Growth-Ring Boundary Anatomy and Dendrochronological

Potential in a Moist Tropical Forest in Northeastern

Bangladesh
Author(s): Mahmuda Islam, Mizanur Rahman, and Achim Bräuning
Source: Tree-Ring Research, 74(1):76-93.
Published By: Tree-Ring Society
https://doi.org/10.3959/1536-1098-74.1.76
URL: http://www.bioone.org/doi/full/10.3959/1536-1098-74.1.76

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the

biological, ecological, and environmental sciences. BioOne provides a sustainable online
platform for over 170 journals and books published by nonprofit societies, associations,
museums, institutions, and presses.
Your use of this PDF, the BioOne Web site, and all posted and associated content
indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/
terms_of_use.
Usage of BioOne content is strictly limited to personal, educational, and non-commercial
use. Commercial inquiries or rights and permissions requests should be directed to the
individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit
publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to
critical research.
TREE-RING RESEARCH, Vol. 74(1), 2018, pp. 76–93
DOI: http://dx.doi.org/10.3959/1536-1098-74.1.76

GROWTH-RING BOUNDARY ANATOMY AND

DENDROCHRONOLOGICAL POTENTIAL IN A MOIST TROPICAL
FOREST IN NORTHEASTERN BANGLADESH

1,2†
MAHMUDA ISLAM , MIZANUR RAHMAN1,2† *, and ACHIM BRÄUNING1
1 Institute of Geography, University of Erlangen-Nuremberg, Wetterkreuz 15, 91058 Erlangen, Germany
2 Department of Forestry and Environmental Science, Shahjalal University of Science and Technology, Sylhet 3114,
Bangladesh

ABSTRACT
We present the microscopic analysis of growth-ring boundary anatomy of 27 tree species from
a moist tropical forest in Bangladesh and evaluate their dendrochronological potential. We observed
high inter-species variability in the anatomical features that define growth-ring boundaries. Marginal
parenchyma, fibre zones, and thick-walled latewood fibres were identified as the dominant anatomical
features delineating growth-ring boundaries. The evaluation of growth-ring boundary distinctness in
thin-sections and scanned images revealed that 25 out of 27 studied species (93%) showed distinct to
fairly distinct growth-ring boundaries. Cluster analysis of wood anatomical features was used to select
an additional 5 species for further investigation. Ring-width series of these species crossdated well within
the same tree. Between trees crossdating was also successful in all 5 species. The Gleichläufigkeit (GLK;
i.e. the proportion of agreement/disagreement of inter-annual growth tendencies among the trees) var-
ied among species between 0.55 and 0.71. A strong synchronization of tree-ring series between trees
suggests that growth rings are annual and influenced by common environmental factors. The derived
tree-ring series, estimated tree ages, and growth trajectories underline the high potential of our study
area for answering a variety of climatological, ecological and archaeological questions by applying den-
drochronology. This study will therefore provide a new endeavor in tropical dendrochronology of South
Asian moist tropical forests.
Keywords: moist tropical forest, tree-ring analysis, Bangladesh, growth-ring boundary, wood
anatomy, boundary distinctness, tree ages, growth trajectory, dendrochronological potential.

INTRODUCTION growth rates, tree ages, regeneration and survival,

which are required for planning sustainable forest
Tree-ring analysis is a powerful tool to recon-
management, can be obtained from tree rings rela-
struct historical growth and physiology of trees,
tively faster and with longer time-scale than alter-
which can be used as key ecological indicators
native sources of information like Permanent Sam-
of environmental and climatic changes (e.g. Fritts
ple Plots (PSPs). Meanwhile, around 230 tropical
1976; McCarroll and Loader 2004; Zhang 2015;
tree species forming annual tree rings have been
Fan and Bräuning 2017). Such information is of
identified (Brienen et al. 2016). Hence, tropical den-
great importance particularly in the tropics, not
drochronology is widely used to reconstruct past
only for understanding past climate variability, but
climate variability, to analyze life-time growth pat-
also for modeling tropical forest response to fu-
terns, to estimate tropical tree age, to predict tropi-
ture climate changes (Hiltner et al. 2016) as well as
cal forest response to future climate changes, and to
for taking management interventions. For example,
answer other ecological questions and their applica-
the basic ecological information such as diameter
tion in forest management (Rozendal and Zuidema
† Authors affirm equal contribution and shared first 2011). Nevertheless, the tropics are still strongly
authorship. understudied considering their high tree biodiver-
*Corresponding author: mizanur.rahman@fau.de sity and complexity of regional climatic regimes,

