pure.southwales.ac.uk€¦ · Web view1 Lee et al.: Habitat utilisation by sun bears *...
Transcript of pure.southwales.ac.uk€¦ · Web view1 Lee et al.: Habitat utilisation by sun bears *...
1 Lee et al.: Habitat utilisation by sun bears
Understanding landscape and plot-scale habitat utilisation by Malayan sun bear (Helarctos
malayanus) in degraded lowland forest
David C. Lee a, b, *, Victoria J. Powell c, Jeremy A. Lindsell b, d
a School of Applied Sciences, University of South Wales, Pontypridd CF37 4BD, UK.
b RSPB Centre for Conservation Science, Royal Society for the Protection of Birds, Sandy SG19
2DL, UK.
c Harapan Rainforest, Jambi, Sumatra, Indonesia.
d A Rocha International, David Attenborough Building, Pembroke Street, Cambridge CB2 3QZ,
UK.
* Corresponding author. E-mail address: [email protected] (D. Lee).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
2 Lee et al.: Habitat utilisation by sun bears
ABSTRACT
Malayan sun bear (Helarctos malayanus) is a forest-dependent species globally threatened by
loss of suitable habitat and hunting. Understanding how sun bears utilise habitat in more
degraded landscapes is increasingly important for the effective conservation of the species. We
studied how landscape and plot attributes affect sun bear habitat use along a gradient of logging
disturbance in a lowland forest site of Sumatra. We conducted surveys of bear claw marks to
indicate sun bear habitat use at plot and landscape scales, and inform forest restoration strategies
that benefit the conservation management of the species. We recorded 12 habitat features and the
presence/absence of claw marks in 262 plots in four different habitat types. We reduced the
number of habitat variables using Principal Components Analysis (PCA), resulting in four
derived habitat factors. We used these factors in a Discriminant Analysis to refine habitat
classifications of plots, and modelled presence/absence of claw marks using the PCA factors in a
binary logistic regression. We inventoried tree species in a subset of randomly selected plots with
claw marks alongside paired control plots with no claw marks. We compared tree community
compositions in these plots using ANOSIM and SIMPER analyses. Based on claw mark signs,
sun bear habitat use appeared to be non-random and was significantly associated with gradients
of increasing habitat intactness, from non-forest habitat to least disturbed forest. Two PCA
factors explained the probability of bears utilising a given habitat, which increased with tree
biomass and decreased with understorey cover. At the plot level, tree family and species
compositions were significantly different between plots without and with claw marks. The
abundance and use of Olacaceae stems was significantly higher in plots with claw marks.
Incorporating forest restoration strategies that enhance or increase more intact forest and the
availability of key tree resources should benefit the conservation of sun bears and encourage
natural forest regeneration in these degraded landscapes. We also emphasise the conservation
* Corresponding author. E-mail address: [email protected] (D. Lee).
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
3 Lee et al.: Habitat utilisation by sun bears
value of degraded forest habitats for this species while ensuring bear movement and connectivity
within modified landscape matrices.
Keywords: Conservation ecology; forest structure; sun bear; tree community composition;
tropical forest restoration.
1. Introduction
Deforestation rates in Sumatra are among the highest in Southeast Asia (Miettinen et al.,
2011). Land conversion has resulted in the loss of nearly half of natural forest cover in the last
25-30 years: in southern Sumatra, 53-69% of forest cover has been lost since 1985 (Uryu et al.,
2010). While large-scale forest loss has generally been prevented within protected areas in
Sumatra (Gaveau et al., 2009), few areas of lowland forest remain outside of these, while large
areas of degraded forest within logging concessions are at risk of permanent conversion to
alternative land-use (Gaveau et al., 2012). Consequently, there is a need to understand what
conservation value these degraded areas may have for key components of forest biodiversity.
The Malayan sun bear (Helarctos malayanus) is a forest-dependent species threatened by
habitat degradation, fragmentation and loss, and poaching (Scotson et al., 2017). It is listed as
Vulnerable by the IUCN due to suspected population declines resulting from these threats
throughout its range in Southeast Asia (Scotson et al., 2017). While sun bears are reported in
secondary and logged forests (Wong, 2002; Wong et al., 2004; Fredriksson, 2005; Meijaard et al.,
2005; Linkie et al., 2007; Lindsell et al., 2013; Sethy and Chauhan, 2016), in the absence of
adjacent forest, it is uncertain how sustainable sun bear populations are in degraded habitats
(Augeri, 2005; Fredriksson, 2012).
* Corresponding author. E-mail address: [email protected] (D. Lee).
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
4 Lee et al.: Habitat utilisation by sun bears
Food availability and diversity influence sun bear ranging behaviour and habitat use,
particularly in relation to fruiting events (McConkey and Galetti, 1999; Wong et al., 2004;
Augeri, 2005; Fredriksson et al., 2006; Fredriksson, 2012). Although sun bears have a varied diet,
which enables them to utilise a range of forest habitats (Linkie et al., 2007), they feed primarily
on insects, honey and fruits (Wong et al., 2002; Augeri, 2005; Fredriksson et al., 2006). In terms
of natural forest regeneration and active restoration, they are considered important seed dispersers
(Leighton, 1990; McConkey and Galetti, 1999). Despite this, effective conservation management
is limited by a lack of ecological information on spatial distribution and habitat preferences
(Servheen, 1999; Wong and Linkie, 2013), particularly when increasing landscape degradation
and fragmentation drive a growing need to understand how sun bears use human-altered forests
and how these can be managed to enhance conservation efforts (Wong and Linkie, 2013).
Indirect signs of sun bears, including scat, footprints, ripped open tree trunks, claw marks,
dug soil and damaged termite nests (Fredriksson, 2012), are useful for confirming and monitoring
the presence of this difficult to observe species (Ngoprasert et al., 2011), and for estimating
relative abundance, habitat use and requirements (Steinmetz and Garshelis, 2010; Fredriksson,
2012). Bear claw marks on trees are particularly conspicuous (Duckworth et al., 1999; Hwang et
al., 2002; Ngoprasert et al., 2011; Steinmetz et al., 2011) and tend to have comparatively slow
decay rates when compared to other types of bear signs (Steinmetz and Garshelis, 2010;
Fredriksson, 2012). These may result from foraging for fruits (Wong et al., 2002; Augeri, 2005;
Fredriksson, 2012) or insects (Payne et al., 1985; Augeri, 2005; Steinmetz, 2011), territory
marking (Augeri, 2005; Fredriksson, 2012), resting (Wong et al., 2002, 2004; Fredriksson, 2012),
or escaping danger (Payne et al., 1985; Lim, 1998; Yasuma and Andau, 2000). Thus, claw marks
can provide evidence of sun bear feeding behaviour (Wong et al., 2002) and habitat selection
(Ngoprasert et al., 2011), whereas other indirect signs, e.g. scat piles and foot prints, may simply
* Corresponding author. E-mail address: [email protected] (D. Lee).
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
5 Lee et al.: Habitat utilisation by sun bears
indicate movement through an area (Augeri, 2005). It is important to note, though, that while
claw marks may be the most prevalent type of sign left by sun bears (Ngoprasert et al., 2011;
Steinmetz, 2011), in Borneo at least, foraging signs for termites appear more abundant, with claw
marks comprising 21-47% of detected signs, depending on habitat disturbance and post-
disturbance recovery time (Fredriksson, 2012).
In this study, we use claw marks and associated signs, specifically ripped open tree trunks, as
an indicator of potential habitat utilisation by sun bears. Specifically, we assess sun bear presence
at the landscape scale using habitat structural attributes, and identify important tree resources for
bears at the plot scale in degraded forest. This will inform and enhance forest restoration
strategies that support the conservation of sun bear in the largest remaining tract of lowland
dryland forest in Sumatra. These approaches and findings may be applicable for conservation
management of degraded landscapes elsewhere within the species’ range.
