https://ntrs.nasa.gov/search.jsp?

R=19780025341 2020-06-14T15:52:16+00:00Z

General Disclaimer

One or more of the Following Statements may affect this Document

This document has been reproduced from the best copy furnished by the
organizational source. It is being released in the interest of making available as
much information as possible.

This document may contain data, which exceeds the sheet parameters. It was
furnished in this condition by the organizational source and is the best copy
available.

This document may contain tone-on-tone or color graphs, charts and/or pictures,
which have been reproduced in black and white.

This document is paginated as submitted by the original source.

Portions of this document are not fully legible due to the historical nature of some
of the material. However, it is the best reproduction available from the original
submission.

Produced by the NASA Center for Aerospace Information (CASI)

 ^iI^

UASA
Technical Memorandum 79606
(NASr. -79r ) UiECT Of S I I> PA,:F 1,;1INES N78 -332Q4
0 10 TF- MTCHC I; ;VF EMISSION FNrM SOTLS (NASA)
30 p itC A03 /"F A01 CSCL 20N I

11nclas l R

c; 3/32 34113
f

Effect of Surface Roughness on the

Microwave Emission from Soils

B. J. Choudhury, T. J. Schmugge,
R. W. Newton and A. Chang

.AUGUST 1978

Nationai Aeronautics and

Space r7dministral,on
T^
i

Goddard Space Flight Center

L7
f
u1 ti I

Greenbelt, Maryland 20-771

-
' -mod
Ir
 f '
1 G T'
I^
rM Not)(, ^I,

' 11-1 : 1 ,1701 : S1 1 K1 ACC R0 11G11NUSS ON TI11- l

\11l'HOWAVI : I'MISSION FROM SOILS

I '

li 1.l'h,^u^lhtiry
hn^u^'^tr
I:. N'. Nr%%It'll
A. Ch.mg

r;

^^u„•.<<.t 1')'8

111111 ► :^t:11 til',^^'!^ I l ll^lll' ('I•:NII^K t

l^rrrnl^rll, ^t.nvlarnt

J ^

^c •
 I

CONlTNrS

Pag e

All l%l RA( r .......... .............................................. .

INTl:O ► PUCIION .................................... ........... .. . i

1 ,11 ORN . .........................................................

f

K.1JI.Itnc rramfor ................................................

Roil l -lulcss t l'Ic: t. ................................................. a

kl SUl FS AND OISCI I SSION ...... ... I()

I ^1^:luncnt.11 Pct.10% .......... ... . ............................ I()

U Iclkl Mcasu It' IIIk , IIts . . . . . . . . . . . . . . . .. ............................. Ill
A11:raft NIcas ► u. , nlcnts ........... ................................. II

CONCLUSIONS .. .................. I .... I .. I .............. 21

ACKN(^11'1. ................... ..... .....................

Itl l t?!tI Nt'! S .......................... ...........................

1':VI;I_t-S

Table

I l'1^1111^,11I^i^tl ^^: :^'^ 11.111 I ,Ilh^l.11i^11 ^^c.l^lllc'lllc'ill^ (►

' C.Ili ul.11i'd 1 1t 's 101 .I S011 I .'n1110.11111C 01'.100k K

l OV11`.111-.011 of 11 ,% IIo111 I lild ^I:.l^llll'llli lit IZ

111
 ^- 5 .. ► `_,Plpqmw
1 Ajq
I

ILLUSTRATIONS

F ►bure _aE
11

I laboratory measurements of dielectric constant for a soil as a Ilmction

of its moisture content. These measurements are for two clay loam
soils with similar textures . ...................................... 7

C;,,culated values o.* the emissivity for a soil usm ,., several values of
roughness parameter h. The calculations use (lie dielectric constants
presented in Figure I for the 21 c ► n wavelength . ..................... 9

3 Comparison of calculated and measured values of T B at X = 21 cm

I'or fields with 3 levels of surface roughness. Measurements made
from truck mounted at Texas A&M University ( Newton, 1978). In
3a the calculations were done for the profiles observekl in all three
fields....................................................... II

4 Aircraft observations ofTg at X = 21 cm during 1972 and 1973 flights

over Phoenix compared with soil moisture. in 3 Depths ol' the soil. The
two dashed curves in 4a indicate the difference produced by using the
72 and 73 temperature profiles ................................... 15

5 Aircraft observations of TB at X = 2I cm during 1975 dawn (6-7 AM) i

!li g hts over Phoenix compared with soil moisture in two depths of a
th%:tioil . .................................................... 18

6 Aircraft observations of "1'g at X = 21 c ►t: during 1975 mid-day, (

2 PM) flights over Phoenix compared with soil moisture in two depths
of' t lie soil . .................................................. 19

7 Aircraft observations ofTg at a = 1.55 cm during 1973 flight over

Phoenix compared with the soil moisture in the surface cm of the soil. .... 20

OKIGINAL PAGE I9
^)F I'^))RQUALI'N

t
I
' iv

l i^
1
 I C

I I'IT(7 01 : SARI CI Ia)11(;IINI : SS ON 1111'

MICROWAVII. .MISSh)N DRUM Soli S
+
H. J. Choudhury M
Colll{ i llte! ^i:lell:e. ('tii'{itilatltill
Stller Spi ni ;, Sit) 20 0 10 y

T. J. Schillugge alld A. Chang

Goddard S{ i .l:e Flight Cente r
Greenli:It.111) _'07"1

K W. Ncwtkin
"Texas ASAI University
l • tilloge Station. IA 77843,

1
ABS`TRAI . 1' ` f

hr eftr:t t,t'surLa: ► • Ikul ►thncss on Ihk • lin{;htnk • ss tens{irratl+;r kit' a

moist terrain has heen studied through the modification k i t Fresnel reflex - I '
r'
tit t tl :tiefflt'Ie111 .Intl using the latlt,ltlk: 11.111stel CkIllAtioll. Illo Illodih:.1-
t
thin im OINT ., ill Irodil, tit'll tit' a .in t 1e {iaranl:ter to ehara: terire Ow t

1,0111 1.1111:.5. It IS sho Il that this h;Il.11lli tcl' tlt'{`O IkIS till Iil +tll Ills• 5111 L C
i
(lell`.11t \ariant y .Intl II1t• horilolltal S:ah• tit tht • rtill1;1111Css.