Copyright 
C 2018 by The Tree-Ring Society 76
 Dendrochronological Potential in a Bangladeshi Moist Tropical Forest
as compared to the temperate and boreal forest
ecosystem, based on the early prevailing, but sub-
sequently disproven assumption that tropical trees
do not form annual growth rings (Fritts 1976; Whit-
more 1998; Worbes 2002; Rozendaal and Zuidema
2011; Pumijumnong 2013).
The search for tropical tree species forming an-
nual growth rings revealed that annually periodic
environmental phenomena could cause annual pe-
riodicity of vegetative growth even in tropical areas
(de Alvim 1964). Annual growth rings are formed
in tropical trees when they experience cambial dor-
mancy caused by unfavorable environmental con-
ditions, such as a dry season in areas with dis-
tinct seasonality (Worbes 1999; Gebrekirstos et al.
2008; Spannl et al. 2016). Distinct seasonality in
climate with a marked dry season (monthly rain-
fall <100 mm) for at least five months triggers an-
nual growth-ring formation in tropical deciduous
and semi-deciduous forests (Gentry 1995; Graham
and Dilcher 1995). However, even a short dry pe-
riod may trigger annual ring formation in tropical
evergreen forests (Dünisch et al. 2003; Fichtler et
al. 2003; Volland-Voigt et al. 2011). It is therefore
of high relevance to test if trees growing in a moist
tropical forest with mean annual precipitation ex-
ceeding 2000 mm and a marked dry season form
annual growth rings.
Although the history of tropical den-
drochronology dates back to early decades of
the 20th Century, the dendrochronological po-
tential of most tropical tree species remains
unexplored (Worbes 2002). Some basic studies
reported growth rings in tropical trees and their
potential for tropical dendrochronology (Maingi
2006; Worbes and Fichtler 2011; Rozendaal and
Zuidema 2011; Pumijumnong 2013; Gebrekirstos
et al. 2014). During the last two decades, tree-ring
studies distinguishably expanded to tropical South
America (Worbes et al. 1999; Bräuning et al. 2009),
subtropical to semi-arid Africa (Gebrekirstos et al.
2008; Trouet et al. 2010; Krepkowski et al. 2012;
David et al. 2014) and Southeast Asia (Berlage et
al. 1931; Pumijumnong et al. 1995, 2001, 2013;
Vlam et al. 2014). However, most of these studies
were carried out in dry tropical forests with mean
annual precipitation below 2000 mm.
In moist tropical forests, tree-ring studies are
still scarce. The species studied so far include but are
77
not restricted to Tectona grandis in India, Thailand
and Myanmar (Bhattacharyya et al. 1992; Pumi-
jumnong et al. 2001; Borgaonkar et al. 2007), Swi-
etenia macrophyla and Cedrela odorata in Brazil-
ian Amazon (Dünisch et al. 2003), Cedrela mon-
tana in Ecuador (Bräuning et al. 2009; Spannl et
al. 2012) and several species in Bolivia and Costa
Rica (Fichtler et al. 2003; Lopez et al. 2012). Re-
cently, Groenendijk et al. (2014) studied the den-
drochronological potential of 22 commercial tree
species from a wet tropical forest (mean annual
rainfall >4000 mm) in Cameroon and found that 14
out of 22 species had clear rings, although crossdat-
ing was difficult for some species. This study clearly
indicates that many more tropical species might
have dendrochronological potential, even though
they are growing in high precipitation areas within
the tropics. Identification of those species is crucial
for further development of tropical dendrochronol-
ogy and its application in multiple fields of research
(Babst et al. 2017). In this article, we investigated the
dendrochronological potential of 27 tree species in
a moist tropical forest in Bangladesh, a tropical re-
gion in South Asia influenced by the Asian mon-
soon climate.
Growth-ring boundaries in the tropics vary
widely in terms of wood anatomy, visibility, and
clarity. Distinct, indistinct, and absent growth-
ring boundaries were reported across the tropics
(Worbes 2002). Climatic fluctuations within the sea-
sons can cause intra-annual density fluctuations
and formation of false rings (Priya and Bhat 1998).
In addition, insufficient light conditions and growth
under competition can result in the formation of
wedging rings (Worbes 2002). Detecting growth-
ring boundaries in tropical trees is thus often not
straightforward. Different approaches are used in
detecting growth-ring boundaries in tropical tree
species. The microscopic analysis of wood anatomy
in thin sections can help detect growth-ring bound-
ary (Pumijumnong 2013), even in tree species grow-
ing in areas with high annual precipitation (>2300
mm). In the present study, we investigated the
wood anatomical nature of growth-ring bound-
aries of 27 tree species growing in a moist trop-
ical forest in Bangladesh. Furthermore, we eval-
uated the dendrochronological potential of some
common and important tree species and provide
the first dendrochronological information from this
78 ISLAM, RAHMAN, and BRÄUNING

Figure 1. (a) Map of Bangladesh showing the location of the study site (Rema-Kalenga Wildlife Sanctuary). (b) Map of Rema-Kalenga
Wildlife Sanctuary (24◦ 06 −24◦ 14 N and 91◦ 36 −91◦ 39 E). (c) Walter and Leith climate diagram based on data from 1950 to 2015
recorded at the nearest weather station (Sreemangal, 23.5 km away from the study site).

forest type of Bangladesh. Specifically, the objec- MATERIALS AND METHODS

tives of the current study include (1) characterizing
the distinctness of growth-ring boundaries by an-
alyzing wood anatomical features; (2) grouping of
species based on wood anatomical features defin-
ing growth-ring boundaries, and evaluating the rel-
ative importance of wood anatomical features in
detecting growth-ring boundaries; (3) testing the
crossdating potential of tree species having dis-
tinct growth-ring boundaries to establish correctly
dated tree-ring chronologies; and (4) assessing the
potential of tree-ring analysis to determine tree
ages, growth-climate relationships, and productiv-
ity in a moist tropical forest in monsoonal South
Asia.
Study Site and Climate
Samples were collected in the Rema-Kalenga
Wildlife Sanctuary, a nature reserve located in
the northeastern region of Bangladesh between
24◦ 06 −24◦ 14 N and 91◦ 36 −91◦ 39 E. The study
site is a part of the Tarap mountain system,
bounded on the east and south by the Tripura
state of India (Figure 1) and situated in the bio-
ecological zone of the Sylhet Hills (Nishat et al.
2002) along the Indo-Burma biodiversity hotspot.
The total area of the sanctuary is 1795 ha and
consists of several hills of different elevations and
low-lying valleys, with the highest altitude of 67 m
 Dendrochronological Potential in a Bangladeshi Moist Tropical Forest
a.s.l. Soils of the sanctuary vary from clay loam
on the relatively level ground or valleys to sandy
loam on the hills (Hassan 1994). Mean annual rain-
fall and mean annual temperature recorded from
nearby weather station Sreemangal (24◦ 30 N and
91◦ 39 E) are 2370 mm and 24.8◦ C, respectively.
There is a distinct seasonality with a long wet
season (April–October) and a relatively short dry
season (November–February) with mean monthly
rainfall less than 100 mm (Figure 2). Wet-season
precipitation contributes 85% of the total annual
precipitation. Temperature and precipitation data
were obtained from the Bangladesh Meteorological
Department (BMD).
According to the life-zone classification of
Holdridge (1967), the forest is classified as moist
tropical forest, containing both deciduous canopy
and evergreen understorey species. The dominant
tree species include Dipterocarpus turbinatus, Tec-
tona grandis, Chukrasia tabularis, Toona ciliata,
Syzygium grandis and Salmalia sp. The under-
growth comprises various types of bamboo species
(Bambusa sp.; Islam et al. 2006). The middle canopy
is mainly composed of Terminalia sp., Lager-
stroemia sp., Dillenia pentagyna as well as different
species of Ficus and Albizia.
Sample Material and Preparation
We studied wood anatomy and crossdating po-
tential of 27 tree species representing 17 families
(Table S1). The selection of species was based on the
abundance of species (Sobuj and Rahman 2011).
Species distribution, main use of wood, phenol-
ogy and other ecological characteristics of studied
species are described in Table S1. Fieldwork was
conducted in the 1795-ha Rema-Kalenga Wildlife
Sanctuary, Bangladesh. We extracted 420 increment
cores from 210 trees; at least 2 trees (4 cores) from
each species were collected. Cores were extracted
at breast height (1.3 m) with a 5.0 mm diameter
Suunto (Vantaa, Finland) increment borer. Trees
with highly irregular stems were avoided because
growth-ring detection, measurement, and crossdat-
ing might be more difficult. Cores were stored
in plastic core holders in the field and then air
dried for 24 hours in order to avoid attack by
stain fungi. The increment cores were mounted on
wooden holders and sanded with increasingly fine
79
grained sandpaper (80–2000 grit) following Stokes
and Smiley (1968). Cores were then scanned on a
high-resolution flatbed scanner (Epson Expression
10000XL, Long Beach, CA USA) at 1600 dpi. To
examine the growth-ring anatomy, microscopic thin
sections of 20–30 μm thickness were prepared us-
ing a GSL-1 sledge microtome. Thin sections were
stained with safranin red following a standard tech-
nique developed for microscopic preparation for
plant stem analysis (Gärtner and Schweingruber
2013).
Evaluation of Tree-Ring Boundary Anatomy
The micro slides were examined under a digi-
tal microscope Zeiss Smartzoom 5 (Carl Zeiss Mi-
croscopy GmbH 2014, Jena, Germany). Growth-
ring boundaries were tested by the presence or
absence of several wood anatomical features as
recommended by the International Association of
Wood Anatomists (Wheeler et al. 1989; Table 1).
Each species was assigned a qualitative score for a
particular anatomical feature based on the visual
characterization of the overall distinctness of re-
spective anatomical feature (i.e. distinct = 1, less
distinct = 0.5, absent = 0). These scores were fur-
ther analyzed to determine the relative importance
of anatomical features in the detection of growth-
ring boundaries.
Crossdating
Tree-ring width (TRW) measurements were
made from the scans with a precision of 0.01 mm
with the program CooRecorder (Cybis Elektronik
& Data AB, Saltsjöbaden, Sweden) with a cross
check by a high magnification light microscope.
Tree-ring series were visually and statistically cross-
dated by comparing measurements from two cores
of the same tree by using the computer program
TSAP-Win (RINNTECH, Heidelberg, Germany).
Crossdating within a tree allows synchronizing the
rings of identical age between the different radii of
an individual and identifying wedging rings (rings
that merge on certain parts of the circumference
of the tree) and ‘false’ ring structures (intra-annual
growth variations; Stokes and Smiley 1968). For
crossdating between trees, we first selected a tree
that showed good correlation and correct crossdat-
80 ISLAM, RAHMAN, and BRÄUNING