2. Methods
2.1. Study area and site management
This study was conducted in Harapan Rainforest (HRF), which is situated in the dry
lowlands of southern Sumatra, Indonesia (103°17'49" E, -02°12'94" S). HRF covers 984.6 km2 of
previously logged forest characterised by a largely flat topography and ranging in elevation from
30 to 120 m a.s.l.. Mean annual rainfall is 2,390 mm, with a pronounced dry season from June to
August and highest rainfall in December.
Up to 2006, the site was managed commercially as two production forest concessions for 20-
30 years, with most areas having gone through two logging cycles. These activities have left a
mosaic of largely secondary forest habitats (totalling 933.3 km2; Schweter, 2009; Fig. 1) with the
* Corresponding author. E-mail address: [email protected] (D. Lee).
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
6 Lee et al.: Habitat utilisation by sun bears
remainder comprising oil palm, rubber and old Acacia mangium plantations, agricultural land and
scrub. Forest habitats are categorised as ‘high cover secondary forest’ (HSF; covering 37.0% of
the site), ‘medium cover secondary forest’ (MSF; 15.2%), and ‘low cover secondary forest’
(LSF; 42.5%; Schweter, 2009). The least disturbed HSF is characterised by a closed canopy with
a mixed tree species composition lacking in Dipterocarpaceae, but otherwise floristically
characteristic of dry lowland tropical forest in the region. The most degraded LSF typically has
an open canopy largely dominated by pioneer species, including Macaranga spp. and the
invasive non-native Bellucia pentamera, and a dense herbaceous understorey of Marantaceae and
Zingerberaceae (Lee and Lindsell, 2011; Briggs et al., 2012).
Since 2007, the site has been managed as an ecosystem restoration concession, and under
license decreed in response to continual large-scale national forest loss, and a need to protect and
restore degraded forests (Sheil and Meijaard, 2010; Harrison, 2011). This study supported the
management objectives of identifying the habitat requirements of key forest species, particularly
large seed dispersers, and using this empirical information to guide restoration strategies that
support effective species conservation alongside other ecosystem co-benefits.
* Corresponding author. E-mail address: [email protected] (D. Lee).
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
7 Lee et al.: Habitat utilisation by sun bears
Fig. 1. Land-cover map of the study site in 2009. ‘Non-forest habitats’ comprise small
monoculture plantations, agricultural land and scrub. ‘Plantation’ areas consist of overgrown
Acacia mangium (Schweter, 2009).
2.2. Landscape-scale habitat surveys
Between October 2009 and April 2010, we completed habitat surveys in 262 plots of 25 m
radius (0.2 ha). Plots were positioned every 200 m along parallel transects spaced 1 km apart
using a site-wide systematic sampling design (Lee and Lindsell, 2011). In total, we surveyed 24
transects of 1.8 to 2.4 km length, covering approximately 9% of the site, and proportionally
representing the main habitat types. We categorised plots in the field as HSF, MSF, LSF or ‘non-
forest habitat’ (NFH, which included scrub and cleared areas).
We estimated or counted a number of habitat attributes in each plot quarter:
Percentage cover of the herbaceous layer using a randomly positioned 1 m2 quadrat;
* Corresponding author. E-mail address: [email protected] (D. Lee).
137
138
139
140
141
142
143
144
145
146
147
148
149
150
8 Lee et al.: Habitat utilisation by sun bears
Understorey cover using a vertical 1 m2 quadrat of 25 points spaced 20 cm apart
(adapted from Bullock, 1996) held at 1.3 m height and 12.5 m from the central point of
the plot;
Canopy openness using a canopy-scope (Brown et al., 2000) and
Numbers of rattans and saplings (<5 cm dbh), and climbers (0, 1, 2-5, and >5 plants).
We recorded the diameter at breast height (dbh; at 1.3 m height), and tree and canopy heights
for the ten largest trees in each plot. We used these metrics to calculate the density (ha-1), basal
area (m2 ha-1), and a measure of above ground biomass (m3 ha-1) of large trees. We checked all
trunk surfaces for bear claw marks and associated signs, including ripped open tree trunks
resulting from foraging for honey (Fredriksson, 2012). Claw mark age, season and tree species
are unlikely to have influenced our assessment of habitat use: ‘old’ signs would represent bear
activity 1-2 years after commercial logging had ceased in the site in 2006; while season and tree
species only influence claw mark decay rates by <1 month (Steinmetz and Garshelis, 2010). We
did not include scat piles or footprints, as they do not necessarily reflect direct dependence upon
the immediate habitat (Augeri, 2005). We also excluded terrestrial feeding signs, as they can be
confused with other mammals (Wong et al., 2002; Fredriksson, 2012) and, compared to claw
marks, can quickly decay beyond accurate identification (Fredriksson, 2012).
2.3. Tree inventory plots
In November-December 2010, we randomly selected 24 forest plots (hereafter referred to as
“bear plots”) across two focal areas (covering 5.3 and 7.0 km2) in which we recorded bear claw
marks during the landscape-scale habitat surveys (see Section 2.2.). We paired each bear plot
with a control plot (n = 24) containing no claw-marked trees or other obvious signs of bear
habitat use. We positioned control plots 100 m away from each bear plot in a random direction.
* Corresponding author. E-mail address: [email protected] (D. Lee).
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
9 Lee et al.: Habitat utilisation by sun bears
We searched each control plot thoroughly to confirm with absolute certainty the absence of claw-
marked trees and associated signs. If we found that a control plot contained claw-marked trees,
we randomly selected a different direction and repeated the process until we found a control plot
location with no bear claw marks present.
We tagged all trees >20 cm dbh, which is a general minimum tree size climbed by sun bears
(Augeri, 2005), in each 50 m x 50 m (0.25 ha) tree inventory plot. We identified each tagged tree
to at least genus level, and often to species level with botanical support from the Indonesian
Institute of Sciences (Lembaga Ilmu Pengetahuan Indonesia, LIPI). Tree taxonomy follows that
of The Plant List (2013).
2.4. Landscape-scale habitat analysis
We explored the habitat survey data for outliers, collinearity and missing values (Zuur et al.,
2010), and to test the normality of distributions (Kolmogorov-Smirnov test) and homogeneity of
variances across groups (Levene's test). Where appropriate, we removed extreme outliers and
transformed those variables significantly different from a normal distribution (P < 0.05) to
improve uni- and multivariate normality, and the linearity of any associations between variables.
We used a Principal Components Analysis (PCA) to reduce data dimensionality from the 12
habitat variables into a set of derived factors, and to correct for multicolinearity between the
explanatory habitat variables. We only considered factors with eigenvalues of >1.0 in the final
analysis. We used a Discriminant Analysis (DA) to refine the original habitat classifications of
the plots made in the field. For this, we entered together the PCA factor scores as predictor
variables and grouped plots by the habitat type categories assigned during the habitat surveys.
The predictive value of the DA was cross-validated using a leave-one-out method of k-fold
partitioning. Since the original (subjective) habitat classifications of plots did not necessarily
* Corresponding author. E-mail address: [email protected] (D. Lee).
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
10 Lee et al.: Habitat utilisation by sun bears
confer correct classification, we reclassified plots using the predicted habitat group memberships
from the DA. We modelled the presence/absence of bear claw marks in habitat plots as a
binomial response against the retained PCA factors (predictor variables) using binary logistic
regression, with model fit assessed using a Hosmer-Lemeshow goodness-of-fit test (Hosmer and
Lemeshow, 2000). To test univariate hypotheses, we used two-sample tests (for differences
between habitat plots with and without bear signs), and four-way tests of difference (ANOVA;
between habitats) and association (preferential habitat selection/avoidance across habitats). All
analyses were carried out using SPSS v24.0 (IBM Corp., 2016).
