u een sland , (see T a b le 11:3)
78 Outline of F o r e s t M en su ratio n
T a b le 11:3
Stand ta b le for an u n e v e n - a g e d stand of virgin ra in -fo r e s t,
A th erto n T a b le l a n d , N orth Q ueensland
G. b. h. o. b.
class (in) 0 -1 2 12-24 2 4 -3 6 3 6 -4 8 48-60 6 0 -7 2 7 2 -8 4 8 4 -9 6 961-
N um ber per
acre 445 208 82 47 23 10 7 3 10
Stand B asal A r e a
Stand b a s a l a r e a is the sum of the sectio n al a r e a s a t b r e a s t height
of all t r e e s in a stand and is u s u a lly e x p r e s s e d on a unit a r e a b a s is ,
fo r exam ple p e r a c r e . J u s t a s the b a s a l a r e a of a t r e e is an im p o rta n t
r e f e r e n c e v a r ia b le fo r t r e e s , so the b asal a r e a of a stan d is an
im p o rta n t r e f e r e n c e v a r ia b le fo r s tan d s and is useful in quantitative
d e s c rip tio n . It is usually d e r iv e d o v e r - b a r k . G e n e ra lly , the basal
a r e a of a stand is d e riv e d d ir e c tly by su m m in g the b a s a l a r e a s of
individual t r e e s . Depending on the n u m b e r of t r e e s , the technique
v a r i e s . F o r a few t r e e s , c o rre s p o n d in g v alu es of b a s a l a r e a from
p r e p a r e d ta b le s such as A. F .R . I. (1964) m ay be s u m m e d . F o r a
la r g e n u m b e r of individual e n t r i e s of d ia m e te r , the s q u a r e s of
d ia m e te r m ay be acc u m u la te d and m u ltip lied by an a p p r o p r ia te f a c to r .
F o r ex am p le, if the d ia m e te r is m e a s u r e d in in c h es and b a s a l a r e a is
r e q u ir e d in s q u a r e feet, as is conventional in A u s t r a li a , the f a c to r is
0-0054. If a stand ta b le has been co m p iled , the b a s a l a r e a
c o rre s p o n d in g to the m id -v a lu e of each c l a s s is m u ltip lied by the c la s s
freq u en cy and th e s e p ro d u c ts a r e su m m e d . If the c l a s s in te rv a l is wide
and th e r e a r e few t r e e s p e r c l a s s , the a s s u m p tio n th a t the m ean b a s a l
a r e a of the t r e e s in the c l a s s c o r r e s p o n d s to th a t of the m iddle of the
c l a s s m ay give anom alous r e s u l t s .
The stand m ean b asal a r e a is a useful v a r ia b le and is d e riv e d sim ply
by dividing the stand basal a r e a by the n u m b e r of t r e e s in the stand.
Stand m e an d . b . h . o . b . is u su ally d e r iv e d as the d . b . h . o . b . equivalent
to stand m ean b asal a r e a r a t h e r than as the a v e r a g e of the d ia m e te r s
except when dealing with a d ia m e te r d is trib u tio n . The q u a d ra tic m ean
is p r e f e r r e d to the a r it h m e t ic m ean b e c a u se of the additional weight
given to the l a r g e r d ia m e t e r s .
Angle Count Method

B asis and Background
A v aria b le such as stand basal a re a is e x p re sse d on a unit a re a
b a sis , that is as a ra te , for two m ain re a so n s: it is a convenient wa}r
of thinking; and the b asal a re a of a stand can be obtained read ily as the
product of the ra te and the a re a of the stand. If the b asal a re a of a
sam pling unit with defined boundaries is d eterm in ed by m easu rin g the
b asal a re a of each tre e , an e stim a te of the b asal a re a of the stand as
a ra te can be obtained if the a re a of land within the boundaries of the
sam pling unit is m easu red . A s im ila r sam pling e stim ate of stand
b asal a re a as a ra te is obtained by the angle count method without the
need to m easu re the basal a re a of each tr e e , o r the a re a of land of
the sam pling unit.
In 1947, an A ustrian fo r e s te r, W. B itte rlic h , who, according to
Thomson and Deitschm ann (1959), had a rriv e d at the b asic m ethod in
1931, published the f ir s t of a s e r ie s of p a p ers on ’angle count sam p lin g '.
He proposed that e stim a te s of stand b asal a re a p er unit land a re a could
be determ ined by an o p erato r standing at each of a num ber of points
selected by som e sam pling p ro c e s s , counting within a com plete c irc le
those tr e e s whose d ia m e te rs at b re a s t height appeared la r g e r than a
c e rta in re fe re n c e angle p ro jected by the o p e ra to r, and m ultiplying the
num ber of tre e s so counted by a facto r ap p ro p riate to the re fe re n c e
angle. The sim ple (though not wholly satisfacto ry ) explanation of the
p rin cip le which follows is due to Keen (1950). R eference to the
explanations by C rom er (1952), Kendall and Sayn- W ittgenstein (1959),
and Dawkins (1965) will suitably am plify it. Suppose one stands at a
point O with a piece of m a te ria l 2 L in length held h o rizo n tally a
distance y in front of one eye a s in Fig 11:5. A ssum e the eye is 4 ft
a / 2a
Fig. 11:5 P rin c ip le of a n g le c o u n t m e th o d (a fte r K een, 1950)

3 in above ground level and the whole ground a re a is level. A ssum e
th at the piece of m a te ria l, held at rig h t angles to a line of sight from
the eye to the cen tre of a v e rtic a l tr e e at b re a s t height a distance
ft away, ju st o b scu res the tr e e , th at is the lin es of sight from the eye
p a st the ends of the m a te ria l a re ju st tangential to the tr e e . A ssum e
the rad iu s of the tr e e is r^ ft. We accept as a c rite rio n th at a tr e e of
rad iu s rj_ w ill be counted if it subtends at the o p e ra to r's eye the angle
2a o r an angle la rg e r than th is - th at is it can be any d istan ce up to
Y l from O but no fu rth er; so th at within a c irc le of rad iu s Y i we sh all
count all tre e s of rad iu s r^ . The diagram shows th is lim iting condition.
7T.Yn
The a re a of the c irc le of acceptance of tre e s r^ in rad iu s is
43560
a cres. Let us assu m e th e re a re n^ tr e e s of ra d iu s r^ which w ill be
tt. Y ^
counted. If n^ tre e s a re counted in an a re a of a c r e s , then the
43560
num ber of tr e e s of ra d iu s r^ p e r a c re is given by — ------ =
TT. Y ± T T . Yi
43560
The b asal a re a of each such tre e is tt. r 1zsq. ft. The b a sa l a re a (sq
n x . 43560
ft p er acre) of the tre e s of ra d iu s r^ ft w ill then be 71.
2
TT. Y 1
nl
(ri r 43560 = nx s in 2a. 43560 since sin a.
(Yx2) Yl
We can reaso n in a s im ila r way fo r a tre e of any ra d iu s. F o r each
p a rtic u la r size tre e th e re w ill be an ap p ro p riate m axim um d istan ce
from O within which the tr e e m ust lie to be counted - tre e s of ra d iu s
T2 m ust lie within a distance Y£, and so on. The b asal a re a p e r a c re
fo r each size of tr e e w ill be the num ber of tr e e s of th at size which have
been counted, m ultiplied by a constant s in 2 a. 43560, since the ra tio
of tre e rad iu s to tre e distance £ i , £2- , J p t and so on m ust alw ays
equal sin a. The total b a sa l a re a p e r a c re w ill then be given by

(n^ + ii2 + n 3 + .........). s in 2 a. 43560 = N s in 2 a. 43560 = N x b asal a re a
fa c to r w here N is the to tal num ber of tre e s counted and the b asal a re a
facto r is s in 2 a. 43560, the value of a depending on the length of the
m a te ria l 2 L and the d istance y it is held from the eye.
In the f ir s t of sev e ra l p ap ers on the sub ject, G rosenbaugh (1952)
pointed out that an angle gauge, the in stru m en t used to se t up the fixed
re fe re n c e angle, is a 'tool for selecting sam ple tre e s with probability
p ro p o rtio n al to . . . s iz e ’ and he developed the ap p ro p riate th eo ry also
outlined by Husch (1963).
S tarting with B itte rlic h 's dem onstration of h is method to the T hird
W orld F o re s try Conference at H elsinki in 1949, and the publication of
what was apparently the f ir s t p ap er in English on the method by Keen
(1950), application of the prin cip le by fo r e s te rs in w estern Europe was
slow ly recognised. A fter propagation of its u sefu ln ess by C ro m er
(1952) in A u stra lia and by G rosenbaugh (1952) in the U .S. A. , the method
g rad u ally becam e popular. By 1959 Thom son and D eitschm ann (1959)
w ere able to lis t over one hundred re fe re n c e s to investigations into the
advantages and lim itations of the m ethod for determ in in g stand basal
a re a under various conditions, and p a rtic u la rly to adaptations of the
m ethod to the m easu rem en t of other stand v a ria b le s which w ere
pio n eered by B itte rlic h and trie d and extended by w o rk e rs in other
c o u n trie s, m ost prom inently w e ste rn E urope. N orth A m erica, and
Japan. As these authors say 'fo re s te rs seldom have the opportunity
to w itness the developm ent of a p ra c tic a l technique from its b asic
th eo ry to its widely accepted application in so sh o rt a tim e '.
A pplication of the method in A u stra lia has been lim ited to estim atin g
stand b a sa l a re a and only th is w ill be d iscu ssed h e re . Several
com prehensive accounts a re av ailab le, such as th at by Finch (1957a).
With so many fo re s te rs in different co u n tries w orking with the method,
it w as inevitable it would be r e fe rre d to by v ario u s nam es such as
W inkelzahlprobe (angle count sam pling), Winke lzahlm es sung (angle
count m easu rem en t), angle count m ethod, v a riab le plot m ethod, p lo tless
en u m eratio n , plo tless c ru isin g , B itte rlic h c ru isin g , point sam pling,
and wedge p rism method. The m ost common te rm in A u stra lia is
angle count m ethod, the sam pling point being re fe rre d to as an angle
count s p o t.
In stru m en ts
The object of an angle gauge is to provide at a sam pling point a
c e rta in re fe re n c e angle against which can be com pared the angle
subtended a t the sam pling point by the sid es of each tre e at b re a s t
height. The original gauge designed h}*- B itte rlic h , the 'R elask o p ',
co n sisted of a m etal blade two c e n tim e tre s wide m ounted at one end
of a rod one m e tre long which was held ho rizo n tally with the o th er end
ju st below the o b s e r v e r’s eye. The re fe re n c e angle was given by the
lines of sight from the o b s e rv e r's eye to the sid es of the blade. The
opaque blade made it difficult to com pare re lia b ly the width of blade
and tre e d iam eter but the in stru m e n t was sim ple and re a d ily m ade and
s im ila r in stru m en ts a re still in use. C o rrectio n of the b asal a re a on
slopes to horizontal equivalent land a re a was m ade by an o v erall
c o rre c tio n for average slope. In 1948, B itte rlic h produced a
'P en d elrela sk o p ' which in co rp o rated a pendulum device th at reduced
the width of the blade to account for the effect of slope in re sp e c t of
each tre e . C ro m er (1952) in A u stra lia com bined the p rin cip le of the
R elaskop with the optical p ro p e rtie s of a re fle c to r gunsight in an
in stru m en t he called the 'R e fle c to rs c o p e '. The R elaskop blade w as
rep laced by an illum inated im age of a g raticu le wTh ich , by a sy stem of
lens and re fle c to r, was seen su p er-im p o sed on the stem viewed
sim ultaneously through the eyepiece. The m ain advantages w ere th at
the position of the eye could be v a rie d without introducing p a ra lla x o r
alte rin g the re fe re n c e angle, and a tra n s p a re n t g ra tic u le made
com parison of tre e d ia m e ter and re fe re n c e angle m o re efficient than
an opaque blade. C o rrectio n for slope was m ade by c o rre c tio n to the
b asal a re a e stim ate for average ruling slope. The b a sa l a re a facto r
was 4-356. G rosenbaugh (1952) d iscu ssed v ario u s c ritic a l angles and
im provem ents to the R elaskop. About 1952, B itte rlic h produced the
'S p ie g e l-re la s k o p ', a com pact optical in stru m en t using a g ra tic u le of
converging lines fo r v ario u s b asal a re a fa c to rs which was fre e swinging
and so autom atically c o rre c te d for slope in re s p e c t of each tr e e . As
v e rsio n s of the blade type of in stru m e n t, Carow and Stage (1953)
suggested calib ratio n of the thumb and fin g e rs, p ieces of p la stic and
the like held a specific distance from the o b s e r v e r's eye.
A rad ical change in th ese designs was proposed by M uller (1953) in
the "B aum zahlrohr' (tree count tube), using the optical p ro p e rtie s of
a w edge-shaped optical p rism mounted at the end of a tube. When the
stem of a tre e is viewed through a p rism of th is type, the im age of the
stem is displaced la te ra lly by an amount depending on the d istan ce to
the tre e and the re fra c tiv e p ro p e rtie s of the p ris m . A tr e e is counted
only when the la te ra l displacem ent of the im age a p p ea rs le s s than the
width of the stem at b re a s t height. By using p ris m s of d ifferen t optical
strength as m easu red in d io p tre s , one can have a ran g e of b a sa l a re a
fa c to rs . The introduction of th is prin cip le prom pted a s e r ie s of
m odifications and developm ents. One was to use s trip s of wedge so that
instead of the two p a rts of the stem above and below b re a s t height being
re la tiv e ly d isplaced, only a section at b re a s t height was displaced.
Wedges w ere fitted to b in o cu lars to make e a s ie r the decision as to
w hether a tre e should be counted or not. V arious suggestions w ere
m ade for incorporating m echanism s to c o rre c t for slope in re s p e c t of
each tr e e viewed. Finch (1957a) has a com prehensive d iscu ssio n of
th ese v a rio u s in stru m e n ts, and techniques of calib ratin g them a re
outlined by G rosenbaugh (1952, 1958), Garow and Stage (1953), and
Husch (1963). Benson (1961b) d iscu ssed supply of wedges in A u stra lia
and th e ir c alib ratio n .
P ra c tic a l A spects
The m ethod gives th e o re tic ally exact e stim a te s which can be made
re la tiv e ly e a sily and quickly. T here a re , how ever, se v e ra l im portant
a sp e c ts which need co n sideration to e n su re th at the application of the
m ethod is as efficient and re lia b le in p ra c tic e as the th eo ry im p lies.
One of th ese concerns the b o rd e rlin e c a se s - those tr e e s which the
o p e ra to r cannot confidently decide should be counted o r not. A rough
solution is to count ev ery second such tr e e , o r to tal them and include
half the to tal in the final count. T his estim ate m ay suffer from an
o p e ra to r b ia s. The only re lia b le solution is to m e asu re the d ia m e ter
of the tr e e and to c alcu late, fo r the fa c to r being used, the m axim um
d istan ce from the sam pling spot w ithin which a tr e e of th at d ia m e ter
should be counted, and then to com pare th is distance with the actual
d istan ce from sam pling spot to tre e c e n tre . This p ro ced u re is , of
c o u rse , som ew hat tedious, but if tab les o r graphs of d ia m e te rs and
equivalent m axim um d istan ces fo r v ario u s b asal a re a fa c to rs a re made
up beforehand, they can be applied fa irly quickly.
A nother aspect is that a tre e w rongly counted co n trib u tes an amount
of b asal a re a equal to the m ultiplying fa c to r. Some co m p ro m ise is
n e c e s s a ry between the use of a fa c to r which gives a few tr e e s with a
m inim um likelihood of wrong counting but a re la tiv e ly high e r r o r from
a w rong count, and the use of a fa c to r which gives a la rg e num ber of
tr e e s with a higher likelihood of wrong counting but a re la tiv e ly low
e r r o r from a w rong count. T his m a tte r h as been investigated by v ario u s
w o rk e rs under v ario u s conditions and th e re a re divergent view s on it.
While the p ro c e ss can be c a rrie d out by one p e rso n , th e re is usually
a p ra c tic a l advantage in having an a s s is ta n t stand behind each tre e
(except those tr e e s v e ry close to the in stru m en t o p e ra to r fo r which the
decision is easy) with an a p p ro p riately coloured b o ard to indicate
b re a s t height and sharpen the tre e im age, and in having the a s s is ta n t
keep the tally.
T h eo retically in stru m en ts should be designed to allow fo r slope in
re s p e c t of each tr e e , to c o rre c t for a slope distance being used ra th e r
than a horizontal distan ce, otherw ise an e r r o r is in c u rre d . In stru m en ts
without such autom atic adjustm ent a re probably the m o re com m on,
c o rre c tio n s to h o rizontal equivalent being made by m ultiplying the b asal
a re a estim ate by an ap p ro p riate fa c to r. The facto r com m only used is
the secant of the g en eral angle of slope. Hodge (1965) a rg u e s th at the
secant of the m axim um angle of slope should be used. As an indication
of the extent of the c o rre c tio n the following secan t values a re given as
p ercentages:
Angle of slope 5° 10° 15° 20° 25° 30°
C orrection fa c to r(%) 100-4 101-5 103-5 106• 4 110-3 115‘5
Since the p ro c e ss norm ally involves m aking a 360° sw eep, b ias is
likely if positions a re re je c te d w here full c irc le sw eeps cannot be m ade,
as on the edge of a stand. On a la rg e a re a , re je ctin g such spots may
be unim portant; on a sm all a re a , it m ight be quite im p o rtan t. A
solution is half or q u a rte r sw eeps w here n e c e s sa ry and w eighting the
estim ate accordingly. This 'slo p o v e r' problem is d iscu ssed a t length
by G rosenbaugh (1958).
The p ro c e ss a ssu m e s th at the stem s a re v e rtic a l and the c ro s s -s e c tio n
is c irc u la r. U nless the lean is g re a t its effect on the shape of the
c ro s s -s e c tio n of view can be ignored. The effects of e c c e n tric ity may
be m ore serio u s and a re exam ined by G rosenbaugh (1958).
In dense stan d s, tr e e s m asked from view by intervening tr e e s may
p re se n t a p ra c tic a l problem . The o p e ra to r should move to view any
such tre e so th at he m aintains the orig in al d istan ce from the cen tre
to sam pling spot.
A dvantages and D isadvantages

E stim a te s of basal a re a p er a c re of a stand by the angle count method
a re sam pling e stim a te s and s im ila r in th at sen se to those made from
sam pling units such as p lo ts, the difference being th at p lo ts have a
defined boundary and size w h ereas angle count sw eeps do not. For
th at re a so n , to com pare the determ in atio n of stand b asal a re a by angle
count sw eeps with th at by plots re q u ire s a co m p ariso n which tak es into
account the re la tiv e p re c isio n of e stim ate and the re la tiv e tim e and
c o st involved. The p re c isio n of e stim ate depends in both c a se s on the
v a ria tio n of basal a re a per a c re over the stand and on the size of the
sam pling unit, ’s iz e ' in the case of the angle count unit being re fle c ted
in the b asal a re a facto r used. C om parisons of angle count sam pling
and sam pling using plots of various size under v ario u s conditions can
be found in the lite ra tu re alread y cited.
T re e s m ay be counted in such qualitative c a te g o rie s as se p a ra te
sp ec ie s, defective and non-defective. The m ethod does not provide a
d ire c t estim ate of num ber of tre e s p e r a c re nor num bers by size
c la s s e s . An in d irec t estim ate can be provided by an a s s is ta n t
m easu rin g with tape o r c a lip e r the d .b .h . o .b . of ev ery tr e e counted
in the sweep and tallying th ese by a p p ro p riate size c la s s e s . The basal
a re a p e r a c re ra te of each c la ss is then d eterm in ed by m ultiplying the
num ber in each c la ss by the basal a re a fa c to r. The num ber of tre e s
p e r c la s s , on a p e r a c re b a s is , is then derived by dividing the b asal
a re a p e r a c re ra te p e r c la s s by the basal a re a equivalent to the
m id-value of the c la s s .
The m ethod is dependent on the o p e ra to r being able to get an
unim peded view of the d .b .h . o .b . of each tre e he is likely to have to
c o n sid e r. If pruning or clearin g undergrow th has to be done for th is,
a balance m ust be stru c k between the effo rt involved in th is and in
estab lish in g sam pling units with defined bo u n d aries. F o r p erm anent
angle count units, in theory only a point need be m arked perm anently;
m ore than th is is usually needed in p ra c tic e .
12 Stand Height
Stand Mean Height

The arith m e tic average or m ean is com m only used as a condensed
exp ressio n of a set of values which v a ry one from another as do the
heights of individual tre e s in a stand. A m ean of th is kind for height
is of lim ited value if v ariatio n of individual values around the m ean is
ex trem e as is usually the c ase in uneven-aged stands w here individual
tr e e s m ay v ary from 2 in to 200 ft in height. H ow ever, m ean height
has considerable p ra c tica l use as a v a ria b le in even-aged stan d s w here
th e re is usually a re la tiv e ly sm all v a ria tio n in height. Mean height
at a specified age is often used as an e x p re ssio n of the productivity
of a site . T here is abundant evidence th a t, w ithin fa irly b road lim its,
th e re is little c o rre la tio n betw een height growth and stand density.
Height growth is likely to be reduced in ex trem ely open and ex trem ely
dense stands but in m ost n a tu ra l o r a rtific ia lly esta b lish e d stands it
is not affected much by the density of the crop and is m ainly a function
of tim e and the fa c to rs of site . If the stand is thinned, how ever, and
the thinning is d ire c ted p a rtic u la rly at rem oving the ta lle s t tr e e s or
the s h o rte st tr e e s , m ean height is a rtific ia lly affected and is not
effective as an index of site . Stand m ean height is a logical re fe re n c e
v aria b le for stand volum e. The concept of stand volume as the product
of stand b asal a re a , stand m ean height, and som e reducing facto r which
m ight be called stand form fa c to r, is a v e ry old one. Stand mean
height is also used as an independent v a ria b le fo r stand volum e ta b le s,
as d iscu ssed in C hapter 13.
To d erive the m ean height of a stand as an a rith m e tic av erag e by
m easu rin g the height of every individual tr e e in the stand is fa r too
laborious and is u n n e c e s s a ^ for m ost p u rp o se s. What is called m ean
height is n orm ally not the av erag e of individual tre e heights but an
approxim ation to it a sso ciated with the a rith m e tic m ean b asal a re a of
86
Stand Height 87
the stand. It is usually obtained e ith e r as the arith m e tic average of
the heights of a num ber of tr e e s having a b asal a re a equal o r
ap p ro xim ately equal to the a rith m e tic m ean b asal a re a of the stand,
o r by reading from a stand height curve (see p. 88) com piled fo r the
stan d , the height corresponding to the d ia m e ter equivalent to the m ean
b asal a re a of the stand.
A m ethod of deriving an ex p ressio n of stand m ean height based on
the B itte rlic h angle count p rin c ip le was proposed by H irata (1955),
E ssed (1955,1957), and Strand (1957). The re fe re n c e angle is held in
the v e rtic a l plane and com pared with the angle subtended at the
sam pling point by the total length of each tre e sighted in tu rn in a full
c irc le sw eep, those tre e s being counted the tops of which a re above
the elevation of the re fe re n c e angle. Mean height is deriv ed by
su b stituting the count num ber in an equation. The p rin c ip le s and
p ro c e d u re a re outlined by Finch (1957a) and G rosenbaugh (1958). This
m ethod h as not had m ore than v e ry c u rs o ry a p p ra isa l in A u stra lia.
Stand Top Height

F o r even-aged stands, an e x p re ssio n of the heights of the ta lle s t tre e s
is m o re often used as an index of site than an e x p ressio n of mean
height, since the fo rm e r is le s s likely than m ean height to be affected
by thinning under m ost conditions of thinning, and it is a m ore
p ra c tic a l index for classifying site over la rg e a re a s . T here a re
s e v e ra l such ex p re ssio n s in common use with d ifferen t nam es. For
coniferous plantations in A u stra lia , the te rm s top h eig h t, predom inant
h e ig h t, and m ean dom inant height a re used. They a re usually defined
as the arith m e tic m ean of the heights of the ta lle s t tre e s in the stand
at the ra te of about twenty to th irty p e r a c re , but, in one c a se , in
re la tio n to the fifty la rg e s t tre e s p e r a c re by d ia m e te r. In New Zealand,
the te rm m ean top height is the m ost com m on, defined as the height on
a stand height curve corresponding to the mean basal a re a o v e r-b a rk
of the 100 la rg e st d ia m e te rs p e r a c re . B ecause it is , when re la te d to
age, an efficient index of site , top height fo rm s a good re fe re n c e
v a ria b le fo r other stand v a ria b le s (see ch. 13). Depending on the way
in which the stand has been tre a te d , so th ere m ay be a relatio n sh ip
betw een stand m ean height and stand top height which allows convenient
co n v ersion of one to the other. F or exam ple for P inus ra d ia ta stands
as they a re thinned in the A. C. T. , th e re ap p ears to be a reasonably
co n stant difference betw een m ean height and top height of about 6 o r
7 ft irre s p e c tiv e of age and site .
Stand Height C urves

A stand height curve is a re la tio n sh ip between the heights of individual
tre e s in a stand and tlie ir re sp e c tiv e d ia m e te rs (or b asal a re a s ).
Even-aged JStands
If values of total height a re plotted on re c ta n g u la r c o -o rd in a te paper
against the corresponding values of d . b . h . o . b . for the individual tre e s
of an even-aged stand, a tre n d usually em erg es though it m ay not
always be c le a r. Within any d ia m e ter c la s s th e re is v ariatio n in the
heights of individual tre e s since tre e height and tre e d . b . h . o. b. a re
c o rre la te d only to the extent th at each is a re s u lt of tre e growth in the
p a rtic u la r environm ent over the p a rtic u la r p erio d of tim e. T ree height
a t a p a rtic u la r age is p rim a rily a product of s ite , tim e and, to a v ery
sm all extent, stand density. T re e d . b . h . o . b . a t th at age is p rim a rily
a product of stand density on that site over the period of tim e to that
age. H ow ever, if c la s s m ean values of height and d . b . h . o . b . a re
plotted, a c u rv ilin e a r tre n d , concave dow nw ards, is u su ally obtained.
O ver a period of tim e , because of the different re la tiv e grow th of
d . b . h . o . b . and height, the h e ig h t-d .b .h . o. b. relatio n sh ip for an
even-aged stand g en erally behaves as illu s tra te d in Fig. 12:1, that is
r 60
10 11 12 13 14
D.b.h.o.b.
Fig 12:1 S tand h e ig h t c u rv e , e v e n -a g e d P. ra d ia ta , K owen, A . C . T . (p la n te d 1940)
Stand Height 89
it moves upward and to the rig h t, usually in a fa irly sy m m e tric al way.

Though height and d . b . h . o . b . a re only p a rtia lly c o rre la te d , the
relationship between them is of co n sid erab le value. It is of
silv ic u ltu ra l in te re st in that it is indicative of the stand s tru c tu re . It
is of p ra c tic a l value in m easu rem en t because a v ery common method
of estim ating stand volume is based on it. In th is method heights a re
derived from a stand height curve correspo n d in g to m id -v alu es of
d ia m e ter c la s s e s of a stand tab le. F o r each d ia m e ter c la s s , the
volume of a tre e with c la ss d ia m e ter and the co rresponding height is
read from a suitable tre e volum e table. The c la s s average volume is
then m ultiplied by the frequency of the c la s s and the c la s s volume then
added to give stand volum e.
Except for special investigation, the stan d -h eig h t curve is usually
determ ined from a sam ple of the stand. The m ost efficient sam ple is
likely to be a s tra tifie d random sam ple, with d . b . h . o . b . c la s s e s as
s tra ta , the sam ple being allocated to the s tra ta according to re la tiv e
frequency and variation within s tr a ta , but th is method is ra re ly
employed in p ra c tic e . System atic sam pling is com m only em ployed.
T his provides an unbiased e stim a te , in the sen se of an unbiased
estim ate of the m ean, if the s ta r t of the sy ste m a tic sam pling is located
at random . The sy stem atic sam pling is so m etim es done with the
population in the o rd e r in which the individual tr e e s a re located or
m easu red in the field; som etim es it is c a rrie d out on a stand ta b le , in
which case the sam ple d ire c tly re fle c ts the d . b . h . o . b . d istrib u tio n .
Often the sam ple is sele c ted subjectively. In a sm all population with
which the o p erato r is reasonably fa m ilia r, and with lim ited v ariatio n
of height within d . b . h . o . b . c la s s e s , the b ias likely to re s u lt from
subjective selection may be unim portant; in o ther c irc u m sta n c e s it
m a y b e quite im portant. Since an e stim ate of e r r o r of e stim ate is
ra re ly made in p ra c tic e , the size of sam ple is v e ry often determ in ed
subjectively as one which the o p e ra to r's ex p erien ce su g g ests will give
what is som etim es called a 're aso n ab ly re lia b le r e s u lt'. If stand
height cu rv es a re com piled perio d ically fo r a p a rtic u la r stand, for
exam ple fo r deriving stand volume in periodic m anagem ent inventory,
growth stu d ie s, and experim ental w ork, it is im p o rtan t th at sensible
uniform ity in the position of the cu rv es and in th e ir change of slope be
m aintained. T here m ay be an advantage in having a perm anent sam ple.
Any bias in sam ple selection is m aintained co n sisten tly and a sensible
check can be kept of the tre e height m e asu re m e n t. A p erm an en t sam ple
m ay not, how ever, be at all tim es re p re se n ta tiv e , p a rtic u la rly in the
c a se of thinned stan d s. If a d ifferent sam ple is selected on each

occasion the method of selection should be o bjective.
If a m athem atical model can be assum ed fo r the stand height cu rv e,
o p e ra to r bias in fitting the relatio n sh ip by eye is avoided, uniform ity
can be m aintained over periodic m e asu rem en t, and calcu latio n s can be
done m echanically. Many attem p ts have been m ade to find a m odel
which d e sc rib e s the relatio n sh ip s a tisfa c to rily yet is sufficiently sim ple
to be calculated re a d ily . The relatio n sh ip usually a p p ea rs as p a rt of
a conic section - a p a rab o la, e llip se , o r hyperbola - and many curve
fo rm s of this kind have been suggested. T ro re y (1932), fo r exam ple,
found that for many stands in Canada the form H = a + bD + CD^ w here
H = tre e height and D = tre e d ia m e te r, was sa tisfa c to ry . He fu rth e r
assum ed that since the d .b .h .o .b . of a tre e of height equal to b re a s t
height is z ero , then the constant a of the equation could be assigned
the value of b re a s t height. S taebler (1954) also recom m ended th is
model for c e rta in sp ecies in N orth A m erica but w ithout the assum ption
that the value of the constant a should equal b re a s t height. Many
people have looked fo r ways of tra n sfo rm in g the c u rv ilin e a r relatio n sh ip
into a lin e a r one. One way is to re a rra n g e the v a ria b le s , co n stan ts,
and coefficients in the equation as exem plified by K er and Smith (1955)
who te ste d a num ber of m odels, concluded that the p arabolic equation
suggested by T ro re y was the m ost su itab le, and recom m ended
tra n sfo rm a tio n of it to a lin e a r equivalent ^ = b +cD (b rea st
height being 4 ft 6 in ). A nother way of tra n sfo rm in g the c u rv ilin e a r
relationship is a logarithm ic tra n sfo rm a tio n ; the p arab o lic form
H = aD^ becom es log H = log a + b log D. H enricksen (1950), w orking
on Norway spruce and beech, recom m ended H = a + b log D. Stoeffels
and van Soest (1953), working with Jap an ese la rc h and Douglas fir in
the N etherlands, a fte r try in g v a rio u s fo rm u lae, p re fe rre d H e n ric k se n 's
equation. Prodan (1965) liste d a num ber of equations and
tra n sfo rm a tio n s which have been proposed.
Kopezky and G ehrhardt (quoted by P ro d an , 1965) suggested th at a
lin e a r relatio n sh ip between tre e height and the re c ip ro c a l of tre e b asal
a re a follows from the lin e a rity of the tre e v o lu m e -tre e b asal a re a
relatio n sh ip in even-aged stan d s, i.e . H = a + *~2 (som etim es called
the G ehrhardt equation). As d iscu ssed in C hapter 13, the v o lu m e-b asal
a re a relatio n sh ip m ay be lin e a r provided the re la tio n sh ip s of tre e
height and tre e d .b .h .o .b . , and tre e height and tre e form fa c to r, take
p a rtic u la r fo rm s. A lin e a r relatio n sh ip betw een tre e height and the
Stand Height 91
re c ip ro c a l of tre e basal a re a , and a constant form fa c to r over the
range of d .b .h .o .b . , co m p rises one of the su itable com binations for
the lin e a rity of the v o lum e-basal a re a re la tio n sh ip . T akata (1962) and
K ajih ara (1966) have te sted the su itab ility of the G eh rh ard t equation,
as w ell as a num ber of the c u rv ilin e a r form s and th e ir tra n sfo rm a tio n s,
for som e Japanese sp ec ie s.
The num ber of m odels proposed confirm s what observ atio n suggests
- th at the relationship probably cannot be d e sc rib e d by one curve form
which m e re ly changes co nstants and coefficients according to sp ec ie s,
lo cality , stand condition, and tim e but that it d iffers in curve form
with tim e for a single stand and from stand to stand at a p a rtic u la r
tim e . Under th ese c irc u m sta n c e s, no single m odel is likely to be
u n iv e rsa lly satisfa c to ry . T h is, allied to the fact th at even if a model
has been found sa tisfa c to ry it m ay involve co n sid erab le calculation,
h as prom oted subjective fitting of the stand height curve in many c ase s
such a s those outlined by Hummel et a l . (1959) in G reat B ritain and by
Duff (1960) in New Zealand.
N e v e rth e le ss, experience shows th at the stand height curve of a
p a rtic u la r even-aged stand is likely to change its form and quantitative
n a tu re in a fa irly re g u la r way as illu s tra te d in Fig. 12:1, and that a
s e r ie s of standard (unit) height c u rv e s, can be estab lish ed to re p re se n t
the likely relatio n sh ip s of height with d .b .h .o .b . fo r a stand over its
life h isto ry from the inform ation of rep eated m e asu re m e n t, o r to
re p re s e n t the likely height-d. b. h. o .b . rela tio n sh ip s of a ll the v ario u sly
aged stands of an even-aged fo re s t from the inform ation of a single
inventory. Such a s e rie s is of p a rtic u la r value in applying the method
of d eterm ining stand volume outlined above, that is by re fe re n c e of a
stand table to a tre e volume table through a stand height cu rv e. The
stand height curve for a p a rtic u la r stand is nom inated from the s e r ie s
in stead of being com piled d ire c tly from sele c ted and m easu red sam ple
tr e e s , through an established c o rre la tio n of s ta tis tic s of the stand
height cu rv es and easily m e asu re d stand v a ria b le s . Work on th is
asp e c t for som e A ustralian stands is going on at p re se n t.
U neven-aged Stands
The relatio n sh ip of total height and d ia m e ter in uneven-aged stands
depends on the species com position of the stand and the tre a tm e n t given
to it. F or an a re a of uniform site in an intensively m anaged selectio n
fo re s t of a single coniferous s p e c ie s, for exam ple, th e re m ay be a
reaso n ab ly close relatio n sh ip which can be e x p re ssed as a slightly
sigm oidal curve constant with tim e , th at is the conditions of the stand
and the growth habits of the sp ecies a re such th at c e rta in heights a re
usually asso c iate d with c e rta in d ia m e te rs as illu s tra te d by P rodan
(1965a) fo r a plenterw ald of sp ru c e. It is to th is s o rt of stand th at a
single o n e-v ariab le tre e volum e table may be applied throughout the
life of the stand. By c o n tra st, it m ay be difficult to e sta b lish any such
relatio n sh ip in the case of an A u stralian uneven-aged fo re s t of m ixed
Eucalyptus sp ecies m anaged extensively under the group selectio n
s y s te m , and any re la tio n sh ip th at is estab lish ed m ay ap p ear to behave
som ewhat e rra tic a lly with tim e , depending on the re la tiv e h isto ry of
the individual tr e e s which make up the sam ple sele c ted to e sta b lish
the relatio n sh ip . F ortunately in th is c a s e , because it is of much m ore
value in p ra c tic e , a relatio n sh ip may be m o re re a d ily estab lish ed
between m erchantable height and d ia m e te r, as shown in F ig u re 12:2.
BLACKBUTT
ALL SPECIES
D b . h o b. (in)
Fig. 12:2 Stand h e ig h t curv e, u n e v e n - a g e d e u c a l y p t forest (m ix e d species), Pine C re e k
S tate Forest, N . S . W .
D espite co n siderable v ariatio n w ithin d ia m e te r c la s s e s , the average

values for d ia m e te r c la s s e s fit a curve fa irly w ell. If such a
relatio n sh ip can be accepted as constant over tim e , height need not be
re m e a su re d in periodic m anagem ent inventory, which obviates e r r o r s
in the e stim ate of volume in crem en t due to e r r o r s in height m easu rem en t
and difference of opinion of o p e ra to rs as to the m erch an tab le lim it,
and allow s the use of a o n e -v a ria b le volume table at all tim e s.
13 Stand Volume
D eterm ining the volume of fo re s ts of v ario u s s tru c tu re s , q u a litie s,

conditions, and value has been reco g n ised as im p o rtan t for about two
hundred y e a rs . During th at tim e many m ethods, and v ario u s techniques
fo r im plem enting them , have been developed and it is difficult to
c a te g o rise them for purposes of d escrip tio n in a c le a r- c u t way. The
c lassific atio n adopted h e re has been found convenient for academ ic
p u rp o ses.
Sampling is an e sse n tial p a rt of all m ethods of stand volume
estim atio n and in many c a se s the conversion fa c to r is v ery la rg e indeed.
In som e m ethods, the final e stim ate of the volume of a fo re s t is based
on m easu rem en ts at a sam ple of positions on individual tre e s which
to g eth er form a sam ple of the tr e e s in a sam pling unit, one of a num ber
of such units form ing a sam ple of the stand which its e lf is one unit in
a sam ple of the stands which to g eth er com pose the fo re s t. The fo re s te r
h as to keep in p ersp ectiv e the lim itatio n s to the in feren ces he can draw
about the population from the inform ation of a v e ry sm all sam p le, and
the effect of sam pling e r r o r and b ias at each stage of sam pling, in
expanding the estim ate from sam ple to population. The d iscu ssio n
h e re will be confined to the estim ation of stand volum e from the ground.
SAMPLE TREE METHODS
T his approach to stand volume estim atio n , in the form of the arith m e tic
m ean volume tre e m ethod, is one of the old est in fo re s try . The
intention is to determ ine the a rith m e tic av erag e of the volum es of the
tr e e s in the stand as sim ply as po ssib le and m ultiply it by the num ber
of tre e s in the stand. This obviously c alls fo r sam pling. F o r the
method to give a reasonable e stim a te w ithout an unduly la rg e sam ple,
93
the stand has to be fa irly hom ogeneous in volum e. The method is
th e re fo re likely to be r e s tric te d in p ra c tic e to even-aged stands of one
sp ec ie s. Even for a sm all plot of an even-aged stand of one sp ecies
we can anticipate having to m e a su re the volum es of se v e ra l tre e s to
get a reasonable e stim ate of m ean volume if we sele c t them by an
objective p ro c e ss . If we can 'sa m p le ’ the v ariatio n by eye, how ever,
we m ight get ju st as good a re s u lt with few er sam ple tr e e s ; at b e st we
m ight need to sele c t only one tr e e . By such reaso n in g fo r e s te rs
developed the p ra c tic e of subjectively selectin g only one tre e .
B ecause the volum es of individual tre e s do v ary one needs clues to
aid in the selection of the m ean volum e tre e . Depending on how much
inform ation is a lre a d y available - fo r exam ple, u sually a stand table
would be com piled in any case and so m ean b a sa l a re a would be
available - and on how much inform ation it is w orth while determ ining
esp e c ially fo r the purpose - fo r exam ple inform ation on height, ta p e r,
and b ark thickness - so one could have v ario u s clu es to aid selection.
As C haturvedi (1926) pointed out, the m ean volum e tre e w ill not have
stand arith m e tic m ean b asal a re a , height, and form fa c to r, for
m ath em atical re a so n s. This suggests a lim it to the tim e involved in
getting refined e stim a te s of th e se values. The m ain question is w hether
th e re actually e x ists in the stand a tre e with the a rith m e tic m ean volum e
of the stand - th e re is no reaso n to expect one (since the arith m e tic
average of som e quantitative m e asu re of a population does not
n e c e s s a rity exist) and experience suggests th e re ra re ly is one. This
suggests that two o r th re e tr e e s with d .b .h . o .b . , height, and form
fa c to r approxim ating stand m ean d .b .h . o .b . , m ean height, and m ean
form facto r would be m ore suitable; or p erh ap s m ore than th re e tr e e s .
The la rg e r the sam ple, the m ore the e sse n tia l sim p licity of the m ethod
is lo st and the m ore inform ation one should expect to get from it.
V ariations of the sim ple approach developed in w e ste rn Europe. The
m ain one was to form groups (stra ta ) on the b a sis of d ia m e ter c la s s e s ,
b a sa l a re a o r equal num ber of tr e e s and to apply the m ean tre e concept
to each group independently. Some of th ese group m ethods, re fe rre d
to by the nam es of th e ir proponents such as U rich, H a rtig , Schwappach,
w ere common in many p a rts of w e ste rn E urope, and in o th er co u n tries
influenced by E uropean p ra c tic e , up to about tw enty-five y e a rs ago, but
have since been abandoned.
The sim ple method is s till som etim es used in m odern p ra c tic e ,
p a rtic u la rly in re s o u rc e s inventory of even-aged stan d s. One exam ple
of th is is the re s o u rc e s inventory of the ra d ia ta pine plantations in the
Stand Volume 95
A. C .T . as discussed by C arro n (1955) and by C ro m er and Brown (1956).
In the N etherlands in re c e n t y e a r s , the m ethod has been used for both
re s o u rc e and p re lim in a ry w orking plan inventory of even-aged stands
of single sp ecies m ore than fifty y e a rs old.
TREE VOLUME TABLES
A tr e e volume table is a statem en t of the expected volum e of a tre e of

p a rtic u la r dim ensions o r of a tr e e in a stand of a p a rtic u la r kind. The
table may be for one sp ecies under p a rtic u la r growing conditions or
for m o re than one species under different growing conditions, depending
on the independent v a ria b les to which volum e is re fe rre d (d .b .h . o .b . ,
height, an expression of ta p e r, and an ex p ressio n of b a rk thickness)
and on the extent to which v ariatio n in tre e volume is accounted fo r by
those v a ria b le s. A tre e volum e table is com piled by p ro c e d u res
involving graphs or m athem atical an aly sis from the data of sam ple
tr e e s . The statem ent m ay be made in ta b u la r o r g rap h ical fo rm , both
of which allow d ire c t reading; or it m ay be e x p re sse d as an equation,
the volum e of a tre e being calculated when re q u ire d by substitu tin g in
the equation the ap p ro p riate values of the dim ensions of the tre e which
have been used as v a ria b le s in the equation. The m ain purpose of a
tre e volum e table is the e stim atio n of volum e of standing tr e e s . Though
tr e e volum e tables m ay be used to e stim a te volum es of felled tr e e s ,
th e se a re m ore often determ ined by d ire c t m ethods. A volum e table
is expected to provide a re lia b le estim ate of the av erag e volum e of a
num ber of tre e s in a d ia m e ter c la ss; it is not expected to provide the
volum e of an individual tre e to the sam e level of re lia b ility as bj" d ire c t
m e asu re m e n t, though the m ore the independent v a ria b le s used the
c lo s e r should be the volume table e stim a te to the actual volum e.
V olum e, the dependent v a ria b le , m ay be to tal volum e o r volum e to
som e d ia m e ter lim it, o v e r- o r u n d e r-b a rk , with o r without an
allow ance for stum p, in any p a rtic u la r unit such as cubic foot, su p er
foot o r Hoppus super foot. The independent v a ria b le s m ust be such
th a t tre e volume is c o rre la te d with them ; and fo r purpose of application
they m ust be read ily identified and re lia b ly m e asu re d on standing tr e e s .
The independent v a ria b le s a re th e re fo re chosen from an ex p ressio n of
siz e o v e r-b a rk at b re a s t height; an e x p re ssio n of height such as to tal
height or height to an easily identified lim it; an e x p re ssio n of ta p e r
th a t can be read ily m easu red o r re lia b ly estim ated ; and, for u n d e r-b a rk
volum e, an e x p re ssio n of b a rk th ick n ess in the vicinity of b re a s t
height.
It is m ost convenient to d isc u ss volum e ta b le s acco rd in g to the
num ber of independent v a ria b le s involved. T h e re is a conventional
nom enclature in which such te rm s as lo cal, stan d a rd , reg io n al,
g en eral, and u n iv e rsal a re m eant to indicate the scope of application
and to im ply th ereb y the num ber of independent v a ria b le s involved.
Scope of application and num ber of independent v a ria b le s a re , how ever,
not n e c e s sa rily connected so th at th is nom en clatu re is not convenient.
If the volum e statem en t is in ta b u la r o r g rap h ical fo rm , volum es a re
re a d in one way (direction) from a table with one independent v a ria b le,
in a tw o-w ay fashion from a tab le with two independent v a ria b le s , and
so on. T his leads to a convenient nom en clatu re - one-w ay volume
ta b le s, tw o-w ay volum e tab les and so on, which w ill be used h e re .
One-W ay T ree Volume T ables