Figure 2. Wood anatomy and growth-ring boundaries of tropical tree species from a moist tropical forest in Bangladesh. Solid triangles
indicate growth boundaries. Scale bar is 1 mm in Figure 2d, f, j, l, and n, and 500 μm in other figures.
 Dendrochronological Potential in a Bangladeshi Moist Tropical Forest 81

Table 1. List of wood anatomical features examined in thin sections of 27 tree species in a moist tropical forest in Bangladesh.

Serial No. Anatomical Feature Code

1 Thick-walled latewood fibres FWT
2 Radially flattened latewood fibres RFF
3 Marked differences in vessel size between latewood and earlywood VS
4 Marked differences in vessel frequency between latewood and earlywood VF
5 Marginal parenchyma band (terminal or initial) MP
6 Narrow vessel elements very numerous or forming the ground
tissue of the latewood NV
7 Distinct fibre zones caused by decreasing frequency of parenchyma bands towards the latewood FZ
8 Porosity P
9 Distended rays DR

ing between two increment cores taken from dif-
ferent directions of the stem. Taking this tree as a
reference, we crossdated subsequent cores of other
trees. The quality of crossdating was verified by
considering statistical parameters such as t-test and
GLK values (sign test) that specify the proportion
of agreement/disagreement of inter-annual growth
tendencies among the trees (Eckstein and Bauch
1969). Crossdated tree-ring series were used to es-
tablish correctly dated tree-ring chronologies. In or-
der to remove biological growth trends from derived
tree-ring chronologies, all individual tree-ring series
were detrended by fitting a cubic smoothing spline
curve with a 60% frequency response (p = 0.6). The
statistical quality check of the crossdated ring width
series was performed using the Dendrochronology
Program Library in R (dplR; Bunn 2010) within the
R statistical programming environment (R Devel-
opment Core Team 2016).
Statistical Analyses
In order to group species that are anatomi-
cally homogeneous, we performed a cluster analy-
sis by using the ward.D2 algorithm. Five clusters
were subsequently defined because this grouping al-
lowed the best separation of species in anatomi-
cally homogeneous groups. Trends in growth-ring
boundary anatomy of the samples within these dif-
ferent groups were analyzed by Principal Compo-
nent Analysis (PCA). PCA also allowed determina-
tion of how anatomical features are inter-correlated
and the relative importance of those anatomical
features in growth-ring boundary detection. The
analyses were performed using vegan, permute, and
gplots packages within the R statistical program-
ming environment (R Development Core Team
2016).
RESULTS
Variability in Growth-Ring Boundary Anatomy
Growth-ring boundaries of several tree species
were mainly demarcated by a marginal parenchyma
band, and in some species in association with thick-
walled and tangentially flattened latewood fibres
and (or) the presence of larger vessels at the be-
ginning of the growth ring (Figure 2). Growth-ring
boundaries of certain species were only determined
by marginal parenchyma band (Figure 1a–f, m and
n). In Dipterocarpus turbinatus and Shorea robusta,
growth boundaries were characterized by a combi-
nation of marginal parenchyma and a narrow ves-
sel zone (Figure 2g and h). Growth-ring boundaries
in Toona ciliata were distinguished by marginal
parenchyma and larger vessels in the earlywood
zone, whereas those in Lagerstroemia speciosa were
delineated by multiple wood anatomical features
such as marginal parenchyma, larger vessels in the
earlywood zone, and thick-walled fibres (Figure 2).
Growth-ring boundaries of some species were
mainly characterized by thick-walled latewood fi-
bres and fibre zones with few or no parenchyma
cells (Figure 3). A fibre zone was the main
anatomical feature delineating growth-ring bound-
aries in Aphanamixis polystachya, Castanopsis ar-
mata, Hopea scaphula, Lagerstroemia parviflora,
Syzygium cumini and Syzygium grande. However,
Aphanamixis polystachya was also characterized
by parenchyma band density variations, showing
82 ISLAM, RAHMAN, and BRÄUNING