2.5. Tree community analysis
We analysed tree community data in the paired control and bear plots at species, genus and
family levels in PRIMER v6.0 (Clarke and Gorley, 2006). The techniques and their application in
PRIMER are described fully in Clarke (1993), and Clarke and Warwick (2001). For bear and
control plots, we calculated diversity (logeH') indices, and estimated predicted species richness
from accumulation plots with 999 permutations using a Jack-knife 2 non-parametric richness
estimator (Colwell and Coddington, 1994). We used two-sample tests to investigate univariate
hypotheses (tree diversity and taxonomic abundances between bear and control plots). Mean
values are reported with S.E., unless otherwise stated.
We factorised each plot in the community data as either a bear or control plot. We pre-treated
tree abundances with a square root transformation to down-weight the importance of the most
abundant tree species (Clarke and Warwick, 2001). We constructed similarity matrices at each
taxonomic level using the Bray-Curtis coefficient on the transformed tree abundance data. We
performed analyses of similarity (ANOSIM), which tested permutations of the similarity
matrices, to determine whether there were any spatial differences between tree communities of
* Corresponding author. E-mail address: [email protected] (D. Lee).
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
11 Lee et al.: Habitat utilisation by sun bears
bear and control plots. If an ANOSIM was significant, we carried out a similarity of percentages
analysis (SIMPER) to identify which trees contributed most to any differences in tree
communities between the two plot types.
3. Results
3.1. Habitat classification
Four factors extracted using PCA accounted for 70.0% of the original variation in habitat
data (Table 1). Habitat plots with high scores along Factor 1 (F1; 32.9% of the original variability
in habitat data) were characteristic of high biomass forest with a high density of large and tall
trees. Factor 2 (F2; 16.0%) described the relationship between numbers of understorey stems and
the presence of large trees. High scores represented plots dominated by a high density of
comparatively short trees with an open understorey of few saplings and rattans. These plots were
indicative of previously logged forest where a few large trees remained standing. Factor 3 (F3;
12.8%) described forest increasing in structural complexity and density, being dominated by high
numbers of saplings and climbers, and with little herbaceous cover. Factor 4 (F4; 8.3%) described
increasingly sparse ground cover and fewer rattan stems, and an associated increase in
understorey cover as the density of any larger trees decreased.
Table 1. Extraction results of PCA of the habitat variables (Rotation method: varimax with Kaiser
normalisation). Only factor loadings (measure of the correlation between the variable and the
factor) > 0.2 are displayed.
Variable Factor 1 Factor 2 Factor 3 Factor 4
* Corresponding author. E-mail address: [email protected] (D. Lee).
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
12 Lee et al.: Habitat utilisation by sun bears
Eigenvalue 3.95 1.92 1.54 1.01
Percentage variance 32.9 16.0 12.8 8.3
Large tree above ground biomass +0.465
Basal area of large trees +0.435 +0.225 +0.200
Mean tree height +0.363 -0.202 -0.425
Density of large trees +0.356 +0.255 +0.319 -0.264
Mean dbh +0.353 -0.319
Mean canopy height +0.352 -0.267 -0.356
Canopy openness +0.381 -0.231
Understorey cover +0.344 +0.540
Herbaceous ground cover +0.200 -0.348 -0.601
Number of rattans -0.422 -0.425
Number of saplings -0.517 +0.246
Number of climbers +0.423
After cross-validation, DA correctly classified 77.5% (n = 203) of the original habitat
classifications using the four PCA factors as predictor variables (Table 2). No habitat plots were
misclassified by more than one habitat category (disturbance or intactness level). We reclassified
the remaining 59 plots using the predicted habitat group memberships from the DA. This resulted
in 22 plots classified as NFH, 67 as LSF, 99 as MSF, and 74 as HSF.
Table 2. Summary matrix of cross-validated DA classification results. Percentages of cases are in
parentheses.
Original
membership
Originally
assigned
habitat
type
Predicted group membership
TotalNFH LSF MSF HSF
* Corresponding author. E-mail address: [email protected] (D. Lee).
246
247
248
249
250
251
252
253
254
255
13 Lee et al.: Habitat utilisation by sun bears
Group
NFH 14 (73.7) 5 (26.3) 0 0 19
LSF 8 (10.9) 51 (69.9) 14 (19.2) 0 73
MSF 0 11 (10.5) 79 (75.2) 15 (14.3) 105
HSF 0 0 6 (9.2) 59 (90.8) 65
Total 22 (8.4) 67 (25.6) 99 (37.8) 74 (28.2) 262
There were significant differences in scores of all four PCA factors across habitat types
(One-way ANOVA: F(3, 251) F1 = 130.51; F(3, 255) F2 = 38.422; F(3, 256) F3 = 10.152; F(3, 256) F4 = 15.633;
all at P < 0.001; Fig. 2). Increasing habitat intactness was characterised by significantly higher F1
and F2 scores compared to areas that were more disturbed. LSF had significantly higher F3
scores than HSF, while MSF had significantly lower F4 scores than all other habitat types
(Games Howell post-hoc test, P < 0.02).
* Corresponding author. E-mail address: [email protected] (D. Lee).
256
257
258
259
260
261
262
263
14 Lee et al.: Habitat utilisation by sun bears
HabitatHSFMSFLSFNFH
Fact
or sc
ore
4.5
3.0
1.5
0.0
-1.5
-3.0
-4.5
-6.0
Factor 4Factor 3Factor 2Factor 1
Fig. 2. Boxplots of factor scores across the four habitat types, as reclassified by DA.
3.2. Habitat use at the landscape-scale
We recorded sun bear claw marks in 69 (26.3%) of the 262 plots surveyed. Based on the DA
habitat classification derived from the PCA factors, 37.8% (n = 28) of HSF plots, 24.2% (n = 24)
of MSF plots, and 25.4% (n = 17) of LSF plots contained claw marks (Fig. 3). Habitat plots with
bear claw marks had significantly higher F1 scores and lower F4 scores than those plots without
(t-test: t258, F1 = 3.341, P = 0.001; t258, F4 = 1.991, P = 0.048). F2 and F3 did not explain any
differences between habitat plots with or without claw marks.
* Corresponding author. E-mail address: [email protected] (D. Lee).
264
265
266
267
268
269
270
271
272
273
274
15 Lee et al.: Habitat utilisation by sun bears
Factor 1 scores7.56.04.53.01.5.0-1.5-3.0-4.5-6.0
Fact
or 2
scor
e s
3.0
2.0
1.0
.0
-1.0
-2.0
-3.0
-4.0
-5.0
HSF (+)HSF (-)MSF (+)MSF (-)LSF (+)LSF (-)NFH (+)NFH (-)Habitat
Fig. 3. PCA ordination of DA habitat plot classifications based on their F1 and F2 scores. The
presence (+) or absence (-) of claw marks are included for each habitat plot.
Sun bear habitat use, based on the presence of claw marks, was significantly associated with
broad habitat type (X23 = 13.810, P = 0.003). Using percentage deviations, the observed
frequency of HSF plots with claw marks present was 43.7% greater than expected, with an
absence of claw marks 15.6% less than expected, suggesting preferential selection of this habitat.
Sun bears appeared to disproportionality avoid NFH, with no claw marks present in any NFH
plots (an absence of claw marks had an observed count 35.8% more than expected). Percentage
deviations for LSF and MSF did not vary by more than ± 7%. Phi and Kendall’s tau-b correlation
coefficients were similar (Phi = 0.230, P = 0.003; tau-b = 0.174, P = 0.001), suggesting an
ordinal correlation between habitat type and the presence of claw marks.