If th e re is to be only one independent v aria b le fo r a volum e tab le, the
obvious choice is d . b . h . o. b. ( g . b . h . o . b . , b . a . o . b . ) . Since the volume
of a tre e is a function of its d . b . h . o. b. , height, and form fa c to r, a
relatio n sh ip of volume with d . b . h . o. b. can be e stab lish ed fo r a range
of d . b . h . o. b. in a stand only if th e re is a relatio n sh ip of height and form
facto r with d. b. h. o. b. for th at range of d. b. h. o. b. in the stand. The
tre n d of height and form fa c to r with d . b . h . o . b . w ill influence the
relatio n sh ip between volum e and d . b . h . o. b. The re lia b ility of the
e stim ate of volume of a tre e of a c e rta in d . b . h . o. b. re a d from a
one-w ay tab le w ill depend on how clo sely the actual height and form
facto r values of the tre e ag ree with those im plied in the table for that
d.b.h.o.b.
Even-aged Stands
The volum e c u rv e . We have a lre ad y seen th at for an even-aged stand

th e re is a reaso n ab ly well defined relatio n sh ip between tre e height and
d . b . h . o . b . a t any tim e and that if we plot th is re la tio n sh ip on
re c tan g u la r co -o rd in ate paper at v ario u s ages of the stand the
relatio n sh ip 'm oves' upw ards and to the rig h t with tim e (see ch. 12).
The rela tio n sh ip betw een form fa c to r and d . b . h . o. b. at a p a rtic u la r
tim e is not alw ays so well defined but in m ost c a s e s som e tren d is
evident. W here th e re a re tre n d s of both height and form facto r with
Stand Volume 97
d . b . h . o . b . , the relatio n sh ip of volume and d . b. h. o . b. is likely to be
a curve concave upw ards. The curve for the stand can be estim ated
by selecting ap p ro p riate sam ple tr e e s , m easu rin g th e ir volum es,
plotting the volum e of each tre e against its d. b. h. o. b. on co -o rd in ate
paper and fitting the curve of b e st fit by eye o r calculation. The volume
of a tre e of any d ia m e ter can be read from the curve and stand volume
derived in this way through a stand table. This p ro c e d u re, known as
the volume curve m ethod, has been in use in w estern Europe for about
a hundred and fifty y e a r s . It is not in use in A u stra lia and w ill not be
d iscu ssed fu rth e r.
The volume lin e . European fo r e s te rs reco g n ised th at the volume curve

often appeared to be p a ra b o lic . This suggested a tra n sfo rm a tio n of the
curve to a stra ig h t line by re la tin g volume to b asal a re a ra th e r than to
d . b . h . o . b . The idea of this v o lum e-basal a re a lin e , o r volume line
a s it has com e to be called , seem s to have been proposed f ir s t by
B ehringer in w e ste rn Europe in 1894 (C haturvedi, 1926). In 1899 it
was investigated fu rth e r by Kopezky and G eh rh ard t, with whose nam es
it is now m ostly asso c iate d in Europe. Since then it has been
investigated and used in G reat B ritain and A u stra lia and recen tly
in te re st has extended to c e rta in p a rts of N orth A m erica. The a ttractio n
of the volume line over the volume curve is its lin e a rity - it is so
much e a s ie r to fit a stra ig h t line to data, by calculation o r by eye,
than any other form of curve, and few er ob serv atio n s a re needed to
estab lish it.
Exam ples of volum e lines a re shown in Fig. 13:1. The volume line
is a sim ple lin e a r re g re ss io n of volume on b asal a re a which can be
e x p re ssed by the g en eral equation
Y = a + bX
w here Y_ re p re s e n ts tre e volum e, X re p re se n ts tr e e b a sa l a re a , a is
the re g re ssio n constant (the in te rc e p t of the volume line on the Y
axis) and b is the re g re ss io n coefficient (the ’tan g en t’ of the angle made
by the re g re ss io n line with the X axis). The re g re ss io n is calculated
by the method of le a s t sq u a re s which is d e sc rib e d in stan d ard tex ts on
sta tistic a l m ethods. The v a ria b le s used in Fig. 13:1 a re volume
u n d e r-b a rk from ground to 4 in and b a sa l a re a o v e r-b a rk .
W hether a lin e a r relatio n sh ip between tre e volum e and tre e b asal
a re a fo r an even-aged stand can be assu m ed , and under what
c irc u m sta n c e s, h as been a m a tte r of co n sid erab le in te re s t since it
was firs t proposed. G eneral experience in A u stra lia has been that the
b . a. o . b. ( s q ft)
Fig. 13:1 V o l ume - ba s a l area lines for an e v e n - a g e d stand of P. radiata, Kowen, A . C . T .
planted 1940 (see Fig. 12:1)
1955 v o l . = - 1 - 0 3 + 1 7 - 2 7 b.a.o.b.
1958 v o l . - 1 • 55 + 2 1 - 2 4 b.a.o.b.
1962 v o l . = - 2 - 0 6 + 25 • 49 b. a. o. b.
1967 v ol . = - 3 - 2 5 + 2 8 - 1 4 b. a . o . b .
relatio n sh ip betw een total volume u n d e r-b a rk and b asal a re a o v e r-b a rk
for even-aged stands of conifers such as P inus ra d ia ta is usually lin e a r
under m ost c irc u m sta n c e s and th is is also tru e for volum e u n d e r-b a rk
to a sm all end d ia m e ter u n d e r-b a rk lim it of the o rd e r of 6 in. F or
young stands, p a rtic u la rly those of poor site quality, for old unthinned
stan d s, and fo r volum e to sm all end d ia m e ter u n d e r-b a rk ( s . e . d . u . b . )
lim its g re a te r than about 6 in, the relatio n sh ip is likely to be
c u rv ilin e a r. The investigation by Keeves (1961) fo r P . ra d ia ta in
South A u stra lia supports th is.
The relatio n sh ip of tre e volum e with tre e b a sa l a re a at any tim e
re fle c ts the re la tio n sh ip s of tr e e height with tre e b asal a re a and tre e
form facto r with tre e basal a re a at th at tim e . The behaviour of the
volum e line with tim e re fle c ts the behaviour of th ese o th er relatio n sh ip s
over th at period. A num ber of com binations of these re la tio n sh ip s of
tre e height and tr e e form fa c to r with b a sa l a re a w ill produce a lin e a r
relatio n sh ip between volum e and basal a re a . F o r exam ple, one which
Stand Volume 99
has been dem onstrated for P. ra d ia ta on many occasions has been
re fe rre d to in C hapter 12. H ow ever, as volume to p ro g re ssiv e ly
g re a te r s . e . d . u . b . lim its is used, the relatio n sh ip of th is volume to
basal a re a becom es p ro g re ssiv e ly m ore c u rv ilin e a r. T e sts a re re a d ily
available to d eterm ine the significance of d e p artu re from lin e a rity .
In p ra c tic e an e stim ate of the volum e line for a stand is m ade from
sam ple tr e e s . To avoid o p e ra to r b ia s, the sam ple should be sele c ted
objectively. V arious m ethods of selection a re av ailab le, the choice
being influenced by w hether lin e a rity can be assum ed o r has to be
te sted . The size of sam ple w ill depend on the method of selectio n , the
v ariation of volume fo r a p a rtic u la r basal a re a , the p re c isio n of e stim a te
of volume of the individual sam ple tr e e s , and the p re c isio n of estim ate
of volume re q u ire d from the volum e line. T h ere is co n sid erab le
v a rie ty in the m anner in which the method is applied - in the num ber
and kind of sam ple tr e e s , in the method of selectin g th em , in the method
of determ ining th e ir volum es, in the method of determ in in g the volume
line and in deriving stand volum e from it. Subjective selectio n of
sam ple tr e e s has been used extensively with the idea of reducing the
size of the sam ple to a m inim um ; the v a ria tio n of volume within
d . b . h . o . b . c la s s e s is a s s e s s e d by eye and av erag e tr e e s chosen.
Hummel (1952), and M cIntyre and C arro n (1954), have indicated the
bias to which such selection is liab le but it can be kept under reaso n ab le
control if the s e le c to r is v ery fa m ilia r with the population. If the
volum es of the sam ple tr e e s a re determ ined from a volume table ra th e r
than by a d ire c t m ethod of m e asu re m e n t, the size of the sam ple m ust
be in c re ase d to obtain the sam e re lia b ility of e stim a te . The line of
b e st fit is often drawn by subjective judgm ent, ra th e r than calcu lated ,
as a check in the field on the fit of the line at a p re se n t tim e re la tiv e
to that at a previous tim e , o r to save the tim e and tro u b le of
calculation. The bias from th is so u rce v a rie s according to the
v a ria b ility of the data and the experience of the o p e ra to r. P e rh a p s
the m ost suitable method is to plot the values of the sam ple tre e s on
co -o rd inate paper in the field as a check on the assum ption of lin e a rity ,
on atypical sam ple tre e s and on e r r o r s and to calcu late the equation to
the line in the office la te r.
The volume line for a p a rtic u la r stand re p re s e n ts a one-w ay volum e
table for the stand. The volum e of any tre e can be obtained by
substituting its basal a re a o v e r-b a rk in the equation o r it can be read
off the graph if the line has been determ ined g rap h ically . The volume
of the stand can be derived bjr adding the volum es of the individual
tre e s so determ ined. F ro m the n atu re of the equation, how ever, stand
volume can be obtained v e ry re a d ily as
S(Y) = a .N + b.S(X)
w here S(Y) = the sum of the individual tre e volum es and thus stand
volum e, N = the num ber of tre e s in the stand, and S(X) = the sum of
the tre e b asal a re a s o v e r-b a rk .
The volume line m ethod has been used extensively in som e p a rts of
A u stra lia . G ray (1966) has review ed its introduction to A u stra lia and
developed s e v e ra l concepts of it. Jolly (1950), Lewis (1954). and
Keeves-(1961) have su m m a rised the co n sid erab le experience of South
A u stralian f o r e s te rs . Hummel (1955) investigated the m ethod in G reat
B ritain and its use by the F o re s try C om m ission was d iscu ssed by
Hummel et a l . (1959).
T a riffs. It has been pointed out th a t the volume curve o r volum e line
for an even-aged stand is a reflectio n of the tre e height and tr e e form
fa c to r re la tio n sh ip s with tre e b asal a re a and th at the behaviour of the
volum e relatio n sh ip with tim e is a reflectio n of the behaviour of th ese
o th er re la tio n sh ip s with tim e. The behaviour of the stand height curve
a p p ea rs to be of p a rtic u la r im portance. The graphical p ro g re s sio n of
the stand height curve with tim e is upw ards to the rig h t (see ch. 12).
The graphical p ro g re ssio n of the volume curve o r volume line follows
the sam e p a tte rn as shown in Fig. 13:1, w here is illu s tra te d the
p ro g re ssio n of the volum e lines equivalent to the stand height c u rv e s
of Fig. 12:1. F o r such a s e r ie s of volume cu rv es o r volum e lin e s the
nam e 'ta r iffs ' has come to be used in the English m ensuration lite ra tu re
a fte r the use of th is te rm in w e ste rn Europe for a volume ta b le in
which volume is e x p re sse d as a function of d .b .h . only. The v a rio u s
kinds of ta riffs have been su m m arised sy ste m a tic ally by P ro d an (1965a)
who p re fe rs the te rm 'ta r iff sy ste m ' for a s e r ie s of them (P ro d an , 1965b)
The g re a t value of ta riffs is th at if ta riffs covering the developm ent
of a stand with tim e can be com piled and the p a rtic u la r ta riff applicable
to the stand at a p a rtic u la r tim e can be nom inated in som e re a d y way.
the need to com pile the ap p ro p ria te volume curve o r volume lin e
d ire c tly is obviated. F o r e s te rs have th e re fo re looked for w ays of
com piling and nom inating ta riffs for v ario u s kinds of stands in v ario u s
p a rts of the w orld.
One way is to com pile from experience a s e r ie s of stand height cu rv es
and the accom panying ta riffs and nom inate the ta riff fo r a stan d at a
p a rtic u la r tim e according to its actual stand height curve at th a t tim e ,
Stand Volume 101
the stand height curve being e a s ie r to d eriv e than a volume line. This
approach has been used extensively in w e ste rn E urope, for exam ple
by K ru tzsch -L o etsch (1938). A lternatively , the ta riffs a re com piled
d ire c tly from experience and the ta riff fo r a stand at a p a rtic u la r tim e
is nom inated by the m inim um num ber of sam ple tr e e s . This was the
b a sis of the Algan ta riffs of F ran ce (Algan, 1902). Jo lly (1950) found
th at fo r ra d ia ta pine in South A u stra lia , volum e lin es of volume
u n d e r-b a rk to 4 in d .u .b . over a wide range of age and stand conditions
tended to converge to a value O’ 09 sq ft on the b . a . o . b . ax is. This
led to the m ethod, used under c e rta in c irc u m sta n c e s in South A u stra lia
for som e tim e , of selecting sam ple tre e s of m ean volume and draw ing
a line on graph paper from the plotted position of th e ir m ean to the
0*09 sq ft b a sa l a re a value, equivalent to nom inating one of a s e r ie s
of ta riffs rad iatin g from the 0* 09 value.
A nother way is to com pile the ta riffs fo r stan d s d ire c tly and then
re la te s ta tis tic s of these ta riffs to stand v a ria b le s . If the volume line
of a stand is a consequence of the stand height c u rv e , the s ta tis tic s of
the volum e line should be re la te d to w hatever c o n tro ls the form of the
stand height curve. If th is w ere site quality and age, for exam ple,
then stand top height m ight be a logical index of the s ta tis tic s of the
volume lin e , and once th e ir in te rre la tio n sh ip w as estab lish ed one
would then need only to determ ine stand top height to have the
a p p ro p ria te ta riff. In the 1930s G ray d e m o n strated th is was so for
ra d ia ta pine in South A u stra lia and la te r stim u la te d the production of
s im ila r ta riffs for coniferous plantations in v a rio u s p a rts of A u stra lia
(G ray, 1966). During the 1940s, the Q ueensland F o re s try D epartm ent
developed 'predom inant height volum e ta b le s ' of th is kind by g rap h ical
techniques, a set of ta riffs com piled from and fo r application to a re a s
of c e rta in stand top (predom inant) height.
Hummel (1955) investigated the c o rre la tio n betw een the s ta tis tic s
of the volum e line and stand indexes for v a rio u s sp ecies on v ario u s
s ite s in G re a t B ritain . O ver a la rg e ran g e of s p e c ie s, age, s ite , and
tre a tm e n t th e re was a common lin e a r c o rre la tio n of the re g re ss io n
coefficients of volume lin es with stand top height and the s e r ie s of
volume lin e s approxim ately converged on a value of 0* 03 sq ft on the
basal a re a a x is. This led to the production of 'g e n e ra l ta riff ta b le s ',
a s e r ie s of lines rad iatin g from 0 ‘ 03 sq ft b a sa l a re a through su cc e ssiv e
values of volum e, with one Hoppus foot in te rv a l betw een them a t 1 sq ft
b asal a re a and identified by num bers equivalent to those values of
volum e. The*tariffs w ere used to e stim a te volum es of thinnings, being
nom inated from the volum es of felled sam ple tr e e s . Finch (1957b)
rep o rted an investigation into the application of the tab les to a stand
through the c o rre la tio n of ta riff num ber and stand top height, and to
the main crop of a stand through a c o rre la tio n betw een ta riff num bers
for thinnings and m ain c ro p s. The m odern p ro ced u re for the use of
the ta riffs is outlined by Hummel et al. (1962) and the p rin c ip le s,
applications, and lim itations d iscu ssed .
C rom er and C arro n (1957) c a r rie d out a s im ila r investigation for
ra d ia ta pine in the A. C. T. and e stab lish ed c o rre la tio n s between the
re g re ss io n coefficients of volum e lines and stand top height, and
between the re g re ss io n constants of the volume lin e s and stand top
height and stand m ean b a sa l a re a . A ccepting a lin e a r c o rre la tio n of
re g re ss io n coefficients and constants with stand top height, the N. S. W.
F o re s try C om m ission (H enry. 1960) developed a system of harm o n ised
ta riffs by calculating a volume table based on the m ultiple re g re ss io n
V = a + bB + cH + d B . H
in which V = tre e volum e. B = tre e b asal a re a o v e r-b a rk , and
H = predom inant height of stand.
This equation can be w ritten
V = (a + cH) -f- (b + dH)B
Since for an a re a of a c e rta in predom inant height H_ is constant, the
equation can be w ritten
V = a' + b'B
a ' and b' having a lin e a r c o rre la tio n with H. This sam e m ultiple
re g re ss io n is also used to com pile tw o-w ay tre e volume ta b le s, as
d iscu ssed la te r, in which the height te rm , H, is the height of an
individual tr e e , the equation giving the volum e of a tre e of a p a rtic u la r
height and d . b . h . o . b . These two uses of the sam e m ultiple re g re ss io n
should not be confused. F o r ta riff s , the re g re ss io n is used to h arm o n ise
a s e r ie s of v olum e-basal a re a lines for stands; for tw o-w ay tre e volume
ta b le s, the re g re ss io n is used to harm o n ise a s e r ie s of v o lu m e-b asal
a re a re la tio n sh ip s each of which is for individual tre e s of the sam e
height. The Q ueensland D epartm ent of F o re s try used the sam e method
to im prove the ta riffs com piled p revio u sly by g rap h ical m ethods
(private com m unication).
Irre sp e c tiv e of how a ta riff s e r ie s is com piled and how the ta riff for
a p a rtic u la r stand at a p a rtic u la r tim e is sele c ted , once selected stand
volume is calculated from it in the sam e way as d e sc rib e d above for a
volume line.
Stand Volume 103
U neven-aged Stands
If the stand height c u rv e fo r an un ev en -a g ed stand r e m a in s co n stan t
with tim e , the v o lu m e -b . a. o. b. r e la tio n s h ip r e m a in s co n s ta n t with
tim e (assu m in g t r e e fo rm f a c to r r e m a in s co nstant). F o r total height
the stan d height c u rv e is u s u a lly such th a t the v o lu m e -b . a. o. b.
r e la tio n s h ip is c u r v il in e a r . When m e rc h a n ta b le height can be a s s u m e d
co n sta n t fo r all d. b. h. o. b. c l a s s e s g r e a t e r than a c e r t a in d. b. h. o. b.
with w hich crow n b r e a k is a s s o c ia te d , an as s u m p tio n which ap p lies
re a s o n a b ly w ell in the c a s e of som e A u s tr a lia n eucalypt f o r e s t s , the
v o l u m e - b . a . o . b . re la tio n s h ip w ill be l i n e a r fo r th o se c l a s s e s . F o r
ex a m p le , a m e rc h a n ta b le height of about 50 ft m ig h t be a s s u m e d for
all s i z e s of blackbutt above 14 in d . b . h . o . b . fro m the evidence of
Fig. 12:2. If the stand height c u rv e is not constant, with tim e , it is
n e c e s s a r y to apply tw o-w ay t r e e volum e ta b le s .
Tw o-W ay T r e e Volume T a b le s
C om pilation
The logical independent v a r ia b le s fo r a tw o-w ay volum e tab le a r e
d . b . h . o . b . and an e x p r e s s io n of height. The tab le is com piled from
the in fo rm atio n of sam p le t r e e s . The n u m b e r of s a m p le t r e e s which
should be used depends on the v a r ia b ility of the m a t e r i a l , the p re c is io n
of the e s tim a te r e q u ir e d , and the m ethod of co m pilation. T ab les have
been com piled in G e rm a n y using data fro m 30,000 t r e e s a c c u m u la ted
o v e r m any y e a r s . They have been com piled for b r o a d - le a v e d s p e c ie s
in A u s tr a lia fro m as few a s 30 t r e e s u sing the m ethod p ro p o sed by
Unwin and Bowling (1951). The m ethod of s elec tio n should be o b jective,
though th e r e m ay be a rg u m e n t fo r su b jectiv e s e le c tio n if the sam p le
m u s t be s m a ll. The population fro m w hich the s a m p le is to be draw n
should be specified c l e a r ly so the s a m p le w ill be r e p r e s e n t a tiv e . The
population to which the ta b le is m e an t to apply should a lso be specified
c l e a r t y so the u s e r is not m is le d . The volum es of the sam p le t r e e s
w ill n o rm a lly be d e riv e d by a d ir e c t m ethod of m e a s u r e m e n t; a
g ra p h ic a l m ethod h as the advantage th a t t r e e volum e can re a d ily be
d e riv e d in a s s o r t m e n t s which m a y be useful at a l a t e r tim e if not at
the p r e s e n t.
C om pilation by g ra p h ic a l m e th o d s . The volum e of a t r e e can be

e x p r e s s e d as V = B. H. F. w h e re V = t r e e volum e. B ~ b. a. o .b . ,
H = height, and F = form fa c to r. F ro m this one would expect a volume
table in which d . b . h . o . b . and height a re independent v a ria b le s to
c o m p rise a s e rie s of values which in c re a se in a re g u la r way with
in c re a se of d . b . h . o . b . for a p a rtic u la r height, and which in c re a se in
a re g u la r way with in c re a s e of height fo r a p a rtic u la r d . b . h . o . b . The
re g u la rity of this in c re a se in two d irectio n s will depend on the way
form fa c to r behaves. F o r a p a rtic u la r height, volume will b e a r a
parabolic relationship to d . b . h . o . b . and a lin e a r relatio n sh ip to
b . a . o . b . , if form fa c to r is constant o v er the range of d. b. h. o. b. If
form fa c to r is re la te d in som e o th e r but n e v erth ele ss re g u la r way to
d . b . h . o . b . , the volum e-d. b. h. o. b. relatio n sh ip may s till be a sim ple
cu rv e, though it m ay not tra n sfo rm to lin e a rity between volum e and
b asal a re a . If form fa c to r v a rie s e rra tic a lly with d . b . h . o . b . , the
c u rv ilin e a r relationship of volum e and d . b . h . o . b . may be com plex.
It m ay be reasonable fo r p ra c tic a l p urp o ses in this case to assu m e a
constant form fa c to r. The way the v o lu m e-d .b . h. o. b. relatio n sh ip s
fo r each height a re re la te d to each o th er depends on how form fa c to r is
re la te d to height. T here m ay be a sy m m e tric al change of form fa c to rs ,
asso c iate d with each d . b . h . o . b . , with in creasin g height. The change
m ay be e rra tic so that th e re is no sym m etry from one volume -
d . b . h . o . b . relationship to another. It may be reaso n ab le to ignore the
e rra tic v ariatio n of form fa c to r in the in te re s ts of p ra c tic a l com pilation
of the table.
The m ain concern in com piling a tw o-w ay tre e volume table is to
d iscover the p a ttern of change of volum e with change of d . b . h . o . b .
and height and then to e x p re ss it in the sim p lest but m ost efficient way.
One method is to plot the values on graph p aper and fit the re la tio n sh ip s
m ost suitable to the data by g rap h ical m ethods. This is the oldest
approach. It was used in w e ste rn Europe o v er a hundred y e a rs ago
and is s till used in many p laces today. G raphical m ethods have c e rta in
advantages; little m athem atical knowledge is re q u ire d w h ereas som e
m athem atical background is re q u ire d to solve equations; the
re la tio n sh ip s a re determ ined from the graphical evidence and no
assum ption about the kind of re la tio n sh ip , or testin g of the m ost likely
re la tio n sh ip s, is n e c e s sa ry as is the case with equations. On the other
hand, the m ethod is subjective and no two co m p ilers n e c e s s a rily derive
exactly the sam e table from the one se t of data, though if they a re
reaso n ab ly experienced the d ifferen ces between th e ir tab les would
usually be sm all enough not to invalidate the use of e ith e r tab le; and
for subjective fitting of re la tio n sh ip s of this kind to be sa tisfa c to ry a
Stand Volume 105
la rg e amount of data is usually re q u ire d . This may be difficult o r
expensive to c o llect, p a rtic u la rly if the sam ple tre e s a re standing.
Most of the grap hical m ethods go under the g en eral nam e harm onised
curve m ethod. The basic approach ap p ears to have been introduced to
N orth A m erica many y e a rs ago by G raves following ex p erien ce in
Europe and is often re fe rre d to under the nam es of the v ario u s authors
who propagated it in text books. The m ain difficulty with the method
is in try in g to e sta b lish a th re e -d im e n sio n al relatio n sh ip on two-
dim ensional pap er. A lso, th e re is no p ra c tica b le a lte rn ativ e to form ing
the sam ple tre e data into c la s s e s of d .b .h .o .b . and height, deriving
a v erag es, and plotting th ese to esta b lish the relatio n sh ip . U nless
th e re are v ery many sam ple tr e e s , th ese av erag es w ill not n e c e s sa rily
be the cen tral values of the c la ss in te rv a ls so that to the n atu ral
v ariab ility of volum e due to v aria tio n in form fa c to r is added th is
a rtific ia l v a ria b ility which fu rth e r com plicates the com pilation.
V arious m odifications have been proposed to overcom e th ese
difficulties. The sim p le st and m ost common m ethod is to a rra n g e the
sam ple tre e data into height c la s s e s within d .b .h .o .b . c la s s e s , and
derive average volum e, height, and d .b .h .o .b . for each c la ss (cell).
At an appropriate scale on re c tan g u la r co -o rd in ate p ap er, average
volume is plotted against average d .b .h .o .b . fo r each height c la ss
sep arately and c u rv e s fo r each height c la ss a re fitted by eye, the fact
th at the values of av erag e height for the c e lls m ay not be the cen tral
values of the c la s s being te m p o ra rily ignored. F o r re fe re n c e p u rp o se s,
th is graph can be called G raph 1. From Graph 1, volum es for the
c en tral value of a d .b .h .o .b . c la s s a re re a d in tu rn fo r each height
c la s s and these a re plotted on co -o rd in ate p ap er ag ain st the average
heights of the sam ple tre e data fo r the a p p ro p riate c e lls. This is done
in tu rn for each d .b .h .o .b . c la s s and cu rv es a re fitted by eye. This
form s Graph 2. F rom G raph 2, volum es for c e n tra l values of d .b .h .o .b .
c la s s are read off and plotted against th e se v alu es, fo r each height
c la s s in tu rn , and c u rv e s fitted by eye. This fo rm s G raph 3 and
constitutes the final volum e ’ta b le ' from which values can be read and
presented in ta b u la r form .
If the tre n d of form fa c to r with d .b .h .o .b . w ithin a height c la ss is
such that volume plotted against basal a re a gives a lin e a r relatio n sh ip ,
b asal area would norm ally be used instead of d ia m e ter throughout. If
the trend of volume with height w ithin d .b .h .o .b . c la s s is thought to
be m ore re a d ily definable than the tre n d of volum e with d .b .h . o .b .
w ithin a height c la s s , volum e is plotted ag ain st height for d .b .h .o .b .
c la s s e s in Graph 1. and G raph 2 becom es the relatio n sh ip of volume
against d. b. h. o .b . accordingly. The m ost common m odification of
the method is th at d escrib ed by G riffith and P ra s a d (1949). proposed
o rig in ally by Dwight (1937). In th is . G raph 1 is com piled as p rev io u sly ,
and, for each height c la s s in tu rn , average height is plotted against
average d .b .h .o .b . in each cell and cu rv es a re fitted to th ese points.
The equivalent to G raph 2 is com piled using volum es fo r the c en tral
value of a d .b .h .o .b . c la s s for each height c la ss in tu rn as read from
G raph 1, but plotted against heights read from the heig h t-d . b. h. o. b.
cu rv es; that is volum es a re plotted against sm oothed values of the
o riginal cell heights ra th e r than the o riginal cell heights th em selv es.
G raph 3 is com piled as before. A nother m odification was suggested
by Bowling (1951) in connection with the com pilation by g raphical
m ethods of volume tables for regrow th eucalypts in T asm ania.
A num ber of volum e ta b le s, com bined with ta p e r ta b le s, fo r v ario u s
exotic co n ifers in New Zealand have been com piled bjr Duff (1954),
Duff and B u rstall (1955). and B u rsta ll (1957). The method used is
d e sc rib e d as
. . . a graphical form fa c to r m ethod based on d .b .h .o .b . and total
height. The av erag e form fa c to rs (calculated from tre e volume
u n d e r-b a rk as a proportion of an equivalent cy lin d er o v er-b ark ) a re
plotted against d .b .h .o .b . A sm ooth curve is draw n through the
points thus providing the f ir s t estim ate volume tab le. Where
n e c e s sa ry , height c o rre c tio n s a re applied to th is p re lim in a ry table.
In c ase s w here the m ethod fails to give s a tisfa c to ry re s u lts it is
n e c e ssa ry to draw s e p a ra te form fa c to r cu rv es for d ifferen t height
c la s s e s .
F or tab les com piled g raphically it is conventional to c a r ry out two
te s ts of com pilation. By ap p ro p riate in terp o latio n , volume is read
from Graph 3 for each sam ple tre e according to its actual d .b .h .o .b .
and height. The difference betw een the sum of th ese volum es and the
sum of the actual volum es of the sam ple tre e s is e x p re ssed as a
p ercentage of the la tte r. Conventionally, th is aggregate difference
p e r cent should not exceed about 1 p er cent. The d ifference betw een
volume table volume and actual volum e for an av erag e tre e of each
cell is e x p re ssed a s a p ercentage of actual volum e. These values of
cell deviation p er cent a re tabulated and checked for unreasonably
la rg e deviations, bias in p a rtic u la r sectio n s of the ta b le , o r ex trem e
com pensations. These deviations may suggest re fittin g of som e of the
c u rv e s. When the checks a re sa tisfa c to rily com pleted, v alu es of
Stand Volume 107
volum e for ap p ro p riate values of d .b .h .o .b . and height a re re a d from
G raph 3 and a rra n g e d in ta b u la r form as shown in Table 13:1. The
in te rn a l fram e superim posed on the ta b u la r layout shows the extent of
co v erage of sam ple tr e e s . V alues outside th is fram e a re derived from
ex trap olation of the cu rv es and a re included in the table only as
ex p erience suggests such tre e s ex ist and th e ir volum es may be re q u ire d .
E xtrapolation is usually lim ited to one c ell in each d irectio n and the
v alu es should be used with caution.
C om pilation by calculating equations. To d e sc rib e a relatio n sh ip

am ongst v a ria b le s by an equation, the kind of ex p ressio n which d e sc rib e s
Table 13:1
Tree v ol u m e table for P. radiata, Mt Stromlo, all ages and sites; v o l u me under-bark
from ground to 4 in d . u . b . , in c u ft: based on 250 sample trees measured standing by the
s ect i onal method; c o mp i l e d by A. Forester, 1968 by the harmonised curve method;
aggregate difference O’ 8 per c e n t (for illustration only and not for use)
D.b.h.o.b. Total height (ft)

(in) 40 50 60 70 80 90 100
5-0 1-4 2-5 3-5 4-6
6-0 2-2 3-5 4-7 5-9 7-1
7-0 3-2 4-6 6-0 7-4 8-8
8-0 4-3 5-9 7-5 9- 1 10-7 12- 3
9-0 5-5 7-4 9-2 11-0 12-9 14-7
10-0 6-9 9-0 11-2 13-3 15-4 17-5
11- 0 8-5 10-9 13*3 15-7 18- 1 20 6
12-0 10- 1 12-9 15-6 18- 3 21- 1 23-8
13-0 15- 1 18-2 21-3 24-4 27-5
14-0 17-4 20-9 24-4 27-9 31 - 4 34 - 8
15-0 20-0 23-9 27-7 31-6 35-5 39-4
16-0 22-7 27-0 3 1- 3 35-7 40-0 44-3
17-0 35-1 39-9 44- 7 49-5
18-0 44-5 49-8 55- 1
19-0 60- 7 67-0

108 O utline of F o r e s t M ensuration
the re la tio n s h ip b e s t should f i r s t be d e te r m in e d , then the co n stan ts
and co efficien ts of the b e s t fit of th a t e x p r e s s io n to the p a r ti c u l a r data
c a lc u la te d . C o n sid e ra b le investigation has been c a r r i e d out in v a r io u s
p a r t s of the w o rld into the kind of equation a p p r o p r ia te to d e s c r ib e the
re la tio n s h ip of volum e with d . b . h . o . b . and height. Many equations
have been suggested: no one p a r t i c u l a r equation has been found su ita b le
fo r all s p e c ie s u n d er all conditions. A part from the use of d iffere n t
s ta n d a r d s of volum e, the m o st likely r e a s o n s fo r th is a r e the d iffere n t
p a tte r n of v a ria tio n of form fa c to r within height and d . b . h . o . b . c l a s s e s
within a s p e c ie s betw een p la c e s , and between s p e c ie s within p la ces:
and v aria tio n in b a r k th ic k n e s s , p a r ti c u la r ly at b r e a s t height.
A n a tu ra l s ta r t in g point in the s e a r c h for an a p p r o p r ia te equation w as
the evidence, fro m g ra p h ic a l m ethods of com piling volum e ta b le s , of
the re la tio n s h ip between volum e and b asal a r e a within height c l a s s e s
and that between volum e and height within b a s a l a r e a c l a s s e s . T hese
r e la tio n s h ip s w e r e often lin e a r o r n e a r ly so. and this su g g ested
com puting them s e p a r a t e ly and then h a r m o n is in g th e se single
r e g r e s s i o n s by g rap h ic al sm oothing of the c o n s ta n ts and co efficien ts
of th e se r e g r e s s i o n s . T his w as the b a s is of the e a r l y in v e stig atio n s
of the F o r e s t r y C o m m is s io n of G reat B rita in (H um m el. 1955) and of
v a r io u s f o r e s t s e r v i c e s in A u s t r a li a at about the s a m e tim e , for
exam ple the T a s m a n ia n F o r e s t r y C o m m issio n (1953) and the New South
Wales F o r e s t r y C o m m is s io n (H enry, 1955). The technique is often
r e f e r r e d to a s com p ilatio n by h a r m o n is e d r e g r e s s i o n s e r i e s .
The next step w as to in v e stig ate the r e la tio n s h ip between height and
the co n stan ts and co efficien ts of the r e g r e s s i o n s of volum e on d . b . h . o . b .
fo r height c l a s s e s , and the r e la tio n s h ip betw een d . b . h . o . b . and
the c o n s ta n ts and co e ffic ie n ts of the r e g r e s s i o n s of volum e on height
for d . b . h . o . b . c l a s s e s . The fo rm of th e se r e la tio n s h ip s in d icate s the
m o s t s a tis f a c to r y m u ltip le r e g r e s s i o n . F o r m any s p e c ie s in v a rio u s
p a r t s of the w o rld , the c o n s ta n ts and co efficien ts of the s im p le lin e a r
r e g r e s s i o n s of volum e on b a s a l a r e a for s e p a r a t e height c l a s s e s have
been shown to be lin e a r ly r e la te d to height c l a s s : a = a j + b. H and
and b = a 2 + b 2 H. Substituting fo r _a and b in V = a + b B, w hich is the
r e g r e s s i o n of volum e on b a s a l a r e a fo r height c l a s s , we have
V = (a^ + b^ H) + (a2 + b 2 H)B, th a t is V = a^ + a 2 B + b^ H + b 2 B. H
w h e re V_ = t r e e v o lu m e , H = t r e e height and B = t r e e b . a . o . b . This
is the A u s tr a lia n e q u a tio n , s o - c a ll e d by S p u rr (1952) with r e f e r e n c e to
the fact that T . N . Stoate of W este rn A u s tr a lia had su g g ested it as a
Stand V olum e 109
lik ely fo rm fo r volum e ta b le s (S toate, 1945), as a m e an s of id en tify in g
it in a c o m p a ra tiv e stu d y S p u rr c a r r ie d out on v a rio u s eq u atio n s. A
re d u c e d fo rm of th is eq u atio n , V = a + b B .H , r e f e r r e d to by S p u rr fo r
id e n tific a tio n p u rp o s e s a s the com bined v a ria b le eq u atio n , h as a lso been
found in m any c a s e s to give a s a tis f a c to r y fit. It im p lie s a h y p erb o lic
re la tio n s h ip betw een fo rm fa c to r and th e p ro d u c t of b a s a l a r e a and h eig h t.
If fo rm fa c to r is c o n s ta n t, th e re is a sim p le lin e a r re la tio n s h ip betw een
v olum e and the p ro d u c t of b a s a l a r e a and height: V = aB . H, s o m e tim e s
r e f e r r e d to a s th e fo rm fa c to r e q u a tio n . T h is a v e ry old fo rm u la fo r
co m p ilin g volum e ta b le s and w as u sed in w e s te rn E urope in th e la s t c e n tu ry .
If th e re la tio n s h ip of volum e on d . b . h . o . b . and heig h t is c u r v ilin e a r , o r th e
re la tio n s h ip of the c o n s ta n ts and c o e ffic ie n ts of r e g r e s s io n s of volum e on
b . a . o . b . in h eig h t c la s s e s is c u r v ilin e a r , m u ltip le r e g r e s s io n s m o re
co m p lex than th e A u s tra lia n eq u atio n a r e needed and v a rio u s eq u atio n s
h av e b een tr ie d in c o rp o ra tin g f u r th e r p ro d u c ts of d . b . h . o . b . and h eig h t,
and p o w ers of heigh t fo r th e se co n d itio n s. T h ese in clude lo g a rith m ic
e x p re s s io n s w hich have been p a r tic u la r ly p o p u la r in N o rth A m e ric a .
H usch (1963), follow ing S p u rr (1952), li s t s and c o m m en ts on th e v a rio u s
e q u a tio n s, a rith m e tic and lo g a rith m ic , w hich have been su g g ested .
B eca u se th e u se of tw o-w ay tr e e v olum e ta b le s is v e ry com m on in
J a p a n , th e re h a s been c o n s id e ra b le in v e stig a tio n into eq u atio n s su ita b le
fo r s p e c ie s in th a t co u n try , su ch as th o se by O htom o (p riv ate
com m u n icatio n ) and T ak a ta (1962).
The in tro d u ctio n of high sp eed c o m p u te rs h as in c re a s e d en o rm o u sly
the sco p e of in v e stig a tio n into s u ita b le eq u atio n s. F o r p a r tic u la r d ata,
a n u m b e r of lik e ly eq u atio n s can be c a lc u la te d and th e ir effic ien cy
c o m p a re d v e ry quickly. The m o s t s a tis f a c to r y a p p ro a c h , h o w ev er, is
lik e ly to be th a t p ro p o se d by L aw ren ce (1965) in w hich eq u atio n s a r e
b u ilt up, new v a ria b le s being added as c a lc u la tio n show s them to be
effectiv e in red u cin g re s id u a l v a ria tio n .
W ork in A u s tra lia s u g g e s ts th a t, fo r p la n tatio n c o n ife rs , the
A u s tra lia n eq uation g iv es a s a tis f a c to r y fit fo r to ta l volum e and fo r
volum e to s m a ll d ia m e te r lim its su ch a s 3 in and 4 in, p ro v id ed th e
ra n g e of d . b . h . o . b . is lim ite d . The co m b in ed v a ria b le eq u atio n is
often a lm o s t a s s a tis f a c to r y . If the ran g e of d . b . h . o . b . is ex ten siv e
o r th e m e rc h a n ta b le lim it is lo w e r down th e tr e e , fo r ex am p le 6 in o r
8 in , the c u rv ilin e a r ity is su ch th a t a m o re co m p lex eq u atio n is n e c e s s a r y .
A lte rn a tiv e ly , te ch n iq u es m ay be u sed such a s th a t of C ro m e r, M cIn ty re,
and L ew is (1955) in co m p ilin g a G e n e ra l V olum e T ab le fo r P . r a d ia ta ,
w h ere s e p a ra te A u s tra lia n eq u atio n s w e re c a lc u la te d fo r th e sam p le tr e e
d a ta divided into th r e e o v erla p p in g s e c tio n s .
M ethods of com pilation fo r native b ro ad -leav ed sp ecies v a ry but

m ost com m only involve fitting suitable re g re ss io n s of volume on
d .b .h .o .b . o r b .a .o .b . in m erchantab le height (log length) c la s s e s , or
volume on m erchantable height in d .b .h .o .b . o r b . a . o . b . c la s s e s ,
e ith e r by subjective graphical m ethods o r by calcu latio n , then plotting
constants and coefficients of th ese re g re s s io n s ag ain st height o r
d .b .h .o .b . to h arm o n ise them . Many tab les have been com piled by the
method proposed by Unwin and Bowling (1951) o r a m odification of it.
Som etim es the com bined v a ria b le equation o r the form facto r equation
has been co n sid ered to give a s a tisfa c to ry fit. As fo r c o n ife rs, the
relatio n sh ip s a re probably b a sic a lly c u rv ilin e a r and ap p ear to be lin e a r
only if the range of the data is lim ited . The technique of subdividing
the sam ple tre e data and com puting lin e a r re la tio n sh ip s fo r each section
of what is o v erall a c u rv ilin e a r relatio n sh ip has also been used in
th ese c a se s as outlined by C ro m er (1959).
Com pilation by alignm ent c h a rt m eth o d s. Many N orth A m erican

textbooks, for exam ple B ruce and Schum acher (1950), re fe r to the use
of an alignm ent c h a rt to com pile a volume tab le. An alignm ent c h a r t,
o r nom o g ram , is a graphical re p re se n ta tio n of a relatio n sh ip o r function
in which the axes re p re se n tin g the v a ria b le s a re se p a ra te stra ig h t lin es
o r c u rv e s and do not in te rs e c t as do the axes on c o -o rd in a te p ap er.
The sca le s on the a x es, and the re la tiv e position of the a x es, a re so
arra n g ed that a connection line a c ro s s the c h a rt in te rs e c ts the axes
at p a rtic u la r values which sa tisfy the function illu s tra te d . The theory
and practical construction of such c h a rts a re outlined in v ario u s texts
on m ath em atics. The common use of such a c h a rt is to display a
relatio n sh ip a lre ad y estab lish ed by o th er g rap h ical m ethods o r by
calculating an equation. The m ethod of com piling a volume table which
is suggested by the North A m erican tex ts is to p re p a re an alignm ent
c h a rt on the b a sis of an assum ed equation and then gradually shift the
sca le positions of the axes in an ite ra tiv e p ro c e ss until the volum es
estim ate d from the c h a rt a re sufficiently clo se to the actual volum es.
The final c h a rt thus re p re se n ts a m odification of the original equation
but its exact form need not be determ in ed , since the c h a rt fo rm s the
volume table. T his method of com pilation has not been used to any
extent in A u stra lia. The m ethods d iscu ssed previously a re likely to
be m ore efficient.
Stand Volume 111
A pplication
The application of a tw o-w ay tre e volum e table to a stand to d eterm ine
stand volum e depends on the s tru c tu re of the stand, the num ber of tr e e s
w hose volum e is to be e stim ate d , and the stan d ard of re lia b ility of
e stim a te re q u ire d . When th e re a re few tre e s o r the v a ria tio n in height
betw een tre e s within a d .b .h .o .b . c la s s is c o n sid e ra b le , the usual
p ra c tic e is to e stim ate the volume of each tre e individually through its
height o r log length. When th e re a re many tr e e s , o r the v ariatio n in
height w ithin a d .b .h .o .b . c la s s is lim ited , the n o rm al p ra c tic e is to
com pile the stand height cu rv e, read from the volume table the volume
co rresp onding to the tre e of average d .b .h .o .b . and height for each
c la s s , m ultiply this average volum e by the c la s s frequency, and sum
th ese c la s s volum es. Som etim es the c la s s av erag e volum es a re plotted
ag ain st d. b. h. o. b. o r b . a . o . b . and sm oothed g rap h ically b efo re being
m ultiplied by c la s s frequency.
F re e s e (1960) has outlined a m ethod of te stin g the 'a c c u ra c y ' of a
volum e table in application.
A pplicability
T re e s of the sam e d ia m e ter and height differ in volume because of
d ifferen c es in shape and ta p e r. The logical approach to tre e volume
table com pilation and application is to e sta b lish the extent of such
d ifferen ces and the p a ttern of th e ir c o rre la tio n with sp ec ie s, provenance,
and environm ent; to com pile tab les fo r stands of p a rtic u la r
c h a ra c te ris tic s to which the tab les can be indexed, and which have a
specified e r r o r of e stim ate; from a range of such ta b le s, to sele c t one
su itab le to re q u ire m e n ts. T here is an unfortunate tendency for a
volume table to be com piled fo r a fo re s t m erely because it has an
entity, irre s p e c tiv e of w hether the re a so n s fo r this a re geographical,
p o litical, or a d m in istra tiv e . This leads to quite an irra tio n a l
p ro liferatio n of volume ta b le s. S purr (1952) has com m ented on the
position in the U. S. A. It is hoped that a s im ila r condition which ap p ears
to be developing in A u stralia can be avoided.
M alform ed tre e s
In coniferous plantations, stem s which a re divided from ground to
tip . usually called 'd o u b le -le a d e rs' o r 'm u ltip le -le a d e rs ', com m only
o ccu r for various re a so n s. U sually the volume of such a tr e e is derived
112 O utline of F o r e s t M en su ratio n
by applying an a p p r o p r ia te c o r r e c t io n to the volum e r e a d fro m a volum e
ta b le for a n o r m a l t r e e of the s a m e d . b . h . o . b , and to tal height.
Depending on the s iz e of the t r e e , the n u m b e r of l e a d e r s , and the lim it
of u tilis a tio n , so the volum e of such a t r e e is m o r e o r l e s s than that
of a n o rm a l t r e e . Duff (1950) h a s developed c o r r e c t i o n f a c to r s for
r a d ia ta pine in New Zealand which seem to apply quite s a tis f a c to r ily
to plantation c o n ife rs in A u s tr a lia .
Volume T a b le s with M ore than Two Independent V a ria b le s

F r o m the p r e v io u s d is c u s s io n it w ill be a p p a re n t th a t, to im p ro v e the
e s tim a te of t r e e volum e given by a tw o-w ay volum e ta b le , an e x p r e s s io n
of t r e e shape o r ta p e r m u st be included a s a f u r th e r independent
v a r ia b le . If this e x p r e s s io n is to be effective as an independent v a r ia b le ,
v a r ia tio n in volum e m u s t be c o r r e l a t e d with v a r ia tio n in the valu es of
the e x p r e s s io n , and the e x p r e s s io n m u s t be re a d ily identifiable and
capable of r e li a b le m e a s u r e m e n t on standing t r e e s . The difficulty is
to find an e x p r e s s io n which s a ti s f i e s th e s e r e q u ir e m e n t s . T his m ight
be f o r e c a s t fro m a c o n s id e r a tio n of the v a r ia tio n s in the ch anges in
ta p e r o v er the length of individual t r e e s , p a r t i c u l a r l y am ong t r e e s of
the s a m e s p e c ie s o th e rw is e identical in d . b . h . o . b . and height.
T ow ards the end of l a s t c e n tu ry , the concept of fo rm quotient was
introduced in w e s te r n E u ro p e. Defined in d iffere n t w ays o v er the
y e a r s it is now a lm o s t u n iv e rs a lly accep ted as the r a t i o , u su ally in
the fo rm of a d e c im a l, for e x am p le -65. of the d ia m e te r at half the
total height above b r e a s t height to the d ia m e te r at b r e a s t height, the
d ia m e te r s being v a r io u s ly o v e r - o r u n d e r - b a r k . T r e e s having the
s a m e form quotient a r e grouped into a fo rm c l a s s in d icated by the
c o r r e s p o n d in g whole n u m b e r, 65 in this c a s e . The concept can only
be applied effectiv ely to t r e e s w hich have a single stem fro m ground
to tip. J o n s o n 's s te m p ro file equation (see ch. 6) in c o rp o r a te d
c o n s ta n ts r e la te d to form quotient. Given fo rm quotient, re la tiv e
v a lu e s of d i a m e te r at p e r c e n tile points on the s te m could be obtained
fro m the equation; fo rm f a c to r could be obtained fro m the r e la tiv e
values of d ia m e te r ; and volum e could be obtained fro m d . b . h . o . b . ,
height and fo rm f a c to r . The Jo n so n ste m p ro file equation thus provided
both a t a p e r table and a volum e ta b le . Since d e te rm in in g fo rm quotient
by d ir e c t m e a s u r e m e n t involved c o n s id e r a b le p r a c t ic a l difficulty on a
standing t r e e , Jonson sought indexes to fo rm quotient. F o r s ta n d s , a
c o r r e l a ti o n betw een stand d e n s ity and a v e r a g e fo rm c l a s s f o r the stand
Stand Volume 113
w as developed. F o r individual t r e e s , a c o r r e l a ti o n w as found betw een
fo rm quotient and form p o in t. the re la tiv e height of the c e n t r e of wind
p r e s s u r e on the crow n on the b a s is of M e tz g e r 's th e o ry of t r e e shape,
the position of the fo rm point being e s s e n tia lly d e te r m in e d by eye. The
e n o r m o u s amount of r e s e a r c h into th is m a t t e r in E urope and N orth
A m e r ic a w hich J o n s o n 's w ork s tim u la te d h a s a lre a d y been r e f e r r e d to.
I n te r e s t in the concept has p e r s i s t e d to so m e extent but its application
is lim ite d b e c a u se of the p r a c tic a l difficulty of d e te rm in in g fo rm
quotient d ir e c tly , and the u n reliab ility of e s tim a tin g it in d ir e c tly fo r
an individual t r e e through a subjectively e s tim a te d fo rm point, and for
s ta n d s th ro u g h such indexes as stand density.
R e c e n t in t e r e s t h a s been c o n c e n tra te d on e x p r e s s io n s of t a p e r , o r
indexes c o r r e l a t e d with t a p e r , which a r e m o r e p r a c t ic a l than fo rm
quotient. S p u rr (1952) d is c u s s e s the v a r io u s r a ti o s su g g ested fo r use
in the U. S. A. , p a r ti c u la r ly 'G i r a r d fo rm c l a s s ' . V olume ta b le s have
been developed in Sweden b ased on the r a tio of the d ia m e t e r s at 1 m e t r e
above b r e a s t height and a t b r e a s t height, and also b a s e d on the height
above ground of the low er lim it of the g re e n crow n a s an index of ta p e r .
I lv e s s a lo (1947) has com piled ta b le s using a s a m e a s u r e of ta p e r the
d iffe re n c e betw een the d ia m e te r s at 6 m e t r e s above ground and at
b r e a s t h eight, using pole c a l i p e r s for the f o r m e r m e a s u r e m e n t .
S im ila r ly , C r o m e r . M cIn ty re, and Lew is (1955) in the G e n e ra l Volume
T able fo r P . r a d ia ta u s e the d iffe re n c e betw een the d i a m e te r s
u n d e r - b a r k a t 15 ft and 5 ft above ground le v e l, the 15 ft point being
r e a c h e d by la d d e r. This ta b le in c o r p o r a te s a m e a s u r e of b a r k th ic k n e s s
as a f u r th e r independent v a r ia b le to p ro v id e e s t i m a t e s of volum e
u n d e r - b a r k . It was found to have a b ia s in c e r t a in s e c tio n s when applied
to P . r a d ia ta in South A u s tr a lia and a R egional Volum e Table for use
in South A u s tr a lia w as co m piled (Lewis and M cIn ty re. 1963). During
the com p ilatio n of the G e n e ra l Volume T able it had been noted th a t,
f o r the s a m e t r e e height and d . b . h . o . b . , d if f e r e n c e s in b a r k th ic k n e s s
had on the a v e r a g e m o r e effect than ta p e r d iffe re n c e on the e s tim a te
of u n d e r - b a r k volum e, and th a t u sing u n d e r - b a r k ta p e r betw een 5 ft
and 25 ft heig h t offered so m e gain in p r e c is io n of volum e e s tim a te
o v e r u se of 5 to 15 ft ta p e r . Two R egional Volume T a b le s w e re
c o m p iled of the s a m e kind a s the G e n e ra l Volume T able but using as
e n t r i e s t r e e height, d ia m e te r u n d e r - b a r k at 5 ft, and t a p e r u n d e r - b a r k
fro m 5 to 15 ft. and fro m 5 to 25 ft r e s p e c tiv e ly . The r e a s o n fo r using
d . u . b . a t 5 ft a s an e n tr y to the ta b le r a t h e r than a conventional b r e a s t
height m e a s u r e m e n t w as th a t t h e r e w as a la rg e n u m b e r of s a m p le t r e e s
with u n d e r-b a rk m e asu re m e n t at 5 ft for the sectio n al m ethod used
for tre e volume d eriv atio n , but v ery few tre e s for which d . u . b . at
b re a s t height had been m easu red .
Com pilation
In the stem p rofile equation approach, the volume table is in effect
com piled at the sam e tim e a s the stem p ro file equation because the
form quotient indicates the re la tiv e d ia m e te rs, th ese indicate the form
fa c to r, and th is has only to be m ultiplied by the b . a . o . b . and height of
the tre e for volume to be derived. H ow ever, form quotient can be used,
as can any other m e asu re of ta p e r re fe rre d to above, d ire c tly as an
independent v a ria b le , quite independently of any stem p ro file equation,
in a suitable equation re la tin g volume to d . b . h . o . b . , height, and form
quotient. The m ost com m only used equations a re those of the
a rith m etic type like the A u stralian equation and com bined v ariab le
equation with the e x p ressio n of form included. L ogarithm ic types
have also been trie d . B ecause of the m agnitude of the calcu latio n s,
the use of high speed com puters is advantageous.
A pplication
The inclusion of a m easu re of ta p e r as a v ariab le in the volume
table is intended to im prove the prec ision of e stim ate of volume of a
tre e . The extent to which it does so depends m ainly on how well the
m easu re of ta p e r used d e sc rib e s the effect of the o v erall ta p e r of the
stem on the volume of the stem . F or p ra c tic a l p u rp o ses, the m e asu re
of ta p er used is a sim ple one of ra tio o r difference of d ia m e te rs at
two places reasonably close to the ground. T ree boles a re for the
m ost p art irre g u la r and the low er p a rt of the bole is subject to
b u tt-sw ell. Such a sim ple m e asu re of ta p e r may o r m ay not re la te to
the effect of the o v erall ta p e r on the volume of the tr e e , and the
estim ate of the volume of an individual tre e may o r m ay not be an
im provem ent over that from a tw o-w ay volume table - gen eral
experience is that it is likely to be an im provem ent. How'ever, g en eral
experience also indicates that it is difficult to im prove the p re c isio n of
e stim ate for a stand over that from a tw o-w ay table by assu m in g an
average ta p e r fo r the stand from sam pling o r assum ing a p a rtic u la r
tre n d of ta p e r over d . b . h . o . b . c la s s e s . This method frequently leads
to biased e stim a te s.
The determ ination of an e x p re ssio n of ta p e r which does re la te to the
Stand Volume 115
effect of the o v erall ta p e r on the volum e of a tre e and which can be
m e asu re d re a d ily is one of the m ost p re ssin g m en su ratio n p ro b lem s.
STAND VOLUME TABLES
The concept of deriving stand volum e d ire c tly from stand v a ria b le s ,
in stead of through individual tre e s as in sam ple tre e m ethods and tre e
volum e ta b le s, has been ap p reciated and applied in w e ste rn Europe for
m o re than a century through the use of stand form fa c to r. This is an
a b s tra c t value, not capable of d ire c t m e asu rem en t and analogous to
tre e form fa c to r, re p re se n tin g a c o rre c tio n fa c to r to red u ce the product
of stand b a sa l a re a and an e x p re ssio n of stand height to stand volum e.
The stand form facto r for a p a rtic u la r stand h as to be determ in ed
in d irec tly from c o rre la tio n s of stand form fa c to rs with indexes of
stand s tru c tu re and stand density. E stim a te s of stand volume by
p h o to g ram m etric determ in atio n s of stand b a sa l a re a and height on
a e ria l photographs, and c o rre la tio n s of stand form facto r with stand
density, w ere made in w e ste rn Europe and N orth A m erica over th irty
y e a rs ago.
R ecently attem pts have been made to com pile stand volume tab les
by estab lish in g the functional re la tio n sh ip s betw een stand volume and
v a ria b le s such as stand b a sa l a re a and stand m ean height, analogous
to the re la tio n sh ip s between tre e volum e and tre e v a ria b le s such as
b asal a re a and height, by g raphical solution o r calculating ap p ro p riate
equations. Such a stand volum e table has been found to be v ery useful
fo r estim atin g stand volume from the ground, p a rtic u la rly w here b asal
a re a e stim a te s can be made re a d ily by angle count m ethod. Stand
volume ta b le s, based on in d irec t e stim a te s of stand b asal a re a through
e stim a te s of such things as crow n density, a re also v ery useful for
estim atin g stand volum es from a ir photographs. P robably the m ost
com prehensive study of the re la tiv e efficiency of stand volume tab les
com piled by v ario u s m ethods and incorporatin g v ario u s stand v a ria b les
is that m ade by Spurr (1952). He found th at o v er a wide range of
stands - even-aged and uneven-aged, of co n ifers o r b ro ad -leav ed
sp ec ie s, pu re or m ixed, good site s o r poor s ite s , heavy o r light
stocking - e stim a te s of stand volum e could be made quite sa tisfa c to rily
in general by a re g re s s io n such as the A u stralian equation:
V - a + bB + cH + d B. H
w here V = to tal stem volum e p e r a c r e . B = basal a re a p e r a c re and
H = average height p e r a c re , the re g re s s io n s of volume on b asal a re a
in height c la s s e s being lin e a r and the constants and coefficients of
th ese re g re s s io n s being lin e a rly re la te d to height. He found the
combined v ariab le equation:
V = a + bB .H
was alm ost as s a tisfa c to ry . He s tre s s e d the likely c u rv ilin e a rity of
m erchantable volum e on b asal a re a and suggested deriving m erchantable
volume by applying ap p ro p riate conversion fa c to rs to e stim a te s of
total volume from a stand volum e table.
Stand volume table com pilation in A u stra lia has been alm o st confined
to even-aged stands of P . ra d ia ta . In 1953, a stand volume table fo r
P . ra d ia ta was com piled by graphical m ethods by the New South Wales
F o re s try C om m ission, using data from 556 sam ple plots in v ario u s
N. S. W. plantations, of volum e u n d e r-b a rk p e r a c re to 4 in d. u .b . ,
b asal a re a o v e r-b a rk p er a c re , and dom inant height (the av erag e
height of the ta lle s t 20 tr e e s p e r a c re ). T h ere was a fa irly gentle
c u rv ilin e a rity in the relatio n sh ip s of volume on b asal a re a fo r s e p a ra te
height c la s s e s over the whole range of the data but lin e a rity o v er the
m ain cen tral p a rt. C rom er and C arro n (1956) com piled a stand volume
table for P . ra d ia ta for the Kowen a re a in the A. C. T. on the b a sis of
the A ustralian equation, p re lim in a ry investigation having shown th at
lin e a r re g re s s io n s of volume on b asal a re a in height c la s s e s w ere a
sa tisfa c to ry fit and th at the coefficients and constants of th e se
re g re ss io n s showed a strong lin e a r tre n d with height c la s s . The
com bined v aria b le equation was alm o st as sa tisfa c to ry . C ro m e r and
Brown (1956) com piled sim ila r stand volum e ta b le s for o th er a re a s of
P . ra d ia ta in the A . C . T .
If volume p e r a c re is e x p re ssed as the product of b asal a re a p er
a c re , stand m ean height and stand form facto r:
V - B. H. F.
then volume p er unit basal a re a is given as
TB-= H .F
This relatio n sh ip has been investigated for v ario u s sp ecies in v ario u s
c o u n trie s. Krenn (quoted by P ro d an , 1965b) developed m ean tre e
ta riffs , which, in the p re se n t context, a re stand volume ta b le s from
relatio n sh ip s of form height with stand m ean height. Form height for
a stand being determ ined from stand m ean height, its product with
m ean b a sa l a re a gives m ean tr e e volume w hich, when m ultiplied by
num ber of tr e e s , gives stand volum e. Lew is (1954) com piled a
Stand Volume 117
g raphical relatio n sh ip oi to tal volume p e r unit b asal a re a on stand top
height fo r unthinned P . ra d ia ta in New Z ealand which was alm o st lin e a r
and form ed the b a sis of a v a riab le density yield tab le from which p re se n t
and future volume could be d erived. A v a ria b le density yield table fo r
P . ra d ia ta in the A. C. T. was com piled by the sam e method (C arron,
1967). B edell and B e rry (1955) estab lish ed a lin e a r relatio n sh ip : ^
B
a + bH w here V = volume from a 12 in stum p to a 3 in top d iam eter
for all tr e e s 4 in d .b .h .o .b . and o v e r, and H= m ean stand height, for
a v a riety of Canadian coniferous and b ro a d -le a v ed sp ecies com bined.
C ro m er (1961) c a rrie d out an extensive investigation fo r P . ra d ia ta
from se v e ra l p a rts of A u stra lia . He showed th at to tal volum e u n d e r-b a rk
p e r square foot of b asal a re a o v e r-b a rk b o re a lin e a r re la tio n sh ip to
stand top height, beyond a stand top height of about 40 ft, of the form
— = a + bH in which the value of a could for all p ra c tic a l pu rp o ses be
B
taken a s z ero and in which b was approxim ately the sam e over a wide
range of lo cality , age, site quality, and density. He suggested that
e stim a te s of m erchantable volum e be made from e stim a te s of total
volume by the use of c o rre c tio n fa c to rs re la te d to stand m ean d .b .h .o .b .
Law rence (1963) has d iscu ssed the com pilation of stand volum e tab les
by m ultiple re g re s s io n m ethods for Eucalyptus obliqua and E. regnans
in T asm ania.
Application
A stand volum e table is a convenient and p ra c tica b le tool fo r re s o u rc e
inventory and m ay provide sufficiently re lia b le e stim a te s for periodic
m anagem ent inventory under c e rta in c irc u m sta n c e s. Where the reg im e
of tre a tm e n t is reasonably stable and future stand b a sa l a re a and height
can be fo re c a s t with som e re lia b ility , futu re stand volum e can be
fo rec ast through stand volume ta b le s. The lim its of application of
stand volum e tab les to single sm all stands should be reco g n ised .
Ju st as a tr e e volum e table should not be expected to give re lia b le
re s u lts for an individual tr e e , so a stand volum e tab le should not be
expected to give re lia b le re s u lts fo r an individual sm all stand.
YIELD TABLES
A yield tab le is a statem en t of v a ria b le s such as volum e, b asal a re a ,