Mtikfel :al: ul.11ltin. are in 1^oo l kluanllt.Inve 41:rnu • nt k1it11 Iht'

kit
obsel-veil dt• I`t'I1t1011:C ' lilt' H-i"..11111esS 1:111I`t'r;ltllrC t i ll 111t' Moisture

colitt• nt Ill (he mut"I e I:ner. Data from titiA nitnn111Cd Mid airhkirnc

17.41101110 C17% .Irk 1i1: x'111: tl tkil C01111`.Ir1St i 11. l he 11• \Illts Illkll.',Ilt' 111,11 the

row ,.hness effects are tWatcst f0t wet skills itlit• R• f is kllttert• n:e lieltiOCII I ^,
t
,
smooth and rt`u+,h .ulta,:..an he is ►treat as ^,11h.
1

t ^

^ I

.^" _
1. ' _ • ____ — _ ___^ . ^..`.d
 r

.d

I % 1.

EFFECT OF SURFACE ROUGHN! SS ON THE

MICROWAVE EMISSION FROM SOILS

INTRODUCTION

There have been s:-veral recent papers presenting theoretical models for tier micro-

wave emission from soils (e.g. Njoku and Kong, 1977; Wilhcit, 1978; Burke et al., 1978;

and England, 1977). These models considered the emission from the soil for a range of
moisture and temperature profiles and studied the effect of variations of these subsurface
1

properties on the emission from the surface. The effects of surface features, such as,

roueliness were not included. However, when the results of one such set of the calcula-

tions were co • npared with observations by airborne radiometers (Schmugge ct al., 1)76a)

there were rather large differences (^-30K) between the calculated and observed bright-

ness temperature (TB). These differences were .attributed to surt •acc ioUghness. ne

purpose of this paper is to show that surface rou g hness effects can account for these

differences.
j'
The scattering or electromagnetic waves from rough surfaces has been studied by

many investigators (sec Barrick, 1970; Wu and Fling. 1972; Sung and Eberhardt, !978).

These studies show that for a detailed quantitative calculation of the scattering by a rough

surface, the knowledge of the statistical surface parameters are important. The roughness

structure of an agricultural terrain depends upon the cultivation practice of that area. A

typical surface may consist of furrows, clods and irregular, small amplitude undulations

of different spatial dimensions. To study the effects of surface roughness on the observes;

dependence of the brightness temperature on the soil moisture a simplistic model has

been developed. The surface roughness effect has been incorporated into the calculation
 1 t f

by modifying the Fresnel rellectivity. This modification is based upon the theory devel-

oped by Ament (1053) for a conducting surface. The emphasis in this paper is to show
r,

qualitatively the effect of surface roughness on the microwave brightness temperature.

Pic present model is simplistic because it has only one parameter to characterize the

surface, namely, the standard deviation of surface height. We realize that this description

may not be an exact representation of actual soil surfaces. At present the knowledge

of the statistical characteristics of soil surfaces is not sufficient to formulate a model

1
which is not only general enough 'aut also numerically tractable to provide a quantitative

description. It is, therefore, not the intention of this paper to claim that this model will

provide a rigorous, quantitative description of the different aspects o1' the microwave

emission from natural terrains but it will provide a First step for including the effects of

roughness in thf! modeling of the emission from these surfaces.

In the calculation of brightness temperaturc, we hav,: used measured soil moisture

and soil temperature profiles. The calculated values of the brightness temperature are in

good quantitative agreement with the values observed by truck mounted and airborne

radiometers. Tltis agreement has been demonstrated for two different wavelengths. Dc-

tails for the theory and the results of the calculation are given in the following sections.

TIIEORY

Rad'ative Transfer

To describe the microwave emission from the soil, we will consider the radiative

transfer equation (Chandrasekhar. 1960):

dl
t
=—Ke(z)I+S (1)
dZ

ORIGINAL PAGE: k
OF POOR QUALITY

a
 !7
ORIGINAL PAGE, IS
OF I'tkil; ()I1ALITY

where I is the intensity propagating in the direction Z. K e is the total extinction het unit

length and S is the source terns describing the contribution to the intensity due to scatter-

ing and due to the contuluunl 111cimal emission of the soil. In principle one should study

this equation in conjunction with the equation describing the heating of the soil. It is this

latter equation which will provide int'mination alxout the Owrmal hart of the source term.

In this parer we will decouple these two equations in the sense that we will consider a

given temperature distribution.