As predictors of bear habitat use, the PCA factors fitted the data adequately in the logistic
* Corresponding author. E-mail address: [email protected] (D. Lee).
275
276
277
278
279
280
281
282
283
284
285
286
287
288
16 Lee et al.: Habitat utilisation by sun bears
regression model (Hosmer-Lemeshow test: Ĉ8 = 9.834, P = 0.277). Two of the four factors had a
significant effect on the probability of bears utilising the habitat. F1 had a positive effect (ß =
+0.263, Z = 3.29, P = 0.001) and an odds ratio of 1.30 (1.11-1.52 95% CIs): for a one-unit
increase in F1, there was a 30% increase in the likelihood of bears utilising the habitat (Fig. 4).
Based on F1 scores, the mean probability of bears using NFH was 0.047 compared to 0.410 in
HSF. F4 had a negative effect (ß = -0.267, Z = -2.08, P = 0.037) with an odds ratio of 0.77 (0.60-
0.98 95% CIs): a one-unit increase in F4 resulted in a 23% decrease in the odds of bears using
that habitat. F2 and F3 did not influence the probability of bears utilising the habitat (95% CIs of
their odds ratios included one).
Factor 1 scores7.56.04.53.01.5.0-1.5-3.0-4.5-6.0
P (S
un b
ear h
a bita
t use
)
0.6
0.5
0.4
0.3
0.2
0.1
0.0
HSFMSFLSFNFH
Habitat
Fig. 4. Predicted probability of sun bear habitat use in relation to habitat F1, with regression line
fitted.
* Corresponding author. E-mail address: [email protected] (D. Lee).
289
290
291
292
293
294
295
296
297
298
299
300
301
17 Lee et al.: Habitat utilisation by sun bears
3.3. Tree community composition and habitat use
Seven plots (four bear plots, three control plots) were illegally clear-felled during the process
of measuring and identifying tree species so were removed from our analyses. In total, 452 trees
were tagged in the remaining 41 plots (16–64 trees ha-1; mean = 44.0 ± 8.36 ha-1).
Of the 600 tree species from 107 families recorded in the site (Harrison and Swinfield,
2015), 33 tree families comprising 71 genera and 83 species were identified in the plots. Of these
families, five (Apocynaceae, Celastraceae, Centroplacaceae, Oleaceae and Symplocaceae) were
only recorded in control plots, while Cardiopteridaceae was the only family unique to bear plots.
Of the 60 tree species recorded in control plots, 18 (30.0%) were unique to these areas, while 23
(35.4%) of the 65 tree species recorded in bear plots were unique to these plots. Predicted tree
species richness was higher in control plots (SPredicted = 111) than bear plots (SPredicted = 96), and
while tree diversity was lower in control plots (HControl = 1.77 ± 0.412 SD) than in bear plots (HBear
= 1.89 ± 0.549 SD), it was not significantly different across plot types (Paired t-test: t39 = 0.774,
P = 0.444).
We recorded claw marks on 11.7% (n = 53) of trees in bear plots, and representing 51.5% of
families (n = 17) and 28.8% of species (n = 23) recorded. Claw marks were significantly
associated with stems of the family Olacaceae (37.7% of claw-marked stems; Chi-square test for
association: χ21 = 76.673, P < 0.001; Fig. 5). Of the other families, 15.1% (n = 8) of claw-marked
stems were Leguminosae, and 5.7% (n = 3) each from Clusiaceae, Euphorbiaceae and Lauraceae.
For those tree species with >10 stems recorded in plots, the most commonly utilised by bears
were Ochanostachys amentacea (Olacaceae; claw marks were recorded on 74.1% (n = 20) of
stems in this family, and 38.8% of total stems with claw marks) and Falcataria moluccana
(Leguminosae; 37.5% (n = 6) of stems had claw marks).
* Corresponding author. E-mail address: [email protected] (D. Lee).
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
18 Lee et al.: Habitat utilisation by sun bears
Fig. 5. Percentage of total tree stems (n = 452) with and without bear claw marks in the key tree
families. Total counts of stems and dominant genera (>20% of stems by family) are attached to
the bars. Chi-square test results are based on the number of stems of all other families in habitat
plots with and without claw marks; * P < 0.001.
Tree community compositions of bear and control plots were significantly different at the
family and species levels (ANOSIM: RFamily = 0.065, P = 0.049; RSpecies = 0.066, P = 0.048), but
not at the genus level (ANOSIM: RGenus = 0.050, P = 0.080). At the family level, tree communities
of control plots had an average similarity of 26.4%, while bear plots were 29.8% similar. Six of
the 33 families contributed >40% of the 73.6% dissimilarity between the tree family composition
* Corresponding author. E-mail address: [email protected] (D. Lee).
326
327
328
329
330
331
332
333
334
335
336
337
19 Lee et al.: Habitat utilisation by sun bears
of control and bear plots (Table 3). At the species level, tree communities of control plots had an
average similarity of 14.1% and bear plots were 16.1% similar to each other. Macaranga
gigantea (Euphorbiaceae), O. amentacea (Olacaceae) and Callicarpa pentandra (Lamiaceae)
contributed most to the 86.3% dissimilarity in tree species between control and bear plots (Table
3). Of these, the density of Olacaceae stems, represented solely by O. amentacea, was
significantly lower in control plots than bear plots (Mann-Whitney: U39 = 77.50, P < 0.001).
Table 3. Key tree families and species contributing to differences in the community composition
of bear and control plots (SIMPER). Only families and species contributing >5% to the overall
dissimilarity are presented. Cumulative percentage contributions are in parentheses. * P < 0.05;
significant difference in abundance across plot types.
Taxonomic level Mean abundance (± SE) Average
dissimilarity
(± SD)
%
Contributio
n
Control plots Bear plots
Family:
Euphorbiaceae 1.6 ± 0.39 1.5 ± 0.42 5.7 ± 1.16 7.7
Leguminosae 0.7 ± 0.21 1.4 ± 0.76 5.4 ± 0.75 7.4 (15.1)
Lamiaceae 1.3 ± 0.45 0.4 ± 0.27 5.1 ± 0.82 7.0 (22.1)
Olacaceae * 0.1 ± 0.05 1.1 ± 0.23 5.0 ± 1.30 6.8 (28.9)
Dipterocarpaceae 0.7 ± 0.25 0.9 ± 0.26 4.4 ± 1.01 6.0 (34.9)
Burseraceae 0.7 ± 0.19 0.9 ± 0.23 4.3 ± 1.12 5.9 (40.8)
Species:
Macaranga gigantea 1.1 ± 0.41 1.0 ± 0.43 5.2 ± 0.93 6.0
Ochanostachys amentacea * 0.1 ± 0.05 1.1 ± 0.23 4.8 ± 1.30 5.5 (11.5)
Callicarpa pentandra 1.3 ± 0.45 0.4 ± 0.27 4.7 ± 0.77 5.5 (17.0)
* Corresponding author. E-mail address: [email protected] (D. Lee).
338
339
340
341
342
343
344
345
346
347
348
349
350
20 Lee et al.: Habitat utilisation by sun bears
4. Discussion and conclusions
4.1 Landscape-scale habitat use
Sun bear claw marks were recorded in 26.3% of habitat plots, which represented all three
forest types, indicating widespread use of these disturbed habitats (Wong, 2002; Wong et al.,
2004; Fredriksson, 2005; Meijaard et al., 2005; Linkie et al., 2007; Lindsell et al., 2015). Based
on the presence of claw marks and associated signs on trees, evidence of sun bear habitat use
increased with forest intactness or recovery stage post-logging, with bears appearing to select
preferentially the least disturbed forest areas (HSF), and with broadly neutral selection of low and
medium secondary forest types. This reflects those areas that were disturbed least during logging
and/or have had more time to recover to a more mature forest structure after logging (Sethy and
Chauhan, 2016). Claw marks were not recorded in non-forest habitat. While these areas did have
fewer trees than forested habitat, they retained approximately 46% of the large tree density
recorded in the more degraded forest. If sun bears selected these areas similarly to the forested
landscape, we would have expected 5-7 of the 22 non-forest plots to include claw marks. It
appears that sun bears disproportionately avoided utilizing these most heavily impacted habitats,
despite contiguity with forested land (Augeri, 2005).