Figure 3. Wood anatomy and growth-ring boundaries of tropical tree species from a moist tropical forest in Bangladesh. Solid triangles
indicate growth boundaries. Scale bar is 1 mm in Figure 3b, d, e, g, j, and l, and 500 μm in other figures.
 Dendrochronological Potential in a Bangladeshi Moist Tropical Forest
increasing density along the latewood zone (Fig-
ure 3a). In Anthocephalus chinensis and Terminalia
bellirica, growth-ring boundaries were entirely dis-
tinguished by thick-walled latewood fibres (Figure
3b and j). Distended rays were the central determi-
nant of growth-ring boundaries in Bombax ceiba,
whereas smaller size vessels in the latewood zone
differentiated growth-ring boundaries in Pistacia
integerrima and Schima wallichii (Figure 3e and l).
Figure 4 shows high magnification digital micro-
scopic images of 5 species and their growth-ring
boundaries.
Clustering Based on Growth-Ring Boundary
Anatomy
Clustering of species resulted in 5 clusters
(Figure 5). The clustering was based on the indi-
vidual score of each anatomical feature that de-
fined the growth-ring boundary of each species as
illustrated in the heat map (Figure 6). Cluster 1
was characterized by growth-ring boundaries de-
fined mainly by marginal parenchyma bands. Clus-
ter 2 was also characterized by growth-ring bound-
aries delineated by marginal parenchyma, but the
parenchyma bands were relatively narrow com-
pared to those of cluster 1. Marginal parenchyma
in association with changing porosity determined
growth-ring boundaries in the species group rep-
resenting cluster 3 (Figure 2j, k and l). Clusters 4
and 5 were closely related and growth-ring bound-
aries in these clusters were defined mainly by thick-
walled latewood fibres and fibre zones with little
or no parenchyma cells. The presence of distinct
radially flattened latewood fibres (RFF) in clus-
ter 5 differentiated it from cluster 4 (Figures 2
and 6).
Relative Importance of Wood Anatomical Fea-
tures in Boundary Detection
The principal component analysis (PCA) re-
vealed that marginal parenchyma, fibre zone, and
fibre wall thickness contributed together to the first
axis. The second axis mainly represented poros-
ity and changes in vessel size and vessel frequency
(Figure 7). These two axes together explained more
than half (51.09%) of the total variance (Table
S2). Marginal parenchyma, narrow vessels and dis-
83
tended rays were positively correlated and were
dominant factors in defining growth-ring bound-
aries in the species belonging to the first two clus-
ters. On the other hand, fibre wall thickness and
fibre zone contributed mostly to the detection of
growth-ring boundaries in species group cluster 4
and cluster 5. These results were reflected in the
PCA as illustrated in Figure 7. In total, vessel size
and vessel frequency were present in around 74%
of our studied species followed by fibre wall thick-
ness (67%) and marginal parenchyma (52%). In
one-third (33%) of all studied tree species, marginal
parenchyma was the dominating wood anatomi-
cal feature in defining growth-ring boundaries, fol-
lowed by fibre wall thickness and fibre zone, each
of which was dominant in 30% of the species.
Narrow vessel elements (NV) had the least con-
tribution in detecting growth-ring boundaries in
only 2 species of the dipterocarp family and there-
fore was considered a special tree-ring anatomical
feature.
Tree-Ring Boundary Distinctness and Crossdating
The evaluation of growth-ring boundary dis-
tinctness in thin-sections and scanned images re-
vealed that 25 out of 27 studied species (92%)
showed distinct to fairly distinct growth-ring
boundaries, a stunning fact for a moist tropical for-
est (Table 2). Growth curves of these species cross-
dated well within one tree as well as between trees.
An example of synchronization of tree-ring series
from 5 tree species representing five clusters is illus-
trated in Figure 8. The GLK values given in Figure
8 provide statistical proof of the synchronization
of tree-ring series of each species. Only 2 species
(Ilex godajam and Syzygium cumini) had indistinct
growth-ring boundaries and were found to have a
lower potential for crossdating. During ring-width
measurement and crossdating, several difficulties
were encountered. Difficulties in ring-width mea-
surement during slow growth periods (SG) were
common in most of the species and hence imposed
the major difficulty in studying tree rings in our
forest. However, no difficulties were observed dur-
ing ring-width measurement and crossdating in Al-
bizia procera, Dyllenia pentagyna, Mangifera sylvat-
ica and Terminalia bellirica.
84 ISLAM, RAHMAN, and BRÄUNING

Figure 4. High magnification digital microscopic images of 5 selected tree species showing characteristic wood anatomical types of
growth-ring boundaries. Growth direction is from right to left. Solid triangles indicate growth boundaries. Double arrow solid line in
(e) indicates a fibre zone with few or no vessels representing an indistinct growth boundary. Scale bar is 200 μm.