height, and num ber of tre e s p e r unit land a re a for stands of v ario u s
age, on site s of v ario u s productive capacity, of an even-aged fo re st
of one species which has developed n atu rally , o r has been m aintained
a rtific ia lly , at a c o n sisten t density. It is a quantitative ex p ressio n of
the p attern of growth of a p a rtic u la r crop on a p a rtic u la r site . It is
com piled from re la tio n sh ip s, estab lish ed g rap h ically o r b 3r calculation,
betw een the stand v a ria b le s as dependent v a ria b le s and age and an
ex p ressio n of the productive capacity of the site as independent
v a ria b le s. It re p re s e n ts a h isto ry of developm ent of an even-aged
fo re s t and for th at re a so n is d iscu ssed in d etail in C hapter 17. It is
included h e re as a m eans of estim atin g stand volume because it is a
kind of stand volum e table in which the independent v a ria b le s a re age
and site . Once the mapping of site quality has been c a rrie d out,
e stim a te s of stand volume at any age, p re se n t o r fu tu re, can be made
without any fu rth e r field w ork. If e stim a te s of volume a re re q u ire d
fo r stands of a density other than th at on which the table is b ased ,
c o rre c tio n s to table volum es have to be m ade through som e m e asu re
of stand density. B e tte r re s u lts may be obtained from a yield table
which in c o rp o ra te s a th ird independent v aria b le ex p ressin g stand
density. The nam e v a riab le density yield table has been given to this
in N orth A m erica, w here its com pilation and application have receiv ed
p a rtic u la r attention (S purr, 1952; H usch. 1963).
Yield tables have been used fo r e stim a te s of stand volume in re s o u rc e
inventory in G reat B rita in , West G erm any, and the N etherlands and
also in m anagem ent inventory in som e in stan ces in w e ste rn Europe.
N. B. Lewis (1957) d is c u sse s the use of a yield table for inventory of
ra d ia ta pine in South A u stra lia.
COMPARISON OF METHODS
Sample tre e m ethods can be applied s a tisfa c to rily only to even-aged

stands. T ree volum e tab les m ay be applied to both even-aged and
uneven-aged stan d s. Stand volum e tab les may be applied to both
even-aged and uneven-aged stands; in p ra c tic e , they a re applied m ost
often to the fo rm e r. Yield ta b le s a re applicable only to even-aged
stands.
Stand volume is determ ined from sam ple tre e m ethods and tre e
volume tables by applying them to individual tr e e s , d ire c tly o r
in d irec tly , usually within sam pling units. Stand volume ta b le s and
Stand Volume 119
yield tab les a re applied d ire c tly to stands, usually through sam pling
u n its.
To com pile tre e volume ta b le s one re q u ire s selectio n and m easu rem en t
of volum e of sam ple tre e s ; fo r stand volum e tab les one re q u ire s
e stim a te s of stand volum e (from sam ple tre e m ethods o r tre e volume
ta b le s), stand b asal a re a , and stand height: for y ield tab les one re q u ire s
e stim a te s of stand volum e (from sam ple tre e m ethods, tre e volume
ta b le s o r stand volum e tables) at v ario u s ages on v ario u s site s
p e rio d ic ally during the developm ent of the crop.
To apply any m ethod, the a re a of the stand m ust be known. If the
e stim a te of stand volume is based on sam pling u n its, the a re a of each
sam pling unit m ust be known except when stand b a sa l a re a is estim ated
by the angle count method.
To apply sam ple tre e m ethods, the d .b .h .o .b . of each tre e in the
stand o r in each sam pling unit m ust be m easu red ; the height and ta p e r
of som e tre e s m ay also be needed; sam ple tre e s m u st then be selected
and th e ir volum es determ ined. To apply one-w ay tr e e volume ta b le s,
the d .b .h .o .b . of each tre e in the stand o r in each sam pling unit m ust
be m easured: for ta riff s , the stand v a ria b le by which the ta riffs a re
indexed m ust be m e asu re d . To apply tw o-w ay tr e e volum e ta b le s, the
d . b .h .o .b . of every tre e m ust be m e asu re d , and e ith e r the height of
ev ery tre e m ust be m easu red o r a stand height curve com piled from
sam ple tr e e s . To apply th re e -w a y volume ta b le s, the d .b .h .o .b . of
ev ery tre e m ust be m e asu re d , e ith e r the height of ev ery tre e m ust be
m e asu re d o r a stand height curve m ust be com piled from sam ple tre e s
and eith e r the expression of ta p e r or form of ev ery tre e m ust be
m e asu re d o r an average determ ined from sam ple tr e e s . To apply a
stand volum e tab le, e stim a te s of stand b asal a re a and stand height a re
re q u ire d on a unit a re a b a s is . To apply yield ta b le s, the age and site
quality (site index) of the stand m ust be known: also the index of density
fo r a v a riab le density yield table.
V ery fewr investigations have been m ade into the re la tiv e re lia b ility
of e stim a te s of the volume of a p a rtic u la r stand by th e se s e v e ra l m ethods.
One of local in te re s t is that of E. R. Lewis (1957) who considered
re la tiv e e r r o r s in vario u s m ethods used by the New Zealand F o re s t
Service in pure even-aged coniferous plantations. F u rth e r com m ents
on the m ethods a re contained in a review by the N. Z. F. R. I. (1965).
STAND VOLUME BY ASSORTMENTS: TAPER TABLES
The m ain stem of a b ro ad -leav ed tre e is co n sid ered to end at the b reak
of crow n. F o r utilisatio n the stem is topped th e re and so volume is
usually estim ate d to th at level. R elatio n sh ip s, so m etim es com plex,
between this 'm e rc h a n ta b le ’ volume and o ther tre e v a ria b les can be
estab lish ed . F o r c o n ife rs, the m ain stem is co n sid ered to extend to
the tip. R elationships between th is to tal volume and other tre e
v a ria b le s can be estab lish ed re a d ily and a re usually sim ple. Total
volume is , how ever, u n re a listic from a u tilisatio n point of view since
conifer stem s a re in p ra c tic e u tilised only up to c e rta in d iam eter lim its.
Since it is difficult to re la te th is ’m erch an tab le' volum e to other tre e
v a ria b le s, as for b ro ad -leav ed sp ec ie s, e stim a te s of m erchantable
volume of co n ifers a re often made by applying c o rre c tio n fa c to rs to
e stim a te s of total volum e. It is found, for exam ple, fo r P . ra d ia ta
in an even-aged stand, that the volume of an individual tre e to a
p a rtic u la r d ia m e ter u n d e r-b a rk lim it can be e x p re ssed as a percentage
of total volum e, the p ercentage varying with tre e d . b . h . o . b . , but not
with height within d . b . h . o . b . , as illu s tra te d in Table 13:2 . It is also
found fo r P . ra d ia ta that the volum e of a stand of tre e s to some upper
d ia m e ter lim it can be e x p re ssed as a p ercen tag e of the total volume
of the stand, this p ercen tag e varying with stand m ean d iam eter
(E. R. L ew is, 1954; C ro m e r, 1961).
T able 13:2
V olum e under-bark from ground to 4 in d . u . b . as a p ercentage of to ta l v olu m e under-bark

for 200 sam ple trees of P. radiata, Kowen Forest, A . C . T .
D.b.h.o.b. T otal h eigh t class (ft) D.b.h.o.b.

class class
(in) 30 40 50 60 70 80 90 average
5 64 65 65
6 77 75 77 70 76
7 87 85 86 85 85
8 90 90 91 90 90
9 92 93 94 94 95 93
10 95 96 96 96 96
11 95 96 96 96 97 96
12 97 98 98 97 98 97
13 97 98 98 98 98
14 99 99
15 99 99 99 99
16 99 99
Stand Volume 121
E stim a te s of stand volume a re of g re a te r value to m anagem ent if
m ade in te rm s of lengths of logs and th e ir d ia m e te rs u n d e r-b a rk at
the sm a ll end of the log. These e stim a te s for a stand may be provided
by a com bination of a stand ta b le , a stand height cu rv e, and a ta p e r
ta b le . A ta p e r table is a ta b u la r statem en t which shows fo r a p a rtic u la r
s p e c ie s, and usually under p a rtic u la r silv ic u ltu ra l conditions or for a
p a rtic u la r locality, the o v e r-b a rk or u n d e r-b a rk d ia m e ter at specified
heights from ground fo r tr e e s of specified d .b .h .o .b . , to tal height, and
p erh ap s form or ta p e r. F or som e p u rp o se s, the height from ground
to a specified d ia m e ter is given. Since the p ro file of the tr e e is defined
by the ta p e r ta b le , so the ta p e r table defines the volum e of the tr e e ,
and ta b le s of volume to vario u s upper d ia m e ter lim its a re usually
com piled in asso ciatio n with ta p e r ta b le s, exem plified by the com bined
ta p e r and volume tab les fo r vario u s exotic co n ifers in New Zealand by
Duff (1954), Duff and B u rstall (1955), and B u rsta ll (1957). T hese
ta b le s w ere com piled by graphical m ethods from data of sam ple tr e e s .
The m eans of u n d e r-b a rk d ia m e te rs at v ario u s specific heights from
ground lev el, within total height c la s s w ithin d. b. h. o . b. c la s s , w ere
plotted against height from ground level on re c ta n g u la r co -o rd in ate
p a p er and tr e e p ro files draw n by freehand cu rv es h arm o n ised within
and betw een d .b . h. o .b . c la s s e s . Two-way tre e volume ta b le s w ere
then com piled from these g e n eralise d 't r e e s '. In th is m ethod of
com piling ta p e r ta b le s, v aria tio n s in ta p e r within height and d .b .h .o .b .
c la s s e s a re sm oothed out by averaging d ia m e te rs along the tre e p ro file
w ithin c la s s e s , and v a riatio n s of th ese av erag e s betw een c la s s e s a re
sm oothed out by harm onising the p ro files betw een c la s s e s .
The tab les of Jonson re fe rre d to previou sly give d ia m e te r at
p e rc en tile points along the s te m , as a percen tag e of d ia m e ter at b re a s t
height, as derived from the stem p ro file equation for v ario u s values
of form quotient. Once such a stem profile equation is estab lish ed ,
the a sso c iate d ta p e r table follows sim ply by substitution of ap p ro p riate
v alu es of the index of form in the equation. T ap er ta b le s based on stem
p ro file equations with form quotient a s such an index w ere com piled in
the 1920s and 1930s, when Jo n so n 's w ork had a ttra c te d in tern atio n al
atten tion, by vario u s w o rk ers such as W right (1927) in Canada and
B ehre (1927) in the U. S. A.
T ap er tab les fo r plantation c o n ifers in A u stra lia have been com piled
by se v e ra l A u stralian fo re s t se rv ic e s but not published. C arro n and
Jaco b s (1964) com piled a ta p e r table fo r P . rad iata which is b ased on
G ra y 's hypothesis th at the m ain stem is a paraboloid of the second
d e g re e (see ch. 8) and w hich is ap p lied to an e v en -a g ed stan d th ro u g h
ta r if f s com piled by C ro m e r and C a rro n (1957). U sing th is ta p e r ta b le
C a rro n (1964), follow ing a p ro c e d u re developed by W a rre n (1959) fo r
P . ra d ia ta in New Z ea lan d , co m p iled a re la tio n s h ip fo r ev en -a g ed
sta n d s of P . ra d ia ta w h ereb y th e p e rc e n ta g e of stan d volum e w ithin
v a rio u s log s m a ll end d ia m e te r c la s s e s can be d e te rm in e d fro m stan d
m ean d .b .h . o. b.
14 Bole Area
The bole a re a of a tre e is its p e rip h e ra l su rface u n d e r-b a rk . F or

p ra c tic a l m ensuration p u rp o se s, the in crem en t of wood volume over
a p erio d can be considered as a la y er placed on th is su rfa c e , a sheath
of wood with an a re a equal to the su rface a re a of the tre e at the
beginning of the period and a thickness which m ay v a ry over the length
and around the c irc u m fe re n ce of the bole depending on w here and to
what extent growth tak es place. The bole a re a of a tre e can be reg ard ed
as the base for potential in c re a se in the volume of the tr e e , so th at the
sum of the bole a re a s of the tre e s in a stand re p re s e n ts the base for
potential in c re a se in the volume of the stand. Since the wood producing
capacity of a stand is obviously im portant to silv ic u ltu re and
m anagem ent, tre e and stand bole a re a a re v a ria b le s of p a rtic u la r
in te re s t. Surprisingly , serio u s attention has been given to the usefulness
and estim ation of th is stand v a ria b le only during re c e n t y e a rs and then
only in a lim ited way.
The f ir s t re fe re n c e in English a p p ea rs to be th at by Lexen (1943) who
pointed out that it was then c u sto m ary in the U. S. A. to d e sc rib e the
in ten sity of selectiv e cutting by the volum e of wood left in the stand and
th a t, while this provided a m e a su re of the wood cap ital left, it did not
indicate the growth potential of the c u t-o v e r stand. He suggested that
th is shortcom ing of volum e could la rg e ly be overcom e if it w ere
supplem ented by an e stim ate of bole a re a and illu s tra te d th is by
n u m erical exam ples. He suggested calculatin g the bole a re a of an
individual tr e e , by analogy with the calculation of volume, by
substituting g irth fo r sectional a re a in the Huber o r Sm alian form ula:
o r by plotting girth at sev e ra l positions along the stem on re c tan g u la r
co -o rd in ate paper and com puting the a re a below a cu rv e fitted to the
plotted points. He com piled a bole a re a table fo r individual tr e e s , in
the form of an e x p ressio n involving d .b .h .o .b . and height for Pinus
123
ponderosa for purposes of illu s tra tio n , and suggested th at an approxim ate
estim ate of bole a re a in square feet of a P . ponderosa could be obtained
by dividing 7- 0 into the product of d .b .h .o .b . (in) and height (ft). (This
fa c to r 7 '0 contains an elem ent analogous to form fa c to r used in reducing
the product of tre e basal a re a and height to tre e volum e. H ere the
value 7- 0 in co rp o rates tt to co n v ert d ia m e te r to g irth , 12 to convert
g irth in inches to feet, and a fa c to r to red u ce the product of tre e g irth
and height to tre e su rfa ce a r e a . ) F ro m th is the bole a re a of a stand
could be derived, e ith e r as the sum of the bole a re a s of individual tre e s
so calculated , or through the bole a re a s of average tr e e s by d iam eter
c la s s e s , o r as the product of the bole a re a of a ’m ean' tre e and the
num ber of tre e s in the stand.
Mulloy (1944) com pared, for a num ber of plots of re d and white pine,
the stand bole a re a calculated as the product of num ber of tre e s and
the bole a re a for a tre e of m ean d ia m e te r and height, with values of
R ein ek e 's stand density index (see ch. 16) derived for the plots and
concluded that 'fo r all p ra c tic a l p u rp o ses they provide the sam e relativ e
m e asu re of density index'. L escaffette (1952) d iscu ssed the concept in
general and the deriv atio n of values of 'su rfa c e g e n e ra tric e ' ap p ro p riate
to a 'ta r if de cubage'. Hummel (1953) co n sid ered bole a re a as an index
of stand dens it y and drew attention to the fact th at 'exam ination of
se v e ra l B ritish and c e n tra l European yield tab les su g g ests th at, above
a c e rta in age, the bole a re a re m a in s m o re o r le ss constant until the
lim it of height grow th is reached and then the bole a re a s ta r ts to
d e c re a se . . . not only fo r stands thinned lightly but also stands
thinned heavily provided the thinning reg im e is c o n siste n t'. However,
he re je cte d bole a re a as an index of stand density, reg ard in g it as
'cum bersom e to ap ply'. B riegleb (1952) also co n sid ered bole a re a as
an index of stand density in com paring stands of Douglas f ir , but found
it unsuitable. Hiley and L ehtpere (1955) used an e x p ressio n of bole
a re a to illu s tra te how coniferous plantations in G reat B ritain might
be thinned to produce a specific num ber of rin g s p er inch. Anuchin
(1962) dem o n strated th at, for a wide range of sp ecies in the U. S. S.R . ,
stand bole a re a a p p ea rs constant w ithin a site quality o v er a wide range
of age. On th is b a s is , he developed from yield tab les for pine stands
a relatio n sh ip w hereby volum e in c re m en t of a stand could be fo re c a st
from a knowledge of site quality, av erag e ra d ia l in crem en t o v er the
previous ten y e a rs , and a m e asu re of stand density.
Table 14:1 is a tr e e bole a re a table com piled by the w rite r from 229
sam ple tre e s of P . ra d ia ta from U ria rra F o re s t. A. C. T. , which w ere
Bole A rea 125
T a b le 14:1
T re e b o le a re a ta b l e for P. r a d i a ta . U riarra Forest, A . C . T . , all ages and sites, based on

229 s a m p le trees; v alues in top le f t c o m e r of c e l l re p resen t bo le a re a (sq ft); v alu es in
b o tt o m r i g h t c o m e r r e p r e s e n t factors w h ereb y b o le are a (sq ft) m a y be d e riv e d by dividing
th e f a c to r into th e p ro d u ct of d . b . h . o . b . (in) and to t a l h e ig h t (ft)
D .b .h . o .b . T o ta l h e ig h t (ft)
(in) 40 50 60 70 80
37 46 54
6
6 -5 6 -6 6 -6
42 52 62
7
6-7 6 -8 6 -8
47 58 70 82
8
6-9 6-9 6 -9 6 -8
51 64 77 91
9
7-0 7 -0 7 -0 6 -9
56 70 84 100 115
10
7 -1 7- 1 7- 1 7-0 6-9
76 92 108 125
11
7 -2 7 -2 7- 1 7- 1
82 99 117 136
12
7- 3 7 -2 7-2 7- 1
126 146
13
7 -2 7* 1
135 156
14
7- 3 7 -2
sele c ted o v er the range of age and s ite , and m e asu re d by the sectional
m ethod, fo r com pilation of a tre e volume tab le. The bole a re a of each
tre e has been calculated as an approxim ation in the way suggested by
Lexen (1943) and the table com piled by the h arm o n ised re g re ss io n
s e rie s m ethod as fo r a tw o-w ay tr e e volume table (see ch. 13). L exen's
'fa c to r' fo r each d . b . h . o . b . -height c la s s is also shown. T here a p p ears
to be a c o rre la tio n of fa c to r with d . b . h . o . b . c la ss but the values a re
reaso n ab ly constant over height c la s s e s and the o v erall v ariatio n is
so re la tiv e ly sm all that a useful approxim ation to the bole a re a (in
sq u are feet) of th ese and s im ila r stands m ight be obtained sim ply by
dividing the product of stand d . b . h . o . b . (in), stand m ean height (ft),
and num ber of tr e e s by 7-0, the sam e fa c to r as Lexen found for P .
ponderosa.
15 Site Quality
A f o r e s te r has a broad silv ic u ltu ra l in te re s t in the productivity of

land - in the a lte rn ativ e kinds of fo re s t an a re a w ill support and in the
fa c to rs which d eterm in e the o c cu rren c e of a p a rtic u la r kind of fo re s t
on an a re a . M ensurational in te re s t in site is much n a rro w e r. The
m ain concern of a fo r e s te r, in a stand m anaged fo r production of wood,
is in its volum e, p re sen t and fu tu re. B ecause the volume of wood in
a stand at any tim e is la rg e lj’ a reflectio n of the site it occupies,
estim atio n of volume is a s s is te d if the site can be c la ssifie d as one or
an o th er site quality c l a s s . As volume is the c rite rio n of site quality,
it is the logical b a sis for the c lassific atio n . A ltern ativ ely , an index to
volum e production m ight be used - a facto r of the environm ent o r a
s ta tis tic of the crop.
Since age has an im portant influence on production, it is convenient
to c o n sid er even-aged fo re s ts of a single sp ecies se p a ra te ly from
o th e rs.
Even-aged F o re s ts of One Species

T h e re a re many fa c to rs of the environm ent which influence the growth
of a tre e crop. They in te ra c t, and it is difficult to co n sid er any one
of them , and the relatio n sh ip betw een it and tre e grow th, in isolation.
F u rth e rm o re , th e re is a continuing reactio n betw een the fa c to rs of the
site and the stand as it develops. The environm ent of a fo re st m ight
be re g a rd e d quite sim ply as the soil and the clim ate of its atm o sp h ere.
C rite ria of clim ate a re used su ccessfu lly fo r broad qualitative
c la ssific a tio n s of fo re st but have not been found su itab le for quantitative
c lassific atio n of stands. Many of the fa c to rs of the environm ent a re
in teg rated in the so il, and because re fe re n c e to soil c h a ra c te ris tic s
is of obvious value for qualitative site c la ssific a tio n , co n sid erab le
127
attention h a s been given to estab lish in g re la tio n sh ip s between tre e
grow th and soil c h a ra c te ris tic s such as te x tu re , and depth and
a v ailab ility of w ater and n u trie n ts. In many c a s e s good c o rre la tio n s
have been obtained. T here is an extensive lite ra tu re on th is subject
which has been su m m arised by Rennie (1963). H ow ever, the use of
soil c h a ra c te ris tic s as an index of productivity for pure even-aged
fo re s ts has been inhibited by the w ork involved in estab lish in g the
im p o rtan t soil c h a ra c te ris tic s for a p a rtic u la r fo re s t, the com plex
n a tu re of the re la tio n sh ip s betw een grow th and soil c h a ra c te ris tic s ,
and the difficulty of evaluating the c h a ra c te ris tic s in a p a rtic u la r c a se .
Good c o rre la tio n s have been estab lish ed between soil c h a ra c te ris tic s
and s ta tis tic a l indexes of site (see p. 129). Work by Pegg (1967) on the
re la tio n of site index of plantations of Pinus ellio ttii in so u th -e ast
Q ueensland to so il, vegetation, and clim ate is of in te re s t.
A ground vegetation of m o s se s , g ra s s e s , fe rn s , flow ering p lan ts,
sh ru b s, and sm all tre e s often accom panies a m ain fo re st crop. It
m ight be expected that d ifferences in site quality would be re fle c ted in
the kind and am ount of th is le s s e r vegetation. If d ifferen ces in
productivity of the m ain cro p could be c o rre la te d with them , these
in d icato r plan ts could be used as the index for stratify in g the main
cro p into site quality c la s s e s . The f ir s t la rg e -s c a le exposition of th is
concept w as p re sen te d by C ajander (1926) fo r Finland. S tatistical
indexes of the m ain cro p such as stand volume and stand height had
been in use in Europe for som e tim e . C ajander conceded these w ere
effective fo r productivity c lassific atio n but he sought an index w hereby
not only could a fo re s t be c la ssifie d into productivity c la s s e s but at
the sam e tim e the land could be c lassifie d according to its potential
productivity under other sp e c ie s. He found th is index in the le s s e r
vegetation. Ilv essalo stim u lated fu rth e r in te re s t in th is sy stem by
w ork in the n o rth e rn h e m isp h ere , and th is use of in d icato r plants as
an index fo r quantitative site classific atio n has since been dem o n strated
in many p a rts of the w orld (Rennie, 1963). However, the system is not
in common use. Lindsay (1939) w as able to make a broad c lassific atio n
of productivity of stands of alpine ash (Eucalyptus gigantea) in southern
N. S. W. on the o c cu rren c e and density of the shrub V eronica d e rw e n tii.
L e s s e r vegetation is often used in A u stra lia as an indication of the
su itab ility of a c e rta in sp ecies fo r affo restin g o r re fo re stin g an a re a
and as a rough guide to the potential productivity of the sp ec ie s. Once
a coniferous plantation is estab lish ed , the re tu rn of the o rig in al le s s e r
vegetation is usually a rtific ia lly discouraged.
Site Q uality 129
The sim p le st, m ost effective way of c lassify in g stands of an
even-aged fo re st of one species into productivity c la s s e s is through a
re la tio n sh ip between s ta tis tic s of the m ain cro p and age over the range
of age and productivity. Two s ta tis tic s have been found useful - stand
volum e and stand height. The relatio n sh ip of each of th ese with age
for a site is a curve which is in general of a sigm oidal n a tu re . F or a
fo re st, the curves over the range of site a re usually anam orphic but
they m ay be polym orphic. F or th is re a so n , and b ecause the exact
n a tu re of any one curve can ra r e ly be determ in ed with any c e rta in ty ,
the c u rv e s a re usually draw n by hand ra th e r than the b e st fit calculated
fo r som e assum ed form . The way in which the cu rv es a re com piled is
s im ila r fo r volume and height and is illu s tra te d fo r height la te r.
Though stand volume is the c rite rio n of site productivity, it is
influenced by stand density and its use for site quality classific atio n is
re s tr ic te d to fully stocked (norm al) stan d s. It is used for site
c la ssific a tio n in some co u n tries of w estern Europe w here intensive
m anagem ent over a long period has en su red full stocking. F o r exam ple,
s ite s a re c la ssifie d according to m ean annual in crem en t of volum e at
100 y e a rs of age. 'Volume p e r a c re which is produced by stands fully
occupying the s ite ' (Lew is, 1957) is used in South A u stra lia fo r site
quality c lassificatio n of ra d ia ta pine plantatio n s, though the actual
m apping of site quality boundaries in the field is b ased on indexes to
volum e production (see p.133).
W here v ariatio n of density in stands of the sam e age on the sam e site
produces v a ria tio n in volum e, a b e tte r c la ssific a tio n of site m ay be
provided by an e x p ressio n of the heights of the ta lle s t tr e e s of the stand,
since th ese a re c o rre la te d with the productive capacity of the site but
a re not affected by varying density within the usual lim its of the range
of density found in stands in fo re s t conditions. V arious ex p re ssio n s
of the ta lle s t heights have been used for th is p u rp o se. F o r coniferous
plantations in A u stra lia stand top height (predom inant h eig h t) is
com m only used, being the average of the heights of the 20 to 30 ta lle s t
tr e e s p e r a c re . F or even-aged stands of eucal^ypts in T asm an ia, C ro m er
and Bowling (1961) used m ean height of dom inants, the av erag e of the
heights of the ta lle s t 12 tr e e s p e r a c re at the ra te of one tr e e p e r 1/12
a c r e . L aw rence (1966) also used th is as m ean dom inant h eig h t. In
G re at B rita in , top height is used, being the av erag e of the heights of
the 40 tr e e s of la rg e st girth p e r a c re (Johnston and B rad ley , 1963).
It is quoted as 'co rresp o n d in g conveniently with the In tern atio n al Union
of F o re s t R ese arch O rg an isatio n 's recom m endation of 100 p e r h e c ta re
which is a lread y in use in p a rts of E u ro p e'. Mean top height has
com m only been used in New Zealand for coniferous plantations, being
the height on a stand height curve equivalent to the m ean b asal a re a of
the 100 la rg e st d ia m e te r tr e e s p e r a c re . In N orth A m erica, the av erag e
of the heights of dom inants o r dom inants and codom inants is commonly
used (Spurr,- 1952; Husch, 1963). Spurr has d iscu ssed the disadvantages
of th is.
T here a re v ario u s ways of com piling the h eig h t-ag e c u rv e s, depending
on c irc u m sta n c e s. T here a re th re e likely c a s e s . A ssum e a coniferous
plantation in A u stra lia in which stands v a ry in age from 1 to 40 y e a rs .
T em p o rary sam pling units a re estab lish ed over the range of site and
age (down to say 8 y e a rs , since below that the fa c to rs of the site will
probably not have found tru e expression) and stand top height is
determ ined fo r each unit. Age and re la te d top height values a re plotted
on re c tan g u la r c o -o rd in a te p ap er giving a c o m e t-ta il s c a tte r of points.
F itting cu rv es to this s c a tte r is not sa tisfa c to ry because th e re is no
way of knowing the tren d of height with age. N e v e rth e le ss, height-age
c u rv e s may have to be draw n from this s o rt of data in the p re lim in a ry
stag es of m anagem ent, as in w estern Europe about tw o -th ird s of the
way through la s t century and in re c en t y e a rs in New Zealand (Lew is,
1954). If. on the other hand, perm an en t sam pling units had been
estab lish ed as the stands w ere planted and top height continually
m easu red , this would provide a com plete h eig h t-ag e stand h isto ry for
som e stands with v a rio u sly incom plete h isto ry for the o th e rs. The
tre n d of height-age would then be c le a r. Com plete h isto ry of th is kind
may also be provided from individual tre e s if they have annual rin g s
and it can be assum ed th at the p re se n t ta lle s t tr e e s have alw ays been
the ta lle s t. A ltern ativ ely , stem an aly sis inform ation can be used to
support periodic m e asu re m e n t. In the th ird and m ost common
c irc u m sta n c e , in term ed iate to the two e x tre m e s d iscu ssed above,
sam pling units a re estab lish ed over the range of age and site when the
oldest stand is say 30 y e a rs and top height is m easu red sev e ra l tim es
in 10 y e a rs . The re su ltin g 1 0 -y ear segm ents of the height-age h isto ry
of the stan d s, though not as good as com plete h isto ry , provide b e tte r
evidence than one m e asu rem en t on which to gauge the tre n d in
height-age over the com plete range of age. W hatever the d ata, c la s s e s
have to be form ed of an in te rv a l which depends on the num ber of site
quality c la s s e s re q u ire d , and on the range of height for the range of
site , usually the range at what is called stan d ard a g e . T his age has to
be such that each site has had opportunity to e x p re ss its e lf, yet not so
Site Quality 131
advanced that the available data w ill be too r e s tric te d , and such that
all s ite s a re re p re se n te d . If the c la s s e s a re defined bjTheight values
and can be thought of in te rm s of m ean annual in crem en t in height, it
is a rith m e tic a lly advantageous to have an age which is a sim ple facto r
of the height values: fo r exam ple, 20 or 25 y e a rs is commonly used in
coniferous plantations in A u stra lia . The age usually recom m ended is
about tw o -th ird s of rotation age. An age of 80-100 y e a rs is com m only
used in Europe. In som e c a s e s , the range of height at stan d ard age is
divided into the re q u ire d num ber of c la s s e s with an equal in te rv a l, the
c la s s e s being labelled with Roman n u m erals (I usually the b e st), and
re fe rre d to as site quality I, site quality II, and so on. In South
A u stra lia , fo r exam ple, seven such site qu alities a re recognised for
P. ra d ia ta (Lew is, 1963b), five of them being thinnable. A ltern ativ ely ,
in te rv a ls of som e specific width such as 10, 15, o r 20 feet a re
e sta b lish e d , the c la s s e s then being called site index c la s s e s . The
c la s s e s may have m id-points such as 60 feet, 70 feet, 80 feet, with
lim its 55-65, 65-75, 75-85 and called site index 60. site index 70, site
index 80; o r m id-points 65 feet, 75 fe e t. 85 feet with lim its 60-70, 70-80,
80-90 and called site index 65 and so on - except th at in Q ueensland
they a re called site index 6, site index 7. A stan d ard age of 20 y e a rs
h as been adopted for P. ra d ia ta in the A. C. T. (C arro n , 1955). F o r
plantation co n ifers in Q ueensland, a stan d ard age of 25 y e a rs has been
adopted (Queensland D epartm ent of F o re s try , p riv ate com m unication).
The m eans or lim its of the c la s s e s then form nom inated points on
the height axis of the c o -o rd in ate p ap er at stan d ard age through which
the cu rv es of height-age m ust p a s s , the tre n d of the cu rv es over the
range of age being d eterm ined from the height-age d ata. Two m ethods
a re com m only used to fit the cu rv es by hand. One m ethod is to draw
two cu rves based on the upper and low er lim its of the data resp ec tiv e ly
and then draw a fam ily of cu rv es in te rm e d ia te to th ese in position and
shape passing through the obligatory points on the height axis at
stan d ard age. The shape of the fam ily of cu rv es is thus determ in ed from
two lim iting cu rv es based on data which, re p re se n tin g only the b est
and w orst s ite s , a re likely to be le a s t in amount and low est in
re lia b ility , while the m ore num erous and m ore re lia b le data of the
m ost common s ite s a re ignored. G ray (1945) recom m ended this method
(which is often re fe rre d to as B a u r's m ethod, the lim iting curve m ethod,
o r the s trip m ethod and was used in w e ste rn Europe la s t century
(J e rra m , 1939)) as a stop-gap m ethod fo r coniferous plantations in
A u stra lia until som e stand h isto ry had been reco rd ed by periodic
m e a s u re m e n t. He su g g ested g ettin g s te m - a n a ly s is in fo rm atio n fro m
only the b e st and w o rst s ite s to sav e tim e and m oney sin c e any o th e r
in fo rm atio n would not in effect be u sed . In the o th e r m eth o d (often
r e f e r r e d to a s th e guiding cu rv e o r d ire c tin g c u rv e m e th o d ) an a v e ra g e
of the h eig h ts o v e r the ra n g e of s ite at v a rio u s ag es is p lo tte d , a
’m a s t e r ’ c u rv e fitte d to th e s e v a lu e s, and a fam ily of c u rv e s b u ilt up
by p ro p o rtio n . T h at is , the c u rv e s a r e lo c a te d w ith r e s p e c t to the
m a s te r cu rv e at each age in the sam e re la tiv e p o sitio n a s th ey a r e at
s ta n d a rd age. A ll th e d ata th u s c o n trib u te s to th e sh ap e of th e c u rv e s .
T h is is the 'h a rm o n ise d o r an am o rp h ic s ite -in d e x te c h n iq u e ' d e s c rib e d
in N orth A m e ric a n tex tb o o k s such as th o se of S p u rr (1952) and H usch
(1963).
In the f ir s t m ethod th e tre n d of th e lim itin g c u rv e s w ill not be
c o r r e c tly d e te rm in e d u n le ss the h is to r y of h e ig h t-a g e fo r the b e s t and
p o o re s t s ite s is c o m p le te . In th e second m eth o d , u n le s s all s ite s a r e
re p re s e n te d eq u ally at a ll a g e s , th e a v e ra g e s of heig h t a t v a rio u s ag es
w ill be b ia se d and a guiding cu rv e fitted to them w ill lead to a fam ily
of c u rv e s b ia sed in both p o sitio n and sh ap e. F o r n a tu ra l f o r e s t, stan d s
on a ll s ite s a t a ll ag es m ay not o c c u r. In th e c a s e of p la n ta tio n s , it
often happens th a t th e b e tte r s ite s a r e p lan ted f i r s t and th e p o o r e r ones
la te r . A lte rn a tiv e ly , p o o r s ite s m ay be plan ted fo r re a s o n s o th e r
than p ro d u ctio n , fo r ex am p le a s a c o n s e rv a tio n m e a s u r e , and th e ir
s u c c e s s p ro m p ts ex ten sio n of th e p lan tatio n into b e tte r a r e a s , a s fo r
a p lan tatio n of r a d ia ta pine in th e A. C .T . , the s ite -in d e x c la s s ific a tio n
of w hich is d e s c rib e d by C a rro n (1967). A h e ig h t-a g e re la tio n s h ip can
be e s ta b lis h e d w ith confidence only when th e re a r e h e ig h t-a g e d ata fo r
a ll s ite s and all a g e s . The v a lid ity of th e re la tio n s h ip fo r a p a r tic u la r
c a s e depends la rg e ly on how w ell th e d a ta m e e ts th is r e q u ire m e n t.
Since th e id e al is r a r e , th e c u rv e s a r e u su a lly fitte d by hand. C om plete
d ata a r e a lso n e c e s s a r y fo r a r e lia b le d e c isio n on w h eth er th e c u rv e s
a r e an am o rp h ic o r p o ly m o rp h ic. A n am o rp h ic c u rv e s a r e often
ac c e p te d m e re ly fo r co n v en ien ce o r th ro u g h ig n o ran c e b e c a u se of
in ad eq u ate data.
O nce the h eig h t-ä g e re la tio n s h ip h as been p r e p a r e d , m apping of s ite
q u ality o r s ite index of a f o r e s t can b e c a r r ie d out. By r e f e r e n c e to
th e c u rv e s , o r a ta b le co m p iled fro m th e m , th e s ite q u ality o r s ite
index of an a r e a can be d e te rm in e d fro m the age and top h eig h t of the
sta n d . One m ethod of m apping is to s u rv e y p a r a lle l lin e s o v e r the
f o r e s t and, w hile w alking along th em and w holly o r p a r tia lly c o v erin g
th e sp ace betw een th e lin e s , to s e le c t th e ta lle s t tr e e s , m e a s u r e th e ir
Site Q uality 133
h eig h ts, locate them on a suitable map such as a com p artm en t plan,
and draw in boundaries to define site c la s s e s accordingly. If th e re a re
fa irly sh a rp changes in slope, a sp e c t, and soil c h a r a c te ris tic s , and
changes in site a re well c o rre la te d with them , the boundaries m ay be
located fa irly readily. O therw ise it m ay be difficult to locate the
bo u n d aries, and various techniques within this g en eral method have
been developed to cope with th is difficulty. One d escrip tio n is given
by C arro n (1955), another by the Q ueensland D epartm ent of F o re s try
(1962).
It is not alw ays convenient to use an e x p re ssio n of the heights of the
ta lle s t tre e s as the index of site even though it is an effective one.
F o r exam ple. Lewis (1954. 1957) d isc u sse s the method of site quality
mapping of ra d ia ta pine plantations in South A u stra lia. Since thinning
of a re a s of high site quality should com m ence at about age 11 y e a rs and
a little tim e is needed to org an ise th is , mapping of site quality m ust
s ta r t at about age y e a rs . The c rite rio n of site quality is volume
p e r a c re , m ost a re a s being silv ic u ltu ra lly fulljr-sto ck ed . Stand top
height is . how ever, well c o rre la te d with volume and can be used as an
index to it for convenience in p ra c tic e , but at age 9^ y e a rs th e re a re
p ra c tic a l difficulties in m easu rin g height in unpruned unthinned stan d s.
Mapping site in such stands is th e re fo re c a rrie d out on the b a sis of
o th er site in d icato rs which a re m ore re a d ily seen and m easu red than
height, such as ’vigour, fo rm , crown density, needle length and colour,
tightness and colour of b a rk , green level and degree of canopy form ation
p a rtic u la rly b efore thinning’ of which height to g reen level is perhaps
the m ost useful in p ra c tic e . The m apping is done by running p a ra lle l
lines 3 chains a p a rt, changes in the site in d icato rs being plotted at a
sca le of 10 chains to 1 inch on s trip -lin e sh eets for about 1 chain each
side of the line with a m inim um distance between changes along the
line of about a h alf-ch ain . The boundaries a re tra n s fe rre d to
com partm ent plans and draw n over the ’blind' section of about 1 chain
width between the s trip -lin e s . To en su re co n sisten cy between o p e ra to rs ,
te m p o ra ry te s t plots a re e sta b lish e d , th e ir volum es d eterm in ed , and
the c h a ra c te ris tic s of the stand a sso c iate d with the site quality as
defined by volume a re noted. T hese plots a re then used as y a rd stick s
in the site a sse ssm e n t.
The conventional p ra c tice of dividing the range of top height into a
num ber of site quality c la s s e s with equal c la s s in te rv a ls of height, or
estab lish in g site index c la s s e s of specific equal width, has the
disadvantage that th ese c la s s e s of equal in terv al of height may not
re p re se n t c la s s e s of equal in terv al of volum e. In som e c a s e s, the
in te rv a ls of height a re a rra n g e d to re p re se n t equal in te rv a ls of volum e.
In the F o re s t M anagem ent T ables of G reat B ritain (B radley et a l . ,
1966) for exam ple, the yield c la s s e s a re b ased on an in terv al of 20
Hoppus feet m axim um m ean annual in crem en t in volume and the top
height-age c u rv e s, on which the yield c la s s e s a re determ in ed in the
field, a re so positioned at unequal in te rv a ls of height to re p re se n t the
equal in te rv a ls of volume MAI.
Site quality boundaries a re usually not m arked in the field.
C om partm ents often assu m e an unjustified im portance as m anagem ent
units in plantations because a c c e ss fo r u tilisatio n and p rotection is
o rganised around th e ir b o undaries. Site q u alities may be m ore
im portant than com partm ents as b asic units and the positions w here
site quality boundaries c ro s s com partm ent boundaries may often with
advantage be m arked in the field.
U neven-aged F o re s ts of Mixed Species