To solve the radiative transfer equation we will consider a semi-infinite medium with

depth dependent temperature and moisture distributions. Since soil is a highly absotbing

material (i.e., large imaginary part of dielectric constant). to a good :approximation, the

brightness temperature TB or the temperature equivalent of the intensity emerging front

the 51111 will Ix' Determined bV Its illefnal teliiherattire dlstribiilion T(!). By integrating

Fq. (1) with the source term as the temperature distribution of the soil one can write

I It - ( 1 tto)) f
/ fo
0
T(z) h.i(z) ext► I

Ka(Z) dr' I d/ (2)

where r(o) is the soil surtace retlectivity at normal incidence and K a (Z) is the absorption

per unit length. These can be dete rill ned from the dielectric constants in the soil (Born

and Woh. 1075). Wilheit (1978) has develo l ► ed a model in which the integral isevaluatcd

by a sum over many homoeeneous layers:

T B = %I*iTi (3)

where t•i is the fraction of the rmh.ltlon incident on the air-soil intert'ace that would be

absorbed in the ith laver and T i is the temperature of t his layer. hhe values of fi are deter-
I
1
mined by applying the electromagnetic t-oundary conditions to deterinme the energy fluxes
I

I
i

1
o+
JL


1 I

entering and leaving each layer. The computations indicate that the radiation Irum the

soil is characterized by two sampling depths: reflective and thermal. The reflectivity is

characterized by changes in the real part of the index )f refraction over it sampling depth:

ar 1 0.1h, where X is the free space wavelength. The thermal Sampling depth is deter-

mined by the losses deeper in the inedium, and is given by

F xiti ti
S p = (4)
r fi

where x i is the depth of the ith l..yer. For a uniform dielectric this reduced to

b T = -- (S)
47t Im( n)

For a low-lo% dry soil, S T will be an order of magnitude larger than S r , while for a wet

soil, it will be only slightly larger. A similar theoretical treatment has been developed by

Tsang et al. ( 1 1) 7 5 ). I his formalism is simpler and has yielded brightness temperature values

which are within I or 2 K of their results for the same moisture and temperature profiles.

Roughness Effects

It has been shown in Tolstoy and Clay (1966) that if the scattering surface is a satis-

tically rough surface such that there is no correlation between the amplitudes of the waves

scattered by two points oil it

surface, then the scattered ► tensity call obtained by

the absclute square of the average scattered amplitude. It has further been shown that if

F,irc represents the scattered amplitude by a perfectly smooth and perfectly reflecting

surface, then the average amplitude that will be specularly scattered at an angle U by a

rough surface is given by

.k
, r

(4c) = ROO inc f a W(Z) c2ik,r di (6) 11

whvic %VW is the height distribution of the face and R O is the reflection coefficient of

a smooth surf,ice. A typical rough %u ► f.ice corresponds to identity mg the slic: ttum with
Gaussian distribution of mo mean and variance o-:

1
WW _: —_._ exl, 1-d/ 20 2 I (7)
o %/ 2 it

For this spectrum, thr average amplitude is given by

(I.- ) Kp(0) h int ex1'I-2o2kij. lR)

Since

k' 1 = ( )-' cos' 0 (9)

Hic scattered intensity obtained by squaring hq. (8) is:

1,(0) = W I R,,(0)1' exp ( - h cos t 0) (10)

w11e1e the roughness paranu • h• r h is given by

I-rom I`y. (10), one can stihuLaic that tlic gross effect of the surface rouglutess on the

sc.iltcred iutcnsity can he incorporated by nu+dtfying the snwoth surface rellectivity r0l,(0)=
1
I
i lK„(t)ll' as

^
rl,(t)) = r,,,,(0) cxp(-It CoS 2 0) (1')
I
:I
Mierc the subscript l, designates the polarization. The surface c missivity is obtained from

(12) by

To verify this result measurcnicnts were made b y Waite et A. ( 1 0 73) of the rellcctivity

I^ for soils with different surface rou g hness conditions. hhcy found that v is not a sufficient

5
1i
i^
Y

him.- \
 1 • .1

ry

? indicator of' the roughness for tlri. model (Hancock, 19761. fable I is a .ummiry of their
.v

results at a book angle of 300 . The ctlective u was Octermined by Irtling the oh-wrved f're-
i
fluency depentleitce of the rctlectivity in each band to that expected from I .q. 12 for a given

o fhc cffectt%c a was always greater than the measured u. They also found that the aulo-

M correlation length of the roughness was also Important. This latter tluantity is essential1% an
I
indicator of' the horizontal scale of the roughness.

t Table I
Comparison of o's Drum Laboratory Waxurcrn.•nt,
l
%
Fffective o Imtn Oas. a I Auto-Corr lation
Surface I- 2 (;H,. 4.^, 7.4 GIV tnurt) I rnglh 11111111

1 5.0 5.0 2.7 $^1

o.0 ^.5 2.2 1'

3 1 6.0 8.0 I 3.6 ^_ I S

The ^aluc of r, , t ,(t► ) can be determined from the Fresnel eiluatiOn for the :ase of a uni- r

firm dielectric, or from the layered modek mentioned earlier fur situation with nun-untt'onn

I dicleclncs. In either case it is necessary to kno%% the striation of dielectric constant for soil t

With its moisture content. This is presented in Figure I li,r a clay loans soil at the wave-

lengths to be considered in this palter. 21 ( Lundien. 1971) :1-u1 1.55 (Wang et al.. 11078) cm.

It can be scan in Figure I that the addition of water has very little effect on the dielectric

properties of the soil a' low-moisture contents K 10 perct —t ,. Presumably this is duc to the
i
1 t
strune interaction of the water molecules with the soil particle. Nhich reduces the polariz-

ability of the ^Natcr m a thin layer .round each particle. A. the %%ater t:mucnt increases the 4

1 water is less ti g ht1v bound and rauscs a greater increase in the dielectric constant for soils.

0
 r
TO
, I

OPIGINAL PACk, l:
(IF IN X )1? ( )F'A;.rrN,

hi
24
-o— k - 21 CM (Lundeen, 1971)
O
--0---,k -1.55 CM (Wanget.ai, 1.978)
22

0
20

18
0 F REAL PART

16
z
a
0 0
Z14
O
U
U
Q: 12
w

LL' 10
0
8 O
i
6
IMAGINARY PART
ii
I '
4 i
-t7 o
O
2
^B o
t
;i
^^ o °D
0 `.old - ff Sd 1

0 10 20 30
SOIL MOISTURE, WEIGHT PERCENT

Figure 1. Laboratory measurements of dielectric constant for a sail as a fullitiOn of its

moisture content. These measurements are for two clay loam sails with similar textures.