The significant effects of habitat F1 and F4, positive and negative respectively, on the
probability of bears utilising less degraded forest support this broad pattern in habitat use or
avoidance. High F1 scores were associated with characteristics of least degraded forest (taller,
more closed canopy, higher density of large trees and saplings, greater prevalence of climbers
less herbaceous ground cover), while higher F4 scores reflected habitat with low densities of
large trees, low ground cover, dense herbaceous understory and few rattans, most indicative of
the non-forest habitat. This is congruous with other studies, which identified greater habitat use
* Corresponding author. E-mail address: [email protected] (D. Lee).
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
21 Lee et al.: Habitat utilisation by sun bears
by bears in older forest, and specifically related to higher canopy, escape and ground covers
(Augeri, 2003; Sethy and Chauhan, 2016), and a low probability of detecting bears in young
disturbed forest (<5 years old; Augeri, 2005), which could be comparable to areas of LSF in our
study site. An absence of bear habitat use in areas with very few rattans (Arecaceae), high F4
scores, may also reflect the dietary importance of their fruit (Fredriksson et al., 2006).
Increasing habitat use by sun bears along a post-logging gradient of forest recovery may be
due to greater availability of key resources (Steinmetz et al., 2011; Sethy and Chauhan, 2016),
with evidence of bear habitat use strongly linked to fruit availability (Wong et al., 2002; Linkie et
al., 2007; Steinmetz et al., 2011; Wong and Linkie, 2013). More intact forest should support
higher densities and productivity of fruiting trees than more disturbed areas (Wong et al., 2012),
providing greater fruit availability and resulting in higher detection probabilities of sun bears
(Steinmetz et al., 2011; Wong et al., 2012). From a species conservation perspective at the global
scale, the impacts of logging on sun bear habitat use may be greater in Borneo than in Sumatra,
where fruit production is higher (Wich et al., 2011). Sun bears are also dependent on large trees
of certain species, e.g. Shorea spp., O. amentacea, for resting or sleeping (Wong et al., 2002;
Meijaard et al., 2005; Padmanaba et al., 2013; see Section 4.2). These are increasingly sparse in
the more degraded forest of the site through more recent or intensive logging activities (Lee and
Lindsell, 2011; Harrison and Swinfield, 2015). Complementary data on sun bear relative
abundance support the spatial relationship between the likelihood of habitat use and gradient of
forest intactness in the site (Lindsell et al., 2015).
4.2 Plot-scale habitat use
Tree community composition was significantly difference between areas where there was no
evidence of bear habitat use and those with clear evidence of habitat use. These differences were
* Corresponding author. E-mail address: [email protected] (D. Lee).
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
22 Lee et al.: Habitat utilisation by sun bears
driven mainly by trees of Euphorbiaceae, Leguminosae, Lamiaceae, Olacaceae, Dipterocarpaceae
and Burseraceae, and the species M. gigantea, O. amentacea and C. pentandra.
Sun bears most frequently climb, for multiple reasons, large trees of Anacardiaceae,
Bignoniaceae, Burseraceae, Combretaceae, Dilleniaceae, Dipterocarpaceae (e.g. Shorea),
Euphorbiaceae, Fagaceae (e.g. Lithocarpus), Labiatae, Lauraceae, Leguminosae, Moraceae,
Myrtaceae (e.g. Eugenia, Syzygium), Sapindaceae and Tiliaceae (Wong et al., 2002; Augeri,
2003, 2005; Steinmetz, 2011). Where present in the study plots, we recorded claw marks on trees
of all these families, except Burseraceae and Sapindaceae.
We found that sun bears in this previously logged, dry lowland forest preferentially select
stems of Fabaceae and Olacaceae. Olacaceae was represented by a single species, O. amentacea,
an animal-dispersed species that is known to fruit throughout the year (World Agroforestry
Centre, 2018), and climbed repeatedly in HRF. Like the Leguminosae genera Dialium and
Koompassia, O. amentacea is also more abundant in less disturbed forest in HRF, indicative of
its late successional condition (Swinfield et al., 2016). There are no records of sun bears feeding
on the fruits of Olacaceae (Fredriksson, 2012), but O. amentacea is a preferred nesting tree
(Padmanaba et al., 2013). Coupled with a naturally scattered distribution (World Agroforestry
Centre, 2018), it appears the importance of O. amentacea for sun bears is related to its nesting
suitability, especially in logged forest with a reduction in large dipterocarps, such as Shorea spp.
which are often selected for nesting sites (Wong et al., 2002).
We recorded claw marks on 15.4% of Moraceae stems (Artocarpus spp., Ficus spp.) and
11.8% of Myrtaceae stems (Syzygium spp.). These families, along with Burseraceae, provide the
majority of fruit in the bear’s diet (McConkey and Galetti, 1999; Wong et al., 2002; Augeri,
2003, 2005; Fredriksson et al., 2006; Steinmetz, 2011), with Ficus species of particular important
(Leighton, 1990; McConkey and Galetti, 1999; Wong et al., 2002; Fredriksson et al., 2006). The
* Corresponding author. E-mail address: [email protected] (D. Lee).
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
23 Lee et al.: Habitat utilisation by sun bears
frequency bears used these stems in this logged forest is commensurate with their recorded use
elsewhere (Wong et al., 2002; Fredriksson et al., 2006), although there appears to be no
preferential selection of these stems.
Of other important tree families, no Burseraceae stems (Dacryodes, Santiria, Canarium spp.)
had claw marks on them. The fact that Dacryodes and Canarium spp. are dioecious (Fern, 2016),
and characteristic of inter-annual phenological variation in the region (Thomas and LaFrankie,
1993), may explain an absence of bear usage in logged forest, while bears do not necessarily
climb Santiria spp. to access the fruit (Wong et al., 2002). Euphorbiaceae appear under selected,
possibly due to higher proportions of pioneer Endospermum and Macaranga trees, and reflecting
a lower likelihood of bears being recorded in these areas.
4.3. Study considerations
While we present strong evidence for landscape and plot-scale differences in sun bear habitat
use in degraded forest based on claw marks, we do not consider possible temporal shifts in
habitat and resource use (Augeri, 2005; Fredriksson et al., 2006) or capture information on other
foraging strategies (Fredriksson et al., 2006) and their associated signs (Fredriksson, 2012). This
may have resulted in an incomplete understanding of habitat use by sun bears (Steinmetz, 2011;
Fredriksson, 2012) across a gradient of forest intactness. In more degraded forest and secondary
growth, where fruit availability is lower, sun bears may spend more time travelling (Wong et al.,
2012) and foraging terrestrially (Wong et al., 2002; Fredriksson, 2012), reducing claw mark
prevalence (Ngoprasert et al., 2011) and evidence of habitat use based on this type of sign. Also,
in the Bornean region of the species’ geographic range at least, claw marks may only represent
21% (to 47%) of signs left by sun bears, depending on habitat disturbance and recovery time,
with the majority of signs (51-71%) associated with foraging for above and belowground termites
* Corresponding author. E-mail address: [email protected] (D. Lee).