A site quality c lassific atio n of an even-aged fo re s t of one sp ecies
p e rm its e stim a te s of p re se n t volume to be m ade m o re efficiently than
otherw ise because v ariatio n in volume in each site quality c la s s wdthin
each age c la ss is reduced to a m inim um , and the m o re homogeneous
an a re a with re g a rd to volum e the m ore sim ple and efficient the method
of estim ation and the m ore re lia b le the statem en t of the e stim a te . A lso,
since the behaviour with age of a stand of a p a rtic u la r site quality can
be fo re c a s t, e stim a te s of future volume on a stand b a sis a re p o ssib le.
F u rth e r, since d ifferent site s have differen t ra te s of grow th, they
usually have to be given d ifferen t silv ic u ltu ra l tre a tm e n t and th is can
be o rganised and effected m ore efficiently if site classific atio n has
been c a rrie d out.
By c o n tra st with an even-aged fo re s t of one sp ecies for which the
efficiency of both estim atin g volume and silv ic u ltu ra l tre a tm e n t is
enhanced by the classific atio n of site quality within age, the m ost
efficient stra tific a tio n fo r volum e estim atio n of an uneven-aged fo re st
of mixed sp ecies is usually by sp ecies o r sp ecies m ix tu re, the
p roportion of vario u s size c la s s e s and height c la s s e s , and re la tiv e
crow n cover, since the influence of sp ec ie s, age, s ite , and density on
volume production a re confounded in a com plex way. F o r silv ic u ltu ra l
p u rp o ses, a stra tific a tio n by fo re st type is usually the m ost efficient.
16 Stand Density
One of the m ost im portant re q u ire m e n ts in the m anagem ent of a fo re st

stand is to en su re that th e re is , at each stage of the developm ent of the
stan d , ju s t the rig h t density of tre e m a te ria l to acco rd with the object
of m anagem ent. It is one of the fundam ental law s of silv icu ltu re that
th e re is a relatio n sh ip between the developm ent of a stand and the
d ensity of the stand. H ow ever, to define in concise te rm s ju st what
we m ean by the density of a stand is ra th e r difficult and p e ru sa l of the
lite ra tu re on stand density suggests that fo r e s te rs a re fa r from
a g reem en t on what the te rm im p lies. The b a sis of our concept of stand
density lie s in the fact that if we look at a fo re st stand we get an
im p re ssio n of the way the tre e m a te ria l fills up an a ir space. We
think of the density of the stand in te rm s of the am ount of tre e m a te ria l
p e r unit a re a or space. This concept of density gives us an a ttrib u te
which im proves the value of our d escrip tio n of the stand but is
valuable fo r p ra c tica l m anagem ent only if we can e x p re ss it in a
quantitative way. If m anagem ent is to en su re the rig h t amount of
m a te ria l a t each stage of stand developm ent, the fo re s te r needs a
schedule of re fe re n c e w hereby he can d eterm in e if the stand has the
rig h t density o r not and reg u late it a rtific ia lly if n e c e s sa ry . An index
of stand density is th e re fo re needed, a m e asu re of som e stand v a riab le
which w ill indicate the amount of tre e m a te ria l in a stand p e r unit a re a
o r space. The regulation of stand density is a concern of silv icu ltu re;
providing an index of stand density tends to be the co ncern of
m en su ratio n . In considering likely v a ria b le s to d e sc rib e stand density
in quantitative te rm s we have to c o n sid er se v e ra l a sp e c ts.
F irs tly , we have to c o n sid er what 'am ount of stand m a te ria l' m eans
because th is will differ depending on what p a rtic u la r p a rt of the stand
one is in te re ste d in and this depends on the kind of stand, the stage of
its developm ent, and the objects of m anagem ent. In te re st in the stand
135
m ay range, for exam ple, from a biological in te re s t in the g ro ss weight
of vegetable m a te ria l above and below ground, to a co m m ercial in te re st
in the quality of a c e rta in p a rt of the tre e bole. We have to be su re
the index im plies the sam e thing to all people o r that the different
am ounts they a re v a rio u sly in te re ste d in can all be derived from the
index.
Secondly, the v a ria b le for the index of density has to be such that if
two stands a re otherw ise the sam e but d iffer in th at v a ria b le, then th is
difference w ill be re fle c ted in a difference in th e ir growth behaviour.
If th is v ariab le is a lte re d then the subsequent growth behaviour of the
stand should re fle c t that a lte ra tio n . This m eans that we shall want to
use the index in a re la tiv e way. We talk , for exam ple, of a stand as
not 'fully occupying the s ite ' when it is not using the fa c to rs of the
environm ent to the fu llest extent. We judge th is by the amount of
m a te ria l on the site o r the way the stand is behaving. To put this in
quantitative te rm s by m eans of the index of density re q u ire s that th is
index be used in a re la tiv e way. This m eans settin g up a stan d ard .
E stablishing th is stan d ard , how ever, m eans defining what we mean by
'full}’' occupying the s ite '. Do we, for exam ple, m ean th at the stand
has m axim um volume o r is producing m aximum in crem en t in volume:
and by 'm axim um v olum e', do we m ean the g ro ss volume of the maximum
num ber of tre e s o r the volume of wood of a p a rtic u la r quality on the
m inim um num ber of tr e e s ? The stan d ard m ay be based on biological
o r m anagem ent grounds.
T hirdly, the v aria b le for the index of density has to be one which
can be m easu red and applied conveniently in p ra c tic e so th at stand
density regulation can be c a rrie d out on the b asis of it; o r some other
v a ria b le , which can be used for th is p u rp o se, h as to be c o rre la te d
with it.
T here a re som e fu rth e r re q u ire m e n ts of an index of density. F or
obvious re a so n s it should be sim ple and objective. It should also be
independent of age and site: the introduction of density as a b a sis for
stra tific a tio n of a stand fu rth e r to that on the b a sis of age and site w ill
not im prove e stim a te s of volume if the index of density is c o rre la te d
with age and site . The index should also be a v ariab le which lends
itse lf to forw ard projection; future yield is likely to be influenced by
the density at the future tim e which will be a reflectio n of the p re sen t
density. Ideally, the index should be applicable to any kind of stand,
from the sim p le st - a stand of one sp ec ie s, age, and s ite , to the m ost
com plex - a stand of mixed sp e c ie s, uneven-aged and with v ariatio n of
site .
Stand D ensity 137
In s u m m a r y , then, the index of density should be sim p le and o b je ctiv e ,

and definable in both ab so lu te and r e la tiv e t e r m s . It should be such
th a t q uantitative d iffe re n c e s in the value r e f le c t d iffe re n t b ehaviour
betw een stan d s at a p a r t i c u l a r tim e , and differe n t beh av io u r within a
stan d d u rin g its developm ent. It should be c o r r e l a t e d with o th e r stand
v a r i a b l e s . It should be independent of age and s ite . It should be
ap p lic a b le to any kind of stand. R e flectin g on the com plex biology of
f o r e s t s and the v a ry in g r e q u ir e m e n t s of m a n a g e m e n t, it is r e a s o n a b le
to a s s u m e th a t no single m e a s u r e w ill m e e t all th e s e r e q u ir e m e n t s
and th a t a n u m b e r of d iffe re n t m e a s u r e s may be n e c e s s a r y to c o v e r the
wide ran g e of conditions. Many m e a s u r e s have been p ro p o sed . B eca u se
they aim to d e s c r ib e the stan d and r e f le c t its grow th b eh av io u r, som e
of the m e a s u r e s p ro p o se d m ight be d e s c r ib e d a s b io lo g ic a l. Since the
r e g u la tio n of stand d en sity h a s to be p r a c t is e d on the b a s is of the index,
s o m e indexes p ro p o s e d m ight be d e s c r ib e d as p r a c t i c a l, such a s those
r e f e r r e d to a s 'm e a s u r e s of thinning g r a d e ' and 'definition of thinning
t r e a t m e n t '. Once a schedule of re g u la tio n of d e n sity of a stand has
been developed, fo r exam ple by e x p e r im e n ta l t r i a l s , to su it a p a r ti c u la r
biological p u rp o s e o r object of m a n a g e m e n t, the d e n sity of the stand
to be left a s the m a in c r o p m a y be d e s c r ib e d in any convenient t e r m s
fo r p r a c t ic a l im p lem en tatio n of the sch ed u le . In e v e n -a g e d stan d s
w h e r e th e t r e e s iz e is r e a s o n a b ly uniform and w h e r e , p a r ti c u la r ly in
th e c a s e of p la n tatio n s, the t r e e s a r e r e a s o n a b ly s y m m e tr ic a lly
d is tr i b u te d , the d e s ir a b le d en sity of a stand of a p a r t i c u l a r age on a
p a r t i c u l a r s ite can be d e s c r ib e d by v a r io u s stand v a r ia b le s , alone o r
in com bination. T his s o r t of index is d e riv e d e s s e n tia lly fo r the
p r a c t i c a l p u rp o s e of im p lem en tin g a thinning schedule.
Indexes of Stand D ensity

Since a stand c o m p r i s e s a n u m b e r of t r e e s of d iffe re n t s iz e d is tr ib u te d
s p a tia lly o v e r an a r e a of land, its d en sity , as am ount of stand m a te r ia l
p e r unit a r e a o r s p a c e , is a function of th r e e e le m e n ts - the n u m b e r
of t r e e s , t h e ir r e s p e c tiv e s iz e , and th e ir s p a tia l d is trib u tio n . To be
c o m p le te a m e a s u r e of d en sity should in c o rp o r a te th e s e e le m e n ts .
N u m b er of t r e e s can be d e te r m in e d , re a d ily o r o th e rw is e , by counting.
Spatial d is trib u tio n of t r e e s is not r e a d ily d e te r m in e d , h ow ever, and
is taken into account in m e a s u r e s of den sity only in d ir e c tly , for exam ple
by a s s u m in g a g e n e r a lis e d r e c t a n g u la r o r t r i a n g u l a r spacing. This is
138 O utline of F o r e s t M e n su ra tio n
p e rh a p s a s e rio u s d e fic ie n c y in m e a s u r e s of d e n sity , but the in clu sio n
of a d ire c t e x p re s s io n of s p a tia l d is trib u tio n h a s been g e n e ra lly avoided
b e c a u se of its co m p lex ity . T h e re a r e a n u m b e r of f a c to r s involved in
individual tr e e s iz e : th e ste m w hich can be c h a r a c te r is e d by d ia m e te r ,
h eig h t, and ta p e r; th e crow n w'hich can be c h a r a c te r is e d by s p re a d and
depth; and the ro o t s y ste m w hich m ig h t s im ila r ly be c h a r a c te r is e d by
its s p re a d and depth ex cep t fo r th e d ifficu lty of m e a s u rin g th em . The
m o s t lik ely tr e e v a r ia b le s fo r in c lu sio n in a m e a s u re of stan d d en sity
a r e th o se w hich a r e r e a d ily m e a s u re d , a r e in te r c o r r e la te d w ith the
o th e r m ain v a ria b le s of s iz e , and in sum o r a v e ra g e r e p r e s e n t a stan d
c h a r a c te r is tic effic ie n tly .
N um ber of T re e s
In m any c a s e s w h ere th e s iz e of th e in d iv id u als can be tak en a s
c o n sta n t o r the d iffe re n c e in s iz e of in d iv id u als can be ig n o red , the
n u m b e r of in d iv id u als s e r v e s w ell as a m e a s u r e of d en sity . F o r
ex am p le, the d e n sity of hum an and an im al p o p u latio n s is u su a lly e x p re sse c
a s n u m b e r of in d iv id u als p e r u n it a r e a ; and fo r m any p u rp o s e s of ecology
th e d e n sity of s m a ll p la n ts is d efined (as by B row n, 1954) in th e sam e
te r m s . In young f o r e s t s ta n d s w h ere the only c o n c e rn is w h eth er the
n u m b er of tr e e s c o n s id e re d n e c e s s a r y fo r the s a tis f a c to r y dev elo p m en t
of the stand is p r e s e n t, d e n sity m ay also be d e s c rib e d by the n u m b er
of in d iv id u als.
N um ber of t r e e s m ay a ls o be u se d a s a p r a c tic a l index of d en sity fo r
im p lem en tin g a sch ed u le of stan d d e n sity re g u la tio n w h ere the d e s ir a b le
n u m b e r of tr e e s , a t th a t age on th a t s ite fo r th a t s p e c ie s , h a s been
decided on. It is to th is co n tex t th a t the te rm sto c k in g , m ean in g n u m b e r
of tr e e s p e r u nit a r e a , is b e s t r e s tr ic te d . In th is c o n tex t, to o , the
d e s c rip tiv e te r m s of a r e la tiv e kind su ch a s u n d e r-s to c k e d , fu lly -sto c k e d
and o v e r-s to c k e d a r e a p p ro p ria te .
F o r m o s t p u rp o s e s , the v a ria tio n in tr e e s iz e is im p o rta n t in the
g e n e ra l co n cep t of d e n s ü y sin c e th e effic ien cy of in dividual tr e e s and
th e ir co n trib u tio n to th e to ta l g row th of th e group u su a lly depends on
th e ir re la tiv e s iz e , and fo r a s a tis f a c to r y m e a s u re of d e n sity , n u m b e r
of tr e e s m u st be com bined w ith t r e e s iz e .
N u m b e r of T r e e s and D . b . h . o . b .
The m o st co m m only quoted index w hich in c o rp o ra te s d ir e c tly the
n u m b e r of tr e e s p e r u nit a r e a and a m e a s u re of d . b . h . o . b . is the
Stand D ensity 139
stand d ensity index of R eineke (1933). R eineke showed th at if values
of num ber of tre e s p e r a c re of even-aged stands 'a t full d en sity ' w ere
plotted on logarithm ic c ro s s -s e c tio n paper ag ain st values of average
d ia m e te r of the stan d s, a stra ig h t line relatio n sh ip could be estab lish ed .
He found th at the slope of this 're fe re n c e c u rv e ' (for m axim um density)
w as id entical for a num ber of sp e c ie s, though the heights of the cu rv es
at a given d ia m e ter v a rie d am ongst the sp ec ie s. He suggested a stand
d ensity index for a stand could be d eterm in ed by 'plotting its num ber of
tr e e s and d ia m e te r, passing a line p a ra lle l to the re fe re n c e curve
through th is point and reading the num ber of tr e e s p e r a c re at its
in te rse c tio n with the 10 inch o rd in ate' and th at th is e x p ressio n m ight
be used 'fo r co m p ariso n s between sp ecies and within a s p e c ie s '. He
found th e re w as 'no significant o r ap preciab le c o rre la tio n betw een age
o r s ite quality and the num ber of tre e s p e r a c re for a given av erag e
d ia m e te r'. R ein ek e 's stand density index o r a re la te d concept has
been used by m any w o rk e rs in N orth A m erica. Spurr (1952) has a
c ritic a l review of it and in sum m ary a rg u e s fo r stand b asal a re a .
Hall (1959) has used stand density index in investigating growth in
volum e in n a tu ra l eucalypt fo re s t and coniferous plantations in V icto ria.
N um ber of tr e e s in re la tio n to d ia m e ter is often used as a p ra c tica l
index in thinning in the form of a 'ru le of th u m b ', the re q u ire d spacing
of a tre e being indicated by tre e d ia m e te r plus or tim e s a constant.
The stand b asal a re a is a function of the d . b . h . o . b . of each tre e in
the stand, being the product of the num ber of tre e s and average b asal
a re a . On a p a rtic u la r s ite , the basal a re a of stands of the sam e age
is likely to v a ry am ongst sp ec ie s. F o r a p a rtic u la r sp e c ie s, the b asal
a re a of stands of the sam e age on different s ite s is also likely to v ary .
But for c e rta in sp ecies the basal a re a s of stands on a p a rtic u la r site
a re likely to be reaso n ab ly constant over a co n sid erab le period of the
developm ent of the stand, p a rtic u la rly tow ard m a tu rity , and stand
b asal a re a p e r unit a re a a p p ea rs to be a good m e asu re of the extent
to which the stand is occupying the site . It is th e re fo re a good m e asu re
of density under these conditions and. because of the g re a te r in te re st
of fo r e s te rs in the m anagem ent of even-aged stan d s, it has been used
m o re than any o th er v a ria b le.
B asal a re a is a p ra c tic a l index of stand density; it can be obtained
re a d ily by d ire c t enum eration of stem s o r by angle count m ethod. Its
use can be objected to on biological grounds in th at it includes the dead
wood of active tr e e s and all the wood of su p p re ssed tre e s which
c o n trib u tes nothing to the stand. It is . how ever, a n a tu ral base for
Standing tr e e v o lu m e, and v olum e in c re m e n t and b a s a l a r e a in c re m e n t
a r e u su a lly w ell c o r r e la te d , so th a t it m ay be s a tis f a c to r y on
m an ag em en t g ro u n d s. Though it h as c o n s id e ra b le lim ita tio n s , b a s a l
a r e a is p ro b ab ly a p p licab le to m o re co n d itio n s than any o th e r m e a su re
of stan d d en sity .
N u m b er of T r e e s and H eight
The indexes of th is c a te g o ry a r e of th e p r a c tic a l kind fo r im p lem en tin g
a sch ed u le of stan d d en sity re g u la tio n , p a r tic u la r ly fo r e v e n -a g e d stan d s
of known age and s ite . The co m m o n est e x p re s s io n is th e r a tio of th e
a v e ra g e sp acin g of the tr e e s to th e stand h eig h t (in th e sa m e u n its).
Such h e ig h t-s p a c in g r a t i o s , a s a fra c tio n o r p e rc e n ta g e , hav e been
d is c u s s e d by B ecking (1953) u sin g tr ia n g u la r sp a c in g , and by W ilson
(1946) and H um m el (1953) u sin g sq u a re sp acin g . The stan d height
u sed m ay be e ith e r m ean o r top height: it should effe c tiv e ly re f le c t
s ite - a g e .
N u m b er of T r e e s and F o rm (T a p e r)
B eca u se the sp acin g of tr e e s m ay have an e ffe c t on th e ir r a te s of
ta p e r , a co m b in atio n of n u m b e r of tr e e s and r a te of ta p e r should
p ro v id e a m e a s u re of d e n sity . The d iffic u ltie s o u tlin ed e ls e w h e re of
ad eq u a tely defining ta p e r of t r e e s and e x p re s s in g th is a s an a v e ra g e
fo r a stan d w ill in d ic a te th a t any such m e a s u re is not lik e ly to be
s u c c e s s fu l, and none is in com m on u se.
N um ber of T r e e s . D ia m e te r, and H eight

B rie g le b (1952) show ed th a t in young D ouglas f ir in w e s te rn E urope
and N o rth A m e ric a , cro w n w idth and length w ere r e la te d to both tr e e
height and tr e e d . b . h . o . b . , in d icatin g th at the sp ace a young D ouglas
f ir can u tilis e is r e la te d to both its d ia m e te r and h eig h t. He s tr e s s e d
th a t th is re la tio n s h ip m u st be re c o g n ise d when d ev isin g d e n s ity s ta n d a rd s
fo r fo r e s ts in w hich heig h t a t a given d . b . h . o . b . m ay v a r y g re a tly .
N um ber of T r e e s . D ia m e te r. H eight, and F o rm

The volum e of a stan d is a function of th e se fo u r sta n d v a r ia b le s .
F o r m any p u rp o s e s , volum e is the e s s e n c e of th e co n cep t of d en sity
and so it should p ro v id e th e id eal index fo r th o se p u rp o s e s . F ro m the
b io lo g ical point of view , h o w ev er, it h as the sa m e d isa d v a n ta g e as
Stand D ensity 141
b a s a l a r e a . It has th e f u r th e r d isad v an tag e th a t it is d iffic u lt to m e a s u r e ,
and a v a ria b le c o r r e la te d w ith volum e p ro v id e s a m o re con v en ien t index
of d e n sity .
The bole a r e a of a stan d is a lso a function of th e se fo u r stan d
v a r ia b le s and so is a m e a s u re of th e am ount of stan d in th e p h y sic a l
s e n s e . Since th e bole a r e a of a tr e e can be re g a rd e d as th e b a se fo r
p o te n tia l in c re a s e in the volum e of th e tr e e , the sum of th e bole a r e a s
of th e tr e e s fo rm in g a stan d r e p r e s e n ts th e b a s e fo r p o te n tia l in c re a s e
in th e volum e of the stan d . As such it h as p a r tic u la r b io lo g ical
s ig n ific a n c e . B eing a m e a s u re of th e am ount of stan d in both the
p h y sic a l and b io lo g ical s e n s e it should be an effectiv e index of d en sity
but the ev id en ce fo r th is is co n flictin g . M ulloy (1944) show ed th a t fo r
a n u m b e r of p lo ts of re d and w hite pine in e v en -a g ed s ta n d s , bole a r e a
and R e in e k e 's stan d d en sity index w e re fo r a ll p r a c tic a l p u rp o s e s the
s a m e . W ilson (1955) d e m o n stra te d a s im ila r a g re e m e n t betw een bole
a r e a and stan d d en sity index b a se d on th e r a tio of a v e ra g e sp acin g on
m ean dom inant heig h t fo r e v e n -a g e d sta n d s of re d p in e. H um m el (1953)
d rew atten tio n to the c o n stan c y of bole a r e a o v er the g r e a te r p a r t of
th e ro ta tio n , irr e s p e c tiv e of kind of thin n in g if th e thin n in g re g im e w as
c o n s is te n t, in the c a s e of s e v e r a l B r itis h and c e n tr a l E u ro p ean y ie ld
ta b le s . Anuchin (1962) a lso d e m o n s tra te d th a t fo r a w ide ran g e of
s p e c ie s in the U!S. S .R . , stan d bole a r e a a p p e a re d to be c o n sta n t w ithin
a s ite quality o v e r a w ide ran g e of ag e. On th e o th e r hand. B rie g le b
(1952) found bole a r e a u n s a tis fa c to ry a s an index of stan d d en sity in
w ell m anaged stan d s of D ouglas f ir in D en m ark .
It h a s been su g g ested th a t, if th e individual t r e e s a r e n e a rly all the
sa m e h eig h t, and if v a ria tio n in ta p e r can be ig n o red , stan d bole a r e a
can be a p p ro x im ate d by the sum of the d ia m e te rs (or c irc u m fe re n c e s ).
T h is should have th e sa m e b io lo g ical sig n ifican ce a s bole a r e a but be
e a s ie r to d e riv e . C ooper (1961) h a s u sed su m s of d ia m e te rs as an
index of d en sity in in v e stig a tin g th e re la tio n s h ip betw een grow th and
stan d v a ria b le s . C u rtin (1962b) show ed a lin e a r c o r r e la tio n of in c re m e n t
in stan d b a sa l a r e a and volum e w ith th e p ro d u ct of sum of d ia m e te rs
and m ean log h eig h t fo r an ev e n -a g e d c o a s ta l eu ca ly p t f o r e s t in N. S. W.
C row n Canopy
Since the v ig o u r of its crow n is known to be one of th e m a jo r fa c to rs
influencing the grow th of a t r e e and o b se rv a tio n of th e cro w n s of m any
s p e c ie s su g g e sts th a t th e ir v ig o u r is r e la te d to a e r ia l grow ing s p a c e ,
the amount of space occupied by crow ns in a stand m ight be expected
to re fle c t the density of a stand. Thus p e r cent crown c lo s u r e , the
horizontal a re a of crow ns as a percen tag e of h o rizo n tal ground a re a ,
has been used as a v a ria b le in a e ria l stand volum e tab les for v ario u s
sp ecies in various c o u n tries and.as an index in stratify in g stands for
ground sam pling in re s o u rc e s inventory, fo r exam ple stands of native
eucalypts in A u stra lia (F. and T .B . , 1951). Its value in th is re g a rd
depends on how sa tisfa c to rily v a ria tio n of volum e in stands of o therw ise
s im ila r com position and s tru c tu re re la te d to v a ria tio n in p er cent of
crow n c lo su re . B ecause ro o t space may be m o re im p o rtan t than crown
space, m e asu re s of stand density based on crow n c lo su re a re often
poorly c o rre la te d with other m e a su re s.
On the assum ption that the a re a occupied by a single tre e is
proportional to its d .b .h .o .b . , Chism an and Schum acher (1940)
developed an equation w hereby ground a re a was re la te d to the num ber
of tre e s and th e ir d ia m e te rs on th at ground a re a from data of sam ple
plots of fully-stocked, even-aged loblolly pine stands in N orth A m erica,
and also fo r uneven-aged mixed conifer and b ro ad -leav ed stan d s.
Applied to the data for any one plot it e x p re ssed the tre e a re a r a tio ,
the ground a re a occupied by the tre e s of the plot as a ra tio of the plot
a re a , and the ground a re a occupied by an individual tre e according to
its d .b .h .o .b . This tre e a re a ra tio was intended as a 'm e a su re of
stocking’, and 'stocking n o rm ' for a given stand type and condition.
That is the density of the tre e s of an a re a is e x p re sse d as a p ercen tag e
of the average stocking of any num ber of sam ple a re a s , in th is case
the sam ple a re a s being chosen for full stocking. The ra tio was
considered to be independent of age and site index. D eetlefs (1953)
em ployed a s im ila r approach, though a m ore com plex equation, for
even-aged stands of Pinus taeda with a b ro ad -leav ed u n d e r-s to re y in
N orth A m erica w hereby the ground a re a occupied by a single tre e o r
by a component of the m ixture can be d eterm in ed . The density of any
stand is determ ined by substituting num ber of tre e s and b a sa l a re a in
the equation: the average density of the stands on which the study was
based being 100 p e r cent, the density of any individual stand may be
g re a te r o r le ss than th is. He has applied this m e asu re of density in
a study of stand density and growth relatio n sh ip s in P. taed a (D eetlefs.
1954).
K rajicek et a l . (1961) developed a crown d ia m e te r-tre e d .b .h .o .b .
relatio n sh ip fo r open-grow n tre e s from which ’maximum crow n a re a '
of a tre e could be d eriv ed . They defined crow n com petition fa c to r as
Stand D ensity 143
the sum of the m axim um crow n a re a s of all the tre e s in a stand p er
a c r e , using it as a m e asu re of density in th at it e stim a te s the a re a
av ailab le to any tre e in a stand in re la tio n to the m aximum it could use
if it w ere open-grow n. Vezina (1962) te ste d th is in stands of v ario u s
sp ec ie s and found it v aried with age and site quality and did not show
any co n siste n t relatio n sh ip with other m e a su re s such as b asal a re a
p e r a c re and R ein ek e's stand density index. C urtin (1964) defined
stand density index in effect as the ra tio of the a re a of the crow ns of
the tre e s on an a c re to that ground a re a , the crow n width of a tre e
being given by a re g re s s io n of crow n width on the d. b. h. o. b. of a tre e
and the square of its height, for even-aged pure stands of Eucalyptus
obliqua in V icto ria, and co n sid ered it was independent of age and site
quality.
Root Spread and Depth

The vigour of the root system of a tre e is as im portant an influence
on the growth of the tre e as the vigour of the crow n, so that the extent
and depth of ro o ts of the stand should re fle c t its density but, as pointed
out p rev io u sly , it is very difficult to m e a su re .
G row th-Stand V ariable R elationships

In re c en t y e a rs many attem p ts have been made to develop equations
to d e sc rib e relatio n sh ip s betw een stand v a ria b le s and th e ir growth
under c e rta in conditions. In these equations, v ario u s stand v a ria b les
have been included to serv e as e x p re ssio n s of stand density, the
efficiency of a v a riab le as a m e asu re of density being judged
m ath em atically by its contribution to the c o rre la tio n .
Lemmon and Schum acher (1962). fo r exam ple, studied the d ia m e ter
and volume growth of P . ponderosa as influenced by s ite index, density,
age, and s iz e , using as ex p re ssio n s of density the b asal a re a of the
surrounding stands and the extent of o v e r-s to re y as m easu red by a
'sp h e ric a l d en sio m ete r' (Lem m on. 1957). They found that o v e r-s to re y
density was significantly c o rre la te d with periodic volum e and d ia m e ter
grow th but b asal a re a of the dom inant portion of the surrounding stand
contributed m ore strongly. W ilson and B re n d er (1963) te ste d four
ex p re ssio n s of stand density fo r loblolly pine volum e grow th p rediction
in m ultiple re g re ss io n in even-aged m anaged stands - S tahelin's
'p e rc e n t of full stocking', total b asal a re a . R ein ek e's stand density
index, and in itial m erchantable cubic foot volum e. They found that
e s s e n tia lly the sa m e am o u n t of v a ria tio n in annual cubic foot grow th
w as explained by each e x p re s s io n , in com b in atio n w ith ag e, s ite , and
c e rta in of th e ir s ig n ific a n t in te ra c tio n s , fro m d ata involving r e p e a t
m e a s u re m e n ts . They co ncluded th a t th e ch o ice am ong the v a r ia b le s
e x p re s s in g stan d d e n sity is p r im a r ily at th e d is c re tio n of the
in v e stig a to r but th a t b a s a l a r e a had th e ad v an tag es of e a s y co m p u tatio n
and ap p licatio n .
P ro v id e d th e im p lic a tio n s of judging a m e a s u re of d en sity p u re ly by
its m a th e m a tic a l co n trib u tio n to a c o r r e la tio n w ith stan d grow th a r e
kept in p e rs p e c tiv e , we m ig h t ex p ect th e g r e a te s t co n trib u tio n to th e
q u estio n of the r e la tiv e e ffic ie n c y of v a rio u s m e a s u re s of d e n sity to
com e fro m th is a p p ro ac h . It is im p o rta n t th a t any m e a s u re p ro p o sed
h av e b io lo g ical sig n ific a n c e and th a t th e re la tio n s h ip m ake b io lo g ical
s e n s e . It is a lso e s s e n tia l, a s s tr e s s e d by C lu tte r (1963) th a t m o d els
e s ta b lis h e d to d e s c rib e gro w th and y ie ld s e p a ra te ly fo r a stan d should
be c o m p atib le.
P o in t D ensity
F o r eco lo g ical o r s ilv ic u ltu ra l p u rp o s e s , it is often d e s ira b le to
e v alu ate the e ffe c ts of d en sity on a sin g le tr e e . S p u rr (1962) h as
p ro p o sed an 'an g le su m m atio n m eth o d ' w hich involves som e of
B itte r lic h 's angle count th e o ry but d iffe re n t co n ce p ts and g iv es an
e s tim a te of stan d b a s a l a r e a p e r a c r e w hich r e f e r s to th e v ic in ity of a
p a r tic u la r tr e e . B row n (1965) s tr e s s e s the need to define and m e a s u r e
the sp ace av a ila b le to individual tr e e s r a th e r than n u m b er of tr e e s p e r
u nit a r e a and s u g g e sts defining the a r e a p o te n tia lly av ailab le to a tr e e
a s the a r e a en clo se d by lin e s w hich b is e c t at r ig h t an g les the lin e s
jo in in g the su b je c t tr e e to th o se su rro u n d in g it.
D en sity and G row th

It is com m on o b s e rv a tio n th a t if the n u m b e r of tr e e s is v a rie d in a
stan d of a p a r tic u la r age o r s tr u c tu r e on a p a r tic u la r s ite , th en th is
v a ria tio n of grow ing sp a c e h a s an effect on th e su b seq u en t g row th in
d . b . h . o . b . of in d iv id u al t r e e s - up to a c e r ta in p o in t, the w id e r th e
sp acin g th e la r g e r th e d . b . h . o . b . It is an eq u ally com m on o b s e rv a tio n
th a t d en sity m u st v a ry c o n s id e ra b ly b e fo re it h as any effect on h eig h t
grow th. T his is th e re a s o n heig h t grow th can be used a s an index of
s ite w hich does affect it ir r e s p e c tiv e of d en sity . H ow ever, w h e th e r
Stand D ensity 145
v ariatio n in density affects d ia m e ter growth along the stem , to what
extent under what c irc u m sta n c e s, is a m a tte r on which th e re is
co n sid erable but conflicting evidence. This is one of the la rg e s t and
m ost c ritic a l gaps in our knowledge of the behaviour of tr e e s .
17 Increment and Yield
In m e n su ra tio n we a r e c o n c e rn e d w ith g row th not as a b io lo g ical

phenom enon but w ith th e q u a n titativ e change in a tr e e o r stan d v a ria b le
w hich is the r e s u lt and e x p re s s io n of g row th, th a t is in c re m e n t; and in
th e accu m u latio n of in c re m e n t av ailab le at a p a r tic u la r tim e fo r a
p a rtic u la r p u rp o se of m an ag em e n t, w hich is y ie ld . We have a lre a d y
r e f e r r e d in C h ap ter 9 to th e v a rio u s e x p re s s io n s of in c re m e n t in
com m on u se fo r in d iv id u al t r e e s - m ean annual in c re m e n t, c u r r e n t
annual in c re m e n t, and p e rio d ic m ean annual in c r e m e n t. The sam e
te rm s a r e used w ith r e f e r e n c e to sta n d s. T h e re a r e m any e x p re s s io n s
of yield such a s an n u al, in te rm e d ia te , fin a l, s u s ta in e d , fin a n c ia l, and
so on w ith sp e c ia l co n n o tatio n s fo r m an ag em e n t. The te rm y ield w ill
be u sed h e re in a v e ry g e n e ra l s e n s e , m e re ly im plying an am ount of
wood m a te ria l a t a p a r tic u la r tim e .
The in c re m e n t of a p a r tic u la r v a ria b le of a stan d is the sum of th e
in c re m e n ts of th a t v a ria b le fo r the individual t r e e s of th e stan d . F o r
a tr e e in is o la tio n , the m ain f a c to r s affectin g its grow th a r e its
g en e tic a l co n stitu tio n (its genotype) and its en v iro n m e n t. When tr e e s
a r e in stan d fo rm a tio n , th e effe c t of th e ir m u tu al co m p etitio n is
su p e rim p o se d . The th r e e f a c to r s fo rm a co m p lex and it is d ifficu lt
fo r m e n su ra tio n p u rp o s e s to s e p a ra te them and g et in d ex es of th e m .
W^e t r y to c la s s ify s ite and to index stan d d e n sity . M anagem ent is
u su ally c o n ce rn ed w ith s ta n d s co m p o sed of a n u m b e r of genotypes and
th e e ffe c ts of g e n e tic a l fa c to r s on stan d grow th a r e u su a lly unavoidably
included w ith e ffe c ts due to en v iro n m en t and c o m p etitio n .
Wdien a s p e c ie s is re la tiv e ly fa s t-g ro w in g and th e p e rio d fo r w hich
in c re m e n t h as been d e te rm in e d is s h o rt, often th e c a s e in A u s tra lia ,
th e f o r e s te r is s o m e tim e s in c lin ed to fo rg e t, in e x tra p o la tin g the
in c re m e n t into th e fu tu re , th a t the w e a th e r o v e r th e p e rio d m ay not
have been re p r e s e n ta tiv e . The annual ra in fa ll fig u re s fo r C a n b e rra
146
Increm ent and Yield 147
T a b le 17:1
A n n u al ra in fa ll (in), C a n b e rra , A . C . T .
(For. and T im b . Bur. )
1957 14 1963 25
1958 30 1964 29
1959 34 1965 16
1960 31 1966 29
1961 32 1967 14
1962 29
in Table 17:1 s e rv e to illu s tra te the point. The fo re s te r whose only

knowledge of C an b e rra was the y e a rs 1958-64 ( a se v e n -y e a r period
which m ight be thought a reasonable length on which to base an e stim ate
of in crem ent) would be m isled u n less he knew th at the lo n g -term average
was 25 in and, in p a rtic u la r, unless he knew of the likelihood of poor
y e a rs such as 1957, 1965, and 1967. An e stim ate of in crem en t m ust
alw ays be qualified by the w eather conditions of the perio d .
If the age of a tr e e o r stand is known at tim e of m e asu re m e n t, we
can re la te size of v a ria b le at p a rtic u la r tim e s of m easu rem en t to ages
at those tim e s and d erive o r e x p re ss in crem en t in d irec tly in this way.
A ltern ativ ely , we can deriv e in crem en t in the v aria b le o v er the p erio d
between m e asu re m e n t d ire c tly and re la te it to the av erag e age of the
p erio d , o r to the size of v a riab le at the beginning o r end of the period.
If age cannot be d eterm in ed , in crem en t m u st be re la te d to size of
v a ria b le . Age is com m only used as an independent v a riab le because
tim e is a common v a ria b le in much of our g en eral thinking. Size of
v a ria b le is , how ever, the re s u lta n t of so m any of the fa c to rs of grow th,
including tim e , that it may be m ore useful as an independent v ariab le
than age fo r many p u rp o se s.
Two p hases of the developm ent of a stand a re of in te re s t - its p re sen t
condition a t a p a rtic u la r tim e and its likely condition at a future tim e.
P a s t in crem en t is used to fo re c a s t the likely future condition of a stand.
The h isto ry of grow th of those tre e s which show annual rin g s can be
d eterm in ed by stem a n aly sis. The h isto ry of growth of any kind of tre e
can be d eterm ined from p eriodic m e a su re m e n t. The fu tu re condition of
any kind of stand can be fo re c a s t by p ro jectin g the p re s e n t siz e s ,
assu m ing c e rta in in crem en ts from the re c o rd of p a st in crem en ts and
assu m ing a c e rta in frequency d istrib u tio n of s iz e s . This is the method
of stand p r o je c tio n , the p ro je c tio n being fu n d am en tally on an individual
tr e e b a s is w ith th e t r e e s gro u p ed into s iz e c la s s e s fo r co n v en ien ce.
F o r ev en -a g ed sta n d s w hich can be c la s s ifie d by age and s ite , th e
h is to ry of the stan d can be co m p iled on a stan d b a s is in the fo rm of a
y ield ta b le , th e tr e n d s of w hich can be p ro je c te d into the fu tu re fo r
fo re c a s tin g th e fu tu re condition of th a t sta n d , o r it can be ap plied in
p re s e n t o r p ro je c te d fo rm to o th e r s ta n d s of a s im ila r n a tu re . T h e re
a r e grow th re la tio n s h ip s in w hich in dependent v a ria b le s o th e r th an age
and s ite a r e u sed w hich can be u sed fo r f o re c a s tin g fu tu re stand
c o n d itio n s .
PAST INCREM EN T FROM PERIODIC MEASUREMENT
P e rio d ic m e a s u re m e n ts a r e c a r r i e d out fo r e x p e rim e n ta l grow th

s tu d ie s , re s o u r c e in v e n to ry , and m an ag em en t in v e n to ry .
T h e re a r e two w ays of d e te rm in in g in c re m e n t depending on w h eth er

th e tr e e s a r e id e n tified in d iv id u ally o r no t, d is c u s s e d la te r .
W hether the t r e e s a r e in d iv id u ally id e n tified o r not, the in c re m e n t
can be e x p re s s e d in two w ay s. It can be c o n s id e re d a s in c re m e n t on
t r e e s of a c e r ta in s iz e at th e beginning of th e p e rio d and can be r e la te d
to th o se s iz e s , u su a lly r e f e r r e d to a s in c re m e n t of the com ing p e r io d .
A lte rn a tiv e ly , it can be c o n s id e re d a s in c re m e n t on tr e e s to m ak e them
a c e r ta in s iz e at th e end of th e p e rio d , and can be re la te d to th o se s iz e s ,
u su a lly r e f e r r e d to a s in c re m e n t of th e p a s t p e r io d . The p u rp o se of
th e d istin c tio n is to allow a ch o ice in p ro je c tin g a stand ta b le . The
in c re m e n t of t r e e s w hich a r e a c e r ta in s iz e now can be a ssu m e d to be,
o v e r a fu tu re length of tim e , th e sa m e a s th e in c re m e n t of tr e e s th a t
s iz e th a t length of tim e ago, th a t is th e in c re m e n t of the com ing p e rio d
is p ro je c te d . A lte rn a tiv e ly , the in c re m e n t of tr e e s w hich a r e a c e r ta in
s iz e now can be a s s u m e d to b e, o v er a fu tu re length of tim e , th e sa m e
a s the in c re m e n t in the r e c e n t p e rio d of th a t length to m ake th em th e ir
p re s e n t s iz e , th a t is th e in c re m e n t of th e p a s t p e rio d is p ro je c te d .
F o r ex am p le, su p p o se tr e e s r e s p e c tiv e ly 8 -0 , 1 0 -0 , and 12- 0 in five
y e a r s ago have now becomfe re s p e c tiv e ly 9 -0 , 12-0 and 1 5 '0 in. Then
th e p ro je c te d in c re m e n t of th e 12 in t r e e fo r th e next five y e a r s could
In crem en t and Yield 149
be re g a rd e d e ith e r as 3 in, this being the in crem en t of a 12 in tre e in
the com ing period of the la s t five y e a rs , o r as 2 in, this being its
in c re m en t in the p a st period. The decision as to which of th ese
e x p re ssio n s should be used for p rojection p u rposes is a silv ic u ltu ra l
one.
D eterm ination of increm ent in d . b . h . o . b . when the tre e s a re identified

in d iv idually. In this c a se , the increm ent on each tre e o v er the period
betw een m e asu re m e n ts is determ ined quite sim ply by su b tractin g the
value of the f ir s t m e asu rem en t from the value of the second
m e asu re m e n t.
D eterm ination of increm ent in d . b . h . o . b . when the tr e e s a re not

identified individually. In this c a s e , the in crem en t on each tre e over
the p eriod betw een m easu rem en ts cannot be derived. T here a re only
two lis ts of tre e size s available - one at the beginning and one at the
end of the p erio d - and it is from th ese that in crem en t m ust be
d eriv ed. F o r this purpose, the lis ts a re usually com piled as stand
ta b le s .
What h a s becom e known as the stan d ard o r c la s s ic a l m ethod of deriving
in c re m en t from stand ta b le s of su cc e ssiv e m e asu re m e n ts was developed
in w e ste rn Europe many y e a rs ago and is outlined by M eyer (1953).
It is illu s tra te d for a sm all plot of P inus ra d ia ta in Table 17:2 w here
two asp e c ts have been om itted for sim p licity - how to deal with n atu ral
and a rtific ia l re m o v a ls, and with ingrow th (recru itm en t); th ese have
been included in another illu stra tio n in Table 17:7 for c y p re ss pine.
The b a sis of the method (Table 17:2) is to d eterm in e for each size
c la s s of the f ir s t inventory how m any tre e s ris e into higher size c la s s e s
in the second inventory and how many rem ain in the c la s s . O r, put
another way, to determ in e from what c la s s e s of the f ir s t inventory the
tre e s in c la s s e s of the second inventory have com e. T re e s ris e
according to the am ount of in crem en t and the width of the c la s s . If
only som e tr e e s r is e , som e a re left in the c la s s , th at is they are
’s ta tio n a ry '. If all the tre e s r is e , th e re a re no sta tio n ary tr e e s . If
the tre e s of a c la s s grow through the next size c la s s th e re will be a
negative num ber of tre e s statio n ary .
The stand tab les a re a rra n g ed in descending o rd e r of c la s s size .
The calculation of the num ber of tre e s risin g and sta tio n a ry begins with
the la rg e s t c la s s . The two 22 in c la ss tre e s of the 1959 inventory a re

assu m ed to have ris e n from the 21 in c la ss of the 1953 inventory.
Since th e re w ere none in the 21 in c lass in 1953 th is m eans -2 w ere
sta tio n ary in th at c la s s over the period 1953-9. The six 21 in c la s s
tr e e s of the 1959 inventory a re assum ed to have ris e n from the 20 in
T ab le 17:2
I n c r e m e n t in d . b . h . o . b . fro m stand tables of successive m e a s u re m e n ts ; stan dard m eth o d ;

P. r a d i a ta , | acre plot, Mt Stromlo, A . C . T .
(1) (2) (3) (4) (5) (6) (7) (8)

D .b . h . o .b . No. of trees No. of trees Trees Trees Double Double D R /D E
class 1953 1959 rising station ary rising eff e c tiv e (increm ent
(in) in v e n to ry in v en to ry (DR) (DE) in units of
class
in terv als for
th e period)
22 0 2 0 2 2 100
2
21 0 6 -2 10 6 1-66
8
20 0 3 -8 19 3 6- 33
11
19 2 10 -9 30 12 2-50
19
18 6 13 -1 3 45 19 2-37
26
17 8 10 -1 8 54 18 3 00
28
16 13 13 -1 5 56 26 2- 16
28
15 13 11 -15 54 24 2 -2 5
26
14 20 6 -6 38 26 1 -46
12
13 9 3 -3 18 12 1 -50
6
12 6 0 0 6 6 100
c la s s of the 1953 inventory; since there w ere -2 tre e s statio n ary in

the 21 in c la s s , a total of 8 tre e s m ust have ris e n from the 20 in c la s s .
Since th e re w ere in fact none in the 20 in c la s s in 1953. th is m eans
th e re w ere -8 sta tio n ary in that c lass over the period. The th re e 20 in
c la s s tre e s of the 1959 inventory a re assum ed to have rise n from the
19 in c la s s of the 1953 inventory; since th e re w ere -8 tre e s statio n ary
in th at c la s s , a total of 11 tr e e s m ust have ris e n from the 19 in c la s s .
Since th e re w ere only 2 tr e e s in that c la s s , th is m eans th e re w ere -9
In c re m e n t and Y ield 151
t r e e s s ta tio n a ry in th e c la s s . And so on.
The v a lu e s in colum n 6, headed double r is in g , fo r each c la s s a r e
d e riv e d by adding th e n u m b e r of tr e e s r is in g out of and into th e c la s s .
F o r e x a m p le , fo r th e 22 in c l a s s , the v alu e 2 in colum n 6 is given by
2 r is in g fro m the 21 in c la s s to th e 22 in c la s s p lu s an im p lied z e ro
r is in g out of the 22 in c la s s ; fo r th e 21 in c l a s s , the valu e 10 in
colum n 6 is given by 8 r is in g fro m the 20 in c la s s to th e 21 in c la s s
plus 2 r is in g out of th e 21 in c la s s ; fo r the 20 in c la s s the valu e 19 in
colum n 6 is given by 11 r is in g fro m th e 19 in c la s s to th e 20 in c la s s
plus 8 r is in g fro m the 20 in to th e 21 in c l a s s . The v alu es in colum n
7, h ead e d double e ffe c tiv e , fo r each c la s s a r e d e riv e d by adding the
n u m b e rs of t r e e s in th e c la s s a t each in v e n to ry . F o r e x am p le, fo r
th e 19 in c l a s s , the value 12 in colum n 7 is given by 2 tr e e s in th e
1953 in v e n to ry plus 10 t r e e s in th e 1959 in v en to ry ; fo r the 18 in c la s s ,
the v a lu e 19 in colum n 7 is given by 6 tr e e s in th e 1953 in v e n to ry plus
13 t r e e s in the 1959 in v e n to ry . And so on.
The v a lu e s in colum n 8, double r is in g (DR) of colum n 6 d ivided by
double e ffe c tiv e (DE) of colum n 7, r e p r e s e n t in c re m e n ts in te r m s of
in te r v a ls of d ia m e te r c la s s fo r the p e rio d . D ividing double r is in g by
double e ffe c tiv e is e q u iv alen t to dividing th e a v e ra g e of th e n u m b e rs
of tr e e s ris in g r e s p e c t i v e ^ into and out of th e c la s s d u rin g th e p e rio d
by th e a v e ra g e of the n u m b e rs of t r e e s re s p e c tiv e ly at th e beginning
and end of th e p e rio d , a d iv is o r of 2 b eing im p lie d in both n u m e ra to r
and d e n o m in a to r of the fra c tio n . P e rio d ic m ean annual in c re m e n t is
given by m u ltip ly in g th e v a lu e s of colum n 9 by th e w idth of the c la s s
in te rv a l (h ere one inch) and dividing by th e len g th of th e p e rio d in y e a r s
(h ere 6 y e a r s ) ; th is h a s n o t been done in T ab le 17:2.
A m o d ificatio n of th is m eth o d , p ro p o se d by S m ith e rs (1949), is
illu s tr a te d in T able 17:3, u sin g th e sa m e d a ta as fo r T ab le 17:2. In
th is m eth o d , the n u m b er of tr e e s r is in g is d e te rm in e d w ithout
d e te rm in in g 't r e e s s ta tio n a r y '. C olum ns 4 and 5 r e p r e s e n t cu m u lativ e
su m s of the n u m b e rs of t r e e s fro m th e la r g e s t s iz e c la s s dow nw ards.
The n u m b e r of tr e e s r is in g into any s iz e c la s s is given by th e d iffe re n c e
betw een the c u m u lativ e n u m b e r of tr e e s shown in the c la s s fo r th e two
in v e n to rie s , fo r ex am p le th e v alu e 26 shown in colum n 6 a s r is in g up
to the 18 inch c la s s is given by 34 in th e 1959 in v en to ry m inus 8 in the
1953 in v e n to ry . The v a lu e s fo r double r is in g , double e ffe c tiv e , and
th en ce p e rio d ic m ean annual in c re m e n t, a r e c a lc u la te d in the sam e
way a s in the sta n d a rd m ethod. T h ese v a lu e s a r e . of c o u r s e , th e sam e
in the tw o m eth o d s.
152 O utline of F o r e s t M en su ra tio n
T a b l e 17:3
In c r e m e n t in d . b . h . o . b fro m stand tab les of successive m e a s u re m e n ts ; S m ithers's

m o d i f i c a ti o n of the stan d ard m eth o d ; P. r a d i a ta , ^ acre plo t, M t Stromlo, A . C . T .
(s am e v a lu e s as T a b le 17:2)
(1) (2) (3) (4) (5) (6) (7) (8) (9)

D .b .h .o .b . No. of No. of C u m . no. C u m no. Trees D ouble Double DR / DE
class trees trees of trees of trees rising rising eff e c tiv e (in c re -
(in) 1953 in - 1959 in- 1953 in - 1959 in - m e n t in
v e n to r y v e n to r y v e n to r y v e n to r y units of
class in
terv als for
the perio d)
22 0 2 0 2 2 2 1-00
2
21 0 6 0 8 10 6 1-66
8
20 0 3 0 11 19 3 6 -33
11
19 2 10 2 21 30 12 2-50
19
18 6 13 8 34 45 19 2 -3 7
26
17 8 10 16 44 54 18 3 -0 0
28
16 13 13 29 57 56 26 2 -1 6
28
15 13 11 42 68 54 24 2 -2 5
26
14 20 6 62 74 38 26 1-46
12
13 9 3 71 77 18 12 1-50
6
12 6 0 77 77 6 6 1 -00
0
F e rg u so n (1952) d rew atten tio n in E n g lish to th e m ethod p ro p o sed

by P ro d a n (1947) w hich d e riv e d co m in g p e rio d and p a s t p e rio d
in c re m e n t s e p a r a te ly , as d is tin c t fro m th e sta n d a rd m ethod which
p ro v id e s a w eighted a v e ra g e of th e two a s outlin ed above. The
m e c h a n ic s of th is m ethod a r e illu s tr a te d in T ab le 17:4 u sin g the sa m e
d ata a s T ab les 17:2 and 17:3. The two stan d ta b le s a r e laid out in
th e fash io n of T ab le 17:4. L in es a r e draw n a c r o s s each colum n of
d ia m e te r c la s s of th e f i r s t in v e n to ry o p p o site a p o sitio n b etw een th e
row of the sam e d ia m e te r c la s s and th e one above it in the second
in v e n to ry , fo r ex am p le a lin e is d raw n in th e 19 in colum n of the 1953
in v e n to ry h o riz o n ta lly o p p o site th e p o sitio n betw een the 19 in and 20
in c la s s of the 1959 in v e n to ry . The c a lc u la tio n of th e m o v em en t of
In c re m e n t and Y ield 153
T a b le 17:4
I n c r e m e n t in d . b . h . o . b . from stand ta b le s of successive m e a s u re m e n ts ; Prodan's m e th o d ;

P. r a d i a t a , \ acre plot, Mt Stromlo, A . C . T . ( s a m e values as T ab les 17:2 and 17:3)
D ia m . dist. D i a m e t e r d istribution 1953 in v en to ry Inert, for past