Estimation of Tree Ages and Biomass Growth
The mean tree ages varied from 27 to 161 years
among the five species studied in detail. The high-
est age was found in Vitex pubescens and lowest age
was observed in Dillenia pentagyna. Mean growth
trajectories of each species clearly showed the vari-
ation in growth rate and biomass growth among
species (Figure 9). The shortest period of time to at-
tain at 25 cm diameter was needed by Dillenia pen-
tagyna, while Vitex pubescens required the longest
time to attain the same size. The other 3 species
showed similar growth patterns throughout their
life.
 Dendrochronological Potential in a Bangladeshi Moist Tropical Forest
Figure 5. Cluster dendrogram of 27 tropical forest tree species representing 17 families from a moist tropical forest in Bangladesh.
Ward.D2 clustering method was used.
DISCUSSION AND CONCLUSIONS
Wood Anatomical Features in Detecting Growth-
Ring Boundary
We analyzed nine anatomical features of wood
as recommended by Wheeler et al. (1989) to de-
fine growth-ring boundaries in a moist tropical
Figure 6. Heat map showing the individual score of wood
anatomical features defining growth-ring boundaries of each
species. See Table 2 for the description of codes used for anatom-
ical features. For species code, see Table 1; the first two letters
represent the genus name and last two letters represent species
name.
85
forest of Bangladesh. Among the anatomical fea-
tures, marginal parenchyma, fibre zones, and thick-
walled latewood fibres had the highest significance
in detecting growth-ring boundaries in our stud-
ied species as also described for other tropical tree
species (Worbes 2002; Marcati et al. 2006; Worbes
and Fichtler 2011; Azim and Ocada 2014; Groe-
nendijk et al. 2014; Nath et al. 2016; Tarelkin et
al. 2016). Changes in vessel frequency, parenchyma
density, and distended rays were reported as more
useful tree-ring markers in tropical trees than
Figure 7. Biplot of first two principal components from PCA of
wood anatomical features defining growth-ring boundaries in 27
tropical tree species from a moist tropical forest in Bangladesh.
See Table 2 for the description of codes used for wood anatomi-
cal features. Each data point represents a species belonging to a
particular cluster.
86 ISLAM, RAHMAN, and BRÄUNING
radially flattened latewood fibres or changes in ves-
sel size (Tarelkin 2016). The variation of vessel size
and/or density is most likely caused by changes
in water availability to trees (e.g. Baas 1976; Baas
et al. 1983; Alves and Angyalossy-Alfonsi 2000;
Ohashi et al. 2009). We observed changes in ves-
sel size and vessel frequency in most of our species
(74%). This is not an unexpected result because our
study area experiences moisture stress during the
dry and pre-monsoon seasons, when mean monthly
precipitation is below 100 mm and evapotranspira-
tion is high. However, vessel frequency contributed
more in detecting growth-ring boundaries (Figures
6 and 7) than vessel size, which is in line with
other studies (Groenendijk et al. 2014; Tarelkin et
al. 2016). Distended rays were mostly associated
with marginal parenchyma and were a dominant
factor in growth-ring boundary detection only in
two species: Garuga pinnata and Gmelina arborea.
Likewise, a continuous narrow vessel zone was only
observed as a boundary detector in Dipterocarpus
turbinatus and Shorea robusta in association with
marginal parenchyma.
We observed a high within- and between-
species variability in wood anatomical features
(Figure 6). We are aware of three studies that
analyzed the anatomy of growth-ring boundaries
of several species in a moist tropical climate,
which overlapped four of our 27 studied species
(Azim and Ocada 2014; Nath et al. 2012, 2016).
Gmelina arborea and Michelia champaca were stud-
ied in Western Ghats, India (Nath et al. 2016)
and also formed distinct growth rings primar-
ily defined by marginal parenchyma. Growth-ring
boundaries in Syzygium cumini are marked by fi-
bre zones and thick-walled latewood fibres (Nath
et al. 2012). In contrast to Nath et al. (2016),
we observed distinct growth-ring boundaries in
Terminalia bellirica defined by thick-walled late-
wood fibres. Likewise, Azim and Ocada (2014)
observed poorly defined growth-ring boundaries
in Gmelina arborea, whereas we observed distinct
growth-ring boundaries in this species delineated by
marginal parenchyma in association with distended
rays. Such variations, however, entail growth-ring
boundary features that can vary strongly between
individuals of a species growing under different cli-
matic or site conditions (Fichtler and Worbes 2012;
Tarelkin et al. 2016).
Inter-species variability in growth-ring bound-
ary anatomy might be linked to leaf phenology and
species stature (Fichtler and Worbes 2012; Nath et
al. 2016). Phylogenetic differences in wood anatomy
might also be attributed to the inter-species variabil-
ity in growth-ring boundary anatomy. For exam-
ple, two species of the genus Syzygium (Syzygium
cumini and Syzygium grande) and Lagerstroemia
(Lagerstroemia speciosa and Lagerstroemia parvi-
flora) were placed at different positions in the clus-
ter dendrogram (Figure 5).
Dendrochronological Potential
Out of the 27 studied moist forest tree species,
25 species (93%) formed distinct to fairly distinct
growth-ring boundaries (Table 2), underlining a
great potential for dendrochronology of the study
region. Ring-width patterns of individuals from
the same tree species were successfully crossdated
(Figure 8), providing evidence that these growth
bands are true annual tree rings. Several experimen-
tal techniques have been developed to prove the an-
nual nature of tree rings in tropical trees, includ-
ing periodic sampling of the cambial zone (Krep-
kowski et al. 2012; Trouet et al. 2012; Volland-Voigt
et al. 2012; Chowdhury et al. 2015), anatomical
studies of trees of known age, radiocarbon dating
(Worbes 1989; Fichtler et al. 2003; Krepkowski et
al. 2012; Groenendijk et al. 2014), and dendrometer
studies (Bräuning 2009; Volland-Voigt et al. 2012;
Spannl et al. 2016). In addition, crossdating of ring-
width time-series between individual trees and be-
tween multiple sites in a region provides strong ev-
idence that the growth rings are indeed annual and
are not arbitrary structural features of the xylem
(Stahle 1999). Crossdating also provides evidence
that the growth of trees is periodical, tied to the
annual seasonality of climate, and synchronized
among trees in a region by the inter-annual vari-
ation in the shared regional climate (Stahle 1999).
The synchronization of tree-ring series and statis-
tical proof of within- and between-tree crossdating
clearly indicates the dendrochronological potential
of our studied species.
It is also possible that, the same tree species
might form annual growth rings depending on the
complexity of local climate and specific site condi-
tions (Wils et al. 2009, 2011). However, our samples
Table 2. Distinctness of growth-ring boundaries and crossdating potential of 27 tree species from a moist tropical forest in Bangladesh.
Main Anatomical
No. of Features Defining
Cores Growth-Ring
Species Name (Disk) Boundarya
Aphanamixis polystachya 4 FZ, FWT, VS, VF
Albizia procera 4 MP
Anthocephalus chinensis 4 FZ, FWT, VS
Artocarpus chaplasha 4 VF, MP
Bombax ceiba 4 RFF, VF, FWT
Bursera serrata 4 FWT, FZ, VS, VF
Castanopsis armata 4 FZ, FWT
Chukrasia tabularis 60 (1) MP
Dillenia pentagyna 4 FWT, VF, VS
Dipterocarpus turbinatus 4 MP, NV
Garuga pinnata Roxb. 4 DR, VF, MP
Gmelina arborea 4 MP, VF, DR
Hopea scaphula 4 FZ, RFF, FWT
Ilex godajam 4 FZ, FWT, VS, VF
Lagerstroemia parviflora 4 FZ, MP, FWT
Lagerstroemia speciosa 70 P, MP, FWT, VS
Distinctness of
Growth-Ring Crossdating Dendrochronological
Boundaryb Potentialc
A HP
A HP
B P
B P
B P
B P
B P
A HP
A HP
B P
B P
B P
B P
C LP
B P
A HP
Potentiald Difficultiese
++ SG
++ N 0
+ SG, VR
+ FS, SG
+ SG
+ VG
+ VG
++ VG, SG, WR
++ N 0
+ SG
+ LR, FS
+ LR
+ SG
− VR
+ SG
SG
++
Previous Tree-Ring
Study In Moist
Previous Tropical Forests
Tree-Ring (Annual Rainfall
Studyf >2200 mm)f
0 0
0
0 0
0 0
0 0
0 0
0 0
1 (Vlam et al. 2014) 0
0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
Dendrochronological Potential in a Bangladeshi Moist Tropical Forest
87
 88

Table 2. Continued.

Previous Tree-Ring
Main Anatomical Study In Moist
No. of Features Defining Distinctness of Previous Tropical Forests
Cores Growth-Ring Growth-Ring Crossdating Dendrochronological Tree-Ring (Annual Rainfall
Species Name (Disk) Boundarya Boundaryb Potentialc Potentiald Difficultiese Studyf >2200 mm)f
Mangifera sylvatica 4 MP A HP ++ N 0 0
Michelia champaca 60 MP A HP ++ VR, SG 0 0
Pistacia integerrima 4 FZ, VS B HP ++ SG 0 0
Schima wallichii 4 VS, FWT B P + SG 0 0
Shorea robusta 4 MP, NV A HP ++ SG 0 0
Swietenia mahagoni 62 MP A HP ++ SG 0 0
Syzygium cumini 4 FZ, FWT, VS C LP − SG 0 0
Syzygium grande 4 FZ B P + SG 0 0
Terminalia bellirica 4 FWT B HP ++ N 0 0
Toona ciliata 80 P, MP, FWT A HP ++ SG, VR 1 (Vlam et al. 2014) 0
Vitex pubescens 4 MP A HP ++ FS, SG 0 0

a
See Table 2 for the description of codes used for anatomical features.
ISLAM, RAHMAN, and BRÄUNING

b
A = fairly distinct, B = distinct, C = indistinct, D = absent.
c
HP = high potential, P = potential, LP = less potential.
d
+ + = high potential, + = potential, – = less potential.
e
SG = ring delineation difficult during the period of slow growth, N = no difficulties, FG = fungal stain makes ring delineation difficult, VR = vagued ring boundaries, LR = light (inconspicuous)
ring, WR = wedging ring.
f
0 = species does not crossdate, or no information on crossdating for this species has been published.
1= species is known to crossdate between cores from the same tree as well as between trees from the same site (Grissino-Mayer 1993).
2= species is known to crossdate between sites in a region (Grissino-Mayer 1993).
 Dendrochronological Potential in a Bangladeshi Moist Tropical Forest 89