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
24 Lee et al.: Habitat utilisation by sun bears
(Fredriksson, 2012). There do appear to be some positive associations between the density of
termite nest signs and claw marks (Fredriksson, 2012), which may further support the use of claw
marks as an overall proxy for habitat use, especially when focusing on presence/absence at the
plot or landscape scale rather than sign density and relative abundance. Overall, our conclusions
on sun bear habitat use in degraded forest landscapes of Sumatra are couched in these limitations.
However, this does not diminish the conservation importance of less disturbed logged forest for
sun bears (Wong et al., 2002; Augeri, 2003; Linkie et al., 2007; Steinmetz et al., 2011; Wong and
Linkie, 2013; Sethy and Chauhan, 2016), and especially in a landscape where degraded forest is
vulnerable to land-use conversion (Gaveau et al., 2012).
4.4. Forest management and restoration implications
Our study supports the 20-year site management plan, helping inform restoration strategies
that benefit a species of conservation concern. It emphasizes the importance of less disturbed
forest habitats for sun bears and a lack of use of more heavily disturbed areas in logged forest. At
the site level, this provides strong evidence for protecting existing least disturbed forest cover to
allow natural regeneration to progress and, through collaboration with local partners and
stakeholders, prevent any further habitat degradation.
Indicators of restoration success should include changes in forest intactness attributes, as
described by F1, the strongest predictor of sun bear habitat use. Management approaches that
encourage structural restoration of NFH, with only a 4.7% average likelihood of bear habitat use,
to forest habitat will, on average, result in 19.7% (LSF), 32.4% (MSF) and 36.3% (HSF)
increases in habitat use by bears across the gradient of forest intactness described at HRF.
Improving LSF to MSF or HSF should increase the likelihood of habitat use by 12.7% and
16.5%, respectively. Marginal gains in bear habitat use are made if MSF returns to HSF (3.9%).
* Corresponding author. E-mail address: [email protected] (D. Lee).
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
25 Lee et al.: Habitat utilisation by sun bears
Consequently, perhaps the most cost and resource-effective approach for enhancing sun bear
habitat use is the restoration of habitat (NFH or LSF) to forest that is structurally similar to MSF.
The benefit to sun bears diminishes with further commitment of resources to actively restoring
MSF to HSF; at this point, natural regeneration becomes more valuable.
Structurally, in disturbed forest with average tree heights <18 m, canopy height <11 m, the
largest trees with girths of <100 cm, basal area <9.5 m2 / ha, and <7.5 rattan stems / ha, the
probability of bear use drops below 0.25. Conversely, the probability of bear use rises to >0.40 in
previously logged forest with average tree heights >26 m, canopy height >17 m, the largest trees
with girths of >115 cm, basal area >13 m2 / ha, and >15 rattan stems / ha. These forest condition
thresholds provide measurable targets against which restoration success can be monitored
alongside the likelihood of bears using the habitat, or against which the impact of any
degradation to bear habitat use can be evaluated (e.g. Augeri, 2004).
Landscape-scale management and restoration approaches that help push habitat quality
further along a post-logging gradient of recovery, e.g. reducing interior edge-area ratios, should
enhance sun bear access to more productive foraging areas, and increase the diversity and
abundance of key resources (Meijaard and Sheil, 2008). Complementary studies to determine
how different restoration strategies may improve fruit availability and distribution would be
beneficial. In turn, as seed dispersers (Leighton, 1990; McConkey and Galetti, 1999), sun bears
are potentially important contributors to natural forest regeneration, benefitting the recovery of
MSF to HSF.
Direct seeding or enrichment planting strategies (e.g. Harrison and Swinfield, 2015) should
consider including tree species preferentially utilised by sun bears in these forests, assuming that
mother trees can be located, and that those seeds/saplings can become established under the
conditions of different levels of forest degradation. In this instance, O. amentacea appears to be a
* Corresponding author. E-mail address: [email protected] (D. Lee).
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
26 Lee et al.: Habitat utilisation by sun bears
particularly important tree species for sun bears, and most likely for resting or nesting
(Padmanaba et al., 2013). While natural regeneration of this slow-growing species tends to be
sparse and scattered (Asian Regional Workshop, 1998), it can be seed-propagated and, assuming
successful germination, would probably take at least 40 years (World Agroforestry Centre, 2018)
to reach a minimum size for sun bears to climb (Augeri, 2005). Direct seeding or planting of
saplings could also help reduce herbaceous cover, enhancing growth of other late successional
species and reducing silviculture costs (World Agroforestry Centre, 2018). Other potentially
important tree species for direct seeding/enrichment planting include late successional species of
Dialium, Falcataria and Koompassia.
Silvicultural trials aimed at assisting natural regeneration (ANR) at HRF have selectively
removed from the understorey two dominant pioneer species, M. gigantea (Euphorbiaceae) and
the invasive, non-native B. pentamera (Melastomataceae), and lianas (Swinfield et al., 2016).
While only over a short time period, the slightly increased canopy openness and increased growth
of late successional stems (≥ 2 cm dbh) at lower intensity thinning (Swinfield et al., 2016) are
promising indicators of natural forest regeneration for sun bears. Such ANR should enhance
habitat suitability for sun bears as the forest understory becomes more open and with an increase
in growth of important late successional species, while M. gigantea is under-selected by bears.
Trialling the removal of pioneer Callicarpa species may also prove beneficial, as they appear to
provide no value to sun bears in this degraded forest.
Other site restoration strategies include planting experiments of Aquilaria malaccensis
(gaharu), a high-value non-timber forest product (Harrison and Swinfield, 2015). Sun bears are
sensitive to, and avoid minor disturbances such as gaharu harvesting (Augeri, 2004). While this
should have minimal impact on bears as it will replace unsuitable Acacia mangium plantation in
the site, it is important to consider how this may displace bears from the immediate area and
* Corresponding author. E-mail address: [email protected] (D. Lee).
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
27 Lee et al.: Habitat utilisation by sun bears
reduce landscape permeability.
We recommend the use of surveys of new sun bear claw marks to monitor changes in sun
bear distribution and occupancy within HRF, particularly alongside active site management. They
are low-cost and easy to conduct (Fredriksson, 2012), while in Sumatra at least and with observer
training, bear claw marks should not be confused with any sympatric tree-climbing species. It is
also worthwhile considering the inclusion of termite feeding signs, as a potentially prevalent and
spatially dependent indicator of habitat use (Fredriksson, 2012). This overall approach could be
used to help validate the most compatible and economically viable large-scale restoration
approach(es) (Harrison and Swinfield, 2015) for effective sun bear conservation management in
recovering forest landscapes. It also lends itself to a standardised approach for monitoring the
population and occupancy of the species throughout its range (Linkie et al. 2007; Steinmetz et al.
2011; Wong et al. 2012).
While sun bears are capable of utilising a range of habitat types, including oil palm
plantations (Servheen, 1999; Augeri, 2002; Normua et al., 2004) and farmland crops (Linkie et
al., 2007), some monoculture landscapes appear to be entirely unsuitable, at least without
proximate forested areas (McShea et al., 2009). If fruit availability is limiting, sun bears may be
forced to forage more beyond forest boundaries (Augeri, 2004). Consequently, meta-population
conservation management of sun bears in modified landscapes, such as southern Sumatra, must
consider movement of individuals through permeable anthropogenic matrices (McShea et al.,
2009) alongside forest management. In the case of HRF, as an exemplar of ecosystem restoration,
increasing site isolation may ultimately undermine any empirically informed forest restoration
strategies for sun bears and, indeed, other large forest mammals.
It is important to emphasise that in a landscape of heavily impacted forest cover (Gaveau et
al., 2009, 2012), areas such as HRF tend to be representative of available lowland forest in
* Corresponding author. E-mail address: [email protected] (D. Lee).