F.


1.
r
}

The dielectric constant variation can hr re presented by the straight lines which are linear l

regression fits in these regions for the two wavelengths.

The effect o f surface roughness on the emissivity of the uiil as a function of soil rnois-
I tire at the ' 1 cm wa%clength is presented in Figure 2 for several v,► lues of h. These values
were calculated assuming uniform moisture and temperature profiles using the Fresnel coef-
ficient for the sutooth surface cast. The curves show a Behavior similar to that of the dielec-
trio constant presented in Figure I, i.e. slow decrease of emissivity for soil moistures below
about Ill ind a much sharper decrease aho y this value. In general the effect of surface

roughness is to increase the emissivity with in •.rease being larger fur the wet soil eau.

The contrast in T f ; between wet and dr% soil is presented in Table ` ,issuming ;I
temperature of 300K. At 0 = 0° the increase in emissivity due to roughness is gi%en by
Ae = r„ p (l erpIAM (14)
When r„ t, is for dry:.soils. Ae will he small, e.g. the range is 0.04 for drN soils which

from Tanie 2 corresponds to 12K range for T,; due to surface roughness. For wet soil r,,p

is larger and the surface roughness effect will he much larger. At the 25' 7 moisture level. the

increase in emissivity .fie = 0.28 corresponding to a 81 K 'I' ll incr ease.

F.rble
Calcul.► ted T B 's for a Soil Temperat tire of 3001:

r„ p I1=0 h-0.3 h
tl
i = 0.c) h=1.0
Dry
SM = 010 0.06 282 288 244 204

^t Wet
SM = 25 0.34 108 201 228 249

-^ hll = 1
11' 11 (dry) T f3 (wet) ► 114 87 ^_ 03 45
I I^

.. I

,, 1 ar .4
 01661NAI, PAGE 1Z
O1'' YtK)R 011ALITY

1.0

^I
N

0.9

i h - 1.0

h - 0.6

0.8

h-0.3

f
w

0.7 h=0

0.6

0.5 ` ` , I i l
0 5 10 15 20 25
SOIL MOISTURE. WEIGHT PER CENT

i 1

Figure 2. Calculated values of the emissivity for a soil using several values of
roughness parameter li. The calculations use the die!ectric constants
presented in Figure 1 for the 21 cm wavelength.

9 i

Yom'"""
j. ' I: _^ " •,4 ^. j-- --^-
 r I

I
f

The n.t effect then is to decrease to the dynamic range of the Tg change with soil moil-
.

lure changes, e.g. a decrease of A"I C h from 1 1 .3 K for smooth surfaces to 63K for a rough sur-

face with h = 0.6. From this discussion it is seen that it is necessary to have some

knowledge of the surface rougimcss to make an unambiguous soil moisture estimate from a

Tg observatit n. By comparing these calculations with radiometric observations for realistic

situations should yield a range of values for file parameter h.

1111

' RFSULfS AND DISCUSSION !

I I ^
Fxperimental Details
+ 1

The experimental results to be discussed in this paper were obtained front a portable

! tower (cherry picker) platform and from an aircraft platform. The tower measurcmcnts

were done at Texas A&M University from a 2 , m height using 21 cnl and 2.8 rill wavelength

radiometers (Newton, 1977). The radiometer measu.ements were supported by observation

of the soil moisture and temperature at several depths down to 15 cm. file surface roughness

profiles were also observed so that values of a'- can be estimated.

r

Field Measurements i

The results from the field measwemems are presented in Figure 3 foi fields with sur-
I
faces having 3 different levels of roughness. A rough plowed field, it uleditim rough field that i

had heen disced-.and a fiel,l that had been dragged smooth. The calculated values were ob-

Pained using the moisture and temperature profiles that were observed at the time of the

measurements. The values of h were selected to yield good agreement with the observed

points for each roughness level. The observed "I *g's for the smooth field are in good agme-

' mem with those listed in Table 2 for It = U. In all three cases the agreement between the
1

10

!
1
 r.
T-1
* 4 ►
J
1 1 r I ^ ^ , r^ . - r ^ ^
Ao
ORIGINAL PAGE IS
OF p; x )R OUALITY
V
M
C v
s
W_

N
0
N

94 Or
v c
•^ O

3
W
^+ u
O O rJ
O an v
v .^
c
an
M M
p O
N
000 O^
S
N N N N N N N N N
S s d O O O 8 O O
00 O O
s00
W .o r
^C
(NIA13)1) 3Hn.LVH3dVY31 SS3NIH9IH9 °C
V L" J
JO
f^
M
Ln

^
_
O
0
E asi

O ^.
v
2
N
J
O O
0 N
1-•
O u
j
m^
C
cc
a Q ad
`
O In
u y
W > ,c O
O "`J a+ ^J
8 N N
N N N N ^y M
N N
^°c
N N N ^ .-
0 g o 0
ro g$ o °rw
.- E
V
SNIAl3I ' 3HnlVW3dW31 SS3NIH9IH9 `~ ^ = v
0
0
M
c OO
s
W
N
O

C/ 'J
c*7 ^,
u •:
N
v O
'n O ,^
C .^
S 0 LA
J
N Ln
v
r
8° N 8 8 0 0 0 0 rv .J.
N N N N N N N N N N .-N r
SNIA13)1 '3univu3dVY31 SS3NIHJIHB
:z. 1
cl
i
1
r
1 I
1 ,..-.^a..-.^,,,.,.^.^y^,.a.^r+^ ►^- ! J ^ __ ^ J • i t ^ .-^. ;^^ ^ - 1^ . . ^.-^1 ^---►— ==--
 I

observed and :alculated values is good at the wet and dry ends but there are differ iccs of f

10 :o I ik at file ntididile moisture levels i.e. about 1 S Recalling Figure I tite muddle

moisture level is the region m Isere "I ll :hangs most rapidly with soil moisture. this also tends

to fit- the region dd here there is the gr: •atest tincertainty ill moislurc deternim alum.