Figure 8. Scanned images and diameter growth anomalies (measured on scanned images) of five tree species from a moist tropical
forest in Bangladesh. Each species is selected from a particular cluster (see cluster dendrogram). Scale bar is 5 mm. Solid triangles
indicate growth boundaries. Arrows indicate the direction of growth.
90 ISLAM, RAHMAN, and BRÄUNING
Figure 9. Mean age-basal area relations for 5 tree species from a
moist tropical forest in Bangladesh. Each species is selected from
a particular cluster (see cluster dendrogram). For species code,
see Table 1; the first two letters represent the genus name and last
two letters represent species name. Horizontal dotted line indi-
cates the variation in age when each species attained a diameter of
25 cm.
were collected from one site that is subject to a ho-
mogeneous climatic condition. In addition to the
crossdating, the evidence for annual tree-ring for-
mation can be strengthened if the derived chronol-
ogy is strongly correlated with time series of in-
strumental climate (Pumijumnong et al. 1995). Our
sample size is not yet high enough to build statis-
tically robust chronologies of most of our stud-
ied tree species, which currently limits our efforts
to perform climate-growth analysis. Therefore, after
the successful identification of tree species with den-
drochronological potential in our study area, a next
step of research will include sampling of additional
tree individuals to establish statistically robust tree-
ring chronologies from multiple species for climate
analyses.
Although crossdating was successful for most
of our studied species, several difficulties were as-
sociated with growth-ring detection if the sam-
ples had only been analyzed from scanned images.
The level of difficulties varied with species. The
most common problem in growth-ring detection
was associated with irregular growth boundaries in
slow growth periods (SG), making correct growth-
ring identification and successful crossdating diffi-
cult. Similar problems were reported in some other
studies (Azim and Ocada 2014; Groenendijk et al.
2014), suggesting that trees growing under sup-
pressed conditions or under strong competition for
resources should only be considered with care for
dendrochronological studies or excluded from fur-
ther analyses.
Tree-Ring Analysis and Carbon Sequestration
We reconstructed lifetime growth trajectories
of 5 species and calculated their basal area incre-
ments as a measure of tree productivity (Figure 9).
Basal area increment per age varied widely among
species, which is in accordance with other stud-
ies (Baker and Bunyavejchewin 2006; Brienen and
Zuidema 2006; Zuidema et al. 2010). We estimated
the highest age for Vitex pubescens (161 years).
These findings suggest that some species could be
used for climatic reconstruction (Pucha-Cofrep et
al. 2015), and calculating biomass accumulation
and carbon sequestration (Simamoto et al. 2014;
Sanogo et al. 2016), as applied in other tropical
forests.
Tree growth rate, forest stand development
and productivity are usually measured in perma-
nent sample plots (PSPs) in annual or 5- to 10-
year intervals. This technique of growth monitoring
was criticized because of its coarse temporal res-
olution, long time needed for obtaining adequate
data for analysis, and its limitation to provide long-
term historic growth data (Metsaranta and Lieffers
2009). Throughout the tropics, PSPs with a long
history of measurements are scarce, and consider-
able time would be required to obtain useful data
from newly established plots. Ages and cumulative
biomass data of trees presented in our study clearly
demonstrate that tree-ring analysis can be used as a
tool to calculate carbon sequestration and tree mor-
tality on an annual basis (Mokria et al. 2015), which
might be used in modelling future productivity
and carbon sequestration in moist tropical forests
(Hiltner et al. 2016).
ACKNOWLEDGMENTS
We are grateful to the Bangladesh Forest De-
partment (BFD) for providing access to the Rema-
Kalenga Wildlife Sanctuary and for the support
during the fieldwork. We sincerely acknowledge the
consistent help of Md. Sumon Reza, Rofiqul Is-
lam, Swapan Miah and several undergraduate stu-
dents of the Department of Forestry and Environ-
mental Science, Sahjalal University of Science and
 Dendrochronological Potential in a Bangladeshi Moist Tropical Forest
Technology, Bangladesh during the fieldwork. We
are thankful to Mahin Rahman Swadesh who as-
sisted in several aspects during the laboratory work.
The research stay of Mizanur Rahman and the
fieldwork were supported by The German Aca-
demic Exchange Service (DAAD).
REFERENCES CITED
Alves, E. S., and V. Angyalossy-Alfonso, 2000. Ecological trends
in the wood anatomy of some Brazilian species. 1. Growth
rings and vessels. IAWA Journal 21(1):3−30.
Azim, A. A. A., and N. Okada, 2014. Occurrence and anatom-
ical features of growth rings in tropical rainforest trees in
Peninsular Malaysia: A preliminary study. Tropics 23(1):
15−31.
Baas, P., E. Werker, and A. Fahn, 1983. Some ecological trends
in vessel characters. IAWA Journal 4(2−3):141−159.
Babst, F., B. Poulter, P. Bodesheim, M. D. Mahecha, and D. C.
Frank, 2017. Improved tree-ring archives will support earth-
system science. Nature Ecology & Evolution 1(2):8.
Baker, P. J., and S. Bunyavejchewin, 2006. Suppression, release
and canopy recruitment in five tree species from a seasonal
tropical forest in Western Thailand. Journal of Tropical Ecol-
ogy 22(2006):521−529.
Berlage, H., 1931. Over het verband tusschen de dikte der jaar-
ringen van djatiboomen (Tectona grandis l.f.). En den regenval
of Java. Tectona 24:939−953. [in Dutch.]
Bhattacharyya, A., R. R. Yadav, H. P. Borgaonkar, and G. B.
Pant, 1992. Growth ring analysis on Indian tropical trees: Den-
droclimatic potential. Current Science 62:736−741.
Borgaonkar, H. P., A. B. Sikder, S. Ram, K. R. Kumar,
and G. B. Pant, 2007. Dendroclimatological investigations
of high altitude Himalayan conifers and tropical teak in
India. The Korean Journal of Quaternary Research 21(1):