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
28 Lee et al.: Habitat utilisation by sun bears
Sumatra. Therefore, while these degraded forests may not provide optimal habitat for sun bears,
our study helps highlight what their adapted distribution is within the habitat available to them
(Wong and Linkie, 2013). Furthermore, degraded lowland forests potentially provide better
habitat quality for sun bears than undisturbed higher elevation forest (Linkie et al., 2007),
highlighting the importance of protecting and enhancing these threatened landscapes (Wong and
Linkie, 2013; Fredriksson, 2012).
Author contributions
DL and VP carried out the field research and collected the data; DL performed the statistical
analyses; DL, VP and JL wrote the paper.
Acknowledgments
Our research supports the forest conservation and restoration activities of HRF, a
collaborative initiative of the Royal Society for the Protection of Birds, BirdLife International
and Burung Indonesia. Funding was provided by a Research and Conservation Grant from the
International Association for Bear Research and Management and the Darwin Initiative of the
Department for Environment, Food and Rural Affairs (Reference no. 162/16/005). Yayasan
Konservasi Ekosistem Hutan Indonesia support of this work was co-financed by the Federal
Republic of Germany within the framework of the International Climate Protection Initiative of
the Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU)
through KfW Development Bank. We thank the research staff of HRF for assisting in the surveys
and Mr D. Girmansyah from the Indonesian Institute of Sciences (LIPI) for his botanical
expertise. We are also grateful to two anonymous reviewers for their valuable comments on this
manuscript.
* Corresponding author. E-mail address: [email protected] (D. Lee).
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
29 Lee et al.: Habitat utilisation by sun bears
References
Asian Regional Workshop, 1998. Ochanostachys amentacea. Conservation & Sustainable
Management of Trees, Vietnam, August 1996. The IUCN Red List of Threatened Species
1998: e.T33238A9770332.
<http://dx.doi.org/10.2305/IUCN.UK.1998.RLTS.T33238A9770332.en>. Downloaded on 17
June 2018.
Augeri, D.M., 2002. Effects on sun bear (Helarctos malayanus) habitat selection, ecology and
landscape use. Paper presented at the International Bear Association Annual Meeting 2002,
Steinkjer, Norway.
Augeri, D.M., 2003. Conservation of the Malayan Sun Bear (Helarctos malayanus) in Indonesia:
Mitigating potential bear/human conflicts and disturbance effects on sun bear ecology and
landscape use. Unpublished report for the Indonesian Institute of Sciences.
Augeri, D.M., 2004. Effects of disturbance on Malayan sun bear habitat use. Paper presented at
the International Conference on Conservation Science, Cambridge, UK.
Augeri, D.M., 2005. On the biogeographic ecology of the Malayan sun bear. PhD Dissertation,
University of Cambridge, Cambridge, UK.
Briggs, M., de Kok, R., Moat, J., Whaley, O., Williams, J., 2012. Vegetation mapping for
reforestation and carbon capture in the Harapan Rainforest. Royal Botanic Gardens, Kew,
London, UK.
Brown, N., Jennings, S., Wheeler, P., Nabe-Nielsen, J., 2000. An improved method for the rapid
assessment of forest under-storey light environments. Journal of Applied Ecology 37, 1044–
1053.
Bullock, J., 1996. Plants. In: Sutherland, W.J. (Ed.), Ecological census techniques – A handbook.
* Corresponding author. E-mail address: [email protected] (D. Lee).
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
30 Lee et al.: Habitat utilisation by sun bears
Pp. 186–213. Cambridge University Press, Cambridge, UK.
Clarke, K.R., 1993. Non-parametric multivariate analyses of changes in community structure.
Australian Journal of Ecology 18, 117–143.
Clarke, K.R., Gorley, R.N., 2006. Primer v6: User Manual/Tutorial. Primer-E, Plymouth, UK.
Clarke, K.R., Warwick, R.M., 2001. Change in marine communities: An approach to statistical
analysis and interpretation. Second edition. PRIMER-E, Plymouth, UK.
Colwell, R.K., Coddington, J.A., 1994. Estimating terrestrial biodiversity through extrapolation.
Philosophical Transactions of the Royal Society B: Biological Sciences 345, 101–118.
Duckworth, J.W., Salter, R.E., Khounboline, K., 1999. Wildlife in Lao PDR: the 1999 status
report. IUCN-The World Conservation Union/Wildlife Conservation Society/Centre for
Protected Areas and Watershed Management, Vientiane, Lao PDR.
Fern, K., 2016. Useful tropical plants database. < http://tropical.theferns.info/>. Last updated on
16 August 2016. Downloaded on 17 June 2018.
Fredriksson, G., 2005. Human-sun bear conflicts in East Kalimantan, Indonesian Borneo. Ursus
16, 130-137.
Fredriksson, G., Wich, S.A, Trisno, 2006. Frugivory in sun bears (Helarctos malayanus) is linked
to El Niño-related fluctuations in fruiting phenology, East Kalimantan, Indonesia. Biological
Journal of the Linnean Society 89, 489–508.
Fredriksson, G., 2012. Effects of El Niño and large-scale forest fires on the ecology and
conservation of Malayan sun bears (Helarctos malayanus) in East Kalimantan, Indonesian
Borneo. PhD Dissertation, University of Amsterdam, Amsterdam, Neth.
Gaveau, D.L.A., Epting, J., Lyne, O., Linkie, M., Kumara, I., Kanninen, M., Leader-Williams,
N., 2009. Evaluating whether protected areas reduce tropical deforestation in Sumatra.
Journal of Biogeography 36, 2165–2175.
* Corresponding author. E-mail address: [email protected] (D. Lee).
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
31 Lee et al.: Habitat utilisation by sun bears
Gaveau D.L.A., Curran, L.M., Paoli, G.D., Carlson, K.M., Wells, P., Besse-Rimba, A., Ratnasari,
D., Leader-Williams, N., 2012. Examining protected area effectiveness in Sumatra:
importance of regulations governing unprotected lands. Conservation Letters 5, 142–148.
Harrison, R.D. 2011. Tropical forests: still vital when degraded. Nature 479, 178–179.
Harrison, R.D., Swinfield, T. 2015. Restoration of logged humid tropical forests: an experimental
programme at Harapan Rainforest, Indonesia. Tropical Conservation Science 8, 4–16.
Hosmer, D.W., Lemeshow, S., 2000. Applied logistic regression. John Wiley & Sons, Inc., New
York, New York, USA.
Hwang, M.-S., Garshelis, D.L., Wang, Y., 2002. Diets of Asiatic black bears in Taiwan, with
methodological and geographical comparisons. Ursus 13, 111–125.
IBM Corp., 2016. SPSS Statistics for Windows, v24.0. IBM Corp., Armonk, NY.
Lee, D.C., Lindsell, J.A. 2011. Biodiversity of Harapan Rainforest: summary report on baseline
surveys of mammals, birds, fish, herptiles, butterflies and habitat. The Royal Society for the
Protection of Birds, Sandy, UK.
Leighton, M. 1990. Seed dispersal syndromes of Bornean fruits: The dependence of plants on
birds and mammals for seed dispersal and forest regeneration. A preliminary report on
research results from Gunung Palung. Center for Research and Development in Biology-LIPI
and the Sub-directorate of Nature Conservation (PHPA), Ministry of Forestry, Jakarta,
Indonesia.
Lindsell, J.A., Lee, D.C., Powell, V.J., Gemita, E., 2015. Availability of large seed-dispersers for
restoration of degraded tropical forest. Tropical Conservation Science 8, 17–27.
Linkie, M., Dinatab, Y., Nugrohob, A., Haidir, I.A., 2007. Estimating occupancy of a data
deficient mammalian species living in tropical rainforests: Sun bears in the Kerinci Seblat
region, Sumatra. Biological Conservation 137, 20–27.
* Corresponding author. E-mail address: [email protected] (D. Lee).