1959 in v e n to rv 19 18 17 16 15 14 13 12 6 - y e a r perio d
Class N o. 2 6 8 13 13 20 9 6 T o t. per A v. per
(in) class class
22 2 2 6 3*00
21 6 6 18 3 -0 0
20 3 ___ 3 9 3 -0 0
19 10 _ 5 5 25 2 -5 0
18 13 ___ 8 5 31 2 -3 8
17 10 _ 8 2 22 2 -2 0
16 13 _ 13 26 2 -0 0
15 11 5 6 17 1-55
14 6 3 3 9 1-50
13 3 3 3 1-00
12 0
Increm ent
for the c o m i n g
6 y e a r perio d
(i) to t. for
class 6 18 19 31 31 37 15 9
(ii) av. for

class 3 -0 30 2- 37 2 38 2 38 1-85 1 66 1-50
t r e e s s t a r t s w ith th e h ig h e st c la s s of the second in v e n to ry . The 2

t r e e s in the 22 in c la s s (1959) a r e a ssu m e d to have com e fro m th e 2
tr e e s in the 19 in c la s s (1953) and a r e p la ced a s show n. Since th ey
a r e th r e e 1 in c la s s e s 'above th e lin e ' th ey r e p r e s e n t th r e e 1 in
in c re m e n ts , a to ta l fo r th e 19 in c la s s of 3 in x 2 shown below a s 6 in
in c re m e n t fo r th e co m in g p e rio d . The 6 tr e e s in th e 21 in c la s s
(1959) a r e a ssu m e d to have com e fro m th e 6 tr e e s in th e 8 in c la s s
(1953) and a r e placed a s show n. Since th ey a r e th r e e 1 in c l a s s e s
'above the lin e ' th ey r e p r e s e n t th r e e 1 in in c re m e n ts , a to ta l fo r the
18 in c la s s of 3 in x 6 shown below a s 18 in in c re m e n t fo r the co m in g
p e rio d . The 3 t r e e s in th e 20 in c la s s (1959) a re a ssu m e d to h av e
com e fro m the 8 tr e e s in th e 17 in c la s s (1953) and a r e p laced a s
shown. T his le a v e s 5 to acco u n t fo r in th e 17 in c la s s (1953). The 10
t r e e s in the 19 in c la s s (1959) a r e a ssu m e d to be th a t 5, and 5 of the
13 tr e e s of the 16 in c la s s (1953), w hich le a v e s 8 of th e 16 in c la s s
(1953) to acco u n t f o r. th e s e b ein g a co n trib u tio n to th e 13 of the 18 in
c la s s (1959); and so on. F o r th e 17 in c la s s (1953). th e re a r e 3 tr e e s
w ith th re e 1 in in c re m e n ts (being 3 sp a c e s o r c la s s e s 'above the lin e ')
and 5 t r e e s w ith two 1 in in c re m e n ts , a to ta l of 19 fo r the c la s s . Any
tr e e s 'below the lin e ' in d icate th ey a r e s ta tio n a ry in th e c la s s (th e re a r e
none in th is illu s tra tio n ).
To e x p re s s in c re m e n t on a p a s t p e rio d b a s is , tak e a s an exam p le th e
18 in c la s s of th e 1959 in v en to ry : th e 13 tr e e s in th e 18 in c la s s (1959)
re p r e s e n t 8 of th e 16 in c la s s (1953) w hich have m oved 2 c la s s e s w hich
eq u als 16 in ch es of in c re m e n t and 5 of th e 15 in c la s s (1953) w hich
have m oved 3 c l a s s e s w hich eq u als 15 in of in c re m e n t, a to ta l of 31 in
of in c re m e n t a s shown in th e seco n d la s t colum n on th e rig h t.
All th e s e m eth o d s a r e b a s e d on th e a s s u m p tio n s th a t the tr e e s a r e
evenly d is trib u te d th ro u g h o u t a d ia m e te r c la s s and th a t th e tr e e s in a
c la s s grow at the sa m e r a te . M ey er (1953) p o in ts out th a t the f i r s t of
th e se a ssu m p tio n s is h a rd ly e v e r fu lfilled and d is c u s s e s an a p p ro p ria te
a d ju stm e n t when it is not fu lfilled . The seco n d a ssu m p tio n also is
r a r e l y fu lfilled but le a d s to v e ry little e r r o r . T h e re is no a lte rn a tiv e
to a ssu m in g th a t th e la r g e s t t r e e s of the seco n d m e a su re m e n t w e re the
la r g e s t of the f i r s t m e a s u re m e n t. The tr e e 'm o v e m e n t' m u st s t a r t
so m ew h ere and. lack in g know ledge of how th e tr e e s a c tu a lly g rew , th is
a ssu m p tio n is the lo g ical one to m ak e. H ow ever, the assu m p tio n m ay
be invalid a s in th e c a s e illu s tr a te d in w hich the t r e e s w ere in fact
id en tified ind iv id u ally in th e field ; th e ir a c tu a l m o v em en t is shown in
T able 17:5 in p a r e n th e s e s fo r c o m p a riso n w ith th e a ssu m p tio n s of
P ro d a n 's m ethod. T he a c tu a l in c re m e n t is c o m p a re d in T able 17:6
w ith th a t d eriv e d by th e s ta n d a rd m ethod and P r o d a n 's m ethod. In each
c a s e th e in c re m e n t is in u n its of c la s s in te rv a ls (1 in) fo r th e p erio d
(6 y e a rs ) w hich can be co n v e rte d to in c re m e n t p e r tr e e p e r y e a r if
re q u ire d .
In p e rio d ic m e a s u re m e n t, a lo w e r lim it to d . b . h . o . b . m e a s u re m e n t
is u su a lly n om in ated fo r p r a c tic a l re a s o n s . T r e e s of a s iz e below th is
lim it a t one m e a s u re m e n t m ay grow to a s iz e above th is lim it by the
next m e a s u re m e n t. T h e se tr e e s a r e c a lle d in g ro w th . B etw een
m e a s u re m e n ts t r e e s m ay d ie - m o rta lity , o r be rem o v e d - re m o v a ls .
The w ay in w hich in grow th and re m o v a ls a r e taken into account is
T a b le 17:5
In c r e m e n t in d . b . h . o . b . from stand tab les of successive m e a s u re m e n ts : v a lu e s deriv ed by

Prodan's m e t h o d (T a b le 17:4) c o m p a r e d w ith a c t u a l v a lu e s (in parentheses)
D i a m . dist. D i a m e t e r distrib u tion 1953 inven to ry

1959 in v en to ry 19 18 17 16 15 14 13 12
c l a ss No. 2 6 8 13 13 20 9 6
(in)
22 2 2 (2)
21 6 (2) 6(3) (1)
20 3 ________ _ 3(3)
19 10 _1LL 5(4) 5(4) (1)
18 13 -
8(9) 5(4)
17 10 8(7) 2(3)
16 13 13(10) (2)
15 11 5(6) 6(4) (1)

14 6 3(3) 3(2)
13 3 J ill
12 0
illu s tra te d in Table 17:7 fo r c y p re ss pine. A m easu rem en t of g .b .h .o .b .

was c a r rie d out in 1946 and in 1958, the num ber of tre e s by 2 in
g .b .h .o .b . c la s s e s being shown in colum ns 2 and 5 re sp ec tiv e ly . The
num ber of rem ovals due to c o m m ercial logging and tre a tm e n t in each
g . b .h .o .b . c la ss during the period betw een the m e asu re m e n ts is shown
in colum n 4. Depending on when the rem o v als a re m ade during the
p erio d betw een m e asu re m e n ts, and when tr e e s a re re c o rd ed as having
died, the stand tab les can be adjusted sensibly for calculation of
in crem en t. Had the rem o v als all been m ade im m ediately a fte r the
f ir s t m easu rem en t they could be su b tracted from the stand table of
th at m easu rem en t. Had they a ll been made im m ediately before the
second m easu rem en t they could be added to the stand table of that
m e asu re m e n t. They w ere made continuously over the p erio d , so it is
reaso n ab le to su b tra ct half the rem o v als p e r g .b .h .o .b . c la s s from
the 1946 stand table and add half to the 1958 stand ta b le . The re su ltin g
stand ta b le s a re shown in colum ns 3 and 6. T hese would then be used
T ab le 17:6
I n c r e m e n t in d . b . h . o . b . from stand ta b le s of successive m e a s u re m e n ts; co m p a riso n of

a c t u a l in c r e m e n t w ith t h a t d e riv e d by stan dard m e t h o d (T a b le 17:2) and Prodan's m e t h o d
(T ab le 17:4)
Inert, stan dard Inert. Prodan A c tu a l in e rt,

D ia m m ethod m e th o d (in) (in)
class (in) C o m in g Past C om in g Past
(in)
p eriod period period period
22 1 -0 0 3 -0 0 3 -9 0
21 1 -66 3 -0 0 2-91
20 6 -3 3 3 -0 0 2 -6 3
19 2-50 3 -0 0 2 -5 0 2 -35 2-49
18 2-37 3 -0 0 2 -3 8 3-12 2-29
17 3 -0 0 2 -3 7 2 -2 0 2 -6 9 2 -2 7
16 2-16 2 -3 8 2 -0 0 2 -2 3 2 -0 8
15 2 -2 5 2 -3 8 1-55 2 -3 0 1-51
14 1 -4 6 1-85 1-50 1-80 1-48
13 1 -5 0 1 -6 6 1-00 1-91 1-13
12 1 -0 0 1 -5 0 1-61
in th e s ta n d a rd m eth o d . S m ith e r s ’s m ethod, o r P r o d a n 's m ethod. The

ingrow th is re p r e s e n te d by th e d iffe re n c e betw een the to ta l n u m b e r of
tr e e s in the two m e a s u r e m e n ts , tak in g re m o v a ls into acco u n t as shown
in T ab le 17:7. In c a lc u la tin g in c re m e n t by any of the m e th o d s, th is
v alu e (2. 017 in th is c a se ) would have to be shown a s th e n u m b e r of
t r e e s of g irth c la s s ’le s s th an 21 in c h e s ’ fo r a b alan ce to be effected .
The e x p re s s io n of in c re m e n t in d . b . h . o . b . W h atev er the m ethod of

c a lc u la tin g in c re m e n t, a re la tio n s h ip is u su a lly so u g h t, by g ra p h ic a l
m e th o d s o r c a lc u la tio n of an a p p ro p ria te eq u atio n , betw een in c re m e n t
and tr e e s iz e (e ith e r at th e beginning o r end of the p e rio d ), o r betw een
tr e e s iz e at beginning and tr e e s iz e a t end of th e p e rio d . T h ese
re la tio n s h ip s a r e v a rio u s ly lin e a r o r c u r v ilin e a r , w ith p o sitiv e o r
n eg ativ e c o r r e la tio n , depending on su ch f a c to rs a s th e grow th h a b its of
th e s p e c ie s , age and s tr u c tu r e of stan d . A s an ex a m p le , d ata of an
unthinned stan d of ra d ia ta pine a t Kowen, A. C. T. , a r e illu s tr a te d in
T a b le 17:7
S tan d ta b le s of su ccessiv e m e a su re m e n ts o f cypress p in e (c o u rtesy Q ld D ep t For. )
(1) (2) (3) (4) (5) (6)

G. b . h . o. b. No. of tre e s N o. of trees R em o v als N o. of tre e s N o. of tre e s
(in) 1946 1946 1946-58 1958 1958
less h a lf of plus h a lf of
re m o v a ls re m o v a ls
21 1 ,4 1 7 1, 399 36 1, 527 1 ,5 4 5
23 1, 276 1, 260 32 1 ,4 5 5 1 ,4 7 1
25 1, 047 1, 032 30 1, 311 1, 326
27 871 860| 21 1 ,0 5 9 1, 0 6 9 4
29 649 639 20 804 814
31 543 534 18 733 742
33 413 404 18 572 581
35 330 3251 9 409 413?
37 231 1 8 l| 99 333 382j
39 140 100 80 170 210
41 93 694 47 115 138?
43 62 45 2 33 80 964
45 40 29 22 48 59
47 32 24 16 31 39
49 23 18 10 21 26
51 10 6 8 12 16
53 6 2j 7 7 io |
55 8 6 4 5 7
57 7 sb 3 4 5?
59
61 4
6 43 3
2 4
5 62
5
63 0 0 2 0 2
65 2 2 0 1 1
67 0 0 0 0 0
69 1 1 1
2 1 0 2
71 0 0 0 0 0
73 0 0 0 0 0
1
75 + 1 2 1 1
7, 212 522 8. 707
T o ta l 1958 8, 707
+ re m o v a ls 522
9, 229
less to ta l 1946 7, 212
2 ,0 1 7 = ingrow th
Figs. 17:1 a and b. The relatio n sh ip s between in crem en t in d . b . h . o . b .

and d . b . h . o . b . at the resp ec tiv e m e a s u re m e n ts, and between
d . b . h . o . b . at each m e asu re m e n t, a re lin e a r with a high positive
c o rre la tio n . As a fu rth e r exam ple, a c u rv ilin e a r relationship of
LO
CD
OJ ] -
Db.ho-b- 1956
Q 4-
D b .h o b 1956
Fig. 17:1 Increm ent in d .b . h. o . b . of P. radiata, Kowen, A . C . T .

(a) Increm ent 1956-1965 (in) = - 0 -7 5 + 0 * 3 6 d . b . h . o . b . 1956 (in)
(b) d . b . h . o . b . 1965 (in) = - 0 * 75 + 1-36 d . b . h. o. b. 1956 (in)
I n c re m e n t and Yield 159
d . b . h . o . b . in c re m e n t and d. b. h. o. b. f o r s e v e r a l eucalypt s p e c ie s
com b in ed (red and white m ahogany, tallow wood, and blue gum) fro m
P in e C r e e k State F o r e s t . N .S . W. is shown in Fig. 17:2.
-O
u
c
D.b.h.o.b. ( in.)
Fig. 17:2 Increment in d . b . h . o . b . of red and white mahogany, tallow wood,

and blue gum, Pine Creek State Forest, N . S . W .
When d . b . h . o . b . in c re m e n t is r e la te d to d. b. h. o. b. on a d ia m e te r
c l a s s b a s is and t r e e s grow m o r e than the width of a c l a s s in te rv a l
du rin g the p e rio d betw een m e a s u r e m e n t s , a d e c isio n h a s to be m ade
on the s iz e c l a s s to which the in c re m e n t should be c r e d ite d . If th e s e
t r e e s a r e c la s s ifie d ac c o rd in g to s iz e at the beginning of the p e rio d
like the r e s t , this keeps the p r o c e d u r e c o n s is te n t, but th e r e is an
a rg u m e n t fo r c la s s ify in g them ac c o rd in g to siz e at the m iddle of the
p e rio d .
D e c re m e n t in d . b . h . o . b .
One would expect a t r e e to i n c r e a s e in d . b . h . o . b . between
m e a s u r e m e n t s except th a t in old age it m ight r e m a in s ta tic .
S o m e tim e s, h o w ev er, the d . b . h . o . b . at a second m e a s u r e m e n t is l e s s
than at the previous m e asu rem en t. F o r plantation co n ifers in A u stra lia ,
th is may be the re s u lt of an actual d e c re a se in the d .b .h . of the wood
due to lack of w ater and the o v e r-b a rk m e asu re m e n t m ay tru ly re fle c t
the p re se n t d .b .h .u .b . The d ifferen ce in the two m easu rem en ts m ust
be re g a rd ed as a negative value in deriving av erag e in crem en t fo r a
c la s s . F o r many eu calypts, the d e c re a se is the re s u lt of lo ss of b a rk
due to annual o r p eriodic shedding and m e asu rem en t m ust be tim ed to
suit th is. It m ay. how ever, be an accid en tal lo ss and nothing can be
done about it. L oss of b a rk due to sudden change of environm ent,
a b rasio n by u n d e rsto re y p lan ts, d e lib e ra te burning of the litte r o r wild
fire m akes it difficult to get re lia b le e stim a te s of in crem en t for many
eucalypts.
Increm ent in B asal A rea

If the in crem en t in b a sa l a re a is re q u ire d by d ia m e te r c la ss o r b asal
a re a c la s s , in crem en t in d .b .h .o .b . is d eriv ed as d iscu ssed above
and then converted to b a sa l a re a in c re m en t. U sually b asal a re a
in crem en t is re q u ire d on a stand b a s is . It is then im m a te ria l w hether
the tre e s a re individually identified o r not, since the stand b asal a re a
is the sum of the b asal a re a of all the tre e s in the stand. It can be
re ad ily derived by determ ining the b. a. o .b . of each tr e e and re la tin g
the sum of the individual b asal a re a s to the siz e of the stand; or on a
p e r a c re b a sis using angle count m ethod. Since it can be determ in ed
sim ply and re lia b ly and fo rm s a good independent v aria b le for
re la tio n sh ip s with other stand v a ria b le s , stand b a sa l a re a is alm o st
invariably m easured in growth stu d ies and for re s o u rc e and m anagem ent
inventory. Increm ent in stand b a sa l a re a is th e re fo re one of the m ost
common ex p re ssio n s of in crem en t.
Increm ent in Height

E stim a te s of height in crem en t a re n e c e s sa ry fo r a study of the growth
p a tte rn of stan d s, th e ir resp o n se to tre a tm e n t, and the effects of site:
also for e stim ate s of future volum e by s e p a ra te p ro jectio n of d ia m e ter
and height. Increm ent in height is usually re la te d to in crem en t in
d .b .h .o .b .
The relatio n sh ip of height and d ia m e te r, the stand height c u rv e , was
d iscu ssed in C hapter 12. In well balanced uneven-aged stands of
co n ifers a stand height curve which does not change with tim e may
som etim es be assum ed. T his m ay also be a re a so n a b le assum ption
fo r a stand height curve of m erchantable height fo r c e rta in uneven-aged

eucalypt fo re s ts in A u stralia. F or th ese c a s e s , a p a rtic u la r height
in crem en t is reg ard ed as asso c iate d with p a rtic u la r changes in
d .b .h .o .b . In even-aged stands the stand height curve is continually
changing and the height in crem en t asso c iate d with in crem en t in d ia m e ter
in tr e e s of a p a rtic u la r size v a rie s according to the age and s tru c tu re
of the stand.
If the tre e s a re identified individually, the height in crem en t of each
tre e can be determ ined and re la te d to it d ire c tly . U sually average
height in crem en t p er d .b .h .o .b . c la ss is calcu lated and re la te d to the
d .b .h .o .b . of the f ir s t or second m e asu re m e n t, depending on w hether
com ing period o r p ast period in crem en t is re q u ire d . G enerally the
in crem ent is determ ined as the difference between heights m easu red
on su cc e ssiv e occasions. Jacobs (1939) d e sc rib e s an in stru m en t for
m easu rin g height increm ent d ire c tly w here th is can be reco g n ised , for
exam ple as an internode or annual shoot on Pinus sp ec ie s.
If the tre e s a re not identified individually, one way of determ ining
in crem en t in height for any d .b .h .o .b . c la ss is as follows:
1. Compile a stand-height curve fo r the f ir s t m easu rem en t.
2. R elate d .b .h . o .b . increm ent to d .b . h. o. b. for the d. b. h. o. b.
c la s s e s of the f ir s t m easu rem en t.
3. Compile a stand-height curve fo r the second m easu rem en t.
4. F o r any p a rtic u la r d .b . h. o. b. :
(a) re a d its height from 1
(b) d eterm ine the corresponding d . b .h .o .b . at tim e of second
m easu rem en t from 2
(c) read the height corresponding to (b) from 3
(d) get the difference between the heights of (a) and (c).
Increm ent in Volume

Increm ent in volume is the difference in volume at su ccessiv e
m e asu re m e n ts.
If a sam ple tre e method is used it is applied d ire c tly and
independently on each occasion.
T h ere a re sev e ra l p o ssib ilitie s if a one-w ay volum e table is used:
eith er a unique volum e-basal a re a line may be com piled independently,
o r an a p p ro p riate ta riff may be used on each occasion, o r a single
ta riff m ay be used as illu stra te d by M eyer (1953) fo r the method of
control.
If two-way volum e tables a re used, volum e is derived each tim e fo r
each tre e through its d .b .h .o .b . and height, or through a stand table
and a stand height curve.
If a stand volume table is used it is applied d ire c tly and independently
on each occasion.
Volume may be estim ated from a yield table each tim e but a yield
table is usually so arra n g ed that the in crem en t in volum e fo r a
p a rtic u la r age and site can be e stim ated d ire c tly .
Yield T ables
N orm al Yield T ables

A yield table (see p. 117) is a statem en t of v a ria b le s such as volume',
basal a re a , height, and num ber of tre e s p e r unit land a re a for stands
of v ario u s ag es, on s ite s of v ario u s productive cap acity , of an
even-aged fo re s t of one sp ecies which has developed n a tu ra lly , o r has
been m aintained a rtific ia lly , at a co n sisten t d en sity . It is a statem en t
of the h isto ry of developm ent of the stands of an even-aged fo re s t of
one sp ec ie s, com piled from re la tio n sh ip s, esta b lish e d g rap h ically o r
by calculation, between the stand v a ria b les as dependent v a ria b le s ,
and age and an ex p ressio n of the productive cap acity of the site as
independent v a ria b le s. The statem en t is usually p re sen te d in ta b u lar
form but since it is derived by graphical m ethods o r form ulae it may
be left in graphical o r alg eb raic form for convenience. Depending on
the am ount of inform ation available for the com pilation of the ta b le , on
the purpose of the table and the stand v a ria b le s co n sid ered of m ost
in te re s t to m anagem ent, so the form of the table v a rie s . It m ay be a
sim ple statem ent of stand volum e p e r a c re produced to v ario u s given
ages on various s ite qualities such as that by Lew is (1957) for ra d ia ta
pine in South A u stra lia . It may be a com prehensive statem en t of
num ber of tr e e s , m ean d .b .h .o .b . , stand b asal a re a , height, volum e
of m ain crop and thinnings, and in crem en t in v a ria b le s by age and site
quality such as th at by Hummel and C h ristie (1953) fo r v ario u s sp ecies
in G reat B ritain .
The f ir s t step in com piling a yield table is to e sta b lish a system of
classific atio n w hereby the fo re s t can be divided into a re a s of d ifferen t
productivity - site quality or site index c la s s e s . Data are*then
obtained from stands on the various s ite s at v ario u s ages fo r the stand
v a ria b le s which a re to be included in the tab le. R elationships betw een
In crem ent and Yield 163
the v a ria b le s and age a re then estab lish ed d ire c tly , o r in d irectly
through top height. The validity and efficiency of th ese relatio n sh ip s
depends on the h is to ric a l extent of the inform ation in the sam e way and
for the sam e re a so n s as the re la tio n sh ip of top height and age as
d iscu ssed in C hapter 15. The m ost s a tisfa c to ry inform ation com es
from p eriodic m easu rem en t, the le a s t s a tisfa c to ry from a single
m easu rem en t. Stem analysis inform ation may be used to sup p o rt e ith e r.
The re la tio n sh ip s a re usually estab lish ed by subjective graphical
m ethods in the sam e way and with the sam e p ro b lem s as the top
h eig h t-ag e re latio n sh ip s d iscu ssed p reviou sly . A good exam ple of this
approach is that d escrib ed by Hummel and C h ristie (1953, 1957). A
nu m b er of attem pts have been made to calcu late re la tio n sh ip s of b e st
fit to form ulae and a re re p o rte d in the lite r a tu r e . They a re not v ery
su cc e ssfu l unless the relatio n sh ip is of a sim ple type - which is not
com m on, and the inform ation is p a rtic u la rly com prehensive - which
is not u sual.
The purpose of a yield table is to provide e stim a te s of p re se n t yield
and future in crem en t and yield. T h ere a re two ways in which a yield
table fo r a p a rtic u la r species can be used to estim ate future yield. If
a sp ec ie s is in the f ir s t ro tatio n in an a re a and has not com pleted that
ro ta tio n , which is the case with m ost plantations of coniferous sp ecies
in A u stra lia , its h isto ry of growth up tp the p re s e n t tim e can be
com piled in yield table form . Likely future yields can be estab lish ed
by extrapolating the relatio n sh ip of volum e on age and site , guiding
th is extrapolation by the form of the p a st tre n d s, o r p erh ap s supported
by som e inform ation on the developm ent of the sp ecies elsew h ere. F o r
safety , such fo re c a sts should be s h o rt-te rm ones, about 5 y e a rs . As
the future becom es the p re se n t the actual developm ent of the fo re s t is
re c o rd ed . This becom es the b a sis fo r future fo re c a s ts by fu rth e r
ex trap olation. Eventually the whole h is to ry is re c o rd ed . If now a
second rotation of the sp ecies is esta b lish e d , the yield table of the
f ir s t ro tatio n can be used for lo n g -term fo re c a s ts of the likely h isto ry
of the second cro p , assum ing th e re is no change in the productivity of
the site , or that the growth tre n d s of the v ario u s s ite q u alities a re not
affected if th e re is a change in productivity.
Yield tab les a re of p ra c tic a l value in m anagem ent as a m eans of
quantitative e stim a te s. Since they afford read y com parison of the
developm ent of a sp ecies under a p a rtic u la r silv ic u ltu ra l reg im e in a
country with its developm ent under s im ila r o r d ifferen t re g im e s in
th at country or elsew h ere, they a re also of in te re s t in national and
international asp e c ts of policy and econom ics. Even if a yield table
has lim itations as a m eans of e stim a te , it may have co n sid erab le value
as a com pact h is to ry with im plications of im portance to silv icu ltu re
and m anagem ent.
Volume p e r a c re on a p a rtic u la r site at a p a rtic u la r age v a rie s
according to stand density. If stand density v a rie s , it is difficult to
e sta b lish re la tio n sh ip s of volum e with age for each productivity c la ss
for the yield tab le, even if the site classific atio n h as been su ccessfu lly
estab lish ed on the b a sis of stand top height - which is not affected by
varying density. Stand b a sa l a re a is of c o u rse also affected by varying
density and, as d iscu ssed in C hapter 16, is used as a m easu re of it.
The solution to th is problem in the e a rly developm ent of yield tab les
in w e ste rn Europe la s t century was to r e s tr ic t the com pilation of a
yield table to stands which w ere fully stocked and to which the nam e
norm al was given - unfortunately, because th is te rm in g en eral usage
im plies the usual condition when in fact fully stocked stands a re ra th e r
unusual. An e stim a te for a stand which was o th er than fully stocked
was then obtained by apptying a c o rre c tio n fa c to r to the 3rield table
according to the density of the stand, b asal a re a com m only being used
as a m e asu re of density for th is purpose. This approach was used fo r
fo recastin g in crem en t of e u c a l ^ t regrow th stands in southern T asm ania
(L aw rence, 1965).
V ariable Density Yield T ables

Except in c e rta in fo re s ts in w estern Europe which w ere fully
stocked a fte r a long h isto ry of intensive m anagem ent, fo re s te rs often
found it difficult to get enough data from fu lly -sto ck ed stands on which
to base a n o rm al yield tab le. O r, if they w ere able to com pile a
n o rm al ta b le , they found that c o rre c tin g the values for d en sities o th er
than th at im plied in the table often led to u n satisfa c to ry re s u lts .
Techniques had also been developed w hereby relatio n sh ip s with m o re
than two independent v a ria b le s (age and site) could re ad ily be handled.
T his led to the developm ent of yield tab les in co rp o ratin g a th ird
independent v a ria b le e x p re ssin g stand density to which the nam e
v a ria b le density yield ta b le s was given (S purr, 1952) and which have
receiv ed p a rtic u la r attention in N orth A m erica. A yield table
in corporating age, site , and density can be com piled for p ra c tic a lly
any kind of even-aged fo re s t. Spurr (1952) and Husch (1963) d iscu ss
the com pilation and application of v a ria b le density yield tab les in the
U .S .A .
Stand b a sa l a re a is the m ost com m only used index of stand density
for ev en-aged stands. A su itab le way of ex p ressin g the volum e-grow th
rela tio n sh ip is a stand volum e table in which stand volum e p er unit
stand b a sa l a re a is re la te d to stand top height. Lew is (1954) com piled
a v a ria b le density yield table on th is b a sis fo r ra d ia ta pine in New
Z ealand. C ro m e r (1961) d isc u sse d the m a tte r for ra d ia ta pine over a
wide ran g e of conditions and suggested 'the growth ra te of Pinus ra d ia ta
stan d s is approxim ately constant under a wide range of s ite , density,
tre a tm e n t and lo cality , when such ra te is e x p re ssed as the in c re a se in
u n d e r-b a rk volum e p e r sq u are foot of b a sa l a re a for unit in c re a s e in
top h e ig h t'. C arro n (1967) com piled a v a riab le d ensity yield table for
a plantation of ra d ia ta pine in the A. C. T. on the b a sis of th is
re la tio n sh ip of volume p er unit b a sa l a re a and stand top height. One
advantage fo r ra d ia ta pine is its apparent lin e a rity over a wide range
of conditions. C ro m er and Bowling (1961) found th at volum e p e r a c re
of ev en-aged stands of Eucalyptus obliqua regrow th in T asm ania could
be s a tis fa c to rily estim ated from a c u rv ilin e a r rela tio n sh ip estab lish ed
betw een volum e p e r unit b a sa l a re a and m ean height of dom inants.
Since m ean height of dom inants by age could also be used to classify
site index, th is allowed e stim a te s of volum e p e r unit b a sa l a re a by
age and site index to be m ade from the relatio n sh ip - in effect, a
v a ria b le density yield tab le.
O ther Solutions for M odification of N orm al Yield T ables

Johnston and B radley (1963) d iscu ss the use of a secondary
c la ssific a tio n of stands into yield c la s s e s according to m axim um MAI
of volum e through top height-age c u rv e s 'to enable local growth
d ifferen c es to be taken into account and th ereb y to co n v ert the g en eral
yield c la ssific a tio n into a local yield c la s s ific a tio n '. F o r each g en eral
yield c la s s th e re a re th re e production c la s s e s , one with the sam e ratin g
as the g en eral yield c la s s , the o th e rs being re sp e c tiv e ly below and
above the g en eral yield c la s s by one yield c la s s . The production c la ss
of a p a rtic u la r stand is a s s e s s e d , a fte r its g en eral yield c la s s has been
d eterm in ed by the top h eight-age c u rv e s, by the relatio n sh ip fo r the
stand betw een top height and (a) to ta l b a sa l a re a production o r (b) to tal
volum e production or (c) the m ean g irth of the 40 la rg e s t girthed tre e s
p e r a c re .
F o recastin g the Stand Table
The v a ria b le s of a yield table a re stand v a ria b le s . The value of a
fo re c a st of future volum e from a yield table m ay be enhanced if it is
supported by a fo re c a s t of the d .b .h .o .b . d istrib u tio n . If the d istrib u tio n
is n o rm al p rojection may not be difficult. If it is skewed and changes
the extent and d irectio n of skew ness with tim e p ro jectio n may be quite
difficult. Lew is (1963a) has re fe rre d to th is for ra d ia ta pine in A u stra lia.
C onsiderable w ork has been done on the m a tte r, som e of which is liste d
by Spurr (1952). In A u stra lia at p re se n t p ro jectio n s a re u su ally made
by subjective graphical m ethods.
The U sefulness of Yield T ables in A u stra lia

N orm al yield tab les a re suitable for m any co n ifer plantations in
A u stra lia . V ariable density yield tab les m ay be n e c e s sa ry for some
tim e fo r plantations w here the effects of e a rly estab lish m en t tro u b le s,
drought, in sect and fungal depredation have caused v ariatio n in density;
w here fluctuating dem and affects the planned thinning schedule; o r w here
thinning schedules cannot yet be defined o r the tre a tm e n t reg im e
stab ilise d . W hatever the kind of yield tab le, it provides a v e ry useful
m eans of estim atin g future in crem en t and y ield , esp ecially for
m anagem ent p urposes and policy planning. Age and site quality a re
the m ost im portant v a ria b le s influencing production and the silv ic u ltu ra l
re g im e s to achieve it. The age of a stand is usually known from the
re c o rd s and the s ite quality of a stand can be read ily d eterm in ed once
a system of site c lassific atio n has been com piled and mapping of site
quality, which becom es a perm anent stra tific a tio n for m anagem ent
p u rp o se s, has been c a rrie d out.
Stand P ro jectio n
To fo re c a st the probable yield from a stand at a future tim e we can
fo re c a st the condition of the stand at th at tim e by stand p ro je c tio n .
By this is m eant determ ining its p re se n t condition by an inventory,
fo rec astin g in crem en t in the future period on the b a sis of in crem en t
in the p a st period which has been derived by p erio d ic m e asu rem en t
o r stem a n aly sis, and adding th is futu re in crem en t to the p re se n t
condition of the stand. W hereas yield ta b le s can be com piled for and
applied to even-aged stands only, stand p ro jectio n can be applied to
any kind of stand.
P ro jectio n of D .b .h .o .b . : Stand Table P ro jectio n
T here a re a num ber of d ifferen t p ro c e d u res fo r stand table p ro jectio n
but with a common prin cip le which involves the com pilation of a stand
table for the p re se n t stand and adding to the d .b .h .o .b . values the
ap p ro priate in crem en ts to fo re c a s t the stand table at the future tim e ,
taking into account ingrow th, a rtific ia l re m o v a ls, and m o rtality . These
step s a re d iscu ssed in d etail by auth o rs such as W ahlenberg (1941) and
a re outlined h e re only in broad te rm s .
The p re sen t stand table is obtained by an inventory. A decision is
needed on the in crem en t values to be used - w hether on a d ia m e ter
o r b asal a re a b a s is , w hether com ing o r p a st p erio d , what fu rth e r
adjustm ents a re n e c e s sa ry . T h ere a re no g en eral ru le s for th is , the
chief facto rs being the stage of stand developm ent, the growth habits of
the species and the effects of a rtific ia l tre a tm e n t. The in crem en t for
each c lass has then to be added to the p re s e n t stand ta b le . The
sim p lest way is to add the d ia m e ter in c re m en t to the m id-points of the
d iam eter c la s s e s form ing the stand tab le. U nless the d iam eter
in crem ents a re exact m ultiples of d ia m e ter c la ss in te rv a l, which is
m o st unlikely, this p rocedure leads to a stand tab le with c la s s e s of
unequal width and m id-points which a re not round n u m b ers. This is
an awkward form of stand table but form ing a conventional stand table
with equal c la s s in te rv a ls and round num ber m id -p o in ts from it may
d estro y its validity.
An alternative is to assu m e th at the tr e e s in c la s s e s a re evenly
d istrib uted throughout the c la s s e s , a c e rta in pro p o rtio n w ill move out
of the class into higher c la s s e s , and a c e rta in p ro p o rtio n w ill stay in
the c la ss, th ese p roportions depending on the ra tio of the in crem en t
and the width of the d ia m e ter c la s s . T his ra tio is called the m ovem ent
fa c to r . F or exam ple, assum e th e re a re 10 tre e s in the 6-0 in c la ss
(5*50-5-49) with d ia m e te rs re sp e c tiv e ly 5* 5, 5 * 6 .........6*3, 6*4;
assum e the in crem en t is 0* 7 in; th e se tr e e s w ill becom e re sp ec tiv e ly
6*2, 6 3, 6*4, 6* 5, 6*6, 6*7, 6*8, 6* 9, 7*0, 7* 1; the ra tio is
0 *7
Y ~ ; th erefo re 70 p er cent should move up 1 in c la s s and 30 p e r cent
stay ir the c la ss - this is so as seen in the actual d istrib u tio n . A ssum e

th e increm ent is 1*4 in with 1 in c la s s e s ; the ra tio is i l l ; then 40
1 0
*
p e r c e it w ill move up two c la s s e s and the rem aining 60 p e r cent w ill

move up one c la s s , as seen from the d istrib u tio n which would be 6*9,
7*0, 7 1 , 7*2, 7*3, 7*4, 7*5, 7*6, 7* 7, 7*8, that is 4 tre e s (40 p er
cent) a re now in the 8 in c la s s , the other 6 a re in the 7 in c la s s . Some

common sense m odification of th is percen tag e has to be adopted when
th e re a re only a few tr e e s p er c la ss o r one tre e p e r c la s s . The use
of this m ovem ent fa c to r is d iscu ssed by W ahlenburg (1941) and M eyer
(1953). Two assum ptions a re involved. One is th at all the tre e s in a
c la s s grow at the sam e r a te , w h ereas th e re is usually a v a ria tio n of
in crem en t within a c la s s . The o ther is th at th e re is an even d istrib u tio n
of tre e s around the c e n tre of the c la s s , w h ereas an even-aged stand
has an approxim ately norm al d istrib u tio n over all the c la s s e s and an
uneven-aged stand has a r e v e r s e - J shaped d istrib u tio n . The effects
of th ese on the validity of the pro jectio n a re d iscu ssed by M eyer (1953).
The m ain problem in fo recastin g a stand tab le is to fo re c a s t what
shape the distrib u tio n is likely to take.
P ro jectio n of B asal A rea

F uture basal a re a can be determ ined in d irectly through d . b . h . o . b .
c la s s e s by stand table p ro jectio n , o r d ire c tly by g rap h ical p ro jectio n
o r m e re ly adding b asal a re a in crem en t to p re se n t b asal a re a .
P ro jectio n of Height
If a stand height curve fo r a future even-aged stand is re q u ire d from
which stand volum e can be determ ined by re fe re n c e to a tw o-w ay
volume tab le, the future height of a d . b . h . o . b . c la ss of a c e rta in size
now can be determ ined by adding a p p ro p riate in crem en t to its p re se n t
height, the future d . b . h . o . b . can be d eterm in ed by stand table
projection, the future height can be plotted ag ain st future d ia m e te r
and a stand height curve fitted. If fu tu re stand m ean height o r top
height is re q u ire d it can be fo re c a st by adding a p p ro p ria te in c re m en t
to p re se n t height.
F o r an uneven-aged stand with a constant stand height c u rv e , p re se n t
d . b . h . o . b . can be p ro jected and the equivalent height re a d from the
stand height curve.
P ro jec tio n of Volume

If a one-w ay volume table is applicable on su cc e ssiv e o ccasio n s,
the volume on su cc e ssiv e occasions can be d eterm in ed by p ro jectin g the
stand table and applying it to the volum e tab le. If d ifferen t one-w ay
tab les a re applicable, p rojection of both the stand table and the
v a ria b le s with which the ta riffs a re c o rre la te d is n e c e s sa ry . To apply
a tw o-w ay volum e table, d . b . h . o . b . and height m ust be p ro jected
se p a ra te ly . F o r a stand volum e tab le, stand b asal a re a and stand
height a re p ro je c ted sep a ra te ly .
Ingrowth
F o r stand table projection, inform ation m ust be obtained about the
frequency and increm ent of c la s s e s ju st below the size c la s s se t as the
low est for enum eration purposes if a re lia b le e stim ate is to be made of
ingrow th.
R em ovals and M ortality

The num ber and size s of a rtific ia l rem o v als can be anticipated from
the intended tre a tm e n t.
Inform ation on p ast m o rta lity m ust be obtained and taken into account
in stand table projection. The g e n eral p ro ced u re is to deduct fo recasted
m o rtality from the p re sen t stand table p rio r to p ro jectin g it o r a fte r
p ro jecting it. The p rocedure is sim p le, the difficulty is to fo re c a st
the m o rta lity .
G row th,P e rc e n tag e

The in crem en t of a tre e o r stand m ay be co n sid ered as a form of
in te re s t on cap ital. It is n a tu ra l to co n sid er in crem en t in th is way as
a b a sis fo r com paring the ra te of in te re s t of production with the ra te
of in te re s t chargeable against m oney invested fo r th at production.
M ethods of calculating in crem en t as in te re s t on a percen tag e b a sis ,
that is as growth p e rc e n ta g e , have been co n sid ered for a long tim e.
The la rg e num ber of form ulae proposed b e a rs w itness to the lim ited
use of the concept. E xpression of in crem en t as growth p ercen tag e
ap p ears to have som e m e rit when it is used for illu s tra tiv e p u rp o ses,
for exam ple to e x p re ss a p a rtic u la r growth condition by a ra te so as
to m ake com parison with som e stan d a rd o r expected ra te m o re c le a r.
It has co n sid erab le lim itations in fo recastin g the fu tu re condition of the
fo re st (see W ahlenburg. 1941 and S p u rr, 1952).
O ther Growth P ro jec tio n s
As an a lte rn a tiv e to stand projection fo r uneven-aged stan d s, th e re is
the p o ssib ility of fo recastin g future volume through equations, using
a s independent v a ria b le s those stand c h a ra c te ris tic s with which volume
is c o rre la te d and which can be pro jected re a d ily and re lia b ly . T u rn er
(1966) developed a m ultiple re g re ss io n in which volum e and b a sa l a re a
growth of an uneven-aged eucalypt fo re s t of m ixed sp ecies on the co ast
of N .S . W. a re fo re c a s t from various stand c h a r a c te ris tic s , and
provided a com prehensive bibliography of w ork of th is kind for both
even-aged and uneven-aged fo re s ts in the n o rth e rn h em isp h ere.
18 Forest Inventory
A fo re st re s o u rc e inventory is a q ualitative and quantitative investigation

and re c o rd of the location, ow nership, extent, n a tu re , condition,
purpose, and capacity of fo re s t e sta te on a bro ad sc a le , for exam ple
an inventory of the fo re s t a re a of a country, a state of the country, o r
p a rt of a sta te . Its purpose is to provide inform ation to the governm ent
for form ulating a land-use and fo re s t policy and for g en eral developm ent,
and the governm ent is re sp o n sib le fo r it. To plan the use of a co u n try ’s
re s o u rc e s and to develop and control tra d e in fo re s t p ro d u cts, the
governm ent needs to know w here the fo re s ts a re , who owns them , how
much th e re is of what type and in what condition, the ro le they play o r
can play in p rotection and production, am ounts of p re s e n t production
and potential production, ra te s of grow th and lo ss. A m anagem ent
inventory is a qualitative and quantitative investigation and re c o rd of
a p a rtic u la r fo re s t sufficiently detailed for efficient planning and
im plem entation of m anagem ent: the ow ner is resp o n sib le for it.
T h ere a re c e rta in s im ila ritie s betw een th ese two kinds of inventory
and in p ra c tice they grade into each o th e r. R eso u rce inventory co v ers
such a com paratively larg e a re a th a t a t le a s t the quantitative
investigation, and usually the qualitative investigation, is b ased on
sam pling. M anagem ent inventory in c e rta in lim ited c a se s is based on
a com plete d .b .h . enum eration of the fo re s t but in m ost c a s e s , as with
re so u rc e inventory, at le a s t the quantitative investigation is based on
sam pling. The sam e kind of sam pling units m ay be used in re s o u rc e
and m anagem ent inventory. If the re s o u rc e inventory is intensive and
the m anagem ent inventory not v e ry d etailed , p ra c tic a lly the sam e kind
of inform ation may be re c o rd ed . A gain, the inform ation from one kind
of inventory m ay be used in place of th at from the o th er because of lack
of funds, staff, or tim e for the o th e r.
The ideal sequence of events in the developm ent of fo re s try in a
171
country would s ta r t with a p re lim in a ry reco n n aissan ce of the location
and extent of the fo re ste d a re a which would indicate the m ost efficient
design of re s o u rc e inventory. A fter th is re s o u rc e inventory had been
done, the a re a s fo r intensive m anagem ent could be nom inated and
m anagem ent inventories of them c a rrie d out. U nfortunately, th e re is
probably no country in the w orld w here th is ideal sequence of events
has been possible and in som e c ase s the inform ation from a re so u rc e
inventory has to se rv e te m p o ra rily for the m anagem ent of an a re a ; in
o th e rs, the inform ation from m anagem ent inventory of typical a re a s
m ay be extrapolated to e stim a te s of the whole country at a re so u rc e
level.
Again, fo r e s te rs in m ost co untries have had to s ta r t m anaging
individual fo re s ts befo re a re s o u rc e inventory of the whole fo re s t estate
of the country has been c a rrie d out. M anagem ent inventory th e re fo re
has a longer h isto ry than re s o u rc e inventory, and som e of the techniques
of re s o u rc e inventory have com e from m anagem ent inventory ra th e r
than the other way around as one might expect. F u rth e r, knowledge
of the fo re s t re s o u rc e s of the country and in ten sity of m anagem ent
v a rie s from country to country, so that the two kinds of inventory may
usually be distinguished fa irly read ily within a country but often appear
indistinguishable betw een c o u n trie s.
F o r th ese re a so n s th e re seem s at tim es to be little difference between
th e se two kinds of inventory. In ideal c irc u m sta n c e s , at le a s t, they
a re e ssen tially differen t. They differ in function, in the kind of
inform ation obtained, and in the size of a re a involved. The p rim a ry
stra tific a tio n for re s o u rc e inventory is by broad fo re s t type, use and
ow nership. F or m anagem ent inventory it is usually by object of
m anagem ent o r the re q u ire m e n ts of silv ic u ltu re . The level of p recisio n
of the quantitative e stim a te s re q u ire d is usually h ig h er for m anagem ent
than re s o u rc e inventory, n e ce ssita tin g m ore intensive sam pling.
B ecause the differen ces a re g re a te r than the s im ila ritie s , the two kinds
of inventory will be d iscu ssed sep a ra te ly .
FOREST RESOURCE INVENTORY
A few c o u n tries of the w orld began org an ised fo re s t re so u rc e inventory

about 1920. Finland has c a r rie d out four su cc e ssiv e in v en to ries and
N orw ay and Sweden have each done th re e (N yssonen, 1964). At that
tim e , an Em pire F o re s try C onference s tre s s e d the need fo r
F o re s t Inventory 173
inventory in Commonwealth c o u n tries such as A u stra lia but little action
followed. An im petus to inventory in m any co u n tries of the w orld,
how ever, o c cu rred about 1946 for re a so n s m ainly connected with World
War II. One re a so n was that a e ria l photographs had played an enorm ous
ro le during the w ar in mapping and in te rp re ta tio n for m ilita ry p u rp o ses.
F o r e s te rs w ere aw are of th e ir value fo r fo re st re s o u rc e study and w ere
able to take advantage of the con sid erab le am ount of a e ria l photographic
co v erage, equipm ent, knowledge, and tra in e d m anpow er w'hich was
available a fte r the w ar. Another re a so n was th at dam age to the fo re s t
e sta te in many countries and cutting fo r the needs of w ar, voluntary
o r under occupation, prom pted national a p p ra isa l for planning p o st-w a r
developm ent. A fu rth e r reaso n w as that the estab lish m en t of the United
N ations O rganization strengthened the concept of fo re s ts as an
in tern ational re s o u rc e and stim ulated fo re s t re s o u rc e stu d ies,
p a rtic u la rly through agencies such as FAO. About th is tim e , and to
som e extent stim ulated by the need for cen su ses of v ario u s kinds
including th at of fo re s t re s o u rc e , extensive developm ents began in the
th eo ry of sam pling.
The decision to c a r ry out a fo re s t re s o u rc e inventory is the
g o v ern m en t's re sp o n sib ility , prom pted by the fo re s t au th o rity , which
is then m ade resp o n sib le fo r its im plem entation. T h ere a re m any
fa c to rs involved in th is decision, but th e se a re not of p re se n t in te re s t
and only the technical a sp ects of the inventory w ill be d iscu ssed .
M odern fo re s t inventory is the application of m ath em atical s ta tis tic s ,
a e ria l photogram m etry and p h o to -in terp retatio n , land surveying, tre e
m e asu re m e n t, and data p ro c e ssin g . L oetsch and H aller (1964) devote
400 pages to the f ir s t two a sp ects only and as they say 'fo re s t inventory
m ethodology .. . |~is now] a sp ec ia lised bran ch of fo re s try s c ie n c e '.
The tre a tm e n t of the subject h e re m ust n e c e s sa rily be both b rie f and
g e n eral, and som e fa m ilia rity with the v ario u s a sp e c ts involved is
assu m ed.
Mapping
The stan d ard of mapping needed fo r re s o u rc e inventory depends on the
scope and objectives of the inventory. P la n im e trie d etail is n e c e s sa ry
for planning the sam pling and executing the field w ork. The addition
of topographic detail m akes the m ap much m ore valuable for both th ese
p u rp o se s. If the esta te is to be c la ssifie d by ow nership, c a d a s tra l
detail is n e c e s sa ry . The map supporting the final statem en t should
include p la n im e tric , topographic, and c a d a s tra l inform ation if at all
p o ssib le. The position w ill v a ry from that w here adequate map coverage
is available to that w here th e re a re no m aps at a ll. Mapping is usually
the resp o n sib ility of sp e c ia list o rg a n isatio n s. If m aps a re not a lre ad y
available fo r fo re s t inventory, they m ay be sp ecially p re p a re d , o r m aps
m ay be provided with a common p la n im e tric , topographic, and c a d a s tra l
b ase to organ isatio n s in te re ste d in re s o u rc e stu d ies such as fo re s try ,
a g ric u ltu re , geology, and hydrology*, each superim posing its own
p a rtic u la r inform ation. In som e c a s e s , fo re s try has to provide its
own m aps for inventory*.
F or general planning, the sca le of m apping is usually* 1 in 250. 000,
1 inch to 4 m iles o r even 1 in 1. 000.000. The w orking m aps on which
the stra tific a tio n is re c o rd ed and the sam pling units selected , and from
which the units a re located in the field, a re usually* at a sca le of 1 in
25. 000 o r 1 inch to 20 chains. The m aps p re sen te d with the final
re p o rt a re usually the com m on civ il or military* map of the country at
a scale of 1 in 50. 000 or 1 inch to 1 m ile.
C lassificatio n and S tratificatio n