An additional factor contributing to the scatter at these middl: moisture levels is the

tact that tile:alculations were pfff'Ormed with a coherent mnodcl. Therefore. there is the

possibility of resonances occurring when a sharp dielectric gradient is present. We heiedr

1
this caused the higher value, of T p (-_'K5K) for tiv three points in Figure 3c at about 10

nntiaure. In the extreme case the dielectric constant char ed from ;; value of o at a depth

of' 1 .0 : m to 1 9.5 at 3.5 :m. A quarter wavelength in the soil for I his Case is given hN

^ rn o
_ - I _
4
_ _', l aft (15)

where X t is the wavelength ill mcdtntm. As ;t result :on"tructive inicit'cicticc occurs

causing an increase ill transmission i l ,rough tic surtace (i.e. emissivity) liar ilu. case. To
t
verify this hypothesis the cal:ulations were perforntedl as a function of wavelength I'rmm 30

to 15 cm, The mli\iimm,. 2 111 5K, occurred at 21 cm wavelen g th and Tit hell off to 851: at

27 ands 15,5 :ill wavelengths. The sharp thcl::tri: gradient ill case was caused by the

transition from values on the low moislure portion ( 10': ) of the dli:lectric curve in Figure I
i
to a point (2017r) on the steep porfidtn of tit: curve in a relatively short distan:c. Therefotr
1
1I' I in apply ing any coherent model one has to he aware of the possibility of these resonances
dot

occurring so that false ntterl rotations from the modlrl can he avoided.
^I t
i In Table 1 the values of a l tidr the fields are listed) along d^tlh those:al:ulIMI

d I from kpiation I I using the observed) dalues of h

!i

i
II I
I
 7 7" 7
^^ t `y `
1

Ito r

Ii
Table 3
Comparison of Ws from Field Measurements

Field h Uffective Measure

Smooth 0 0 0.9 cnl

Med. Rough 0.2 0.1) rill 2.6 rill

Rough 0.5 1.2 rill 4.3 cm

I
1
The paranmter h increases with increasing roughness but does not do so as rapidly as expec-

led front the nlca.uled u's. For exampic. the ratio of the u = fol the rough and rlledmill

rough :axes is 2.7 N%hllc tIlr ratio of the It's is 1.7.

File effeclive a's fill these cases are less Man the liv.Lured values. p his is opposite

from the sitll:ltions { p resented In I .Iblc I for thr Lihormol'y I11r.ISllrel111 • llti. I liS tilllerell:e

probably result. from the different huriruntal s.'ales of the roughness in the two Situations,

:aircraft Measurements

The air:rait results acre obtained during flights with NASA aircraft over the Phoenix, r
1
;Armin, mca and the Imperial Valle\ ofCalifurnia during March 1 0 72 and Fchruary 1173

(Schmugge ct A.. 197(m) and tll t ,hts over onl\ the Phoenix area during 'larch 1075

(Schnrtlggc. 197hb). The aircraft altitude for these tligllls wen 000 ill in 1472 and 1473

and 00 ill in 1075. On board the aircraft were microwave radiometers covering the wave-

lengths range of 0.8 to 21 cm. In this pares only the result, at the ' 1 :ill :slid 1.55 :ill wave-

lel^gtll: will ;.c The _' I ;ill radiometer was nadir dewing wit,, :I 15" 1 - I'' .-adian)

beanlwitilh, therefore its spatial resolution was approxilrlat.U. 1/4 the aircraft altitude. I he

1.55 .,, ill radlomcter is a scanning radumu • tcr whi:h has .In -moular beam %\idtll of 2.80

• 1/20 radial). This sensor wasonl,, used on the 1 0 72 and I'11 '3 missions. 41 1

.._-.
4,,.,, -.1.,,4,.x,.-.../^^ J ,, { _ i . — .^, ,,
•,! .#


i71-11
.I

'r
I'he airCl alt I lcw along flight lutes Centered on t he agricult ur al I adds which were at
I
^ least Icy hectares (40 acres) in area. These fields heneralh had uniform surface and moisture
I
conditions over their total area. All the radiometer data obtained over each ficlt. \\ere used

to obtain the average brightness temperature ( r It ) lot flit• tiCld. I hr Soil moistulc measure-

menu wert made at 4 locations and for several depilis in each field. Hit- \,flues prescntcd

here are the averages for ca_ i fidd. For the 1 4 75 Ilights soil temperature I l rofiles were also

p leasured. Soil te ,,tures tICIC imi.ition were also made tot the sampled fields.

Because of the rattle of soil femme. from ,aridy loam y to Clays, that .Ire present at both

. 1 I these sites it is necessary to account fur the different \rater bolding capacities of these

.'( 1 115. I hi, \\'asdolle by lim111,1111ing the measured sell moistures to (lit' Ifeld r.11,lilt\ Icvds

(FC) for each soil. the amount of water in a soil at FC is that which remains two of three

,lays after (laving been saturated and after free drainage has practically ceased. I his Ic\el is

deter lined to it large eXtrnt b y the soil Ic\turc i.e. particle sire Composition. File value oI

IV for each field was estimated on the hasis o1* the soil textures that were nleasurctl for that

field (Schmuggc. 1978).