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
32 Lee et al.: Habitat utilisation by sun bears
Lim, B.L., 1998. The sun bear. In: Yong H.S. (Ed.), The encyclopaedia of Malaysia-Animals. Pp.
24–25. Archipelago Press, Kuala Lumpur, Malaysia.
McConkey, K., Galetti, M., 1999. Seed dispersal by the sun bear Helarctos malayanus in Central
Borneo. Journal of Tropical Ecology 15, 237–241.
McShea, W.J., Stewart, C., Peterson, L., Erb, P., Stuebing, R., Giman, B., 2009. The importance
of secondary forest blocks for terrestrial mammals within an Acacia/secondary forest matrix
in Sarawak, Malaysia. Biological Conservation 142, 3108–3119.
Meijaard, E., Sheil, D., 2008. The persistence and conservation of Borneo’s mammals in lowland
rain forests managed for timber: observations, overviews and opportunities. Ecological
Research 23, 21–34.
Meijaard, E., Sheil, D., Nasi, R., Augeri, D., Rosenbaum, B., Iskandar, D., Setyawati, T.,
Lammertink, M., Rachmatika, I., Wong, A., Soehartono, T., Stanley, S., O’Brien, T., 2005.
Life after logging: Reconciling wildlife conservation and production forestry in Indonesian
Borneo. The Center for International Forestry Research (CIFOR), Bogor, Indonesia.
Miettinen, J., Shi, C., Liew, S.C., 2011. Deforestation rates in insular Southeast Asia between
2000 and 2010. Global Change Biology 17, 2261–2270.
Ngoprasert, D., Steinmetz, R., Reed, D.H., Savini, T., Gale, G.A., 2011. Influence of fruit on
habitat selection of Asian bears in a tropical forest. Journal of Wildlife Management 75,
588–595.
Normua, F., Higashi, S., Ambu, L., Mohamad, M., 2004. Notes on oil palm plantation use and
seasonal spatial relationships of sun bears in Sabah, Malaysia. Ursus 15, 227-231.
Padmanaba, M., Sheil, D., Basuki, I., Liswanti, N., 2013. Accessing local knowledge to identify
where species of conservation concern occur in a tropical forest landscape. Environmental
Management 52, 348–359.
* Corresponding author. E-mail address: [email protected] (D. Lee).
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
33 Lee et al.: Habitat utilisation by sun bears
Payne, J., Francis, C.M., Phillipps, K., 1985. A field guide to the mammals of Borneo. World
Wildlife Fund Malaysia, Kuala Lumpur, Malaysia.
Schweter, M., 2009. Forest cover analysis – Harapan Rainforest, Sumatra, Indonesia. The Royal
Society for the Protection of Birds, Sandy, UK.
Scotson, L., Fredriksson, G., Augeri, D., Cheah, C., Ngoprasert, D., Wai-Ming, W. 2017.
Helarctos malayanus (errata version published in 2018). The IUCN Red List of Threatened
Species 2017: e.T9760A123798233. http://dx.doi.org/10.2305/IUCN.UK.2017-
3.RLTS.T9760A45033547.en. Downloaded on 10 December 2018.
Sethy, J., Chauhan, N.S., 2016. Assessing habitat use by sun bears in Namdapha Tiger Reserve,
Arunachal Pradesh, India. Applied Ecology and Environmental Research 14, 215–236.
Servheen, C., 1999. Sun bear conservation action plan. In: Herrero, S., Peyton, B., Servheen, C.,
(Eds.), Bears: status survey and conservation action plan. Pp. 219–224. IUCN-World
Conservation Union, Gland, Switzerland.
Sheil, D., Meijaard, E., 2010. Purity and prejudice: deluding ourselves about biodiversity
conservation. Biotropica 42, 566–568.
Steinmetz, R.G., Garshelis, D.L., 2010. Estimating ages of bear claw marks in Southeast Asian
tropical forests as an aid to population monitoring. Ursus 21, 143–153.
Steinmetz, R.G., 2011. Ecology and distribution of sympatric Asiatic black bears and sun bears in
the tropical dry forest ecosystem of Southeast Asia. In: McShea, W., Davies, S.,
Bhumpakphan, N. (Eds.), Dry forests of Asia: conservation and ecology. Pp. 249–273.
Smithsonian Institution Press, Washington, D.C.
Steinmetz, R.G., Garshelis, D.L., Chutipong, W., Seuaturien, N., 2011. The shared preference
niche of sympatric Asiatic black bears and sun bears in a tropical forest mosaic. PLoS ONE
6, e14509. doi:10.1371/journal.pone.0014509.
* Corresponding author. E-mail address: [email protected] (D. Lee).
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
34 Lee et al.: Habitat utilisation by sun bears
Swinfield, T., Afriandi, R., Antoni, F., Harrison, R.D., 2016. Accelerating tropical forest
restoration through the selective removal of pioneer species. Forest Ecology and
Management 381, 209–216.
The Plant List 2013. Version 1.1. Published on the internet; http://www.theplantlist.org/
(accessed 17th June 2018).
Thomas, S.C., LaFrankie, J.V., 1993. Sex, size and inter-year variation in flowering among
dioecious trees of the Malayan rain forest. Ecology 74, 1529–1537.
Uryu, Y., Purastuti, E., Laumonier, Y., Sunarto, Setiabudi, Budiman, A., Yulianto, K., Sudibyo,
A., Hadian, O., Kosasih, D.A., Stüwe, M., 2010. Sumatra’s forests, their wildlife and the
climate. Windows in time: 1985, 1990, 2000 and 2009. Technical Report to the National
Forestry Council (DKN) of Indonesia. WWF-Indonesia, Jakarta, Indonesia.
Wich, S.A., Vogel, E.R., Larsen, M.D., Fredriksson, G., Leighton, M., Yeager, C.P., Brearley,
F.Q., van Schaik, C.P., Marshall, A.J., 2011. Forest fruit production is higher on Sumatra
than on Borneo. PLoS ONE 6(6), e21278. doi:10.1371/journal.pone.0021278.
Wong, S.T., 2002. The ecology of Malayan sun bear (Helarctos malayanus) in the lowland
tropical forest of Borneo. MSc. Thesis, University of Montana
Wong, S.T., Servheen, C., Ambu, L., 2002. Food habit of Malayan sun bears in lowland tropical
forest of Borneo. Ursus 13, 127–136.
Wong, S.T., Servheen, C.W., Ambu, L., 2004. Home range, movement and activity patterns and
bedding sites of Malayan sun bear Helarctos malayanus in the rainforest of Borneo.
Biological Conservation 119, 169-181.
Wong, W-M., Leader-Williams, N., Linkie, M., 2012. Quantifying changes in sun bear
distribution and their forest habitat in Sumatra. Animal Conservation 16, 216–223.
Wong, W-M., Linkie, M., 2013. Managing sun bears in a changing tropical landscape. Diversity
* Corresponding author. E-mail address: [email protected] (D. Lee).
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
35 Lee et al.: Habitat utilisation by sun bears
and Distributions 19, 700–709.
World Agroforestry Centre, 2018. Tree functional attributes and ecological database.
<http://www.worldagroforestry.org/output/tree-functional-and-ecological-databases>.
Downloaded on 19 June 2018.
Yasuma, S., Andau, M., 2000. Mammals of Sabah, Part 2, habitat and ecology. Japan
International Cooperation Agency and Sabah Wildlife Department, Kota Kinabalu, Malaysia.
Zuur, A.F., Ieno, E.N., Elphick, C.S., 2010. A protocol for data exploration to avoid common
statistical problems. Methods in Ecology and Evolution 1, 3–14.
* Corresponding author. E-mail address: [email protected] (D. Lee).
711
712
713
714
715
716
717
718