The main purpose of a re s o u rc e inventory* is to provide a statem en t of
the vario u s kinds of fo re s t in the country, how much of each kind th e re
is , w here it is , who owns it. what condition it is in and what ro le it is
playing o r can play*; and. fo r the production fo re s t, a statem en t of
p re se n t volume and ra te s of growth and lo s s. A sy*stem of c lassific atio n
is needed for th is. T h ere a re strong arg u m en ts fo r adopting an
in ternational c lassific atio n which would provide e stim a te s on an
in ternational b a s is , modifying it if n e c e s s a ry to su it dom estic
classific atio n ra th e r than the re v e rs e p ro c e d u re. The f ir s t step is to
nom inate the ’fo re s t' or 'fo re ste d la n d ', second to classify* it into c la s s e s
of p re se n t or potential use and into c ate g o rie s of ow nership. The th ird
step is a classific atio n into broad fo re s t types which may* be a sim ple
division into b ro ad -leav ed o r coniferous fo re s t, o r a m ore detailed one
to suit the p a rtic u la r conditions of each country* - for exam ple, the
types suggested fo r the national inventory of A u stra lia (F. and T .B .,
1951) a re 'coniferous plantation, c y p re ss pine, ra in fo re s t, mixed
fo re s t, eucalypt fo re s t, m a lle e, and m is c e lla n e o u s '.
F or som e policy* p u rp o se s, the c lassific atio n m ight stop at th is stage
but usually* e stim a te s of volum e of wood of the productive fo re s t a re
re q u ire d and th ese can be made and used m o re efficiently if some s o rt
of stra tific a tio n of broad fo re st types is c a rrie d out. Under c e rta in
c ircu m stan ces of sam pling, even a geom etric stra tific a tio n m ay im prove
the efficiency of the sam pling, but the g re a te s t contribution to efficiency
is usually given by a stra tific a tio n which red u ces the v arian ce of volume
within s tra ta to a m inim um . The b e st b a sis for s tra tific a tio n for this
purpose is g e n erally a n a tu ral one. The b e st stra tific a tio n for
estim ating volum e is often, but not n e c e s sa rily , the b e st for m aking
statem ents about volum e. The stra tific a tio n suggested for the
indigenous eucalypt fo re s t for the A u stra lian national inventory su its
both p u rp o ses, com prising a p re lim in a ry division into stand size c lass
(reg en eration, sap lin g s, poles, m atu re and o v e r-m a tu re ), then stand
height class (170 ft and o ver, 130-169 ft, 90-129 ft, 50-89 ft, 10-49
ft, and under 10 ft) and density c la s s (open, s c a tte re d , m o d erate,
dense).
C onsiderable thought has to be given to the kind and num ber of
c r ite r ia used in the c lassific atio n and the stra tific a tio n . T here should
be enough c la s s e s and s tra ta to be of p ra c tic a l value but not so many
th at tim e is spent unprofitable' delineating them . If the classific atio n
and stra tific a tio n a re based on in te rp re ta tio n of a e ria l photographs, the
c r ite r ia should be re ad ily distinguishable both on a e ria l photographs
and ground. The m inim um a re a of each c la s s and stra tu m which should
be delineated m ust be defined in te rm s of the a re a which w ill re p re se n t
it at the scale of the map fo r the final re c o rd . F o r the A u stralian
national inventory, m inim um a re a s of 80-160 a c re s fo r fo re st types
and 320-340 a c re s fo r density c la s s e s w ere recom m ended fo r final
m aps a t i sca le of 1 m ile to 1 inch. If p o ssib le, the stra tific a tio n
should be capable of expansion fo r subsequent m anagem ent inventory
if so re q iire d .
Sampling fo r Q uantitative E stim a te s

Q uantitative e stim a te s in fo re s t re s o u rc e inventory a re alm ost
invariably d eterm ined by sam pling. E stim a te s of a re a of the vario u s
c la s s e s of fo re s t m ay also in c e rta in c a s e s be d eterm in ed by sam pling.
In any discussion of inventory, a co n sideratio n of the th eo ry and
application of sam pling is th e re fo re n e c e s sa ry . Sampling is an extensive
and complex subject and w ill be d iscu ssed h e re only in elem en tary te rm s .
T here is a v a st lite ra tu re of which Y ates (1949), C ochran (1963), Finney
(1947, 1)48), and L oetsch and H aller (1964) provide good re fe re n c e s .
T h e re a re th re e g eneral objectives in sam pling for e stim ate s of a
v a ria b le such as stand volume in inventory: that the sam ple should
provide an unbiased e stim a te of the m ean of the population; th at th is
estim ate of the population m ean should be as p re c ise as p o ssib le for
the tim e and money expended; and th a t the sam ple should provide an
unbiased estim ate of v arian ce from which the p re c isio n of the e stim ate
(usually the standard e r r o r of the mean) m ay be a s s e s s e d . If the
inventory is to be c a rrie d out p e rio d ic ally to get inform ation on change
(growth and loss) at national lev el, an unbiased and p re c ise estim ate
of change may also be re q u ire d . The re q u ire m e n ts of sam pling for
th is may conflict to som e extent with the re q u ire m e n ts fo r the o ther
objectives.
As a p re lim in a ry c lassific atio n it is convenient to co n sid er sam pling
m ethods, or the m ethods of selectin g a sa m p le , as of e ith e r a subjective
o r objective kind.
Subjective Selection
In subjective, purposive or judgm ent se le c tio n , o r selectiv e sam p lin g ,
the se le c to r studies the population and then by p erso n al judgm ent se le c ts
what he co n sid ers to be ty p ical, re p re se n ta tiv e , o r average units of the
population. This method of selectio n is obviously liable to p erso n al
b ia s, the extent of which is not re a d ily evaluated. Subjective selectio n
may produce unbiased re s u lts when the population is sm all, the v ariatio n
is lim ited , average values e x ist, and the sele c to r is s u ffic ie n t^
experienced with the population to nom inate them . These conditions do
not obtain in inventory. B ecause it does not have the n e c e s sa ry elem ent
of random ness in the selectio n e ss e n tia l to the b asic m ath em atical
p rin c ip le s of sam pling m ethods fo r which th e o rie s have been developed,
'it is not am enable to the developm ent of a sam pling th e o ry ’ (C ochran.
1953) so that no a s s e ss m e n t of p re c isio n of estim ate can be c a r rie d
out. But a knowledge of the p re c isio n of the estim ate is often as
im portant in inventory as the e stim ate its e lf. H aphazard selectio n ,
in which the o p e ra to r does not m ake a d e lib e rate choice but em ploys
devices such as sticking a pin into a map o r a lis t while blindfolded,
has the sam e disadvantages as purposive selectio n . However w ell
intentioned, the selectio n is liab le to the p erso n al b ias of the sele c to r;
and the method is not ’am enable to the developm ent of a sam pling
th e o ry ’.
Subjective and haphazard selectio n a re th e re fo re not common in
inventory but have been used so m etim es with s a tisfa c to ry re s u lts .
Seely (1961, 1964), fo r exam ple, d isc u sse s the advantages,
d isad v antages, te s ts , and re s u lts , of using subjective selectio n of
sam pling units within s tra ta on a ir photographs; and of using the
'se le c te d lin e ' method w hereby 'e a s ily identifiable map fe a tu re s a re
sele c ted a t som e d istance from the stands and s tra ig h t lin es joining
them a re draw n through the stand o r stands to be sam pled. Sample
plots m ay be sele c ted at random on the lines o r m ay be placed at
re g u la r in te rv a ls th e re o n '. The selected line m ethod has been used
in A u stra lia. H a rriso n (1950) d isc u sse s the advantages and
disadvantages of se le c tiv e sam pling of a re a s of a p a rtic u la r
classific atio n on a map com piled from stra tific a tio n of a e ria l photographs
and then sam pling such a re a s by sam pling units.
O bjective Selection
In objective selectio n , the individual sam pling units a re selected by
a method in which chance may o r m ay not play a p a rt but which is not
influenced by the p e rso n al judgm ent of the s e le c to r. This group includes
a ll those m ethods which can be co n sid ered as b a sic a lly e ith e r random
selection o r sy ste m a tic selection.
Random selectio n . If a population fo r inventory is co n sid ered as

having a finite num ber of units, say N, sim p le (u n restricted ) random
sam pling is a m ethod of selectin g n units out of the N units so that
every p o ssible sam ple of n units has an equal chance of being chosen.
In p ra c tic e , the sam pling units a re sele c ted unit by unit independently
with equal p ro b ab lities of selection. The m ost s a tisfa c to ry way is to
assig n to each unit a num ber if the units can be lis te d , o r a set of
re c ta n g u la r c o -o rd in a te s if they com pose an a re a , and then to select
the ap p ro p riate quota of num bers m echanically from a table of random
num bers such as th at of F is h e r and Y ates (1943) which is an a rtific ia lly
com piled population of random ly d istrib u te d elem ents from which
m echanical selection gives a random sam p le. The sam pling units with
num bers co rresponding to those chosen then make up the sam ple. F o r
inventory p urposes the selection is usually done without re p la c e m e n t,
th at is a num ber draw n p reviously is ignored if draw n again
subsequently, on the grounds th at th e re is little point in having the
sam e unit m ore than once in the sam p le. This m ethod of selection
provides an unbiased e stim ate of the m ean and of the v arian ce but
unless the sam ple is well d istrib u ted o v er the population the p re c isio n
of the e stim ate of the mean m ay be low, p a rtic u la rly if th e re is
considerable v a ria b ility in the population and if, as is u su ally the c a se ,
the sam pling inten sity is low.
S tratification of the population and draw ing an ap p ro p riate pro p o rtio n
of the sam ple random ly from each s tra tu m , called s tra tifie d (re s tric te d )
random sam pling, is likely to be m ore efficient than sim ple random
selection from the whole population b ecau se it e n su res a m ore effective
d istribution of the sam pling un its. The stra tific a tio n m ay be based on
a rtific ia l g rid s, geographical divisions o r c a d a stra l b o u n d aries, but
since the efficiency of the sam pling in c re a s e s as the v a ria tio n of volume
within s tra ta is reduced, the m ost effective s tra tific a tio n is u sually one
of a ’n a tu ra l' kind using c r ite r ia with which volume is c o rre la te d . F o r
p recisio n of e stim ate to be a s s e s s e d , th e re m u st be at le a s t two sam pling
units p e r stra tu m , since no e stim a te of v a ria tio n can be m ade from the
evidence of one unit.
A m ajo r advantage of stra tifie d sam pling is that a v a ria b le sam pling
intensity am ongst s tra ta m ay be em ployed, a fractio n of the sam ple
being allotted to each stra tu m by som e sy stem of w eighting. In
proportional a llo c atio n , the sam ple is allo tted to each s tra tu m in the
proportion which the num ber of units in the stra tu m b e a rs to the to tal
num ber of u n its, th at is the sam ple d istrib u tio n is w eighted p u re ly by
a re a . In optimum allo c atio n , the sam ple is allocated so as to give the
sm a lle st stan d ard e r r o r of the e stim a te of the population m ean (the
highest p re c isio n of estim ate) p o ssib le fo r the p a rtic u la r size of sam p le,
the sam ple d istrib u tio n being w eighted by the product of the a re a of
stra tu m and the w ith in -stratu m stan d ard deviation of volum e. T his
ex p ressio n of v a ria tio n w ithin s tra ta is e stim ated from w h atev er evidence
is available, such as previous sam pling of s im ila r populations, o r by
p re lim in a ry sam pling. If co st of sam pling v a rie s from one stra tu m to
another, as is n o rm ally the c a s e , a fu rth e r weighting involving co st
p er sam pling unit in each stra tu m can be used to d eriv e the allocation
which gives m ost inform ation for the m oney to be spent. F o r v a rio u s
re a so n s, o v erall e stim a te s m ay be of le s s im portance than e stim a te s
for p a rtic u la r s tra ta in which case optim um allocation is ignored and
em phasis placed on sam pling those p a rtic u la r s tra ta so as to obtain a
d e sire d degree of p re c isio n of e stim a te fo r them . T h ere a re also
p ra c tic a l advantages in sam pling s tr a ta s e p a ra te ly - p rio rity of w ork
m ay be v a rie d as the inventory p ro ceed s to su it the w eath er, the
av ailability of m en fo r field and office w ork, o r a change of in te re s t
in s tra ta a s inform ation is accum ulated.
System atic s e le c tio n . In sy ste m a tic sam pling, the p a tte rn of the
position of the units of the sam ple re la tiv e to one an o th er is determ ined
in advance. U sually the p a tte rn is a re g u la r one, the units being spaced
an equal d istan ce a p a rt in one-dim ensional p a tte rn on a lis t o r in a line,
o r spaced in a sq u are or re c tan g u la r grid system in a tw o-dim ensional
p a tte rn . By the n atu re of such a p a tte rn , the position of ev ery unit is
d eterm ined once the position of one unit is d eterm in ed . It is th e re fo re
e sse n tia l th at the fram ew ork of the sam ple be placed by random selectio n
from all p o ssible positions (that is , the sy stem atic selectio n should
have a 'random s ta r t') otherw ise the e stim ate of the m ean is liable to
b ias.
If a stra tifie d random sam ple m ay be expected to show an im provem ent
in p recisio n over a sim ple random sam ple because of the im proved
d istrib u tio n of the sam pling u n its, it m ight then be expected th at a
sy stem atic distrib u tio n of units would also show an im provem ent, at
le a s t over sim ple random sam pling, and perhaps would be as p re c ise
as a corresponding stra tifie d random sam ple with one unit p e r s tr a tu m .
T his has been investigated fo r a rtific ia l and n atu ral populations as
d iscu ssed by Cochran (1963). N e v erth ele ss, w hatever the re la tiv e
p re c isio n of the e stim ate of the m ean from sy stem atic sam pling as
com pared with that from sim ple o r stra tifie d random sam pling, a
single sy stem atic sam ple does not provide an unbiased estim ate of
v arian ce and so the p re c isio n of the e stim a te of the population m ean
cannot be calculated. The conventional m ethod of calculating the e r r o r
of e stim ate re q u ire s the assum ption that every p o ssib le sam ple of the
size chosen h a s an equal chance of being chosen, w h ereas th ere is a
definite re s tric tio n to the rep etitio n of the sy stem atic sam pling p ro c e ss
according to the p a ttern and the sam pling intensity. System atic
sam pling has p ra c tica l advantages and under c e rta in conditions the
p re c isio n of the estim ate of the m ean m ay be re la tiv e ly high.
S tatistician s have th e re fo re investigated ways in which an approxim ate
e stim ate of p re c isio n m ight be made from a single sy stem atic sam ple
and the extent of the bias in an e stim ate of p recisio n calculated from
a sy stem atic sam ple as if the sam ple had been random ly selected
(O sborne, 1946; Y ates, 1949; Finney, 1947, 1948, 1950, 1953; and
M atern, 1962). V arious w o rk e rs such as Shiue (1960) have suggested
estab lish in g a num ber of sy stem atic sam p les, each being reg ard ed as
a c lu ste r of sam pling units, and estim atin g sam pling e r r o r from th ese
c lu s t e r s .
F o r fo re st inventory, s trip s a re usu ally spaced equidistant on a base
lin e and p lo ts are spaced equidistant in te rm s of re c ta n g u la r
c o -o rd in a te s. Seely (1961) d e sc rib e s te s ts in Canada of sam ple plots
equi-spaced in te rm s of p o lar c o -o rd in a te s, the design being ’of m ost
use when it is d e sira b le to p e n etrate the stands as fa r as p o ssib le from
the sta rtin g point [of the day’s w ork] and re tu rn d ire c tly to th at point
instead of making long offsets betw een the rad iatin g location lin es with
consequent lim ited p enetration during a day’s w ork’.
A com parison of stra tifie d random and sy ste m a tic sam p lin g . One of
th ese two m ethods of sam ple selectio n is n o rm ally used in re so u rc e
inventory. Several fa c to rs influence the choice.
In re so u rc e inventory, inform ation on the com position of the fo re st
is usually wanted and som etim es topographic inform ation as w ell. If
a e ria l photographs a re not available o r a re not adequate fo r
in te rp re ta tio n of fo re s t com position and for mapping topography, this
inform ation can b e st be obtained by sy stem atic sam pling. A fter map
com pilation, quantitative e stim a te s m ight be obtained by s tra tifie d
random sam pling but it m ay be m ore econom ic and convenient to make
quantitative e stim a te s while sy ste m a tic ally sam pling the fo re st
com position and topography. Even if photographs a re av ailab le, the
fo re s t m ay not lend its e lf to stra tific a tio n , o r boundaries of s tra ta m ay
not be re a d ily d isce rn ib le on the photographs, o r the e stab lish m en t of
random units within s tra ta m ay be difficult, in which c a se s sy stem atic
sam pling would be m ore p ra c tic a b le .
F o r p ra c tic a l location of units on the ground, the d ire c t n atu re of a
sy stem atic layout m akes it e a s ie r under m ost fo re s t conditions to
p re s c rib e , execute and check the location of u n its, and to re -lo c a te
them . V arious so u rc es of bias in the e s tim a te s , which m ay a ris e due
to problem s of locating random ly sele c ted u n its, m ay be avoided.
Provided the fram ew ork of selectio n is located by random choice of
all p o ssible p o sitio n s, an unbiased e stim ate of the m ean of the
population can be obtained from a sy ste m a tic sam ple as with a s tra tifie d
random sam ple - one of the m a jo r objectives of sam pling.
A second objective, that the evidence of a single sam ple provide an
unbiased estim ate of v a ria n ce from which the p re c isio n of the e stim ate
of tlm population m ean m ay be calcu lated , is m et by stra tifie d random
sam pling but is not m et by sy ste m a tic sam pling, though an
approxim ation to the estim ate of p re c isio n m ay be obtained in c e rta in
c irc u m sta n c e s.
A fu rth e r objective is th at the e stim a te of the population m ean should
be as p re c ise äs po ssib le fo r the tim e and money expended. It has
been shown th at under c e rta in c irc u m sta n c e s the p re c isio n of estim ate
m ay be higher fo r sy stem atic sam pling than stra tifie d random sam pling.
T h ere i s , how ever, no way of fo re c a stin g th ese c irc u m sta n c e s.
Sample S tru ctu re

If each unit in a population can be divided into a num ber of sm a lle r
elem en ts each of which can be fu rth e r subdivided, the sam pling may
be done by m u lti-sta g e sam pling. At the f ir s t stag e, p rim a ry units o r
lo c atio n s a re selected by som e su itable m ethod of selection; then a
sam ple of the seco n d -stag e units of the p rim a ry units is taken by the
sam e o r a differen t m ethod of selection; a sam ple of th ird -sta g e units
of the sec o n d -sta g e units is then selected ; and so on. Yates (1949)
d is c u s s e s the advantages and disadvantages of m u lti-sta g e sam pling.
In som e c a se s the re q u ire d inform ation is d eterm in ed from se v e ra l
sa m p le s. If the f ir s t sam ple pro v id es the b a sis for the re s t, so that
c e rta in inform ation is obtained from all the units of the sam ple, o ther
inform ation from only a sam ple of th e se , and o th er inform ation from
only a sam ple of th ese la tte r, each sam ple re p re s e n ts a phase of the
co llectio n of the wanted inform ation and th is is called m u lti-p h ase
sam pling.
The Sampling Units

Kind of u n it. T h ree m ain kinds of units a re used - the s tr ip , the p lo t,
and the point (spot) of the angle count m ethod. The te rm s trip (s trip lin e ,
tra n s e c t) is given to an a re a enclosed betw een p a ra lle l stra ig h t lin es
stre tc h in g from som e re fe re n c e boundary to an o th er, the distance
betw een the lines (width of strip ) depending on the method of sam pling,
the sam pling inten sity , the n a tu re of the topography and the fo re st
com position, being from i to 2 chains wide. U sually, fo r quantitative
e s tim a te s , the s trip is laid down by e stab lish in g one survey line, then
offsetting the whole p re sc rib e d width on one side o r half the p re s c rib e d
w idth on each side of the line; the offset m u st be exact b ecause the
quantity is re la te d to a re a . F o r qualitative inform ation such as
topographic and fo re s t type boundaries, the su rv ey line m e re ly acts
as a re fe re n c e for location. The s trip is so m etim es broken up into
re c o rd in g units by a b reak in the re c o rd of the inform ation to allow
e a s ie r supervision and checking. T his m ay also a s s is t in mapping
ty p e s. R ectangular plots m ay have e x tern al boundaries c lea re d and
m ark ed . They m ay also be estab lish ed by the sam e technique as a
s trip , that is a single line is surveyed and the boundaries of the plot
offset from it, the conventional lim it to the length being 10-15 chains
- anything longer is called a s trip . C irc u la r plots m ay not have any
ex tern al boundaries c le a re d , though^they m ay be indicated by m ark s
on boundary tr e e s . Often the units a re not re g u la r but of v ary in g size
with irre g u la r boundaries such as sm all rid g e s , c re e k s , tr a c k s , and
rid e s . F o r the angle count m ethod, the sam pling unit is identified
m e re ly by a point o r spot from which the c irc u la r sw eep, c h a ra c te ris tic
of the m ethod, is m ade.
V arious fa c to rs influence the choice of kind of unit. If inform ation
on topography and fo re s t com position cannot be obtained from a e ria l
photography and ground sam pling is n e c e s s a ry , s trip s a re m o st con
venient, the quantitative e stim a te s usually being m ade at the sam e tim e.
B ecause the ra tio of w orking tim e on units to tra v e llin g tim e between
units is g re a te r fo r s trip s than plo ts, a fo re s t can usually be sam pled
to a g re a te r intensity (by area) by s trip s than by plots in a c e rta in
tim e. However, the num ber of u n its, and so the num ber of d e g re es
of freedom in calculating sam pling e r r o r , is likely to be le s s and
th is m ay not be balanced by the reduction in the betw een-unit v a ria tio n .
W here the fo re st com position is such th at esta b lish in g p lo ts, p a rtic u la rly
random ly selected plo ts, involves a d isp ro p o rtio n ate am ount of tim e
and difficult 'navigation’, then s trip s a re g e n erally u sed . A com bination
of s trip lines and plots is som etim es used, topography and type
inform ation being re c o rd e d along sy ste m a tic a lly located s tr ip s , and
quantitative e stim a te s being made at in te rv a ls within p o rtio n s of these
s trip s . The te rm line plot is often used for th is a rra n g e m e n t. The
sam e te rm is som etim es used to indicate a sy ste m a tic location of
plots which is b e tte r d escrib ed as an aligned a rra n g e m e n t of plots.
Size of u n it. Size of unit has an effect on p re c isio n of e stim a te . A

com prom ise is usually n e c e s sa ry between a c e rta in num ber of la rg e
units which would provide a c e rta in v aria n ce betw een the units and a
g re a te r num ber of s m a lle r units which would in c re a s e the v aria n ce
between units but m ight d e c re a se the v a ria n ce of the e stim a te of the
m ean. F or a given num ber of units, the la r g e r the individual unit the
m ore p re c ise is the m ean likely to be on a unit a re a b a s is , but th is
depends on the way the unit is in c re a se d in size - g en erally , in c re a s e
in the long dim ension w ill give the g re a te s t gain, p a rtic u la rly if a
fe rtility gradient e x ists and if the unit is o rien ted along it as it should
be. The size of "the unit should b e a r a sen sib le re la tio n sh ip to the size
of the elem ents of the population. F o r exam ple, a unit of one a c re may
be n e c e s sa ry in a stand of m atu re eucalypts c a rry in g 10-15 tre e s p er
a c r e , but one-tenth a c re m ay be m o re suitable fo r a stand of poles and
a m il-a c re unit m ay be adequate fo r re g e n e ra tio n . In a fo re s t com posed
of m ixed species and range of s iz e s , concen tric units a re com m only
used . Too many concentric units m ay lead to undue com plications in
field w ork and office com putations and about th re e seem s a d e sira b le
lim it in p ra c tic e .
Though it is often stated that in p rin cip le the s m a lle r the unit the
m ore re lia b le m easu rem en t is likely to be through reduced opportunity
fo r m aking m istak es and through m ore intensive application by the
m e a s u re rs , it seem s in p ra c tic e that the unconscientious o p e ra to r is
not affected by the size of the plot and th at the conscientious o p erato r
is only affected a d v erse ly if the size of the unit is such as to prom ote
m ista k es how ever conscientious he m ight be.
The relatio n sh ip betw een size of unit and tim e -c o s t of m easu rem en t
is often but not n e c e s s a rily a d ire c t one. Once o p e ra to rs a re on the
spot and org an ised , in creasin g the amount of m e asu re m e n t u sually does
not p roportionally in c re a s e the co st.
If single plots a re used th e re a re many advantages in handling the
d ata if they a re all the sam e s iz e , and a co m prom ise is often made
betw een the la rg e s t unit which is p ra c tica b le in densely stocked stands
of sm all siz e s and the s m a lle s t w orthw hile unit in s p a rse ly stocked
stands of la rg e s iz e s . T here m ay be p a rtic u la r advantages in the field,
how ever, in having units of varying size and irr e g u la r shape bounded
by n atu ral fe a tu re s, p a rtic u la rly if the units a re p erm an en t, and th ese
advantages m ay outweigh aw kw ardness in handling the data.
Where tim e to get to a unit is disp ro p o rtio n ate to the tim e spent in
estab lish in g and m easu rin g it, c lu s te r sam pling is often em ployed, a
num ber of sub-units being estab lish ed as a group o r c lu s te r around a
p rim a ry sam pling point. T here a re o ther advantages as outlined by
L oetsch and H aller (1964). V arious p a tte rn s a re used. F o r exam ple,
L oetsch (1957) d e sc rib e s the use in Thailand of a num ber of ’c a m p s’,
each cam p com prising a c lu s te r of seven cam p u n its, each cam p unit
co m p risin g a c lu s te r of seven tr a c ts , each tr a c t being a sq u are each-
side of which was form ed of tw elve plots each 0- 05 h e c ta re . T ra c ts
of th is kind had been used in the Swedish National Inventory (M atern,
1962).
Shape and orientation of u n it. Sampling units a re u sually re g u la r in

shape. S trips a re long re c ta n g le s, plots a re usually re c ta n g u la r o r
c irc u la r. The la s t have the le a s t p e rim e te r fo r a sp ecific a re a and
so reduce the likelihood of w rong inclusion o r exclusion of tr e e s on the
boundary. In som e c a se s th e re a re p ra c tic a l advantages in estab lish in g
and identifying c irc u la r ra th e r than re c ta n g u la r p lo ts. W here th e re
a re silv ic u ltu ra l as w ell as p ra c tic a l advantages in identifying units by
topographic and a rtific ia l b o u n d aries, ir r e g u la r shaped units of varying
size a re som etim es used.
W here fe rtility grad ien t can be reco g n ised , units with a long axis in
the d irectio n of the g radient can usu ally be expected to give g re a te r
p re c isio n than plots of other shape and equal a re a . W here fe rtility
g radient cannot be reco g n ised , e stab lish in g sq u are o r c irc u la r plots
avoids the ris k of b ias which m ight o th erw ise a ris e by o rien tin g long
units other than along the grad ien t. S trips fo r qualitative d escrip tio n
a re usually d ire c te d at rig h t angles to the contours of the m ain
topography of an a re a since th e re is usually a good c o rre la tio n betw een
kind of fo re st and its position on a slope. T his usually su its
quantitative d escrip tio n as w ell because of a c o rre la tio n , p ositive o r
negative, between the c arry in g capacity of a fo re s t and its position on
a slope.
P erm an en ce of u n its . Changes in the fo re s t e sta te a re ju s t as im p o rtan t

as its state at any tim e . R eso u rce inventory, to be effectiv e, needs
to be c a rrie d out p erio d ically . W hether the sam pling units should be
perm anent or not fo r th is purpose is of s im ila r but p o ssib ly g re a te r
concern in m anagem ent inventory and is d iscu ssed la te r in th at
connection.
Size of Sample
The size of the sam ple m ay be e x p re sse d e ith e r as the num ber of
u n its, o r as sam pling in tensity - the a re a of the sam ple as a p ercen tag e
of the a re a of the population. The fo rm e r ex p ressio n is g en erally
p re fe ra b le because if two a re a s of differen t size have the sam e m ean
and v a rian ce they w ill re q u ire the sam e num ber of units fo r a p a rtic u la r
p re c isio n of e stim ate but th is w ill re p re s e n t quite d ifferen t sam pling
inten sity .
The v a ria b ility of the v a ria b le being estim ate d , the siz e and shape
of the sam pling u n its, and the m ethod of selectin g them , d eterm in e
the num ber of units needed fo r a p a rtic u la r p re c isio n of e stim a te .
T im e, m oney, and labour available determ in e the num ber of units
which can be estab lish ed . A com prom ise is often n e c e s sa ry between
w hat is d e sira b le and what c irc u m sta n c e s p e rm it.
Standard tex ts on s ta tis tic a l m ethods illu s tra te the method of
calculating the num ber of units needed fo r a p a rtic u la r p re c isio n of
e stim ate if the extent of the v a ria b ility is known. Inform ation about
the v a ria n c e s of v ario u s c la s s e s and s tra ta of an inventory is
accum ulated to th is end. In stra tifie d random sam pling, the b est
allo cation of the sam ple to the v ario u s s tra ta can be d eterm in ed , taking
into account both re la tiv e v a ria b ility and the re la tiv e co st of estab lish in g
units in the v ario u s s tra ta .
A decision has to be made on the p re c isio n re q u ire d . T his is an
ad m in istra tiv e decision and the fa c to rs affecting it w ill not be d iscu ssed
h e re , but the point is made th a t the decision seem s fa r too often to be
m ade without sufficient ratio n ale and a hig h er p re c isio n is called for
o r expected than is re a lly n e c e s s a ry fo r the p urpose of the inventory.
E stablishing the Sampling U nits on the Ground

O p erato r B ias in Locating the Units
In the system atic location of u n its, the random ly chosen sta rtin g
point of the sy ste m a tic p a tte rn is estab lish ed by su rv ey m ethod, using
a s a datum som e point which has been o r can be re fe re n c e d on a map.
S trip s a re set out equidistantly along a surveyed b ase line and plots
a re laid out at p re s c rib e d spacing along surveyed p a ra lle l lin e s. The
p re s c rip tio n s for the location of units a re w ell defined and easy to
follow and th e re is little source of b ias in location.
Units selected random ly a r e usually selected on p ap er (a map or
a e r ia l photograph), and th e ir positions m u st be tra n s la te d into the
field by survey m ethod, using re fe re n c e points which can be identified
both on paper and in the field. In random location th e re a re v ario u s
so u rc e s of b ias. S trips can be located by p re s c rib e d d istan ces along
a surveyed base but exact location of plots is often difficult - m aps
lack sufficient d e ta il, the fo re s t com position p rev en ts location of
fe a tu re s apparent on photographs, the quality of the photographs
p rev en ts re lia b le in te rp re ta tio n of them , o r the fo re s t shows no
distinguishing fe a tu re s. T here a re two common c a s e s . In one, the
o p e ra to r can locate h im self n e a r the re q u ire d position but not exactly
on it. If he then decides to locate the unit at a c e rta in distance in a
c e rta in directio n from w here he stands without d elib erately try in g to
influence the location th ereb y , he can m inim ise subjective bias in
location. In the o ther c a se , location m ust be made by survey method
over a considerable distan ce, from an identifiable fe a tu re , the
p re sc rip tio n for location by bearin g and distance being determ in ed in
advance. If the o p e ra to r ad h eres faithfully to th at p re sc rip tio n he can
avoid subjective b ias even if fa c to rs outside h is control p revent exact
location of the unit.
The p re sc rip tio n s m ust anticipate c irc u m stan c es w here the o p e ra to r
m ay feel prom pted to substitute positions with good intention but which
w ill be a source of b ia s. F or exam ple, a unit m ay fall in a 'm inim um
a r e a ' - an a re a of one kind, within a la rg e r stra tu m of a d ifferen t kind,
which is too sm all to be w orth delineating on a e ria l photograph o r m ap.
The o p erato r should be in stru c te d to locate the unit as planned but to
re c o rd the c irc u m stan c es as an explanation of why the unit m ight ap p ear
atypical. Again, a c c e ss to p riv ate land to estab lish a unit may be
denied, in which c a s e the o p e ra to r would n o rm ally be in stru cted not to
su bstitute a position but to om it it and re c o rd the circu m stan ce as
explanation of the om ission.
Edge Effects
The outer edge of a fo re st stand surrounded by open land has d ifferen t
m ic ro -lo c a lity conditions from those within the stand which may be
re fle c ted in d ifferent sp e c ie s, different growth ra te s of the sam e
sp ec ie s, and d ifferent tre e ta p e r. The change may be of a continuous
o r discontinuous n a tu re and extend ov er a few feet or hundreds of feet.
T his fringe usually cannot be read ily identified as a sub-population for
sam pling purp o ses and the extent to which it should be taken into account
in sam pling has to be considered. In selectin g u n its, som e a re likely
to fall so that they lie p a rtly within and p a rtly outside the stand. If
the frin g e a re a is sm all re la tiv e to the a re a of the m ain population,
these units might be re je c te d without the b ias being se rio u s. To avoid
b ias the units should be estab lish ed up to the ex tern al boundary of the
stand as p a rtia l units and ap p ro p riate adjustm ents made to e stim a te s
in these p a rtia l units to bring them to whole unit equivalents. T his is
not inconvenient if the fringe a re a is re la tiv e ly sm all.
In an extensive inventory of sm all iso la ted stan d s, how ever, the
fringe a re a is likely to re p re s e n t a la rg e proportion of the total a re a
and m ust be taken into account in the sam pling. Units cannot be
re je c te d but estab lish in g p a rtia l units may be inconvenient in both field
and office. In such a case it m ay be d e sira b le to modify the p re sc rip tio n s
of estab lish m en t of units in the p e rim e te r to avoid th is inconvenience
without biasing the e stim a te . The solution to th is problem in the Census
of Woodlands of G reat B rita in , w here it was of p a rtic u la r im portance,
is d iscu ssed by Hummel (1949) and Finney and P alca (1950). The m a tte r
will need attention in a re s o u rc e inventory of A u stralian native fo re s t
which includes p riv ate p ro p e rty .
Units O verlapping S trata B oundaries

A s im ila r problem a r is e s in stra tifie d random sam pling w here units
fall so th a t they overlap the boundary between two d ifferen t s tra ta . A
method of moving such a unit so that it lie s wholly w ithin one stra tu m
with a p p ro p ria te weighting of the quantitative e stim ate was suggested
by G. A. M cIntyre fo r the A u stralian national inventory (F. and T .B .
MSS.).
F e rtility G radient
Units with one dim ension longer than the other should have that
dim ension d ire c ted along the fe rtility gradien t. T his is usually a c ro ss
the line of topography, that is a t rig h t angles to the co n to u rs. When
s trip s a re e stab lish ed so that inform ation on fo re s t types and topography
can be obtained at the sam e tim e as quantitative e s tim a te s , they a re
o rien ted at rig h t angles to the contours to get a re p re se n ta tiv e sam ple
of this inform ation a lso . In larg e a re a s with an irr e g u la r topographic
p a tte rn , th is may n e c e s sita te survey sub-d iv isio n . In any c a se , s trip
length should be r e s tric te d to about one m ile fo r control of d irectio n .
The d irectio n of the s trip s is usually p re s c rib e d beforehand and th e re
is no source of o p e ra to r bias in locating the unit re la tiv e to fe rtility
gradient.
If long n arro w plots a re used purposely to take advantage of th e ir
in c re ase d efficiency if they a re oriented along the fe rtility g radient,
it is e sse n tia l that they be o riented c o rre c tly . It is p re fe ra b le not to
leave it to the o p e ra to r in the field to decide the o rien tatio n . If the
a re a is e sse n tia lly of one topographic fo rm , o r is known to be of
v ariab le form but no d etailed plan is available a m erid io n al d irectio n
common to all units in the a re a is often p re s c rib e d . If the topography
v a rie s by zones, a fixed directio n fo r all units in a p a rtic u la r zone
m ay be p re s c rib e d . The p re sc rip tio n s fo r orientation of units which
w ere used for inventory of an a re a of known broken topography a re
d e sc rib e d by C arro n and Hall (1954).
—
The Q uantitative E stim ate
T ree and Stand V ariab les
The objectives of the inventory w ill d eterm in e what tre e and stand
v a ria b le s a re m e asu re d o r estim ate d . How th is m ight be done fo rm s
the previous p a rt of this text.
The scope of an extensive inventory and the fact that tim e , funds,
and labour a re usually lim ited , often prom pt substitution of ocular
estim ate fo r m e asu re m e n t of tr e e v a ria b le s . O cular estim atio n is
often a serio u s so u rce of b ia s. The capacity of o p e ra to rs to estim ate
tre e dim ensions tends to be g ro s sly ex ag g erated , and if the inventory
is serio u s in intention and not ju s t a c u rs o ry rec o n n aissan c e, o cu lar
estim ation should not be co n sid ered except w here a re g re ss io n of o c u lar
e stim a te s on m e asu re d values is a p a rt of the inventory design.
The A rea of the Unit

Any e r r o r in d eterm ining the a re a of sam pling units has a d ire c t
effect on the o v erall e stim a te . When the com plete boundaries of units
a re estab lish ed by surveying m ethods and m e asu re d , the d eterm in atio n
of a re a is under control. When s trip s o r long plots a re u sed , it is
common to e sta b lish only a c en tre line by survey. In th is c ase the
determ ination of a re a m ay o r m ay not be under s tr ic t co n tro l and th is
so u rce of e r r o r m ay be s e rio u s. F o r n a rro w units (2 m e tre s , 10 ft,
o r a q u a rte r-c h a in in w idth), a pole is often used to d eterm in e from
the c e n tre line w hether tr e e s a re within o r outside the unit. F o r w id er
un its, o p e ra to rs frequently e stim a te by eye w hether tre e s a re 'in ' o r
'out' of the s trip because the difficult fo re s t conditions which prom pt
the choice of such units a re by th e ir v e ry n atu re those which prom pt
estim atio n ra th e r than m e asu rem en t. If s trip s a re used and the
o p e ra to r is collecting qualitative inform ation at the sam e tim e as
quantitative inform ation, he m ay also tend to fo rg et, in collecting the
qualitative inform ation as fa r on each side of the c e n tre line as he can,
that he should co llect quantitative inform ation within only a v ery s tric tly
defined distance of the c en tre line. The argum ent is often advanced th at
a c e rta in amount of e r r o r in s trip width estim atio n can be p e rm itte d
because tre e s w rongly included a re lik ely to be balanced by tre e s
w rongly excluded, p a rtic u la rly over long s trip s .
An investigation into th is m a tte r was c a r rie d out by students of the
A u stra lia n F o re s try School at R oses T ie r, T asm an ia, in 1951. On an
a r e a of 240 a c re s of eucalypt fo re s t (m ainly E. obliqua), 30 contiguous
s tr ip s w ere e stab lish ed , each 40 chains long, making a to tal of 15 m iles
of s trip lin e . The c e n tre lines w ere surveyed and c lea re d and p a rtie s
re q u ire d to a s s e s s 1 chain each side of the c en tre line. The tre e s
w e re re c o rd ed m e re ly by an identifying num ber and th e ir volum es
d e riv e d la te r. The a re a was at an elevation of about 3. 000 ft and v a rie d
fro m flat to steep; the undergrow th was dense in m ost p la ce s, of high
b lad ey g ra s s in the flat dry a re a s and m ixed sp ecies in the w e tter a re a s ,
p a rtic u la rly thickets of m y rtle beech in the g u llies. The w eather
v a rie d from hot and dry to cold and w et, with snow on two occasions
and heavy hail on a n o th e r. The bush w as alw ays wet for the e a rly
p a r t of the m orning’s w ork. In sh o rt, the g en eral conditions w ere
re a so n a b ly typical for strip lin e w ork in many A u stralian fo re s ts . Four
p a r tie s each a ss e ss e d the 15 m iles of s trip lin e , th ree p a rtie s working
in the common fashion of m easu rin g s trip width only when in re a l doubt
about a boundary tre e . The fourth p arty was in stru c te d to check the
s tr ip width except when th e re was no doubt at a ll. and subsequently
w ent over the s trip s again to determ ine the exact population of each
s tr ip and to analyse e r r o r s the p arty had m ade. A tr e e w as co n sid ered
w ithin the s trip if its g eom etric c en tre at ground level was within the
s tr ip . The re s u lts a re shown in T ables 18:1, 18:2, and 18:3.
The g en eral conclusion is that without continuous check of s trip width
by m e asu re m e n t, the e r r o r in the estim ate of volum e may be high
The e r r o r s in individual s trip s a re likely to v ary in size and sign
depending on the o p e ra to rs, as seen in Table 18:1. The su m m ary of
th e se e r r o r s by c la s s e s in Table 18:2 shows the extent to which the
e r r o r s can be reduced by conscientious m easu rem en t of s trip width
(o p e ra to r 4). Table 18:3 shows that th e re was no balance between
tr e e s w rongly included and those w rongly excluded. Even w here the
n u m b e rs of tre e s clo sely agree for o p e ra to r 3, th e re is a v e ry big
d ifferen ce in volum e. A nalysis showed th at th e re was a m ark ed
tendency to include tre e s doubtfully in o r out if they w ere la rg e and to
exclude them if they w ere sm all. O p erato r 4 w rongly included only
6 tr e e s in the 15 m iles of s trip s , all of which w ere due to confident but
w rong e stim a te of s trip width. Of the 49 w rongly excluded (a
s u rp ris in g ly high figure in the c irc u m stan c es), 23 w ere excluded
b e ca u se of confident but w rong estim ate of s trip width; 13 w ere m issed
because of the n atu re of the undergrow th around the tr e e s , w eath er,
and topographical conditions; and 13 w ere m issed for no valid re a so n .
The evidence suggests that to estim ate the contents of a s trip c o rre c tly ,
T a b l e 18:1
Strips, Roses Ti e r ; error as per c e n t of true v o l u m e for striplines by four operators
Strip no. Oper. 1 Oper. 2 Oper. 3 Oper. 4
1 3-4 -4-6 17-5 -2-0

2 4-7 -3-7 5-8 nil
3 12-4 20 - 2 19- 6 -0-4
4 0-7 3-3 7-9 -0-4
5 11 - 0 2-7 15-0 -0-4
6 3-8 -5-3 -1-3 -1-7
7 -0- 1 -0-3 -3-7 ni l
8 3-8 12-6 7-7 -0-5
9 1-5 11-4 - 7- 3 -0-2
10 -0-7 8-8 1-8 -1-7
11 1-0 3-6 -1-8 -0-8
12 -4-4 15-6 -2-8 -1-4
13 -1-7 7- 1 -4*2 0-2
14 -4-2 -2-0 -0-3 nil
15 -4-6 -2-7 -5* 1 -0-8
16 -0- 1 9-5 1-2 -0-8
17 11-2 4-4 7- 2 -3-5
18 7-1 5-9 5*2 0-9
19 6-0 -1*3 -1-2 -1*2
20 6-0 3-3 7-2 -0- 3
21 -0-9 3-0 -2*6 -2*4
22 -0-7 16-9 -3-9 -2-0
23 2-4 4-4 1-3 -1- 3
24 -7-8 3- 3 -4-2 -3-8
25 -7-9 3-5 3-4 1-7
26 -2-2 1-0 -2-6 0*2
27 -2-0 8-1 3-6 ni l
28 -9-8 -9-9 -5-3 -3-0
29 -1-7 -3-5 5-8 -0-7
30 0-6 -6-8 2-5 -0- 1
T a b l e 18:2
Strips, Roses Ti e r ; distri buti on of error as pe r c e n t of true v o l u m e for striplines by

four operators
Class Oper. 1 Oper. 2 Oper. 3 Oper. 4
Nil 0 0 0 4
1* 0 and less 8 2 1 14
1- 0 - 5*0 13 15 17 12
5*0 and g r e a t e r 9 13 12 0
No. of + ve errors 15 20 16 4
No. of - ve errors 15 10 14 22
F o r e s t In v en to ry 191
T a b le 18:3
Strips, Roses T ier; error as n u m b e r of trees and as v o lu m e of w hole a re a for striplines by

four operators
N u m b e r of trees
O per. 1 O p er. 2 Oper. 3 O per. 4
W r o n g l y e x c lu d e d 156 145 162 49

W r o n g l y in c l u d e d 136 185 164 6
D if f e r e n c e -20 40 2 -43
D i f f e r e n c e as per
c e n t of to t a l -0 .7 1.2 0 .0 7 -1 .4
V olume
D i f f e r e n c e ( c u ft) 6406 33866 16696 -1 0 0 2 9
D i f f e r e n c e as per
c e n t of to t a l 0. 6 3 .2 1 .6 -0 .9
c o n s id e ra b le c a r e g e n e ra lly and co n tinual m e a s u re m e n t of s tr ip w idth

is n e c e s s a r y .
S p e c ie s Id en tifica tio n
It is u s u a lly im p o rta n t th a t tr e e s p e c ie s be c o r r e c tly id e n tified in
in v e n to ry and s p e c ia l in s tru c tio n of o p e r a to r s b e fo re an in v e n to ry and
f re q u e n t ch eck in g d u rin g the in v en to ry to e n s u re c o n s is te n t and c o r r e c t
s p e c ie s id e n tific a tio n is g e n e ra lly n e c e s s a r y . T his is p a r tic u la r ly the
c a s e w h e re c e r ta in s p e c ie s have s p e c ia l q u a litie s , t r e e s look alik e but
h a v e wood of q u ite d iffe re n t q u ality and v a lu e , and th e r e a r e a n u m b er
of s p e c ie s not re g a rd e d a s u sefu l at p r e s e n t b ut w hich m ay be so
r e g a r d e d la te r .
P r o c e s s in g and A n alysing the D ata