TI I e St 111:ICC I-OLII_Iiness rhararterlstlt> \`t, Cl IIlose rest IIIiilg from t I w agn it, illttit II prac-
1i
tices of the two area.. The donunant method of irritation is flit• flooded furrow I he furrow

separation waS about one nt etc r :IntI tIic furro\\ height \\ a. about 20 Cult. Stipctnnposed on

these Corrugations \\ cre Clods, \\ 111L. 11 \\ ere gcnC rally less than 5 Cll.

Plots of IF \cntis the soil moisture in various layers for the 1 0, : ,,nil I 07 ? flights .Ire

presentcd in hi!tirc 4. We note Out the range of Fit is not as great as that e\peC fed for a

"Illoolh surface.
1

i# 1.1
i
\

1
1 4
 _3
• 44
Ep
P4
E
C
0 0
O
0 so V
L- i L I I I
•41
0
ir
71
0
If
0 4u
m
0
rL ^, ./ 8
rn
ij
I -1 --t
Ni
SNIAII)i 3uniV'J3dV431 SS3NlH9IW9
11,, . 15
v dl -
t)j,j6jN,%l,
QUAJ.SVY5

OIL VOOK
I ..;

iI '

I ^ •• Ir

% 1

o^

t: Calculations using the layered model (Wilhcit. 1978) were performed using moisture

and temperature profiles measured by the personnel at the U.S. Water Conservation Labor- i

tory at Phoenix (Jackson, 1973). The soil moisture and temperature profiles were observed

at frequent intervals after it irrigation on March2. 1971. These data from the same

area at flu sa •ie time of year were assumed to be reasonable estimates of the situations oc-

curing during the 1972 aircraft overflights. It should be noted that the moisture and tem-

perature profiles had been obtained from a smooth field, while the microwave radiometer

results were obtained from rough-surfaced fields. The moisture profiles were rather uniform

when wet, but dried rapidly at the surface after 5 or 6 clays producing sharp moisture and i y

dielectric constant gradients just helow Vie surface. The calculations were performed for
I
the early afternoon (1 :30 p.m.) profiles of each day. The corresponding temperature profiles ,I

are probably quite representative of the actual situation for the 1971 flights. For the 1973

flights, occurring early, in February, the teniIlerattires were somewhat cooler. The surface
;a
temperatures as observed by the aircraft IR S01150t were found to be about 15K lower than
r
that ohscrved during lurch. The February temperature profiles were then obtained front

the observed March temperature profiles by adjusting the gradient to fit the observed surface i

temperature for February data and assuming two profiles to be equal at about 50 ent.

The dielectric constants used in the calculation were those presented in I igt -c 1. These

values are for soils having texture similar to the Avondale clay loam soil at t h e WaterC'onser-

vation Laboratory.
i
'File solid curves in Fi g ure 4 are the calculated values assuming a sntm,th utrface

(it 0). It is clear that the aircraft F B 's do not get as low as those calculated for the smooth

surface. The form of the calculated curve however does agree with the observations i.e. little
I ,

variation out to about 50% of Ft' and then the ra pid decrease in T,,.
I ^

lo
i^

^^ ^----••sr—^=. ^-.^.^La..^, _ -^;.-.,rte
C
.l E

Pic dashed can-e-,

ill 4a. 1, and c are for h = 0.45 mill it is seen that the range of ! j

calculated I' ll 's is is good agrecn ► ent th0 observed rang .Ind that the roughncss factor has its 1

greAL'%I .'Ilea at the higher levels of mot%turc as predicted h. III. ( la).

In observing the %ariation of T it %%ith soil moisture

ill 3 layers. we not. • the Imear ! ,

decrease 1' H with the `oil moisiur: in the 0-1 cnl laver. hilt for the 0-2.5 and 0-5 cnl layers

there is a region al Io\% soil moisture. for Miich Ihcrc is little variation of T it . Alime this

lc%cl there is a sharp dc.•rca>,c ill I tits hel ► al ior is similar to I l i-it 1)r0s0ntcd tit

and that the location of the hrcak I l k l int tile the :.5 .• m . • one, at approximately 50';• of FC.

is good ar..rccmcnt with I11c location prom-mcd ill igurc `. assuming a FC of 20- 25`^ I'm a
1
day l o ' lnl soil. liccause of this agreement, we ►VIII assunl0 that the radion ► 0tcr is responding

to the n ► oistul'r %anatlons ill top _' or 2.5 cm.

Ili hil tlres J and h the restilts Irom the 1075 flights are pi cst, nted. figure 5 presents tllc

results from the pre-.I.mn Ilights and Figtlre h tronl ill y mid-da% Illhhls. hsentl.11k . Ilic

S.Illll• depelldel ► cl• of 1 11 oil SMI moisture Is ollsmed IUr these Illl'hts .1% for the 1 11 . t Illghts.

Hit . Dallies ► n I1 1 1.:au us0d the moisture mid temperature profiles that were mca-

sured
ill field at the tin g e of the Ilights. Th0 rough surface curves area visual hest fit to

the calculated points. Note that the I' ll dil ' I 'erCllce 11OWTc11 Itit' ANI :Ind 1'M Ilights due to

Soli tl• nlp0rit Ill k • changes is ex1+1ain0d h\ the calculations.

Tic hest agrccinctlt is obtained ill else for It 0.0 l ► hich is slighth larger than the

result fo g the 72 73 da1.1. I he reason for this dllfcrencr is unknown. and because of , the

scaticr ill dat'i ma\ . indicale 1110 un6.0rtainl\ of our esomatcs for the \clue of Ii. Th0 scat-

ter in 1110 Lalcu1,11ed points ahout 1110 curves is due 10 two causes: first. the variations in the
1
tillll Ilrioislllre profiles having the 5.1111: a%efagC U _' :I11 or t) 5 illl aVel':Igl• lll0iStIIrC Icvels:

.11ld secondly . the Variations in soil temperature.