15−26.
Bräuning, A., F. Volland-Voigt, I. Burchardt, G. Oswaldo, N.
Thomas, P. Thorsten, 2009. Climatic control of radial growth
of Cedrela montana in a humid mountain rainforest in South-
ern Ecuador. Erdkunde 63(4):337−345.
Brienen, R. J. W., J. Schöngart, and P. A. Zuidema, 2016. Tree
rings in the tropics: Insights into the ecology and climate sen-
sitivity of tropical trees. In Tropical Tree Physiology, Vol-
ume 6, edited by Goldstein G., and L. Santiago, pp. 439–461.
Springer, Berlin.
Brienen, R. J. W., and P. A. Zuidema, 2006. Lifetime growth pat-
terns and ages of Bolivian rain forest trees obtained by tree ring
analysis. Journal of Ecology 94(2):481−493.
Bunn, A. G., 2008. A Dendrochronology Program Library in R
(dplR). Dendrochronologia 26(2):115−124.
Chowdhury, M. Q., N. Schmitz, A. Verheyden, U. Sass-Klaassen,
N. Koedam, and H. Beeckman, 2008. Nature and periodicity
of growth rings in two Bangladeshi mangrove species. IAWA
Journal 29(3):265−276.
David, E. T., S. Chhin, and D. Skole, 2014. Dendrochronological
potential and productivity of tropical tree species in Western
Kenya. Tree-Ring Research 70:119−135.
91
de Alvim, P. T., 1964. Tree growth periodicity in tropical climates.
In The Formation of Wood in Forest Trees, edited by M. Zim-
merman, pp. 479–495. Academic Press, New York, NY, USA.
Détienne, P., 1989. Appearance and periodicity of growth rings
in some tropical woods. IAWA Bulletin 10(2):123−132.
Dünisch, O., V. R. Montóia, and J. Bauch, 2003. Dendroecolog-
ical investigations on Swietenia macrophylla King and Cedrela
odorata l. (Meliaceae) in the central Amazon. Trees – Structure
and Function 17(3):244−250.
Eckstein, D., and J. Bauch, 1969. Beitrag zur rationalisierung
eines dendrochronologischen verfahrens und zur analyse
seiner aussagesicherheit. Forstwissenschaftliches Centralblatt
88:230−250. [in German.]
Fan, Z., and A. Bräuning, 2017. Tree-ring evidence for the his-
torical cyclic defoliator outbreaks on Larix potaninii in the
Central Hengduan mountains, SW China. Ecological Indica-
tors 74:160−171.
Fichtler, E., D. A. Clark, and M. Worbes, 2003. Age and
long-term growth of trees in an old-growth tropical rain
forest, based on analyses of tree rings and 14 C. Biotropica
35(3):306−17.
Fichtler, E., and M. Worbes, 2012. Wood anatomical variables in
tropical trees and their relation to site conditions and individ-
ual tree morphology. IAWA Journal 33(2):119−140.
Fritts, H. C., 1976. Tree Rings and Climate. Academic Press Inc,
London.
Gärtner, H., and F. H. Schweingruber, 2013. Microscopic Prepa-
ration Techniques for Plant Stem Analysis. Kessel Publishing
House, Remagen, Germany.
Gebrekirstos, A., A. Bräuning, U. Sass-Klaassen, and C. Mbow,
2014. Opportunities and applications of dendrochronology
in Africa. Current Opinion in Environmental Sustainability
6(1):48−53.
Gebrekirstos, A., R. Mitlöhner, D. Teketay, and M. Worbes,
2008. Climate-growth relationships of the dominant tree
species from semi-arid savanna woodland in Ethiopia. Trees
– Structure and Function 22(5):631−641.
Gentry, A. H., 1995. Diversity and floristic composition of
neotropical dry forests. In Seasonally Dry Tropical Forests,
edited by S. Bullock, H. Mooney, and E. Medina, pp. 146−94.
Cambridge University Press, Cambridge.
Graham, A., and D. Dilcher, 1995. The Cenozoic record of trop-
ical dry forest in northern Latin America and the southern
United States. In Seasonally Dry Tropical Forests, edited by S.
Bullock, H. Mooney, and E. Medina, pp. 124−45. Cambridge
University Press, Cambridge.
Groenendijk, P., U. Sass-Klaassen, F. Bongers, and P. A.
Zuidema, 2014. Potential of tree-ring analysis in a wet tropical
forest: A case study on 22 commercial tree species in Central
Africa. Forest Ecology and Management 323:65−68.
Hassan, M. M., 1994. Forest soils of Bangladesh. Journal of For-
est Science 23:1−11.
Hiltner, U., A. Bräuning, A. Gebrekirstos, A. Huth, and R.
Fischer, 2016. Impacts of precipitation variability on the dy-
namics of a dry tropical montane forest. Ecological Modelling
320:92−101.
Holdridge, L. R., 1967. Life Zone Ecology. Tropical Science Cen-
ter, San Jose, Costa Rica.
92 ISLAM, RAHMAN, and BRÄUNING
Krepkowski, J., A. Bräuning, and A. Gebrekirstos, 2012. Growth
dynamics and potential for crossdating and multi-century cli-
mate reconstruction of Podocarpus falcatus in Ethiopia. Den-
drochronologia 30(4):257−265.
Lopez, L., R. Villalba, and M. Peña-Claros, 2012. Determining
the annual periodicity of growth rings in seven tree species of
a tropical moist forest in Santa Cruz, Bolivia. Forest Systems
21(3):508−514.
Maingi, J. K., 2006. Growth rings in tree species from the Tana
River floodplain, Kenya. Journal of East African Natural His-
tory 95(2):181−211.
Marcati, C. R., J. S. Oliveira, and S. R. Machado, 2006. Growth
rings in cerrado woody species: Occurrence and anatomical
markers. Biota Neotropica 6(3):1–31.
McCarroll, D., and N. J. Loader, 2004. Stable isotopes in tree-
rings. Quaternary Science Reviews 23(7–8):771–801.
Metsaranta, J. M., and V. J. Lieffers, 2009. Using dendrochronol-
ogy to obtain annual data for modelling stand develop-
ment: A supplement to permanent sample plots. Forestry
82(2):163−173.
Mokria, M., A. Gebrekirstos, E. Aynekulu, and A. Bräuning,
2015. Tree dieback affects climate change mitigation potential
of a dry afromontane forest in northern Ethiopia. Forest Ecol-
ogy and Management 344:73−83.
Nath, C. D., A. Boura, D. de Franceschi, and R. Pélissier, 2012.
Assessing the utility of direct and indirect methods for esti-
mating tropical tree age in the Western Ghats, India. Trees –
Structure and Function 26(3):1017−1029.
Nath, C. D., F. Munoz, R. Pélissier, D. F. R. P. Burslem, and
G. Muthusankar, 2016. Growth rings in tropical trees: Role of
functional traits, environment, and phylogeny. Trees – Struc-
ture and Function 30(6):2153−2175.
Nishat, A., S. M. I. Huq, S. P. Barua, A. H. M. A. Reza, and A.
S. M. Khan (eds.), 2002. Bio-Ecological Zones of Bangldeesh.
IUCN Bangladesh country office, Dhaka.
Ohashi, S., N. Okada, T. Nobuchi, S. Siripatanadilok, and