B e c a u s e th e am oun t of d ata c o lle c te d is u su a lly su ch th a t it p o s e s g re a t
d iffic u lty in h an d lin g it m an u ally , m o st m o d e rn in v e n to ry in v o lv es the
u se of a u to m atic d a ta p ro c e s s in g m eth o d s. T his is a s p e c ia lis t field
of s tu d y and a p p licatio n and no a tte m p t wall be m ad e to d is c u s s it h e r e .
T h e r e is e x te n siv e lite r a tu r e on th e su b je c t b e c a u se of th e f a s t ad v an ces
in th e d ev elo p m en t and ap p licatio n of au to m atic eq u ip m en t in th e la s t
d e c a d e . F u rn iv a l (1962) and Husch (1963) p ro v id e good o u tlin es from
w h ich the f o r e s te r can obtain b ackground of a g e n e ra l n a tu re . The
a u th o rity o rg a n is in g an in v e n to ry w ill e ith e r have its own s p e c ia lis t
staff to deal with p ro cessin g the data, o r w ill have it done by another
organisation. In e ith e r case it m ust be reco g n ised th at the p ro c e ssin g
of the data is an e sse n tia l p a rt of the planning of the inventory a s a
whole; the re q u ire m e n ts of p ro cessin g w ill influence what data a re
collected and how they a re collected. Advice on d ata p ro c e ssin g m ust
be sought before an inventory, not afte rw ard s.
The natu re of the s ta tis tic a l an alysis for e stim a te s of m eans and
sam pling e r r o r s depends on the method of sam pling. T his is covered
in standard tex ts on the subject.
F o re s t R esource Inventory in A u stra lia
A ustralian fo re s try a u th o rities have agreed on the d e sira b ility of an

inventory of the indigenous fo re s t of the country on a national scale
but for various reaso n s no com prehensive inventory has been c a rrie d
out.
The general p a tte rn in A u stra lia up to the late 1940s was a ss e ss m e n t
of a re a s of local m anagem ent in te re st by ground survejdng. Sampling
units w ere usually s trip s from which fo re s t type and topographical
m aps w ere p re p a red and e stim a te s made of q uantities of wood available,
usually including only the species then u tilisab le and by the c r ite r ia
of economic a c c e ssib ility and u tilisatio n of the tim e . Some State fo re s t
s e rv ic e s had by that tim e mapped and a s s e s s e d a la rg e proportion of
the fo re s ts of the State of in te re st to m anagem ent on th is b a sis . O ther
State se rv ic e s had covered only a sm all p ro p o rtio n .
A fter World War II, in te re st in a re s o u rc e inventory on a national
scale was stim ulated by vario u s fo re s te rs for s e v e ra l re a so n s. Some
had been involved in the use of a e ria l photographs during the w ar and
the application of a e ria l photographs to fo re st inventory had alread y
been confirm ed before and during the w ar; th e re was a co n sid erab le
r e s e rv o ir of m aps, photographs, and e x p ertise available which had not
been available before; autom atic data p ro cessin g equipm ent, by which
the v ast amount of data collected in inventory could be conveniently
handled, was becom ing available and exciting new developm ents w ere
prom ised; other c o u n tries w ere com m encing com prehensive in v en to ries
and the sense of in tern atio n alism which was being g enerated a fte r the
w ar prom pted a sen se of national re sp o n sib ility to survey n atu ral
re s o u rc e s .
Those State se rv ic e s with a la rg e proportion of th e ir fo re s ts alread y
a s s e s s e d began to use a e ria l photographs to m odify a sse ssm e n t
in fo rm atio n from low in tensity and faulty sam pling and to enhance
f u r th e r s trip a s s e s s m e n ts . O ther State se rv ic e s began to use a e ria l
p h o to g raphs in a com prehensive p ro g ram of inventory. T asm an ia, for
e x a m p le , had a lre ad y investigated the u sefu ln ess of a e ria l photographs
b e fo re World War II and H em m ings (1949) was able to re p o rt a
c o n sid e ra b le am ount of re s e a rc h into in te rp re ta tio n for fo re s t
c la s s ific a tio n and stra tific a tio n . C ro m er (1949) outlined the p ro g re s s
th a t had been made in photo in te rp re ta tio n of v ario u s fo re s t types in
A u s tra lia but em phasised the n e c e ssity for photographs taken esp ecially
fo r f o r e s tr y p u rp o ses. He also em phasised that b asic m apping, on
w hich fo r e s te r s could su p erim p o se inform ation of sp ecial in te re s t to
th e m , w as inadequate and m ight have to be provided by the fo re s t
s e r v ic e s th e m selv e s.
hi 1951, a conference was held of Commonwealth and State fo re s try
a u th o ritie s and a fram ew ork which was suitable to a national inventory
was a g re e d on (F. and T. B. MSS. , 1951). Definitions w ere adopted,
m a in ly following those of H a rriso n (1950) which had been proposed as
an in te rn a tio n a l stan d ard . It w as recom m ended th at m apping be on
the b a s is of 1 m ile to 1 inch m ilita ry sheets with a c la ssific a tio n of
land and fo re ste d land as d iscu ssed e a r lie r in th is c h ap te r. It was
su g g este d th at, w herever p o ssib le, the estim atio n of volume be c a rrie d
out by s tra tifie d random sam pling using plots as sam pling u n its. The
sam p lin g in tensity of the s tra ta was to depend on the a re a and im portance
of th e s tr a ta and the p re c isio n d e sire d , and tentative suggestions w ere
m ade a s to the ap p ro p riate p re c isio n for e stim a te s of g ro ss bole volume
and c o m m e rcial volume for v ario u s p u rp o ses and s tr a ta . It was
reco m m en d ed that c e rta in plots be made perm anent for determ ination
of in c re m e n t. To te st the p ra c tic a b ility of the p re s c rip tio n s, they
w e re applied by students of the A u stralian F o re s try School to an a re a
in V ic to ria as re p o rte d by C arro n and Hall (1954). and an inventory of
the A . C . T . w as also c a r rie d out (R odger, 1953).
H ow ever, the inventory on a national b a sis did not eventuate and the
v a rio u s State s e rv ic e s have continued independent in v en to ries which
v a ry fro m re s o u rc e type to m anagem ent type, with v ario u s lev els of
c la s s ific a tio n and stra tific a tio n based on various c r ite r ia , and at
v a ry in g in tensity of sam pling. The 'fo re s t inventory s u rv e y s ’ of State
f o r e s t a re a s of Q ueensland a re intended as m anagem ent inventories
but the a re a s of the w estern c y p re ss pine-eu caly p t and some of the
c o a s ta l eucalypt fo re sts a re so la rg e th at th ese in v en to ries a re alm o st
r e s o u r c e type inventories in scope. In New South W ales, 'c la ssific a tio n
s u rv e y s ’ have been c a rrie d out over a c o n sid e ra b le p a rt of the State
based on a classific atio n such as outlined by B yles (1958). T hese a re
re so u rc e type inventories in scope and p u rp o se. Though the main
purpose of the a s s e s s m e n ts c a r rie d out in T asm an ia, as re p o rte d by
L aw rence and W alker (1954) and L aw rence (1963), has been fo r
m anagem ent planning, they have followed fa irly clo sely the p re s c rip tio n s
fo r national fo re s t inventory as have the a ss e ss m e n ts of W estern
A u stra lia as d iscu ssed by Nunn (1959). Some a u th o ritie s have claim ed
that the vario u s in v entories c a r rie d out in the States have sufficient in
common for s a tisfa c to ry national e stim a te s to be derived from them .
O ther au th o ritie s such as C ro m er (1960) deny th is.
C rom er (1949) has draw n attention to the lim itatio n s to the
application of photo-m ensuration techniques to the eucaljqDt fo re s t of
A u stra lia. F o r m any y e a rs these techniques have form ed the b a sis of
much of the re s o u rc e inventor}7, of fo re s ts of the n o rth e rn h em isp h ere.
The h isto ry of a considerable p a rt of the eucalypt fo re s t is one of
selectiv e logging, and the p re se n t fo re s t has a legacy of la rg e faulty
tr e e s . Many sp ecies also develop the ty p ical c e n tra l pipe which is
hollow o r not u tilisa b le , o r have been subject to in sect and fungal
attack p a rtic u la rly a fte r fire . The o c cu rren c e o r extent of one of th ese
faults cannot be d eterm ined by a e ria l photograph in te rp re ta tio n - even
on the ground estim atio n of th ese faults is notoriously u n reliab le. The
m ajo r contribution of a e ria l photographs to inventory in the eucalypt
fo re s t is to allow fa st and re lia b le m apping of the location and extent
of fo reste d a re a s , c lassific atio n into productive and non-productive
fo re ste d land, and stra tific a tio n into v a rio u s s tra ta which im prove the
efficiency and p ra c tic a b ility of sam pling for quantitative e stim a te s on
the ground. The efficiency of this kind of stra tific a tio n was te ste d in
T asm ania as rep o rted by L aw rence (1957). A ttention has been given
to p h o to -in terp retatio n techniques, p a rtic u la rly the use of ste re o g ra m s
and density sc a le s (Aitken and H all, 1951), and to the c au se , effects,
and avoidance of 'shadow p o int’ (Sim s, 1954; F. and T .B . , 1955).
Sims and Hall (1955) considered v ario u s re q u ire m e n ts for
in te rp re ta tio n of a e ria l photographs in A u stra lia and in p a rtic u la r
outlined p re sc rip tio n s and c r ite r ia for the selectio n and tra in in g of
a e ria l photograph in te rp re te rs which have been applied to a la rg e
num ber of people undertaking c o u rse s in in te rp re ta tio n at C an b erra
in the la st decade.
F o re s t R esource Inventory E lsew here
Much h a s been w ritten on fo re st re s o u rc e inventory in v ario u s co u n tries
of th e w o rld . T here is no point in su m m arisin g any of it h e re . The
r e a d e r ’s in te re s t w ill be b e tte r se rv e d by d ire c t re fe re n c e to the
lite r a tu r e liste d below, which has been chosen because it explains the
in ten tio n s of the inventory and m ethods and techniques used (ra th e r
than s ta te m e n ts of re so u rc e s) or has itse lf the m ost com prehensive
b ib lio g rap h y on the p a rtic u la r inventory.
A s t a r t was made on a fo re st re s o u rc e s inventory of Papua-N ew
G uinea during World War II for m ilita ry purp o ses and for planned
p o s t-w a r developm ent. V arious a sp e c ts of the m ilita ry su rv ey w ere
o u tlin ed by Shillinglaw (1945) and Luke (1946). P o s t-w a r inventory is
d is c u s s e d by W om ersley and McAdam (1957).
A d e sc rip tio n of the national fo re s t inventory of New Zealand which
had th e n re c en tly com m enced was given by Thom son (1946). A gen eral
outline w as given by M a ste rs, Holloway, and McKelvey (1957).
T h e re is a long h isto ry of re s o u rc e inventory in Canada, w here many
a s p e c ts of the use of a e ria l photographs w ere pioneered. Seely (1957)
has given a sum m ary of the in v entories of the v ario u s p ro v in ces, and
also an outline of technical asp e c ts (Seely, 1960). T h ere is a s im ila rly
long h is to ry of re s o u rc e inventory in the U. S. A. S purr (1952, 1960)
has p ro v id ed an extensive bibliography on the v ario u s m ethods and
tech n iq u es; Wilson (1962) lis ts ’c r ite r ia useful in ap p raisin g the re la tiv e
m e r its of a su rv ey design' and evaluates a num ber of the designs used
in N o rth A m erica on the b asis of them . Inventories of South A m erican
c o u n trie s a re much m ore re c en t and lim ited . H einsdijk (1961, 1962)
has d is c u s s e d inventories in B ra z il.
In E urope, the Scandinavian co u n tries a re noted for the num ber of
su c c e ss iv e national fo re s t inventories they have c a r rie d out, b ased on
field o b serv atio n s only. The inventory of Sweden was outlined by
H agberg (1957), that of Norway by L a n g sae te r (1949), and that of
F in lan d by Ilvessalo (1952). In o ther co u n tries of E urope, the in v en to ries
a re b a s ic a lly of m anagem ent ra th e r than re s o u rc e s type.
The m ethods and techniques of the C ensus of Woodlands of G reat
B rita in w ere d escrib ed by Hummel (1949) and fu rth e r by the F o re s try
C o m m ission (1952, 1953).
F o r A sian c o u n trie s, M athauda (1957) outlined m ethods used for
in v e n to rie s in India, b asic re s e a rc h on m ethods of sam pling fo re s ts
in In d ia having been re p o rte d by G riffith (1945-6). de R osayro (1962)
d is c u s s e d in v e n to ry in C eylon and r e f e r s to in v e n to rie s in o th e r
tro p ic a l f o r e s ts . L o e tsc h (1957) d e s c rib e d a f o r e s t in v e n to ry in
T hailand. F r a n c is and Wood (1955) d is c u s s e d a c la s s ific a tio n of
v eg etatio n in N o rth B o rn eo fro m a e r ia l p h o to g ra p h s. B ru n ig (1963)
has rev ie w e d th e h is to r y of f o r e s t in v e n to ry in S araw ak . K inashi
(1954) and K in ash i et a l . (1962) d is c u s s e d th e d ev elo p m en t of sam p lin g
d esig n s fo r f o r e s t in v e n to rie s in Ja p a n .
N yssonen (1961) h as d is c u s s e d in v e n to rie s of tro p ic a l f o r e s ts in
g e n e ra l and p a r tic u la r .
FOR EST MANAGEMENT INVENTORY
T he m e a s u re m e n t o r e s tim a te of stan d v a r ia b le s o v e r the w hole o r a

sam p le of a f o r e s t o r stan d so th a t p r e s e n t y ie ld can be d e te rm in e d
and its u tilis a tio n re g u la te d h a s alw ay s been th e b a s is fo r m an ag em en t
of the f o r e s t o r stan d . It h as th e re f o r e been c a r r i e d out in so m e p a r ts
of the w o rld fo r a v e ry long tim e . H is to ric a lly , th is kind of in v en to ry
p re c e d e d r e s o u r c e in v e n to ry , w hich b o rro w e d man}" of the m eth o d s
and te ch n iq u es of m an ag em en t a s s e s s m e n t. M uch of w hat h a s been
sa id about m eth o d s of m aking q u a n tita tiv e e s tim a te s u n d er th e h ead in g
of re s o u r c e in v e n to ry is ap p licab le h e r e and n eed not be re p e a te d .
The m ain d iffe re n c e s b etw een r e s o u r c e and m an ag em en t in v e n to rie s ,
a lre a d y m en tio n ed , a r e th e b a s is of c la s s ific a tio n and s tr a tif ic a tio n ,
th e kind of in fo rm a tio n c o lle c te d , and th e p re c is io n of the q u an titativ e
e s tim a te s re q u ir e d . C h a r a c te r is tic a lly th is kind of in v e n to ry of an
a r e a is c a r r i e d out o nce, the in fo rm a tio n o b tain ed by a sin g le in v e n to ry
being r e g a rd e d a s su fficie n t fo r im p lem en tin g a m a n a g e ria l c o n tro l
o v e r o p e ra tio n s on the a r e a .
T h e re h a s been a long h is to ry of such a s s e s s m e n t in A u s tr a lia , th e
m ain function of w hich in e a r l i e r tim e s w as to p ro v id e in fo rm a tio n fo r
tim b e r o p e ra tio n s to m e e t th e d em an d s of an e s ta b lis h e d and expanding
in d u s try , but w hich in la te r tim e s w as to p ro v id e in fo rm a tio n fo r
m a n ag em e n t in th e s t r i c t e r s e n s e . The h is to r y of a s s e s s m e n ts of
th is kind in New South W ales f o r e s ts w as o u tlin ed by H enry (1955)
and in W este rn A u s tra lia n f o r e s ts by Nunn (1957). R ecen t a s s e s s m e n ts
of eu caly p t f o r e s ts in T a s m a n ia fo r m a n ag em e n t planning w e re d e s c rib e d
by L aw ren ce (1963).
T h e re is an e x ten siv e li te r a t u r e on a s s e s s m e n ts in v a rio u s c o u n trie s
of the w o rld . Some of th e m o re in te r e s tin g r e f e r e n c e s a r e th o se
d e sc rib in g investigations in tro p ic a l high fo re s t in A frica by Dawkins
(1952, 1957).
P eriodic M anagem ent Inventory

If e s tim a te s of future yield a re re q u ire d , the e stim a te s of a single
in v en to ry can be adjusted by inform ation about likely growth in the
fu tu re p erio d draw n from special investigatio n s. T here a re , how ever,
m any advantages in repeating an inventory of a fo re s t so th at the rep eated
in v e n to rie s provide inform ation about yield available at the tim e of
in v en to ry and also about changes during the period betw een in v en to ries
th a t a r e d ire c tly re la te d to the fo re s t and so provide a b e tte r b a sis for
ex tra p o latio n . T his system of rep eated inventory is known by vario u s
n a m e s , among them periodic m anagem ent in v en to ry , w orking plan
in v e n to ry , continuous fo rest inventory and fo re st inventory s u rv e y .
T he e a r lie s t exam ple of im portance is the periodic m anagem ent
in v en to ry of the control method which was f ir s t applied in 'w estern
E urope over seventy y e a rs ago. T h ere has been effective periodic
in v en to ry , on which m anagem ent has been b ased , in som e A u stralian
fo r e s ts for a little over th irty y e a rs . In the la st ten o r fifteen y e a rs
th e re h as been an in c re ase d in te re s t, both public and p riv a te , in
re p e a te d inventory in many p a rts of the w orld. This is due to both an
in c re a s in g aw aren ess of the need for m anagem ent and so of the need
fo r quantitative data on which to base it and to the developm ent of data
p ro c e ss in g , which has given the fo r e s te r m eans to handle the enorm ous
am ount of data usually involved. T his expansion of in te re s t, with
concom itant spate of lite ra tu re , has been p a rtic u la rly m arked in the
U. S. A.
T he g en eral p rin cip les of p eriodic m anagem ent inventory can be
applied to any fo re s t or stand w hether even-aged o r uneven-aged, of
co n ifero u s o r non-coniferous s p e c ie s, pure o r m ixed, virgin or
c u t-o v e r, though the method and technique of applying the p rin cip les
d iffe r according to the objective of m anagem ent and the kind of fo re s t.
The inform ation req u ired for the p a rtic u la r p u rp o ses of m anagem ent,
and the re q u ire d le v els of re lia b ility of the quantitative e stim a te s , a re
nom in ated. This inform ation is so m etim es obtained by com plete
en u m eratio n s but m o re often by sam pling, which involves co n sid eratio n
of th e size of the sam ple, the m ethod of selectin g the sam pling units,
n u m b e r, siz e , shape and o rientation of the sam pling u n its, th e ir
e sta b lish m e n t in the field, th e ir m aintenance if p erm an en t, and th e ir
re p re se n ta tiv e n e ss at all tim e s. The stand v a ria b le s a re m easu red
o r e stim ate d on the sam pling units and a field re c o rd m ade. The data
is then analysed, m anually o r by som e system of data p ro c e ssin g , to
provide inform ation on the p re se n t stand condition and changes in the
stand over the perio d between m easu rem en t. On the b a sis of th is
inform ation, m anagem ent decisions a re m ade.
The purpose of a m anagem ent inventory being to provide the m anager
with inform ation to make ratio n al decisions on the m anagem ent of a
fo re s t, the logical sta rtin g point of any m anagem ent inventory is for
the m anager to nom inate what inform ation he n eed s, and at what level
of re lia b ility . T his m ay re q u ire som e p re lim in a ry investigation.
U sually the m anager w ill know enough about a fo re s t to decide the so rt
of inform ation he needs to choose the objects of m anagem ent. If, having
decided on the object of m anagem ent, he needs fu rth e r inform ation to
im plem ent it, this m ay have to be determ in ed by e x tra w ork. U sually
the inventory can be a rra n g ed to obtain not only the rig h t kind of
inform ation on which to decide on the b e st object of m anagem ent but
also the kind of inform ation needed to im plem ent it. If the inventory
cannot be so a rra n g ed it is fa r b e tte r to c a r r y out an ap p ro p riate
p re lim in a ry investigation, on which an efficient inventory can be based,
than to c a r ry out an inventory which gives m ore inform ation than is
needed to decide on the object of m anagem ent but le s s than is needed
to im plem ent it.
The level of re lia b ility at which the m anager needs the inform ation
is im portant. To a la rg e degree it d e term in e s the m ethod and in ten sity
of sam pling and so the amount of w ork to be done in the inventory. In
deciding th is level the m anager should be m ainly co ncerned with
balancing it against the consequence of a wrong d ecisio n . The usual
position in which the m anager finds h im self is that the lev el of
re lia b ility he would like to have c o sts m ore than the revenue of the
fo re s t (or other so u rc es of funds) can support and the inventory which
the fo re st can support w ill not give him a level of re lia b ility which w ill
enable him to make his decisions with a p p ro p riately com forting
confidence. He m ust effect what to him is an uneasy co m p ro m ise
betw een these two. One of the m ain difficulties in th is is th at, w h ereas
confidence lim its can be re a d ily se t to quantitative e s tim a te s , it is
not com parably easy to nom inate ’consequence lim its ' in decision
m aking. F o re s t m an ag ers do not appear to have given sufficient
attention to th is la tte r asp ect and as a consequence the level of
re lia b ility re q u ire d fo r m ost m anagem ent in v en to ries, like th at for
m o s t re so u rc e in v en to ries, is ra re ly nom inated in a ra tio n a l way and
v e ry often is likely to be higher than is n e c e s sa ry .
The design of a m anagem ent inventory m ay be co n sid erab ly affected
if the inventory is to provide inform ation on w here cutting o perations
sh o u ld be c a r rie d out at a p a rtic u la r tim e as well as inform ation on
ho»w much is available for cutting at that tim e. A se p a ra te operational
in yen to r 7 m ay be n e c e s sa ry for the fo rm e r.
B ecause inventory is an in fo rm atio n -g ath erin g p ro ced u re and
m anagem ent involves devising and controlling o p eratio n s, th e re is a
ten d en cy to think of it as som ething a p a rt from m anagem ent. Inventory
is an esse n tial p a rt of m anagem ent. It is b e st designed and su p erv ised
by th e m anager, o r, if m anagem ent and inventory re q u ire d ifferent
s k ills , they should be in com plete sym pathy with each o th er.
P a in e (1966) has illu s tra te d the use of lin e a r p ro g ram m in g in
m anagem ent planning of a eucalypt fo re s t in V icto ria. M cG rath and
C a rro n (1966) have suggested a m ethod of determ ining the yield of an
uneven-£ged m ixed eucalypt fo re s t on the c o ast of New South W ales,
th e inform ation n e c e s sa ry fo r th is, and the p ro b lem s of obtaining the
in fo rm ation by p eriodic inventory
C om plete E num erations

A s in te re s t in m anagem ent inventory has developed in re c en t y ears
SO' h a s referen ce to the control m ethod (methode du co n tr8 le,
K ontrollem ethode) in c re a se d in the lite ra tu re . The te rm is asso ciated
w ith the m anagem ent of som e fo re s ts in w e ste rn E urope, p a rtic u la rly in
S w itzerland, along lin es suggested by Gurnaud in F ran ce about 1878
and fu rth er developed by B iolley in Sw itzerland about 1890. D etailed
o u tlin e s of the method w ere given by Schaeffer et a l . (1930) and Knuchel
(1953). T here h as been co n sid erab le d iscu ssio n at v ario u s tim e s on
w h e th er the control m ethod is a m ethod of m anagem ent, a silv ic u ltu ra l
s y s te m , o r a method of m en su ratio n . It is convenient fo r p re se n t
p u rp o se s to re g a rd it as of th re e com ponents: c a rry in g out periodic
in v en tory from which the p re s e n t state of the stand and p a st in crem en t
a r e determ ined; p re sc rib in g p re se n t and future yield on th is b asis;
obtaining th is yield through a p a rtic u la r silv ic u ltu ra l sy stem . The
re a s o n for the p a rtic u la r m ensuration in te re s t in the control method
is th a t, when the p ro ced u re for the inventory of the p a rtic u la r fo re s ts
w a s estab lish ed over seventy y e a rs ago, fa r-sig h te d attem p ts w ere
m ad e to ensure that the su cc e ssiv e in v en to ries would be com parable
so as to give re lia b le e stim a te s of in crem en t. This co m p arab ility has
been ensured and the re c o rd s , kept m eticulously o v er this long p erio d ,
a re of p a rtic u la r in te re s t as an exam ple of a lo n g -term inventory
p ro je c t.
The se v e ra l im portant fe a tu re s a re b rie fly as follow s. The
d .b .h .o .b . of ever}1- tre e in the stand on each occasion is m easu red
by c a lip e r, and m easu red con sisten tly by placing the beam of the
c a lip e r in a perm anent scrib e m ark. The tr e e s a re not identified
individually. They a re ta llie d in fa irly wide (for exam ple, 5 cm)
c la s s e s ; th e re is a low er lim it to the enum eration (for exam ple, 20
cm ). D .b .h .o .b . c la s s volum es a re d eriv ed by re fe re n c e to a ta riff
which is used on each occasion of m e asu rem en t to keep the e stim a te s
of volume com parable, the unit of volume of the ta riff being called a
s ilv e . The volume of tre e s h a rv e ste d is d erived both in silv es and in
te rm s of the utilisatio n conditions of the tim e. T his provides a facto r
w hereby the standing volume in silv es can be c o rre c te d to equivalent
p re se n t com m ercial volum e. Increm ent is deriv ed from the d ifferen ces
of the e stim a te s at the periodic m e a su re m e n ts, taking into account
n atu ral m o rtality and a rtific ia l re m o v a ls, and ingrow th. Growth is
usually computed fo r the stand and by d . b . h . o . b . c la s s e s which a re
often sim plified into th re e broad c la s s e s - sm all, m edium , and la rg e .
Complete enum eration is c a r rie d out in m anagem ent inventory in
som e o ther fo re s ts of w estern Europe not m anaged by the control
m ethod. Where the stands a re re la tiv e ly sm all and m anagem ent very
intensive, the cut and tre a tm e n t a re p re s c rib e d by co m p artm en ts, the
tim b e r has a high value, and the silv ic u ltu re is such that a continually
changing stand com position p re s e n ts p ra c tic a l problem s in sam pling.
A tra d itio n al re s ista n c e to sam pling and the ready av ailab ility of a
lo w -sa la rie d but w e ll-tra in e d labour fo rc e a re fu rth e r reaso n s in som e
c a s e s.
Sampling
In m ost countries p eriodic m anagem ent inventory is c a rrie d out by
sam pling. In A u stra lia , fo r exam ple, the fo re s ts a re too la rg e in
extent for com plete enum erations to be p ra c tica b le ; because of
re la tiv e ly fa st growth ra te s frequent in v en to ries a re req u ired and the
fo re s t revenue could not support frequent full enum erations. In any
c a s e , the inform ation req u ired fo r m anagem ent can be obtained at an
a p p ro p riate level of re lia b ility by p ro p e rly designed sam pling - which
is a lso the c ase for the fo re s ts under the control method except w here
co ntinual change in stand com position m akes sam pling difficult.
The Sample
The usual m ethod of selectin g the sam ple is s tra tifie d random or
s y ste m a tic . The sam pling units a re usually plots o r angle count spots;
s tr ip s a re not com m only used. A spects of siz e , shape, and orientation
of u n it d iscu ssed under re s o u rc e inventory also apply to m anagem ent
in v en to ry . Sampling for m anagem ent inventory is usually of much
g r e a te r intensity than fo r re s o u rc e inventory. The intensity fo r re so u rc e
in v en to ry is usually le s s than one p e r cent and m ay be as low as one
ten th of one p e r cent. F or m anagem ent it is usually g re a te r than one
p e r cent and may be as high as ten p e r cent.
Sam pling on Successive O ccasions

T h e re a re two e sse n tia l functions of periodic m anagem ent inventory:
to r e c o r d the condition of the fo re s t at each inventory and from these
re c o rd s to d eterm in e the changes over the p erio d betw een in v en to ries.
F o r exam ple, stand volum e m ay be d eterm in ed on each occasion and
net in c re m en t for the period in betw een. F o r such sam pling on
s u c c e ssiv e occasions th e re a re se v e ra l a lte rn ativ e designs. These
a re d iscu ssed in stan d ard te x ts, such as Yates (1949), b ecause they
a re applicable to any kind of census or inventory which is c a rrie d out
to give inform ation on both the state of a population on occasions and
the changes in the population in betw een those o ccasio n s. In the p re se n t
co n tex t, for a sam ple of units estab lish ed in a fo re st and m easu red
on one occasion, th e re a re th re e p o ssib ilitie s for m easu rem en t on a
subsequent occasion. The f ir s t p o ssib ility is to re m e a s u re all the
u n its; the units a re p e rm a n e n t, the sam p les a re dependent o r the
sam p le is fixed; 'th is is form ally equivalent to the observation of
d ifferen t c h a ra c te rs (variables) on the sam e sam p le' (Y ates, 1949).
The second p o ssib ility is to re m e a s u re none of the units; the units a re
te m p o ra ry , a new sam ple is chosen on each occasion, the sam ples
a re independent, 'th is is form ally equivalent to in te rp e n etratin g
s a m p le s ' (Y ates, 1949). The th ird , in term ed iate p o ssib ility is (a) to
re m e a s u re onty som e of the u n its, som e of the o rig in al units having
b een m ade p erm anent and the o th ers te m p o ra ry ; th is is su b -sa m p lin g ,
'fo rm a lly equivalent to a tw o-phase sa m p le ' (Y ates, 1949); o r (b) to
re m e a s u re only som e of the u n its, som e of the orig in al units having
been made perm anent and the o th e rs te m p o ra ry , but to re p la c e the
te m p o ra ry ones not re m e a su re d w ith new units (which m ay be te m p o ra ry
o r perm anent fo r a subsequent occasion). This is sam pling w ith p a rtia l
rep lacem en t and 'has no fo rm al equivalent . . . . T hese equivalences
a re of im portance in that the m ethods of estim atio n w ill be the sam e
for form ally equivalent sam pling p ro c e s s e s ' (Y ates, 1949). In choosing
one or another of th e se designs, the m anager m u st c o n sid er s e v e ra l
fa c to rs.
One facto r involves s ta tis tic a l a sp e c ts. F o r e s te rs have ap p re cia te d
for a long tim e th at the p re c isio n of e stim ate of in c re m en t in a stand
v aria b le was higher (other things being equal) fo r dependent sam p les
than independent sam ples fo r re a so n s em p h asised by Hall (1959) and
th is has been one of the im portant s ta tis tic a l a sp e c ts influencing them
in deciding which of the two a lte rn a tiv e designs to u se. In re c e n t y e a r s ,
th e re have been se v e ra l p ap ers on the subject of sam pling w ith p a rtia l
rep lacem en t of units in fo re st m anagem ent inventory in N orth A m e ric a,
p a rtic u la rly by Bickford (1959. 1963), B ickford, M ayer, and W are
(1963), W are and Cunia (1962) and Cunia (1964, 1965). The alte rn ativ e
designs w ill all provide unbiased e stim a te s of stand v a ria b le s and of
sam pling e r r o r s a sso c iate d with those e s tim a te s , but one of the d esigns
w ill provide b e st e stim a te s at le a s t co st (usually the m ajo r c rite rio n
of efficiency of sam pling design). T his w ill depend on the m ethod of
sam ple selection, the extent of the c o rre la tio n betw een the e stim a te s
of the stand v ariab le on the su cc e ssiv e o ccasio n s, the actual stand
v a ria b les e stim ated , the re la tiv e p re c isio n re q u ire d of the e stim a te of
the stand v ariab le and the e stim ate of the change in the stand v a ria b le ,
the re la tiv e cost of estab lish in g new units and of finding and re m e a su rin g
perm anent units and the p ercentage of p erm anent units which cannot be
found.
A nother facto r is d . b . h . o . b . in crem en t. It can be calcu lated re lia b ly
only when tre e s a re identified individually and th is is fe a sib le only on
perm anent un its. In many kinds of fo re s t, too, the m ost re lia b le guide
to silv ic u ltu ra l tre a tm e n t is from the p erfo rm an ce of individual tr e e s
and this can only be studied on perm an en t units. A gain, it is e sse n tia l
that a com prehensive and re lia b le re c o rd of tre a tm e n t, re m o v a ls, and
m o rtality be kept between m e asu re m e n ts and th is is u su ally much m ore
p ra c tic a l on perm anent units. A gainst th is , changes of a qualitative
kind may b e tte r be brought to notice through a d ifferen t sam ple.
A th ird facto r is the re p re se n ta tiv e n e ss of the sam p le. T h e re a re
two m ain points fo r concern. One is that p a rtic u la r sam pling units do
re p re s e n t what they p u rp o rt to re p re s e n t - that they do not get biased
tre a tm e n t. This m ay p re s e n t d ifficu lties. If units a re p erm an en t they
m u st be so identified that exact relocation does not cause any p ro b lem s.
H ow ever, if they a re c le a rly m ark ed , th e re is the danger th at o p e ra to rs
w ill give them biased tre a tm e n t and te m p o ra ry units may be m ore
sa tisfa c to ry . The other point of concern is th at the re la tin g of sam pling
u n its to sub-populations be c o rre c t. In stands of even-aged coniferous
plantations in A u stra lia , fo r exam ple, thinning is usually c a rrie d out
on a com partm ent b a sis or w ithin site quality c la s s w ithin a com partm ent.
The a re a s to which sam pling units r e f e r can be fa irly re a d ily defined
and m easu red . On the other hand, in those native eucalypt fo re s ts
w here a silv icu ltu ral system such as the A u stralian group selectio n
system is applied, it m ay be quite difficult to define the stand a re a s
to which p a rtic u la r units r e f e r and quite im p racticab le to m e asu re
them . Nor is it p ra c tica b le to c a r r y out com plete en u m eratio n s, the
solution re s o rte d to for s im ila r p roblem s in c e rta in p a rts of w estern
E u ro p e.
A nother facto r is the effect of m ista k e s and b ia se s in m easu rem en t
of u n its. P erm anent units give both o p e ra to rs and s u p e rv iso r the
opportunity to pick up m ista k es and b ia s e s , and the fact that they can
be picked up m ay prom pt o p e ra to rs to be conscientious. An im p o rtan t
fa c to r h e re is ready identification of sam pling unit and individual tr e e s
so that th e re is no problem in reconciling the re c o rd . It is so m etim es
argued that the difficulties of reco n ciliatio n and the e r r o r s d isco v ered
in re m e asu re m e n t a re such th at it is b e tte r to have te m p o ra ry units -
which is equivalent to arguing th a t e r r o r s a re acceptable if we a re
ignorant of them .
P ra c tic a l A spects
In a rtific ia lly e stab lish ed plan tatio n s, perm an en t units can be
estab lish ed at any tim e a fte r e stab lish m en t. U nless inform ation is
re q u ire d on e arly growth o r th e re a re p ra c tica b le advantages in
establishing the units before estab lish in g the cro p , a su itable tim e is
ju st p rio r to f ir s t thinning. If the sam ple is to be a s tra tifie d random
one, using site s as s tr a ta , e stab lish m en t of units m ust aw ait the
m apping of site quality b o u n d aries, which cannot be done until the crop
has settled down a fte r e sta b lish m e n t tro u b les and its condition is tru ly
reflectin g the site . Even fo r a sy ste m a tic sam p le, it m ay be w ise r
to w ait som e y e a rs so that the population can be c le a rly defined. In
n atu ral stands, units a re e stab lish ed as soon as m anagem ent in te re s t
re q u ire s it.
Frequency of m easu rem en t depends on the v a ria b le s being m easu red ;
the stage in the crop rotation, and ra te of grow th of the crop; the
relationship between tim e of m e asu re m e n t and cutting cycle, thinning,
working plan rev isio n ; staff and funds av ailab le. F o r f ir s t ro tatio n
conifers in A u stra lia , a common p a ttern of m e asu re m e n t has been
every 2-3 y e a rs up to 20-30 y e a rs and about e v ery 5 y e a rs th e re a fte r.
In n a tu ral fo re s ts in A u stra lia , re m e a su re m e n t about ev ery 5 y e a rs
is m ore a p p ro p ria te . Once sufficient in c re m en t data a re av ailab le,
m easu rin g for w orking plan rev isio n about e v ery 10 y e a rs m ay be
sufficient.
If the units a re to be te m p o ra ry , only the m inim um tro u b le is
usually taken in identifying the plot location. F o r p erm an en t u n its,
the maximum trouble should be taken. E very atten tio n m ust be paid
to ensure easy re -lo c a tio n of units by a p p ro p ria te unit location plans
and location m a rk s. If individual tr e e s a re to be identified, they can
be num bered by painting num bers on the tr e e s , by a sy stem of tagging,
o r they can be identified by som e a p p ro p ria te sy stem of re c o rd in g such
as s e ria lly from a specified sta rtin g point in a sp ecified d irectio n in
plantations, or by tr e e location plans using m e asu re d c o -o rd in a te s
from som e defined axis of c o -o rd in a te s in n a tu ra l fo re s t. C le ar,
concise p ra c tic a l p re s c rip tio n s for the e sta b lish m e n t, m aintenance,
and m easurem ent of the sam pling units a re e s s e n tia l, p a rtic u la rly for
con sisten t p ro ced u re. The o rig in a to rs of the co n tro l m ethod reco g n ised
th is and it has been the stre n g th of the m ethod. Since m ethods and
techniques of m easu rem en t a re continually being im proved, som e
com prom ise has to be found betw een continuing to use a p rim itiv e
m ethod which gives c o n sisten t e stim a te s and changing the p ro ced u re
ev ery tim e an im provem ent is p o ssib le , with consequent likelihood
of bias in the e stim a te of in c re m en t. The way in which the Code of
Sample Plot P ro c e d u re of the F o re s try C om m ission (Hummel et a l . ,
1959) was p re p a re d m ight w ell se rv e as a m odel.
P e rio d ic M anagem ent Inventory in A u stra lia

The main asp ects of periodic m anagem ent in v en to ries c a r rie d out in
coniferous plantations by the State F o re s t S erv ices have been
su m m arised by Lewis (1963a).
The p re s c rip tio n s of the fo re s t inventory su rv ey s of the coniferous
p la n ta tio n s, ra in fo re s t a re a s , c o asta l eucalypt fo re s ts , and the c y p re ss
p in e -e u ca ly p t fo re s ts of the State a re outlined by the Q ueensland
D ep artm en t of F o re s try (1962). In each case the sam ple c o m p rises
s y ste m a tic a lly located plots but th e ir size and sam ple in ten sity v a rie s
w ith kind of fo re s t. Some of th is inventory has been c a rrie d out for
o v e r th irty y e a rs .
'C ontinuous fo re s t inventory' has been im plem ented by the New South
W ales F o r e s try C om m ission in re c e n t y e a rs in v ario u s native fo re s ts
as d e sc rib e d by F o r re s t (1961), C urtin (1962b), and T u rn er (1966), as
w ell a s in the coniferous plan tatio n s. As a re s u lt of considerable
in v estig atio n in e a r lie r a s s e s s m e n ts into the re la tiv e efficiency of
v a rio u s m ethods of sam ple sele c tio n , sam pling in te n sitie s, and siz e s
and shapes of units in various kinds of fo re s t, th e re is no single stan d ard
p ro c e d u re . The use of a e ria l photographs is an in teg ral p a rt of the
in v en to ry . H um phreys (1961) has d isc u sse d the re lia b ility of and lim its
to p h o to -in te rp re ta tio n of sp ec ie s a sso c iatio n s for m anagem ent planning
in eu calypt fo re s ts of the north c o a st of New South W ales.
A p erio d ic inventory of the plantations of P, ra d ia ta in the A. C. T.
is b a se d on a s tra tifie d random sam ple of angle count spots with age
and site index as s tra ta .
H all and N eal (1966) have outlined the m ethod of p ro c e ssin g data fo r
m anagem ent and re s e a rc h from p erio d ic inventory used by A u stralian
P a p e r M anufacturers L im ited fo r both coniferous plantations and native
eucalypt fo re s t, m ainly in V icto ria. The F o re s ts C om m ission of
V ic to ria c a r r ie s out periodic m anagem ent inventory in the coniferous
p lan tatio n s, the o v erm atu re and second-grow th eucalypt fo re s ts , and
the foothill fo re s ts .
The Woods and F o re s ts D epartm ent of South A u stra lia has a larg e
n u m b er of sem i-p erm an e n t 'yield p lo ts' in the coniferous plantations
which a re m e asu re d perio d ically to provide inform ation ap p ro p riate
to d eterm in in g m a te ria l available from thinnings and c le a r fellings.
Lew is (1963a) has d iscu ssed the p ro b lem s of fo rec astin g future yield
fro m th e se in ventories.
P e rio d ic m anagem ent inventory is c a r rie d out by the F o re s try
C om m ission of T asm ania in coniferous plantations.
Nunn (1957) re fe rs to the p eriodic inventory c a r rie d out in W estern
A u stra lia in a general outline of the h isto ry of a ss e ss m e n ts in that
S ta te .
P e rio d ic M anagement Inventory E lsew here
The vario u s c o n serv an cies of the New Zealand F o re st Service c a r ry
out periodic m anagem ent inventory in coniferous plantations as do many
of the p riv ate fo re s try o rg an isatio n s. The m ethods a r e su m m a rised by
N. Z. F . R. I. (1965).
Much has been w ritten on the subject for the n o rth e rn h e m isp h ere .
Prodan (1965a) d isc u sse s it fo r v ario u s European c o u n trie s. The
F o re s try C om m ission of G reat B ritain has sc ru tin ise d the usefulness
of periodic inventory to m anagem ent fa irly thoroughly and d iscu ssio n s
by Johnston (1960). W aters (1961), and Johnston and B rad ley (1964)
a re of p a rtic u la r in te re s t. Jack (1962) outlines the m ethod used in
N orthern Ireland. Some of the g re a t in te re s t in the m a tte r in N orth
A m erica is outlined by Purdue U niv ersity (1960) and P utm an, F u m iv al,
and McKnight (1960).
References
A. F. R . I . (A u stra lia n F o r e s t R e s e a r c h Institute) (1964). F o r e s t r y

T a b le s P a r t I. M e n s u ra tio n . F o r e s t r y and T i m b e r B u reau ,
C anberra.
A itken. J . D. and H all. N. (1951). A note on the in t e r p r e ta tio n of f o r e s t
in fo rm a tio n fro m a i r p h o to g rap h s. A ust. F o r . . 15(1): 9-16.
A lgan, H. (1902). T a r i f s de cubage. Bull. Soc. F o r . F r a n c h e - C o m te ,
6: 123-30.
A nuchin, N . P . (1962). D e term in atio n of the c u r r e n t a c c r e tio n of
s ta n d s by the l a t e r a l s u r f a c e of t r e e s . P r o c . Fifth World F o r .
C o n g r . , U . S- A. , 1960.
B a k e r , F . S . (1950). P r in c i p l e s of s i l v i c u l t u r e . New York.
B ecking, J . H . (1953). Thinning r e s e a r c h in f o r e s t r y . Neth. J o u r .
A g ric . Sei. , 1(2): 122-9.
B e d e ll. G. and B e r r y . A. (1955). A m ethod of d e te rm in in g ap p ro x im ate
m e r c h a n ta b le v o lu m e s . Can. F o r . R e s . Div. Tech. Note No. 14. Ottawa.
B e h r e . C. E. (1927). F o r m c l a s s t a p e r c u r v e s and volum e ta b le s
and t h e i r ap p licatio n . J o u r . A gr, R es. . 35: 673-744.
B en so n . M . L . (1961a). T he Suunto c lin o m e te r . A ust. F o r . , 25(2):
122- 6 .
---- (1961b). O ptical wedge p r i s m s . In st. F o r. A ust. N e w s l e t t e r , 2(13):
7-10.
---- (1964). Volum e ta b le co m pilation u sing e le c tro n ic c o m p u te r s .
P a p e r p r e s e n te d to ANZAAS C o n g r e s s , C a n b e r r a .
---- and Slinn. R. J , (1964). The P e n ta p r i s m t r e e c a l ip e r . A ust. F o r .
R e s . , 1(1): 34-42.
B . F . A . ( B r itis h F o r e s t r y A sso ciatio n ) (1953). B r it is h C om m onw ealth
F o r e s t r y T e rm in o lo g y . London.
B ic k fo rd , C . A . (1959). A t e s t of continuous in v e n to ry fo r N ational
F o r e s t M anagem ent based upon a e r i a l p h o to g rap h s, double sam pling
207
208 R e fe re n c e s
and re m e a s u re d p lo ts . P ro c . Soc. A m e r. F o r. , 1959: 143-8.
---- (1963). On s u c c e s s iv e f o r e s t in v e n to rie s . P r o c . Soc. A m e r. F o r. .
1963: 25-30.
.---- , M ay er, C . E . , and W are, K. D. (1963). An e ffic ie n t sam p lin g d esig n
fo r fo re s t in v en to ry ; th e N o rth E a s te rn f o r e s t r e s u r v e y . J o u r, F o r . ,
61: 826-33.
Bow ling, P . J . (1951). V olum e ta b le fo r re g ro w th of so m e T a sm a n ia n
e u c a ly p ts. A u st. F o r . , 15(2): 8 7 -9 5 .
---- (1954). U se of s c ru b s p e c ie s in age d e te rm in a tio n . In st. F o r . A u s t.
N e w sle tte r, 10:1.
Boyd, J . (1950a). T r e e grow th s t r e s s e s . I G row th s t r e s s ev alu atio n .
A ust. J o u r. Sei. R e s . , B3: 270-93.
---- (1950b). T r e e grow th s t r e s s e s . II The d ev elo p m en t of sh a k e s and
o th e r v isu a l f a ilu r e s in tim b e r. A ust. J o u r . App. Sei. , 1(3): 296-312.
---- (1950c). T re e g row th s t r e s s e s . Ill T he o rig in of g row th s t r e s s e s .
A ust. J o u r. Sei. R e s . , B3: 294-309.
B rad ley , R . T . , C h r is tie , J . M . , and Jo h n so n , D. R. (1966). F o r e s t
M anagem ent T a b le s . F o r. C o m m . B ooklet No. 16, London.
B rie g le b , P . A. (1952). An ap p ro ac h to d e n s ity m e a s u re m e n t in D ouglas
f ir . J o u r. F o r. , 50(7): 529-36.
B row n. D. (1954). M ethods of su rv e y in g and m e a s u rin g v eg etatio n .
C o m m . A g ric . B u r. B ull. 42, London.
B row n. G. S. (1965). P o in t d en sity in s te m s p e r a c r e . N . Z . F o r.
R es. Note No. 38. W ellington.
B row n, J . K . (1963). C row n w eig h ts in re d pine p la n ta tio n s . U. S.
F o r .S e r v . L ake S tate F o r. Expt. Stn. P a p e r No. L S -1 9 .
B ru c e , D. and S ch u m a ch er, F . X. (1950). F o r e s t m e n s u ra tio n . 3rd
ed. New Y ork.
B ru n ig , E . F . (1963). H isto ry of f o r e s t in v e n to ry in S araw ak . M a la y .
F o r. , 34(3): 1 4 1 -5 8 .'
B u rs ta ll, S. (1957). C om bined ta p e r and volum e ta b le s f o r E uropean
la rc h , New Z ea lan d , a ll sta n d s 1 956. N. Z . F o r . R e s. N ote No. 14,
W ellington.
B u rt, E . A . P . (1888). R ailw ay r a te s s ta n d a rd tim b e r m e a s u r e s .
London.
B usgen, M. , M unch. E. , and T h o m so n , T. (1929). T he s tr u c tu r e and
life of fo r e s t t r e e s . London.
B y le s, B. U. (1958). The re s p o n s ib ility of th e f o r e s tr y p ro fe s s io n
to w a rd s th e m an ag em en t of fo re s te d lan d s fo r th e p u rp o s e of
p ro te c tio n . A u st. F o r. , 22(2): 5 5 -7 .
R eferences 209
C a ja n d e r , A. K. (1926). T h e o ry of f o r e s t t y p e s . H elsinki.
C a ro w , J . and Stage, A. (1953). B a s a l a r e a by the thum b. J o u r . F o r . ,
51(7): 512.
C a r r o n , L . T . (1953). M e a s u re m e n t of s ta c k e d firew ood. A ust. F o r . ,
17(2): 4 4-8.
---- (1955). Site c la s s ific a tio n and volum e a s s e s s m e n t of a P inus r a d ia ta
p la n ta tio n . A ust. F o r . , 19(2): 141-5.
---- (1964). The p ro p o rtio n s of stand m e rc h a n ta b le volum e by log s iz e
c l a s s e s in P in u s r a d i a t a . Aust. F o r . , 28(4): 258-68.
---- (1967). A v a r ia b le d en sity yield ta b le fo r a plantation of P in u s
r a d i a t a . A ust. F o r . , 31(1): 10-18.
---- and H all, N. (1954). National f o r e s t in v e n to ry B eech F o r e s t
M ili ta r y Sheet. A ust. F o r . , 18(2): 128-40.
---- and J a c o b s , M .R . (1964). An a r ti f ic i a l ta p e r ta b le for e v en -a g ed
s ta n d s of P in u s r a d ia ta . F o r . a n d T i m b . B u r . L e a fle t No. 90,
C anberra.
---- and M cIn ty re. G .A . (1959). T he ta p e r lin e (Gray) and Sectional
M ethods fo r m e rc h a n ta b le volum e of P in u s r a d ia ta s te m s . A u s t.
F o r . , 23(1): 50-60.
C hapm an, H .H . and M e y e r, W.H. (1949). F o r e s t M e n s u ra tio n . New
Y ork.
C h a tu rv e d i, M .D . (1926). M e a s u r e m e n ts of the cubical co n ten ts of
f o r e s t c r o p s . O xford F o r e s t r y M e m o ir, Oxford.
C h is m a n , N .H . and S ch u m a ch er, F. X. (1940). On the t r e e a r e a r a tio
and its ap p lic a tio n s . J o u r . F o r. , 38: 311-17.
C lu tte r , J . L. (1963). Compatible grow th and yield m o d els f o r loblolly
pine. F o r . Sei. , 9(3): 354-71.
C o c h ra n , W .G . (1963). Sampling t e c h n iq u e s . 2nd ed. , New Y ork.
C o o p e r, C. F. (1961). Equations f o r the d e s c r ip tio n of p a s t growth
in ev e n -a g e d s ta n d s of p o n d e ro s a pine. F o r . Sei. , 7(1): 72-80.
C o u lte r , F . (1959). D ensity of P inus r a d ia ta logs. N . Z . J o u r . F o r . ,
8(1): 143-7.
C r o m e r , D .A .N . (1949). A e ria l s u r v e y and photo in te r p r e ta tio n in
A u s tr a lia n f o r e s t r y . P r o c . T h ird W orld F o r . Congr. , H elsinki.
---- (1952). B a s a l a r e a p e r a c r e . A ust. F o r . , 16(2): 57-61.
---- (1959). A volum e ta b le p r o g r a m m e . I n s t . F o r . A u s t . N e w s l e t t e r ,
2(2): 8.
---- (1960). A u s tr a lia n f o r e s t r e s o u r c e s and t h e i r a s s e s s m e n t . A u s t .
F o r . , 24(1): 22-4.
---- (1961). New a p p ro a c h e s in f o r e s t m e n s u r a t i o n . F o r . and T i m b .B u r . ,
C anberra.
---- and Bow ling, P . (1961). T he d ev elo p m en t of a y ie ld ta b le fo r
E ucalyptus obliqua re g ro w th in S o u th ern T a s m a n ia . P a p e r p re s e n te d
to 2nd W orld Euc. Conf. , B ra z il.
---- and B row n, A. (1956). P la n ta tio n in v e n to rie s w ith a i r photos and
angle count sam p lin g . F o r. and T im b . B u r. B ull. N o. 34, C a n b e rra .
---- and C a r r on, L . T . (1956). Stand vo lu m e ta b le s - w ith p a r tic u la r
re fe re n c e to P in u s r a d ia ta . F o r. and T im b . B u r. B u ll. No. 35.
C a n b e rra .
---- and ---- (1957). T he volum e lin e w ith re fe r e n c e to P in u s ra d ia ta ;
its d e riv a tio n w ithout th e u se of s a m p le t r e e s . F o r. and T im b . B u r.
L e a fle t No. 76, C a n b e rra .
---- , M cIn ty re, G . A . , and L ew is, N. (1955). G e n e ra l v o lu m e ta b le fo r
P inus ra d ia ta . F o r . and T im b . B u r . B ull. No. 35, C a n b e rra .
C unia. T. (1964). W hat is sam p lin g w ith p a r tia l re p la c e m e n t and why
u se it in continuous f o r e s t in v e n to ry ? P r o c . Soc. A m e r. F o r . , 1964:
207-11.
---- (1965). C ontinuous f o r e s t in v e n to ry , p a r tia l re p la c e m e n t of s a m p le s
and m u ltip le r e g r e s s io n . F o r. Sei. , 11(4): 480-502.
C u rtin , R. A. (1962a). Ideal sto ck in g fo r m ax im u m p ro d u c tio n in an
ir r e g u l a r b la ck b u tt f o r e s t. P a p e r p re s e n te d to ANZAAS C o n g re s s ,
C a n b e rra .
---- (1962b). The a p p licatio n of continuous f o r e s t in v e n to ry to a c o a s ta l
hardw ood fo r e s t. F o r e s tr y C o m m issio n of New South W ales
m a n u sc rip t.
---- (1964). Stand d en sity and th e re la tio n s h ip of crow n w idth to d ia m e te r
and h eig h t in E u caly p tu s o b liq u a . A u st. F o r. , 28(2): 91-105.
D ale, M. E. (1962). An e a s y w ay to c a lc u la te crow n s u rfa c e a r e a .
J . F o r. , 60(11): 826.
D aw kins, H . C . (1952). E x p e rim e n ts in low p e rc e n ta g e e n u m e ra tio n s
of tro p ic a l high f o r e s t w ith s p e c ia l r e f e r e n c e to U ganda. Em p, F o r .
Rev. , 31(2): 131-45.
---- (1957). Some r e s u lts of s tr a tif ie d ran d o m sam p lin g of tro p ic a l
high f o r e s t. P a p e r p re s e n te d to 7th B r it. C om m . F o r. Conf. ,
C a n b e rra and New Z ealan d .
---- (1963). Crow n d ia m e te r s : th e ir r e la tio n to bole d ia m e te r in tro p ic a l
fo r e s t t r e e s . C om m . F o r . R ev. , 42(4): 318-33.
---- (1965). P o in t-s a m p lin g by a n g le -g a u g e : a sim p le a p p ro a c h . F o r .
S o c . J o u r . , 13: 28-35.
D eetlefs, P . P . (1953). M eans of e x p re s s in g and re g u la tin g d e n sity in
fo re s t s ta n d s . J o u r . Sth. A f r , F o r. A s s o c . , 23: 1-11.
R eferen ces 211
---- (1954). The relatio n sh ip betw een stand density, crow n size and
b a s a l a re a grow th in stands of Pinus taed a in the native h ab itat of th is
s p e c ie s. Jo u r, Sth. A fr. F o r. A ss o c ■, 24: 14-28.
de L io co u rt, F. (1898). De l'am en ag em en t des sa p in ie re s . Bull, de
la s o c .f o r .d e F ranche-C om te et B e lfo rt. (Quoted by Schaeffer,
A. et a l . (1930.))
de R o sa y ro , R. A. (1962). A pplication of a e ria l photography to fo re s t
m anagem ent in the tro p ic s. P r o c . Fifth World F o r. C o n g r. , U. S. A.
1960.
Duff, G. (1950). C om parison betw een volum es of forked and n o rm al
tr e e s of Pinus ra d ia ta in K aingaroa F o r e s t. N . Z . F o r. R es. Notes
1(2), W ellington.
---- (1954). Combined ta p e r and volum e tab les fo r P inus tax ifo lia, Pinus
n ig ra v a r. Cal. , and Pinus ra d ia ta. R otorua. N. Z. F o r. R es. Notes
1(2), W ellington.
---- (1959). M ethods of m easu rin g log volum e in New Z ealand. N. Z .
Tim b. Jo u r. , 5(9): 49-50; 5(10): 43-4.
---- (1960). M ethods of m easu rem en t and volume calculation fo r exotic
c o n ife rs . N . Z .F o r. R es. Tech. Pap. No. 29, W ellington.
---- and B u rs ta ll, S. (1955). Combined ta p e r and volum e tab les for
P inus ra d ia ta - Auckland. N elson. C anterb u ry and Southland. N. Z.
F o r . R es. Note No. 1, W ellington.
Dwight, T. W. (1937). The p e rc en t deviation m ethod of co n stru ctin g
volum e ta b le s. F o r. Chron. , 13: 409-16.
E ssed , F . E . (1955). De Kegel m ethode. Ned. Bosch. T ijd s. , 11:
285-7.
---- (1957). E stim ation of standing tim b e r . W ageningen.
FAO (1954). Ja l h y p so m e te r. Food and A g ricu ltu re O rganization,
United N ations. F o r. Eqpt. Note A. 2 .54 .
---- (1958). The L ig n o m ete r. Food and A g ricu ltu re O rganization,
U nited N ations. F o r. Eqpt. Note A. 11.58.
F. a n d T . B . (F o re s try and T im b er Bureau) (1951). D efinitions fo r
c la ssific a tio n of the fo re st e sta te by in te rp re ta tio n of a ir photos:
p re s c rip tio n s of the 1951 C onference on N ational F o re s t Inventory
fo r A u stra lia . M anuscript, C an b erra.
---- (1955). The avoidance of shadow point on a e ria l photographs.
F . a n d T . B . L eaflet No. 74, C an b e rra.
F a lc o n er, F. G. (1931). A m ethod of a cc u ra te height m e asu rem en t
fo r fo re s t tr e e s . Jo u r. F o r. , 29: 742-6.
F erg u son, J . H . A . (1952). D iam eter in c re m en t by d ia m e te r c la s s e s .
T e c to n a , 42: 79-88.
F in ch , H .D .S . (1957a). P lo tle s s e n u m e ra tio n w ith angle gau g es.
F o r e s t r y , 30(2): 173-92.
---- (1957b). New w ays of u sin g the g e n e ra l ta r if f ta b le s fo r c o n i f e r s .
F o r. C o m m .F o r. R ec. No. 32, London.
F in la y so n , W. and A r c h e r , G .R . (1964). L e tte r to C om m . F o r .R e v . ,
43(4): 285.
F inney, D .J . (1947). V olum e e s tim a tio n of stan d in g tim b e r by sa m p lin g .
F o r e s t r y . 21(2): 179-203.
---- (1948). R andom and s y s te m a tic sam p lin g in tim b e r s u rv e y s .
F o r e s t r y . 22(1): 6 4 -9 9 .
---- (1953). The e s tim a tio n of e r r o r in the s y s te m a tic sam p lin g of
f o r e s ts . J o u r. Ind. Soc. A g r. S ta ts. , 5: 6 -16.
---- and P a lc a , H. (1950). The elim in a tio n of b ia s due to edge e ffe c ts
in sam p lin g . F o r e s t r y , 23(1): 3 1-47.
F is h e r , R .A . and Y a te s, F . (1943). S ta tis tic a l ta b le s . London.
F o r e s tr y C o m m issio n (1952). C en su s of W oodlands 1 9 4 7 -9 . W oodlands
5 a c r e s and o v e r . F o r e s tr y C o m m issio n C en su s R e p o rt No. 1,
London.
---- (1953). H edgerow and p a rk tim b e r and w oods u n d e r 5 a c r e s .
F o r e s tr y C o m m issio n C en su s R e p o rt N o .2, London.
F o r r e s t, W. (1961). New te ch n iq u es in continuous in v e n to ry . In s t.
F o r.A u s t. N e w s le tte r , 2(11): 1 9-21.
F o rw a rd , D . F . and N olan, N . J . (1961). G row th and m o rp h o g e n e sis
in the C anadian f o r e s t s p e c ie s . Can. J o u r. B ot. , 39: 4 1 1-36.
F r a n c is , E . C . and Wood, G . H . S . (1955). The c la s s ific a tio n of
v egetation in N o rth B o rn eo fro m a e r ia l p h o to g rap h s. M alay. F o r e s t e r ,
18(1): 38-44.
F r e e s e , F . (1960). T e s tin g a c c u ra c y . F o r. Sei. , 6: 139-45.
F u rn iv a l, G. (1962). M achine p ro c e s s in g of f o r e s t s u rv e y d ata.
P r o c . Fifth W orld F o r. C o n g r. , U . S . A . , 1960.
G a rris o n , G. A. (1949). U se s and m o d ific a tio n s fo r th e m o o se h o rn
crow n c lo s u re e s tim a te s . J o u r. F o r. , 47: 733-5.
G ra v e s , H. S. (1914). F o r e s t m e n su ra tio n . New Y ork.
G ra y , H. R. (1943). V olum e m e a s u re m e n t of f o r e s t c r o p s . A u st. F o r. ,
7: 48-74.
---- (1944). V olum e m e a s u re m e n t of sin g le tr e e s . A u st. F o r. , 8: 44-65.
---- (1945). Site c la s s ific a tio n of co n ife ro u s p la n ta tio n s . A ust. F o r . ,
9: 9-17.
---- (1956). The fo rm and ta p e r of f o r e s t tr e e s te m s . Im p. F o r . In st.
O xford In s t. P a p e r 32, O xford.
---- (1966). P r in c ip le s of f o r e s t tr e e and c ro p volum e g ro w th . F o r.
and T im b . B u r. B u ll. No.42_, C a n b e rra .
G riffith , A .L . (1945-6). The effic ie n c y of e n u m e ra tio n s. Ind. F o r.
S erv . L e a fle ts 8 3 -9 3 , D ehra Dun.
---- and P r a s a d , J . (1949). T he s ilv ic u ltu ra l r e s e a r c h co d e, vol. 3:
T he t r e e and c ro p m e a s u re m e n t m a n u a l. D elhi.
G ro se n b au g h , L .R . (1952). P lo tle s s tim b e r e s tim a te s - new , f a s t,
e a s y . J o u r. F o r . , 50: 3 2 -7 .
---- (1958). T he e lu siv e fo rm u la of b e s t f i t . Sth. F o r. E ^pt. Stn. O cc.
P a p e r N o.158, L o u isia n a .
---- (1963). O p tical d e n d ro m e te rs fo r out of re a c h d ia m e te r s ; a
c o n sp ectu s and so m e new th e o r y . F o r. S e i.M o n o g r. N o .4, W ashington.
H ag b erg , E. (1957). The new Swedish n atio n al f o r e s t s u rv e y . U n a sy lv a ,
11: 3 -8 .
H all, M. J . (1959). T he effect of sto ck in g d en sity , s ite q u ality and age
on the volum e of pulpw ood p ro d u ced p e r a c r e . A. P . P . I. T . A. ,
13(1): 22-9.
---- and N eal, H. (1966). D ata p ro c e s s in g in f o r e s tr y - an ex am p le.
A ust. F o r. , 30(4): 283-92.
H all, O. F . (1959). T h e co n trib u tio n of r e m e a s u re d sa m p le p lo ts to
th e p re c is io n of grow th e s tim a te s . J o u r. F o r. , 57: 807-11.
H anson, A .G . and W ilson, D. (1960). M ethods of log m e a s u re m e n t
in A u s tra lia and T e r r i to r y of P ap u a and New G uinea. A u st. F o r. ,
24(1): 4 6 -8 .
H a rris o n , J .D .B . (1950). P lan n in g a n atio n al f o r e s t in v e n to ry . F .A .O .
F o r. and F o r. P ro d . S tudies No. 1, R om e.
H eg er, L. (1965). M o rp h o g en esis of s te m s of D ouglas f ir . P h .D . th e s is ,
D e p t.F o r. , U n iv e rs ity of B ritis h C olum bia.
H ein sd ijk , D. (1961). F o r e s tr y s u rv e y in the A m azon V alley .
U n asy lv a, 15(4): 167-74.
---- (1962). S urveys ap p licab le to e x te n siv e fo re s t a r e a s in South
A m e ric a . P r o c . F ifth W orld F o r . C o n g r. , U . S . A . , 1960.
H elm s, A. D. (1945). A gian t eu caly p t (E. re g n a n s ). A ust. F o r. , 9(1):
25-8.
H em m ings, W. D. (1949). The u s e of a e r ia l pho to g rap h s in f o r e s tr y .
A u s t. T im b . J o u r ., 15(4): 256-65.
H e n ric k se n , H. A. (1950). H o jd e -d ia m e te r d ia g ra m m ed lo g a ritm is k
d ia m e te r [H e ig h t-d ia m e te r d ia g ra m w ith lo g a rith m ic d ia m e te r ] .
D ansk. Skovforen. T i d e s k r . , 35: 193-202.
214 R eferences
H enry, J . L. (1955). Sustained yield m anagem ent of hardw ood fo re s ts
in N .S .W . A ust. F o r. , 19(1): 45-59.
---- (1960). Volume table p re p a ratio n fo r Pinus ra d ia ta in N. S. W.
P a p e r p resen ted to F ir s t A u stralian C om puter C onference, Sydney.
Hiley, W. E. and L eh tp ere, R. (1955). Thinning g ra d e s based on the
thickness of annual rin g s. F o r e s tr y , 28(1): 17-32.
H irata, T. (1955). Height estim atio n through B itte rlic h 's method:
v e rtic a l angle count sam pling. Jap . J o u r. F o r. , 37: 479-80.
Hodge, J. D. (1965). V ariable plot cru isin g : a s h o rt-c u t slope
c o rre ctio n m ethod. Jo u r. F o r. , 63(3): 176-80.
H um m el, F .C . (1949). The m ethods em ployed in the N ational F o re s t
Survey of Gt. B ritain 1947-49. P a p e r p re sen te d to 3rd World F o r.
Congr. , H elsinki.
---- (1951). In stru m en ts for the m e asu re m e n t of height, d ia m e te r and
ta p e r on standing tr e e s . F o r. A b str. , 12(3): 261-9.
---- (1952). An ex p erim en t on the sam pling of e a rly thinnings. F o r e s tr y ,
25(1): 19-31.
---- (1953). The definition of thinning tre a tm e n ts . P ro c . 11th IUFRO
Congr. , Rome: 582-3.
---- (1955). The volu m e-b asal a re a line. F o r. Comm. B ull.N o. 24,
London.
---- and C h ristie , J . (1953). R evised yield tab les fo r c o n ifers in G reat
B rita in . F o r. Comm. R ec. No. 24, London.
---- and ----- (1957). M ethods used to c o n stru ct the re v ise d yield tab les
fo r conifers in Gt. B rita in . Rep. on F o r. R es. , F o r. Comm. , London.
---- et a l . (1959). Code of sam ple plot p ro c e d u re. F o r. C om m . Bull.
No. 31, London.
---- et a l . (1962). T a riff ta b le s , F o r. C om m . Rec . No. 31, London.
H um phreys, N. (1961). The use of a e r ia l photographs in the
p rep aratio n of m anagem ent plans fo r m ixed eucalypt fo re s ts on the
north coast of New South W ales. A u st.F o r. , 25(2): 92-101.
Husch, B. (1963). F o re s t m en su ratio n and s ta tis tic s . New York.
Ilv essalo , Y. (1947). Volume tab les fo r standing tr e e s . Comm. In s t.
F o r. Fenn. , 34(4): 1-149.
---- (1952). The th ird national survey on the fo re s t of Finland. Comm.
Inst. F o r. Fenn. , 39(3).
Jack , W .H. (1962). The collection of grow th inform ation fo r
m anagem ent. P a p e r p re se n te d to 8th B rit. Comm. F o r. C o n f., E ast
A frica.
Jaco b s, M .R . (1939). An in stru m e n t fo r m e asu rin g the height in crem en t
of p in e s . C o m m . F o r. B u r. L e a fle t No. 51.