17

l _ +l - ^ I I .
1 1
I
 i
L
{A W O I
a W / O
W 7
V D C.
G
N Q n E
^¢ Ip
Q y M O U Q
^ m N
W to ^ d`
/'
•WPM _Y
o
¢
> Qppo
~~_^
gyp]
C { {}
^
k'
C)
^^• r
=W O
^m u^ix oLn a.
LA. 0 do
V^^i^ O
i
p
} C^
C
a a ^ ^
^
10
I
U
^ ^ a
N LL ^`
c) n
° w :3 .^
r -r c w
i
O C11
rP
O II
rc
E
II
n 0 1, o oU
J
o N J^
O cr V O D 0
p _.
4
i
n JU
t
OO O O O w O O O O O A 4 O C1N
Q1 ap 1, tG N ♦ N N r O O, N n t0 ll'1 ^;
N N N N N N N N N N r r r r r
SNIAIIN '3!lniv83dMi SS3N1N',)11f9 -"
f
rteGS l5
Itl
of
'-^-.¢.;.-•rte_ ' ,' • ,

r
l
I
ki _ .., l f
CWGINAI• p A^ ^i. I
0
— T - - T I I I _ 1---T— t N
N ur
WVW c
G.
w CQU
U J^Q O
Q
N
r
a n n^p >7Q
^ C
W fil M H F- _ ^D
t ^ // I
_^ M
^ > O v ^ O c7 o O ^ 1 I
CL.
oz
'O cTM^ O m
v^^ O
Q q OiO O -^
o b c 0 .0 1
o d ^L'4 Y
v 0 0 S H
©0 U
^ ^ Q
^ U "' y
1
V ^
g v
J
_W C
ry
IO 4 0 a a
^_1 ____ .^ 1 1 1 l l L l L . 1 1 - 1 -- N
O
^ O
^, 3
aY S
i N r ^
n
'm O u ,^
i 0 j
^i E
i
orb 1R n r'
O
^ o N v ^,
^o
000
o H00 3
i a ^° a ^ o ^ O
C
f
p,r • ^ • U
a
Or r q 0 N fr
C o
M I
r
C O
iQ
i `I rc)
V
.J
0 V
r 0 ,a
0 0 0 0 0 0 0 0 0 0 0 0 o ^ o
O^ m n ,D eN ♦ rq N C4 r! A m
N N N N N N N N N N
SNIA13N '3wnlVU3dW31 SS3N1H91HO LZ
f
I `)
' f•
 1

300

290

2130
R

I
270

21

N
2

J 2!
W
Y
W
x
D 2e

a
W

2;
W
H

N
W
2.

x
cc

21

!S

11

0
C I u
SOIL MOISTURE, OF FIELD CAPACITY

Figuto Aircraft observation of F I; at = I.5; cm Turin!, fli g ht o%cr I'llocilix

comp,ired with 111: soil I oisluic In the surlac :m of the sell

1(^1 '10

f^
 ,
i

i
i

The values of h obtained for the aircrali data generally fall in between those obtained

for the medium rough and rough cases for flit- field measurrni.-nt results (Figure i). This in-
it

dicates that for the Phoeni y region none of the observed fields approached the smooth cute-

gory and an assumption that agricult ►lral fields are in the medium rough to rough (i.e. h

0.5) may be reasonable fur future calculations at this wavelength,

I he data at 1.55 rill are presented in Figure 7. They are also described by an h of 0.6,

the fact that h does not wale with wavelength is indicative of the shortcomings of the model

In this cast-, it appear% that different portions of the roughness spectrum will contribute at

the different 'Aavelengths. II is also clear that ,erlam of our assumption concerning
f
the roughness are violated.

C'UNC'LUSONS

A one parameter model for estimatin-, the effect of surface roughness on the microWave

emission from soils his been developed and compared H ith radiometer measurements from

both tower and aircraft plall'Orms. B\ a suitable choice of the parameter. h, the model. %% hen

conihincd %k,th a ra.ti.mve transfer model for the %oil. yields Rood agreenient with

the observed brightness temperatures. An et •fective range ti► r the paranieter was found to be

from 0 for a smooth surface to O.h for a rough glowed surfs... From the derivation of the

model the pmametcr It is expected to be proportional to the variance of the surface height.
I
l however when compared with the measured variance for the tower measurements this de-
i!
pendence was not verified. A similar result was found in laboratory measurements of surface

retlectivity (Hancock. 1 0 70. These latter measurennt-nts indicate that the horizontal scale

Of surface roughness, i.e. surface slopes, is also important in determining, flit- magnitude of
1
the par,unrter h. Because of tlu, factor it does not appear possible to extrapolate the ralue

lie."_.


r +^ t t I

of It Itom the nkcjst;tc cnt% 1`li%ellle.t here at the 21 cm %%i%clolglll lt ► othel N.1\i1e11^11N '
I'
Jn.l iddiliolul meJ9urc • n110111% JI other \\ J\elenlah• % 111 he tc.11111C I to Jetellllm.` the .ICIVII

.1en ► e of It on -A j%dength,

AI the 21 on \\ .I%dotoIt ill. \.flue of 0 .5 for It .Il , he.lrs, t. ► t i c rciiicwIlt,lll\e r, ► f the con-

.11tloll ol-w1\.• .I Ill 111i .Itlir.11t J.It.I owl the .11'll^lllluti l .Ili.l .110i111.1 I'lloelll\.