T. Veenin, 2009. Detecting invisible growth rings of trees
in seasonally dry forests in Thailand: Isotopic and wood
anatomical approaches. Trees – Structure and Function 23(4):
813−822.
Oksanen, J., F. G. Blanchet, R. Kindt, P. Legendre, and R. B.
O’Hara, 2016. Vegan: Community ecology package. R package
2.3-3, Available at: https://cran.r-project.org/web/packa.
Priya, P. B., and K. M. Bhat, 1998. False ring formation in teak
(Tectona grandis l.f.) and the influence of environmental fac-
tors. Forest Ecology and Management 108(3):215−222.
Pucha-Cofrep, D., T. Peters, and A. Bräuning, 2015. Wet season
precipitation during the past century reconstructed from tree-
rings of a tropical dry forest in Southern Ecuador. Global and
Planetary Change 133:65−78.
Pumijumnong, N., 2013. Dendrochronology in Southeast Asia.
Trees – Structure and Function 27(2):343−358.
Pumijumnong, N., D. Eckstein, and W. K. Park. 2001. Teak tree-
ring chronologies in Myanmar: A first attempt. Palaeobotanist
50:35−40.
Pumijumnong, N., D. Eckstein, and U. Sass-Klassen, 1995. Tree-
ring research on Tectona grandis in northern Thailand. IAWA
Journal 16(4):385−392.
R Development Core Team, 2016. R: A Language and Envi-
ronment for Statistical Computing. R Foundation for Statis-
tical Computing, Vienna, Austria. Retrieved (http://www.R-
project.org/).
Rozendaal, D. M. A., and P. A. Zuidema, 2011. Dendroecol-
ogy in the tropics: A review. Trees – Structure and Function
25(1):3−16.
Sanogo, K., A. Gebrekirstos, J. Bayala, G. B. Villamor, A. Kalin-
ganire, S. Dodiomon, 2016. Potential of dendrochronology in
assessing carbon sequestration rates of Vitellaria Paradoxa in
Southern Mali, West Africa. Dendrochronologia 40:26−35.
Shimamoto, C. Y., P. C. Botosso, and M. C. M. Marques, 2014.
How much carbon is sequestered during the restoration of
tropical forests? Estimates from tree species in the Brazilian At-
lantic forest. Forest Ecology and Management 329:1−9.
Sobuj, N. A., and M. Rahman. 2011. Comparison of plant di-
versity of natural forest and plantations of Rema-Kalenga
wildlife sanctuary of Bangladesh. Journal of Forest Science
27(3):127−134.
Spannl, S., S. Günter, T. Peters, F. Volland, and A. Bräuning,
2012. Which factors control tree growth in a tropical moun-
tain forest? The case of Cedrela montana in southern Ecuador.
TRACE - Tree Rings in Archaeology, Climatology and Ecology
10:99–105.
Spannl, S., F. Volland, D. Pucha-Cofrep, T. Peters, E. Cueva,
A. Bräuning, 2016. Climate variability, tree increment patterns
and ENSO-related carbon sequestration reduction of the trop-
ical dry forest species Loxopterygium huasango of Southern
Ecuador. Trees – Structure and Function 30:1245−1258.
Stahle, D. W., P. T. Mushove, M. K. Cleaverland, and G. Haynes,
1999. Management implications of annual growth rings in Pte-
rocarpus angolensis from Zimbabwe. Forest Ecology and Man-
agement 124:217−229.
Stokes, M. A., and T. L. Smiley. 1968. An Introduction to Tree-
Ring Dating. University of Chicago Press, Chicago, IL, USA.
Tarelkin, Y., C. Delvaux, M. De Ridder, T. El Berkani, C. De
Cannière, and H. Beeckman, 2016. Growth-ring distinctness
and boundary anatomy variability in tropical trees. IAWA
Journal 37(2):275−S7.
Trouet, V., J. Esper, and H. Beeckman, 2010. Climate/growth re-
lationships of Brachystegia spiciformis from the miombo
woodland in south central Africa. Dendrochronologia
28(3):161−171.
Trouet, V., M. Mukelabai, A. Verheyden, and H. Beeckman,
2012. Cambial growth season of brevi-deciduous Brachyste-
gia spiciformis trees from south central Africa restricted to less
than four months. PLoS ONE 7(10):e47364.
Vlam, M., P. J. Baker, S. Bunyavejchewin, and P. A. Zuidema.
2014. Temperature and rainfall strongly drive temporal
growth variation in Asian tropical forest trees. Oecologia
174(4):1449−1461.
Volland-Voigt, F., A. Bräuning, O. Ganzhi, T. Peters, and H.
Maza, 2011. Radial stem variations of Tabebuia chrysantha
(bignoniaceae) in different tropical forest ecosystems of south-
ern Ecuador. Trees – Structure and Function 25(1):39−48.
Wheeler, E. A., P. Baas, and P. E. Gasson (eds.), 1989. IAWA
list of microscopic features for hardwood identification. IAWA
Bulletin n.s. 10(3):219−332.
 Dendrochronological Potential in a Bangladeshi Moist Tropical Forest
Whitmore T. C., 1998. Potential impact of climatic change on
tropical rain forest seedlings and forest regeneration. Climatic
Change 39(2−3):429−438.
Wils, T. H. G., I. Robertson, Z. Eshetu, U. Sass-Klaassen,
and M. Koprowski, 2009. Periodicity of growth rings in
Juniperus procera from Ethiopia inferred from crossdat-
ing and radiocarbon dating. Dendrochronologia 27:45−58.
DOI:10.1016/j.dendro.2008.08.002.
Wils, T. H. G., I. Robertson, Z. Eshetu, R. Touchan, U. Sass-
Klaassen, and M. Koprowski, 2011. Crossdating Juniperus pro-
cera from North Gondar, Ethiopia. Trees – Structure and Func-
tion 25:71−82, 2011. DOI:10.1007/s00468-010-0475-0.
Worbes, M., 1999. Annual growth rings, rainfall-dependent
growth and long-term growth patterns of tropical trees from
the Caparo forest reserve in Venezuela. Journal of Ecology
87(3):391−403.
Worbes, M., 2002. One hundred years of tree-ring research in the
tropics – A brief history and an outlook to future challenges.
Dendrochronologia 20(1–2):217−231.
93
Worbes, M., and E. Fichtler, 2011. Wood anatomy and tree-ring
structure and their importance for tropical dendrochronology.
In Amazonian Floodplain Forests: Ecophysiology, Biodiversity
and Sustainable Management, edited by Junk, W. J., M. T. F.
Piedade, F. Wittmann, J. Schöngart, and P. Parolin, pp. 329–
346. Springer, Dordrecht, Netherlands.
Zhang, Z., 2015. Tree-rings, a key ecological indicator of
environment and climate change. Ecological Indicators 51:
107−116.
Zuidema, P. A., M. Vlam, and P. D. Chien, 2011. Ages and
long-term growth patterns of four threatened Vietnamese tree
species. Trees-Structure and Function 25(1):29−38.
Received 27 February 2017; accepted 9 June 2017.
Supplementary Material is available at http://
www.treeringsociety.org/TRBTRR/TRBTRR.htm