---- (1955). G row th h a b its of th e e u c a ly p ts . C a n b e rra .
---- (1962). T he e ffe c t of g rad e of thinning on g row th, rin g w idth and
thinn ing y ie ld in p la n tatio n s of P in u s r a d ia ta in A u s tr a lia . F o r. and
T im b .B u r. L e a fle t No. 84. C a n b e rra .
---- (1965). S tr e s s e s and s tr a in s in tr e e tru n k s a s th ey grow in length
and w idth. F o r. and T im b .B u r. L e a fle t N o .96, C a n b e rra .
J e f f e r s , J .N .R . (1955). The B a r r and Stroud D e n d ro m e te r. F o r .
C om m . R e p o rt on F o r. R e s . , 1954-5.
J e n n in g s , L .G . (1965). G lo s sa ry of log and tim b e r m e a s u r e s . W ood,
30(8-12).
J e r r a m , M .R .K . (1939). E le m e n ta ry f o r e s t m e n s u ra tio n . London.
J o h n sto n , D. R . (1960). P ro b le m s of yield co n tro l and in v e n to ry in
B r itis h f o r e s tr y . F o r e s tr y , 33(1): 19-36.
---- and B ra d le y , R .T . (1963). F o r e s t m an ag em en t ta b le s . C o m m .
F o r. R ev. , 42(3): 217-27.
---- and ---- (1964). D evelopm ents in yield c o n tro l and in v e n to ry in
B r itis h f o r e s tr y . F o r e s tr y , 37(1): 21-30.
J o lly , N. W. (1950). The volum e lin e th e o ry in r e la tio n to th e
m e a s u re m e n t of the standing volum e of a f o r e s t. Woods and F o r.
D ept, m a n u s c rip t, A d elaide.
Jo n so n , T . (1910-12). T a x a to ris k a u n d e rso k n in g a r on s k o g strid e n s
fo rm . Skogs. T i d s k r . , 8(11): 285-328; 9(9-10): 285-329: 10(4):
2 3 5 - 75. (E n g lish tr a n s . by A. D. H elm s in F o r. and T im b . B u r . L ib. ,
C a n b e r r a .)
K a jih a ra , M. (1966). E x am in atio n on th e e r r o r of h e ig h t-d ia m e te r
c u rv e . B ull. Kyoto P r e f . Uni. F o r. , 10: 1 -14.
K een, E .A . (1950). The r e la s c o p e . Em p. F o r. Rev. , 29(3): 253-64.
K eev es, A. (1961). A c o n sid e ra tio n of c e rta in m e n su ra tio n m ethods
of in te n siv e m an ag em en t of co n ife ro u s p la n tatio n s in A u s tra lia .
Dip. F o r. th e s is , O xford U n iv ersity .
K endall, R. and S ay n -W ittg en stein , L. (1959). An ev alu atio n of the
re la s c o p e . Can. F o r. R e s . D iv. T e c h . N ote No. 77, O ttaw a.
K e r, J .W . and S m ith, J . H . G . (1955). A dvantages of th e p a ra b o lic
e x p re s s io n of h e ig h t-d ia m e te r re la tio n s h ip s . F o r. C hron. , 31(2):
2 3 6 - 46.
K inashi, K. (1954). F o r e s t in v en to ry by sam p lin g m e th o d s . B ull.
Kyushu U niv. F o r. No. 23.
---- N ish izaw a, M. , and K itagaw a, T. (1962). New d ev elo p m en ts of
sam p lin g d e sig n s in f o r e s t in v e n to rie s. B u ll. K yushu U n iv .F o r .
No. 35.
216 R eferences
Knuchel, H. (1953). Planning and co n tro l in the m anaged fo rest.
London.
Kozlowski, T .T . (1963). Growth c h a ra c te ris tic s of fo re s t tr e e s .
Jo u r. F o r. , 61(9): 655-62.
K rajicek, J . E . , B rinkm an, K .A . , and G ingrich, S. F. (1961). C row n
com petition - a m e asu re of density. F o r. Sei. , 7: 35-42.
K ra m e r, P . J . and Kozlowski, T .T . (1960). Physiology of t r e e s .
New York.
K rutzsch-L oetsch (1938). H olzv o rratsin v en tu r und L eistungsprufung
d er n atu rg em assen W aldw irtschaft. Neudam m . S. 164.
L an g saeter, A. (1949). F o re s t survey s of N orw ay. U nasylva, 3:
107-12.
L arso n , P .R . (1962). Auxin gradien ts and the regulation of cam bial
activity. In Kozlowski, T .T .( e d .) , T re e g ro w th . New York,
97-117.
---- (1963). Stem form developm ent of fo re s t t r e e s . F o r . Sei. M onogr.
No. 5.
L aver, C .F . (1951). P rin c ip le s of log m e a su re m e n t. London.
L aw rence, P .R . (1957). T esting the efficiency of p h o to -in terp retatio n
as an aid to fo re s t inventory. P a p e r p resen ted to 7th B rit. Comm.
F o r. Conf. , C an b erra and New Zealand.
---- (1960). A ra tio n a l approach to tre e volume m e asu rem en t. A u st.
F o r. , 24(2): 107-11.
---- (1963). The b a sis of m anagem ent planning fo r in teg rated u tilisa tio n
in southern T asm ania. A ust. F o r. , 27(1): 15-26.
---- (1965). A re g re s s io n m odel-building com puter p ro g ra m m e . F o r.
and Tim b. B ur. Bull. No. 40, C an b erra.
---- (1966). G irth c la ss unit volume tab les fo r eucalypt regrow th. F o r.
Comm, of T asm ania m an u scrip t, H obart.
---- and W alker, B .B . (1954). Methods and re s u lts of fo re s t a ss e ss m e n t
using random sam pling units in p h o to -in terp reted s tr a ta . A ust. F o r. ,
17(2): 107-27.
L eary , R. and B u m s, T. (1963). M easurem ent of upper stem
d ia m e ters with a tra n s it. Jo u r. F o r. , 61(6): 448-50.
Lem m on, P .E . (1957). A new in stru m en t (the s p h e ric a l d ensiom eter)
fo r m easuring fo re s t o v e r-s to re y . J o u r.F o r. , 55: 667-9.
---- and Schum acher, F .X . (1962). Volume and d ia m e te r growth of
ponderosa pine tr e e s as influenced by site index, density, age and
size. F.........
o r. Sei. , 8: 236-49. ^ ^
L escaffette, J. (1952). Une p ro p rie te des a rb re s et des
p e u p le m e n ts - la su rfa c e g e n e r a tr ic e . S o c .F o r .d e F ra n c h e -C o m te ,
25: 712-25.
L e w is, E .R . (1954). A v a ria b le d e n s ity y ield ta b le fo r P in u s ra d ia ta
in New Z e a la n d . N. Z . F o r. S erv. F o r . R e s .N o te s 1(10).
---- (1957). E r r o r s in stand volum e e s tim a tio n fo r exotic c o n ife rs in
New Z e a la n d . N. Z. F o r. R es. In st. N ote No. 9.
L ew is, N . B . (1953). A handy b ark g au g e. A ust. F o r . , 17(1): 7-8
---- (1954). P e rm a n e n t sam p le plot p r a c tic e in South A u s tra lia . A u s t.
F o r. , 18(1): 21-9.
---- (1957). M anagem ent fo r su sta in e d y ie ld of S tate P . r a d ia ta p la n ta tio n s
in South A u s tra lia . P a p e r p re s e n te d to 7th B r it. C om m . F o r. Conf. ,
C a n b e rra and New Z ealan d .
---- (1963a). The u se of yield plots in c o n ife ro u s p la n tatio n m an ag em en t
in A u s tra lia . C om m . F o r, Rev. , 42(2): 124-8.
— (1963b). O ptim um thinning ran g e of P in u s r a d ia ta in S th .A u st.
A u st. F o r. , 27(2): 113-20.
---- and M cIn ty re, G. A. (1963). R eg io n al volum e ta b le fo r P in u s r a d ia ta
in South A u s tr a lia . A d elaide.
L exen, B. (1943). B ole a r e a as an e x p re s s io n of grow ing sto c k . J o u r .
F o r . , 41: 883-5.
L indsay, A. D. (1939). R ep o rt on g row th stu d ie s of A lpine ash
(E u caly p tu s g ig a n tea) Bago S . F . No. 560. N. S. W .F o r . C o m m .
m a n u s c rip t, Sydney.
L o e tsc h , F . (1957). A f o r e s t in v en to ry in T h ailan d . U n a sy lv a ,
11(4): 174-80.
---- and H a lle r, K. E. (1964). F o r e s t in v e n to ry . M unich.
Luke, R . H . (1946). An e s tim a te of th e tim b e r r e s o u r c e s of New
B rita in . A u st. F o r. , 10: 72-86.
M cG rath, K . P . and C a rro n , L . T . (1966). P r e s c r ib e d y ield fro m
p re v io u s ly unm anaged indigenous f o r e s ts . P a p e r p re s e n te d to 6th
W orld F o r e s tr y C o n g re s s , Spain.
M cIn ty re, G . A . and C a rro n , L . T . (1954). B ias in s e le c tio n of sam p le
t r e e s fo r th e volum e line m ethod. I n s t . F o r. A u st. N e w s le tte r ,
No. 10: 2 -4 .
M a s te rs , S. , H ollow ay, J . , and M cK elvey, P . (1957). N atio n al f o r e s t
in v e n to ry of New Z ealan d . P a p e r p r e s e n te d to 7th B r it. C om m . F o r.
Conf. , C a n b e rra and New Z ealand.
M atern , B. (1962). F o r e s t su rv e y s and th e s ta tis tic a l th e o ry of
sam p lin g - som e re c e n t d ev elo p m en ts. P ro c . F ifth W orld F o r .
C ongr. , U . S . A . , 1960.
218 R eferences
M athauda, J .S . (1957). Methods of fo re s t su rv ey s in India. P a p e r
p resen ted to 7th B rit. Comm. F or. Conf. , C an b erra and New Zealand.
M eyer, H .A . (1953). F o re s t m ensuration. Pennsylvania.
---- and Stevenson, D. (1943). The s tru c tu re and growth of virgin
b e ech -b irch -m ap le-h em lo ck fo re s ts in n orth Pennsylvania. J o u r .
A gr. R es. , 67(12).
M eyer, W .H. (1930). D iam eter distrib u tio n s e r ie s in even-aged fo re s t
s ta n d s . Yale U niversity School of F o r. Bull. No. 28.
M uller, G. (1953). Das B au m zah lro h r. Allg. F o r s tz . , 64(19/20).
Mulloy, G .A . (1944). Stand density v e rsu s stand bole a re a and stand
intensity indices in even aged stands. F o r. Chron. , 20: 167-70.
N ash, A. J. (1948). The Nash scale fo r m easu rin g tre e crown w idths.
F o r. C h ro n ., 24: 117-20.
N ousianen, A. O. (1965). The D iatrom b, a device fo r m easu rin g
d ia m e te rs of standing tr e e s . Jo u r. F o r. , 63(1): 21-2.
Nunn, G. W. M. (1957). F o re s t m anagem ent in the eucalypt fo re s ts
of the south-w est of W estern A u stra lia. P a p e r p resen ted to 7th
B rit. Comm. F o r. Conf. , C anberra and New Zealand.
---- (1959). A u stralian fo re s t re s o u rc e s and th e ir a sse ssm e n t, with
sp ecial re fe re n ce to the fo re s t inventory of W estern A u stra lia.
A ust. F o r. , 23(2): 100-4.
N yssonen, A. (1961). Survey methods of tro p ic a l f o r e s ts . FAO, Rom e.
---- (1964). A nalysis of two a lte rn ativ e m ethods fo r national fo re s t
inventories in n o rth e rn Europe. A cta. F o r. Fenn. , 76 A rt.6.
N. Z .F .R .I . (New Zealand F o re s t R ese arch Institute) (1965).
A ssessm en t m ethods in New Zealand and A u stra lia (Sym posium ,
R otorua, 1964). W ellington.
O sborne, J .G . (1946). Sampling e r r o r s of sy stem atic and random
surveys of co v er type a re a s . Jo u r. A m er. Stat. A ss o c . , 37: 256-64.
P aine, D .W .M . (1966). A nalysis of a fo re s t m anagem ent situation by
lin e a r program m ing. A ust. F or. , 30(4): 293-303.
Pegg, R. E. (1967). R elation of slash pine site index to so il, vegetation
and clim ate in so u th -e a st Q ueensland. Qld Dept. F o r. R es. Notes
No. 19.
P e trin i, S. (1921). Stam form s u n d erso k n in g ar. Medd, from S tatsk o g s,
18(4).
P ro d an , M. (1947). D er Starkezuw achs in P len terb estan d en Schweiz.
Z . f . F orstw . , 98.
---- (1965a). H o lzm essleh re. F rank fu rt.
---- (1965b). A sim plification of the volume ta riff sy ste m s. Conf. Adv.
R eferences 219
G roup F o r . S ta t s . IU F R O S e c .2 5 , Stockholm, 1965.
P u r d u e U n iv e rs ity (1960). P r o c e e d in g s of f o r e s t m a n a g e m e n t c o n tro l
c o n f e r e n c e . Indiana.
P u tm a n , J . A . , F u rn iv a l, G. M. , and Me Knight, J . S. (1960).
M anagem ent and in v e n to ry of s o u th e rn h a r d w o o d s . U .S .D e p t .o f
A g ric . Handbook N o .181, W ashington.
Q u eensland D ep artm en t of F o r e s t r y (1962). F o r e s t su rv ey in g .
Q u eensland Dept, of F o r . m a n u s c r ip t.
R e in e c k e , L .H . (1926). The d e te rm in a tio n of t r e e volum e by p la n im e te r .
J o u r . F o r . , 24: 183-9.
---- (1933). P e r f e c tin g a stand den sity index fo r even aged f o r e s t s .
J o u r . A g r i c . R es . , 46: 627-38.
R ennie, P . J . (1963). Methods of a s s e s s i n g s ite c a p a c ity . C o m m .
F o r . Rev. . 42(4): 306-17.
R obinson, R .W . (1947). An in s tr u m e n t to m e a s u r e f o r e s t crow n
c o v e r . F o r . Chron. , 23: 222-5.
R o d g e r, G. J . (1953). Annual r e p o r t of the F o r e s t r y and T im b e r
B u re a u , C a n b e r r a .
R o m b e r g e r , J . A . (1963). M e r is te m s , grow th and d evelopm ent in
woody p l a n t s . U .S . Dept. A gric. F o r . S erv. T ech. B u ll. 1293.
S . A . F . (Society of A m e r ic a n F o r e s t e r s ) (1950). F o r e s t r y te r m in o lo g y .
W ashington.
S a r v a s , R. (1956). A m ethod of plotting crow n p r o je c tio n s fo r crow n
m apping. P a p e r p r e s e n te d to IUFRO Conf. , Oxford.
S ch aeffer, A. et a l . (1930). Selection w orking by a r e a . E din.U niv.
F o r . D ept. B u ll. N o . 3.
Seely, H .E . (1957). F o r e s t in v e n to ries in C anada. P a p e r p r e s e n te d
to 7th B r it. C om m . F o r . Conf. , C a n b e r r a and New Z ealand.
---- (1961). Some in vestigations of f o r e s t sam p lin g m e th o d s. Can.
Dept. F o r . , F o r . R e s . B r . T e c h . N o t e N o . i l , Ottawa.
---- (1962). A e ria l p h o to g ra m m e try in f o r e s t s u r v e y s . P a p e r p r e s e n te d
to 5th World F o r .C o n g r . , U .S . A. , 1960.
---- (1964). C anadian f o r e s t inventory m e th o d s. Can. Dept. F o r . , F o r .
R e s . B r. Pub. No. 1068, Ottawa.
Shillinglaw, A .W . (1945). The m i lit a r y s u r v e y of the f o r e s t r e s o u r c e s
of th e C om m onw ealth T e r r i t o r i e s . A ust. F o r . , 9(2): 57-9.
Shiue, C h e rn g -J ia n n (1960). S ystem atic s am p lin g with m u ltip le random
s t a r t s . F o r . Sei. , 6: 42-50.
S im s, W .G. (1954). Shadow p o in t. F o r . and T im b . B u r. L eaflet
No. 67, C a n b e r r a .
---- and H all, N. (1955). Some re q u ire m e n ts fo r p h o to -in te rp re ta tio n in
A u s tra lia . A u st. F o r . , 19(2): 120-35.
S m ith e rs , L. (1949). A sim p lific a tio n of th e continuous in v e n to ry
m ethod of c a lc u la tin g d ia m e te r g ro w th . Can. Dorn. F o r. S erv . L ea flet
No. 31.
S p u rr, S .H . (1952). F o r e s t in v e n to ry . New Y ork.
---- (1960). P h o to g ra m m e try and p h o to - in te r p r e ta tio n . New Y ork.
---- (1962). A m e a s u r e of point d en sity . F o r. Sei. , 8(1): 85-96.
S ta e b le r, G. (1954). S tandard co m p u tatio n s fo r p e rm a n e n t sam p le
p lo ts . P u g et Sound R e s. C en tre A dvis. C om m . , P a c . N . W. F o r. R e s .
E x p t. S tn .
S toate, T . N . (1945). The u se of a volum e eq u atio n in pine sta n d s .
A ust. F o r. , 9(1): 48 -5 2 .
S to effels, A. and van S o est, J . (1953). P r in c ip le V raag stu k en bij
p ro e fp a rk e n . N ed. B oschb. T ijd s . , 25: 190-9.
S tran d . L. (1957). 'R e la s k o p is k 'h o y d e -o g kubikk m a s s e b e s te m e ls e .
N o rsk . Skogbruk. , 3: 535-8. (Quoted by H usch, B. (1953).)
S u tte r, H. (1965). D ia m e te r m e a s u re m e n t of la rg e t r e e s w ith a w ide
s c a le re la sk o p . J o u r . F o r. , 63(2): 101-2.
T a k a ta , K. (1962). Study of m ethod of stan d volum e e s tim a tio n by
in te g ra l g e o m e tric a l m e th o d . B ull. N iig ata U niv. F o r. 1.
T a sm a n ia n F o r e s tr y C o m m issio n (1953). S tan d ard volum e ta b le fo r
E ucalyptus obliqua seco n d grow th fro m the s o u th e rn f o r e s ts .
F o r e s tr y C o m m issio n of T a sm a n ia m a n u s c rip t, H o b art.
T hom son, A . P . (1946). D esign fo r a f o r e s t s u rv e y . N . Z . J o u r, of
F o r, , 5(3): 191-9.
T hom son, G. W. and D eitsch m an n , G. H. (1959). B ib lio g rap h y of w orld
li te r a tu r e on th e B itte rlic h m ethod of p lo tle s s c ru is in g 1 9 4 7 -5 9 .
A g ric. and Home E con. Expt. Stn. , Iowa.
T ro re y , I . G. (1932). A m a th e m a tic a l m ethod fo r the c o n stru c tio n of
d ia m e te r-h e ig h t c u rv e s b ased on s ite . F o r. C hron. , 18(2): 3 -1 4 .
T u r n e r , B. J . (1966). An in v e stig atio n into th e grow th of ir r e g u l a r
eu caly p t stan d s in New South W ales. F o r. C o m m . N . S. W. R es. N ote
No. 18, Sydney.
U n iv e rsity of T o ro n to (1967). Wood m e a s u r e m e n t. T e c h .R e p o rt
No. 7, co n fe re n c e p r o c e e d in g s . E d ited by F. B uckingham .
Unwin, P . T . and B ow ling, P . J . (1951). P ro d u c tio n of volum e ta b le s
fo r m a tu re tr e e s in T a sm a n ia . A u st. F o r, , 15(1): 31-41.
Van Schie, R. (1967). M easu rin g stan d in g sa m p le tr e e s w ith a
d e n d ro m e te r in reg ro w th f o r e s ts of so u th e rn T a s m a n ia . A ust. F o r. ,
31(3): 181-92.
R eferences 221
V ezina, P . E . (1962). Crown w id th - d . b . h . r e la tio n s h ip s f o r open

grow n b a lsa m f i r and white s p r u c e in Q uebec. F o r . Chron. , 38(4):
463-73.
W ahlenburg, W. (1941). Methods of f o r e c a s t i n g t i m b e r grow th in
i r r e g u l a r s ta n d s . U. S. Dept. A g ric . T e c h . Bull. 796. W ashington.
W a lte r s , J. and Soos, J . (1962). T he g im b a l sight f o r the p ro je c tio n
of crow n r a d i u s . B rit. Col. Univ. F ac . of F o r . R e s .N o t e No. 39.
W a re ,K . D. and Cunia, T. (1962). Continuous f o r e s t inventory with
p a r ti a l r e p la c e m e n t of s a m p l e s . F o r . Sei. Monog. No. 3.
W a rre n , W .G. (1958). T e s t s of so m e in s tr u m e n ts f o r m e a s u r i n g t r e e
h e ig h t. N. Z. F o r . R es. Inst. R e s .N o t e No. 12.
---- (1959). The d is trib u tio n of stand m e r c h a n ta b l e volum e by log d ia m e te r
c l a s s e s . N . Z . F o r . R e s. I n s t . , Note No. 15.
W a te r s , W. T. (1961). R ecent d e v e lo p m e n ts in f o r e s t inv e n to ry in the
F o r e s t r y C o m m issio n . P a p e r p r e s e n t e d to the B r it. A ss o c , fo r Adv.
Science, S ec.K .
W h ee ler, P . R . (1962). P e n ta p r i s m c a l i p e r for u p p e r ste m d ia m e te r
m e a s u r e m e n t s . J o u r . F o r. , 60(12): 877-8.
Wile, B . C . (1964). Crown s iz e and s te m d ia m e te r in re d s p r u c e and
b a ls a m f i r . Can. Dep. F o r. P u b .N o . 1056, Ottawa.
W ilson, F . G . (1946). N u m e ric a l e x p r e s s i o n of stocking in t e r m s of
height. J o u r . F o r . , 44(10): 758-61.
---- (1955). Evaluation of t h r e e thinnings a t S tar L ake. F o r . Sei. ,
1: 237-41.
Wilson, R . C . (1962). Surveys ap p licab le to ex te n siv e f o r e s t a r e a s
in N o rth A m e r ic a . P r o c . Fifth W orld F o r . C ongr. , U . S . A . , 1960.
Wilson, T. and B r e n d e r , E. V. (1963). C o m p a ris o n of stand den sity
m e a s u r e s fo r loblolly pine cubic foot grow th p re d ic tio n . F o r . Sei. ,
9(1): 8-14.
W o m ersley , J . and M cAdam, J . B . (1957). The f o r e s t s and f o r e s t
conditions in the t e r r i t o r i e s of P a p u a and New Guinea. P a p e r
p r e s e n te d to the 7th B rit. C om m . F o r . Conf. C a n b e r r a and New
Z ealand.
W right, W. G. (1927). T a p e r as a f a c t o r in the m e a s u r e m e n t of
standing t i m b e r . Can. Dorn. F o r . S erv. Bull. No. 79.
Y ates, F . (1949). Sampling m ethods f o r c e n s u s e s and s u r v e y s .
London.
Index
A e ria l photographs in fo re st in v e n to ry , 1 9 2 -4 Error: a c c id e n ta l, 3; in h e ig h t m e a s u re m e n t,

A ge: stan d ard , 130; tr e e , 6 4 -5 3 4 -5 ; m ista k e s, 2; sa m p lin g , 3; sta n d a rd ,
A lig n m e n t c h a rt, 110 3
A llo c a tio n of sam p le: o p tim u m , 178; E stim a te , 2, 3 -4
p ro p o rtio n a l, 178
A ngle c o u n t, 7 9 -8 5 ; p o in t (spot), 181 F ace m e a su re , 7
A n n u al ring, 6 4-5 F e rtility g ra d ie n t, 187
A rith m e tic m e a n v o lu m e tre e m e th o d , 93 F ifth g irth m e th o d , 12
A ssessm ent, 192, 194; m a n a g e m e n t, 196 Forest: d e fin itio n , 69: in v e n to ry survey, 193
Form , 3 6 -4 4 ; cla ss, 41; fa c to r, 40; p o in t, 41;
Bark, 4 5 -8 ; g a u g e, 46; re la tio n sh ip s, 4 7 -8 q u o tie n t, 40
Basal a re a , 20 F u lly sto c k e d (n o rm a l) stands, 129, 164
Bias, 3 F req u en cy : c u rv e , 71; d ia g ra m , 71;
B iltm ore stick , 20 d istrib u tio n , 71; p o ly g o n , 71
Board fo o t, 6
Bole a re a , 1 23-6, 141 G irth , a t b re a s t h e ig h t, 18
B reast h e ig h t, 17-18 G ra p h ic a l m e th o d (stem v o lu m e ), 5 8 -9
Butt, 10; b u tt-s w e ll, 10, 43, 61 G roup m eth o d s, 94
G row th: c h a ra c te ris tic s , 6 6 -7 ; d e fin itio n , 63;
C a lip e r, 19 p e rc e n ta g e , 169; p ro je c tio n s, 170; stand
C o llim a tio n a d ju stm e n t, 30 v a ria b le s , re la tio n sh ip s w ith , 143
C la ss ific a tio n , 174; surveys, 193
C lin o m e te r, 29 H a rm o n ised c u rv e m e th o d , 105
C lu ster sa m p lin g , 183 H a rm o n ise d reg ressio n series, 108
C o m p le te e n u m e ra tio n , 1 9 9 -2 0 0 H eig h t: le a n in g tr e e , 3 1 -4 ; rods (sticks), 27;
C o n tro l m e th o d (m eth o d e du c o n tro le ), 197, sp acin g ra tio , 140; stan d , 1 2 9 -3 0 ;
1 99-200 tre e , 2 6 -3 5
C ord, 7 H isto g ram , 71
Crown: b reak, 10; c a n o p y , 1 4 1 -3 ; clo su re , Hoppus (q u a rte r g irth ): s e c tio n a l a re a , 5;
50, 142; c o m p e titio n fa c to r, 1 4 2 -3 ; v o lu m e , 13
le n g th , 50; ra tio , 48; su rface, 50; H u b er's fo rm u la , 11
v o lu m e , 50; w e ig h t, 50 H y p so m eter, 28
Cun it, 7
Id e n tific a tio n of sp e c ie s, 191
D ata processing, 191 In c re m e n t: b a sa l a re a , 160; b o rer, 66;
D e c re m e n t in d. b. h. o . b . , 1 5 9 -6 0 c u rre n t a n n u a l, 148; cu rv es, 67, 146;
D ensity: and grow th, 1 4 4 -5 ; p o in t, 144; d . b . h . o . b . , 4, (c o m in g p e rio d ) 148,
stand, 135-45, (in d ex es of) 1 3 7-43 (d e te rm in a tio n of, in d iv id u a l tre e s
D ia m e te r: a t b re a st h e ig h t, 1 9 -2 0 ; u p p er id e n tifie d ) 149, (d e te rm in a tio n of,
stem , 22 in d iv id u a l tre e s n o t id e n tifie d ) 14 9 -5 6 ,
D ire c tin g curve m e th o d , 132 (ex p ressio n of) 1 5 6 -9 , (p ast p erio d ) 148;
d e fin itio n , 63; h a m m e r, 66; h e ig h t, 1 6 0-1;
Edge effe c ts, 186-7 m e a n a n n u a l, 67, 146; p e rio d ic m e a n
E n u m eratio n , 71 a n n u a l, 67; stan d , 1 6 0 -1 ; v o lu m e , 16 1 -2
223
224 In d ex
In d ic a to r p la n ts, 128 S e c tio n a l m eth o d , 53
Ingrow th, 154, 169 S ilve, 200
Inventory: c o n tin u o u s, 197; m a n a g e m e n t, S ite in d e x , 131
171, 1 9 6 -2 0 6 ; o p e ra tio n a l, 199; p e rio d ic , Site q u a lity , 1 2 7 -3 4
1 9 7-206; reso u rc e , 171, 1 7 2 -9 6 ; w orking S m a lia n 's fo rm u la , 11
p lan , 197 S ta c k e d m a te ria l, 7, 16
Stand: b a sa l a re a , 78; d e fin itio n of, 69;
L eaning tr e e , h e ig h t m e a s u re m e n t, 3 2 -4 form fa c to r, 86, 115; d e n sity , 1 3 5 -4 5 ,
Load, 6 (in d ex es) 137—43; h e ig h t, 8 6 -9 2 , (cu rv e)
Log rules, 14 87, 8 8 -9 2 ; m e a n b a sa l a re a , 78; m e a n
Line p lo t, 182 d . b . h . o . b . , 78; m e a n d o m in a n t h e ig h t,
87; m e a n h e ig h t, 86; m e a n to p h e ig h t, 87;
M ath eso n 's s c a le (v o lu m e ), 13 p re d o m in a n t h e ig h t, 87; p ro je c tio n , 1 6 6 -9 ;
M o rta lity , 154, 169 ta b le , 71; ta b le p ro je c tio n , 1 6 7 -8 ;
M o v em en t fa c to r, 168 v a ria b le s , 69
S tan d ard ag e, 130
N orm al: c u rv e , 72; d is trib u tio n , 72; y ie ld S tem : an aly sis ( c o m p le te ), 66 (p a rtia l) 66;
ta b le s, 1 6 2 -4 ; (fu lly sto c k e d ) stands, 129, form th e o rie s, 38; p ro file , 4 0 -2 ; v o lu m e ,
164 5 1 -6 2
S to ck in g , 138
O cu lar e s tim a te : of d ia m e te r , 25; of h e ig h t, S tra tific a tio n , 175
35; of sa m p lin g u n it w id th , 188-91 S urvey, fo re st in v e n to ry , 197
Past in c re m e n t, 148-56 T a lly , 72

P e rc e n tile m e th o d , 5 5 -6 T a p e , 18
P e rm an en ce of units, 184, 201 T a p e r, 9'-10, 3 6 -4 4 ; c u rv e , 58; lin e (G ray),
Pipe, 14 5 9 -6 2 ; ta b le s, 120-2
P oint d en sity , 144 T ariffs, 100
P recision, 3 T e m p o ra ry u n its, 2 0 1 -3
Processed m a te r ia l, 8 T on, 7
P ro jectio n : o f b a sa l a re a , 168; of d. b . h. o. b . , T r e e - a r e a ra tio , 142
167-8; h e ig h t, 168; v o lu m e , 1 6 8 -9 T re e v o lu m e ta b le s, 9 5 -1 1 5 ; one in d e p e n d e n t
v a ria b le , 9 6 -1 0 3 ; tw o in d e p e n d e n t
Q u arter g ir th (H oppus): s e c tio n a l a re a , 5; v a ria b le s (a p p lic a b ility ) 111, (a p p lic a tio n )
v o lu m e , 13-14 111, (c o m p ila tio n ) 1 0 3 -1 0 , (m a lfo rm e d
tre e s) 1 1 1 -1 2 ; tw o -p lu s in d e p e n d e n t
R em o v als, 154, 169 v a ria b le s , 112-15
R oot spread, 143
R unning f e e t, 7 U nits, m e a s u re m e n t: of g irth , 5; of hew n
tim b e r, 7; of le n g th , 5; of log v o lu m e ,
S am p le, sa m p lin g : d e p e n d e n t, 201; 6; of p ro cessed m a te r ia l, 8; of round
in d e p e n d e n t, 201; in te n sity of, 1 8 4 -5 ; tim b e r, 7; of saw n tim b e r , 7; of
in te rp e n e tra tin g , 201; on su ccessiv e s ta c k e d m a te r ia l, 7; of tre e v o lu m e , 6
o ccasio n s, 2 0 1 -3 ; s e le c tio n (haphazard)
176, (ju d g m e n t) 1 7 6 -7 , (o b je c tiv e ) V a ria b le d e n sity y ie ld ta b le , 118
177-80, (purpo siv e) 1 7 6 -7 , (sim p le V o lu m e: c u rv e , 97; lin e , 97; log, 9 -1 6 ;
ran d o m ) 177, (s tra tifie d ra n d o m ) 178, stan d , 9 4 -1 1 2 , (by asso rtm e n ts) 1 2 0 -2 ;
(s u b je c tiv e ) 1 7 6 -7 , (s y ste m a tic ) 178 -8 0 ; ste m , 5 1 -6 2 ; ta b le s (stan d ) 1 1 5 -1 7 ,
s e le c tiv e sa m p lin g , 1 7 6 -7 ; size, 18 4 -5 ; (tre e ) 9 5 -1 1 5 ; see also T re e v o lu m e
stru c tu re , 181; s u b -sa m p lin g , 201; tre e ta b le s
m e th o d , 9 3 -5 ; tw o -p h a s e , 201; units
(a re a of) 1 8 8-9 1 , (e s ta b lish m e n t of) W eig h t m e a s u re m e n t, 1 4 -1 6
1 8 5 -8 , (o rie n ta tio n ) 1 8 3 -4 , (p o in t (spot))
1 81-2, (p lo t) 1 8 1 -2 , (sh ap e) 1 8 3 -4 , Y ield : d e fin itio n of, 146; expressions of, 146;
(size) 1 8 2 -3 , (strip) 181—2; w ith p a rtia l ta b le s , 1 1 7 -1 8 , 1 6 2 -6 , (n o rm a l) 1 6 2 -4 ,
r e p la c e m e n t, 202 (v a ria b le d e n sity ) 164-5
S e c tio n a l a re a , 20
Typed on IBM Executive Bold F ace No. 2 ty p e w rite r and printed
by G illingham P r in te r s P ty Ltd, A delaide
P rin te d and m anufactured in A u stra lia
R eg iste re d in A u stra lia fo r tra n s m is sio n by post as a book

Forest mensuration is the starting point
of organised forestry and is essential
to efficient management. This book
outlines the fundamental principles and
their application to practice in Australia.
Many of the methods and techniques
of forest mensuration in Australia have
been adapted from those of northern
hemisphere countries, but they have
been so modified to suit Australian
conditions that the available textbooks
are not suitable for Australian students.
This is the first book to provide a basic
text for a university undergraduate in
Australia. In addition the clear descrip
tions of forest mensuration practice
w ill make it of use as a reference for
practical foresters and forest owners,
in Australia and in other countries with
similar conditions.

Forest Mensuration: An Outline of

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Forest Mensuration: An Outline of

Uploaded by

Copyright:

Available Formats

An Outline of

Australian National University Press

T his book is co p y rig h t in a ll c o u n trie s s u b sc rib in g to th e B ern e

L ib ra ry of C o n g re ss C atalo g C ard no. 68-54900

Text books on fo re s t m ensuration are available fo r many co u n tries

A f o r e s t e r is c o n c e rn e d with m anaging f o r e s t s . I r r e s p e c t i v e of the

T re e s and th e ir products a re m e asu re d in v ario u s units depending on

L ength. In co u n tries using the English sy stem of u n its, such as

G irth, d iam eter and sectional a r e a . In A u stra lia , as in m ost

T ree vo lu m e. In A u stra lia the volume of the standing tre e is m ost

Stacked m a te r ia l. The common unit fo r stacked m a te ria l in

P ro c e s se d m a te ria l. The common unit for plywood, v e n ee r,

A log is a portion of the stem o r of a branch of a tre e p re p a re d for

The Shape of Logs

F o rm u lae to Give T rue Solid Volume

(b) V = s l . L (H uber's form ula)

F orm u lae to Give A pproxim ations to T ru e Solid Volume

^ . 2 su b stitu tes for —- the re s u lt being an o v e r-e stim a te of only

Hoppus M easure: Q u a rte r G irth Method

E stim ates of Log Volume in T erm s of F o rec a ste d P roduct

Weight M easu rem en t

In dealing with a population co m p risin g a num ber of individuals of

M easurem ent of G irth at B re a s t Height

M easurem ent of D iam eter at B re a st Height

Errors in c a lc u la tio n of tree se c tio n a l area

M ethod Area Error %

T hat is , calculatin g the c ro s s -s e c tio n as c irc u la r on the b a sis of the

C om parison of G irth Tape and C alip er

P rovided a tape is c o rre c tly graduated th e re is no in stru m en t

M easurem ent of D iam eter at O ther P o sitio n s on Standing T re e s

The section at b re a s t height is often u n re p re se n ta tiv e due to the

The total height of a tre e is usually defined as the d istan ce from

D irect M ethods of M easurem ent

In d irect M ethods of M easurem ent

In stru m ents Based on T rig o n o m etrical P rin c ip le s

Sighting from O, the o b s e rv e r re a d s the angle AOC to the top of the

The Leaning T ree

A ngle of lean A ngle of observation to tot3 of tree

Fig. 5:2 Leanirtg tree

O cular E stim ate

T h ere a re good re a so n s fo r w anting to know the shape (form) of a tre e

Growth P ro c e s s e s and T ree Shape

the constants C and c having, of c o u rs e , d ifferen t arith m e tic values

The Main P ro p o sitio n s of T re e Shape C om pared

d. b. h. o . b . (in) 2b. t. (in) d. b . h. o. b. (in) 2b. t. (in) d . b . h . o . b . (in) 2b. t. (in)

5-0 0-65 1 0- 0 1-58 15-0 2-20

The d ia m e ters of the crow ns of tr e e s in a stand a re frequently

d . b. h. o . b . Total height (ft)

6 0-52 0-63 0- 72 0-82

If a f o r e s t is m anaged fo r wood p ro d u ctio n , the volum e of wood in the

The D efinition of Stem V olum e

F a c to rs A ffecting M ethods of D irect E stim ation of Stem Volume

R equirem ents of the Ideal Method

Stem volume D erived by Volume of Sections

If a stem is considered to c o n sist of a num ber of sec tio n s, its volume

5 ft 10-6 •613 1-2 9-4 •482

15 ft. 9 -4 •482 0-6 8-8 •4 22

25 ft 9-0 •442 0-5 8 -5 • 394

35 ft 7-6 • 315 0-4 7- 2 •283

45 ft 6 -0 • 196 0- 3 5-7 • 177

D iam . O v e r-b a rk U n d e r-b a rk

ABOVE ODD LOG

This is also an old method which goes by se v e ra l nam es but is