I hereloti \\ IIIt this Illodd lot rotll;hiw%% I.IJ1.111\i tian-iler Illodel ". ak- llJtiol1% --hould \ Idd

liillrJte e^11 ►il.lte^ .`t 111 ` \Jlties .it i It Ior .1 \\ I.le IJlhe . ► f Il omure Jnd (.`nlh iatuli

"on.1 ► t lolls.

l urtha Noll. Atli hr re.lulrm to .le lrrnrine the \ ► .1\.•L•n^th .IcpcnJ.'nie 01 , 11 .InJ to J: -

telmille 11 till• illodel i.11l .liiu1.11e1) 1 1 hAllil 111e 1 1 ol.111IJtloll Jillerellii% i \1 l iiled h► 1 ott-

11 i.\UNi.i Ik^^tt ^I

I lie 1 0-5 data Hire .ioluired h\ the Illi 1111`os. ► I Joint ^% oll M.`mm' I \pC11111i111 \\ 1101

111w1wd 111\ CSItg.Itols tion) A &M ( mwlstl) .Ill.l .`1 \Ikon--as III ,IJJ1lloll

It , \ASA. 111e1r i.`. I ` i i,ltl\ 1 11 Ill Ii A1111; 111CM' MS 1% .IhhliilJli ►I.

1Z I I l I: I V('

Anion t. fit! \ I` l ` : I . o \\ ,Iid% A I hio1\ of I2ellcoton B\ A Rough Silll.lie, o

IRI-, \.+1. 41 , v. I4_'-14c•.

lt.11rick. 11 1 . 1 .1- 0. ul -R.1J.n Ccos: Section H.111.ihook - . \t ► I. _, I'ditor G I Wt1A.

11 '. 11L.111 Pic%%, N .Y.

It"111, N. and I'. \ ,At. 1 0 '^ "Pti i11► lea Chapt:l \III V lg.lmon. \. ) .
 pprr _ 7 ^ ^ r - - --
L

ML L % ►I

Burke, W. 1., T. J. Schinugge and J. F. Paris, 1978, Comparison of 2.8 and 21 cm Microwave

Radiometer Observations ever Soils with Emission Model Calculation, Journal of Geo-

physics: Research, (to be published).

Chandrasekhar, S., 1960. "Radiative Transfer", Dover, N.Y.

England, A. W., 1976, Relative Influence upon Microwave Emissivity of Fine-Scale Strati-

graphy, Internal Scattering and Dielectric Properties, Pure and Applied Geophysics,

vol. 114, p. 287-289.

Ilancock,G. D., 1976, Broad Spectrum Microwave Measurements of Vugh Discontinuous

Surfaces for the Determination of Moisture Content, Department of Electrical Engi-

neering, University of Arkansas, Arkansas 72701.

Jackson, R. D.. 1973. Diurnal soil-water content charges during drying, Field Solar Water

Regime, SSSA Special Publication 5, pp. 37-56, Soil Sciences Society of America,

Madison, Wisconsin.

Lundien, J. R., 1971, Terrain Analysis by Electromagnetic Means, U.S. Army Waterways

Experiment Station, Technical Report No. 3-693, Report 5.

Newton, R. W., 1977, Microwave Remote Sensing and its Application to Soil Moisture De-

tection, Technical R!port RCS-8l of the Renu,te Sensing Center of Teas A&M

University, College Station, Texas, Available from University Microfilms order No. 77-

20-398.

Njoku, E. G. and I. A. Kong, 1977, Theory for Passive Microwave kemote Sensing of Near-

Surface Soil M oiStUre, Journal of Geophysical Research, vol. 82, p. 3108-31 17.
t
Schmugge, T. J., T. T. Wilheit, W. Webster, and P. Glorsen, 1976a, Remote Sensing of Soil

Moisture with Microwave Radiometers 11, NASA Technical Note, TN D-8321.

i
;f

23 i

t ^

5
f
A"
7 ( T •'.^^ .:s;._-
t ^-^
 I
SchniugKe,T. 1., 19761)• Preliminary Results from the March 1975 Soil Moisture Flight,

Goddard Space Flight Center, NASA X-913-76-216.

Schntugge, T., 1978, Effect of Soil Texture on the Microwave Emission of Soils. To be

published.

Sung, C. C and W. D. Eberhardt, 1978, Explanation of the Experimental Results of Light

Backscat(cred from a Very Rough Surface, Journal of Optical Society of America

vol. 68, p. 323 -328.

Tolstoy, 1. and C. S. Clay, 1966, "Ocean Acoustics McGraw-lull, N.Y.

Tsang, L., E. Njoku and J. A. Kong.. 1975, Microwa%c Thermal Emission from a Stratified

Medium with Non-uniform Tempera ► .tre Distribution, Journal of Applied Physics,

Vol. 46, pp. 5127-5133.

Waite, W. P., K. R. Cook and 13. B. Bryan, 1973, Broad Spectrum Microwave System for

Remotely Measuring Soil Moisture Content, Publication No. 18, Water R,sources Re-

search Center, University of Arkansas, Arkansas 72701.

Wang, J., T. Schniugge. and 1). Williams. 1978, Dielectric Constants of Soil Microwave Fre-

quencies - 11, NASA Technical Paper TP-1 238.

Wilheit, T., 1978, Radiative Transfer in a Plane Stratified Dielectric. IEEE Transactions of

Geoscience Electronics, vol, GE:-16, p. 138-143. i

Wu, S. T. and A. K. Fung, 1972, A Non-Coherent Model for Microwave Emissions and

Backscattering from Sea Surface, JoUr:al of Geophysical Research, vol. 77, p. 5917-

5929.

i
i

, i

rk ^