Foraging ecology and coexistence of Asiatic black bears and sun ...

2 downloads 0 Views 472KB Size Report
Feb 25, 2013 - Minnesota Department of Natural Resources, 1201 E Highway 2, Grand Rapids, MN 55744, ... or express the highest population growth rate will eventually .... trunk. Finally, we compared our purported bear foods with bear.
Journal of Mammalogy, 94(1):1–18, 2013

Foraging ecology and coexistence of Asiatic black bears and sun bears in a seasonal tropical forest in Southeast Asia ROBERT STEINMETZ,* DAVID L. GARSHELIS, WANLOP CHUTIPONG,

AND

NARET SEUATURIEN

World Wide Fund for Nature—Thailand, 87 Phaholyothin 5, Samsen Nai, Phayathai, Bankok 10400, Thailand (RS, NS) Minnesota Department of Natural Resources, 1201 E Highway 2, Grand Rapids, MN 55744, USA (DLG) King Monkut’s Institute of Technology Thonburi, 83 Mu 8, Thakham, Bangkhuntien, Bangkok 10150, Thailand (WC) * Correspondent: [email protected] Asiatic black bears (Ursus thibetanus) and sun bears (Helarctos malayanus) are ecologically similar and coexist extensively across Southeast Asia. We used foraging signs identified to bear species to examine their food selection and dietary overlap relative to food abundance, nutrition, and phenology in 3 habitats in Thailand. We posited, based on ecological theory, that coexistence of these 2 species would be explained through resource partitioning; our data, however, did not support this hypothesis. We conducted 71 sign transects and recorded 730 bear signs, mainly claw marks on trees that bears climbed for food. Both species fed predominantly on fruit; we documented 93 plant species from 42 families that bears consumed. Insects were of secondary importance. Bears of the 2 species selected fruit trees of the same families and genera in each habitat, especially lipid-rich Lauraceae and Fagaceae, tracking fruiting phenology through time. Diet overlap was high, even during periods of diminished fruit availability. We propose a number of mechanisms that may have promoted coexistence of these 2 species. For example, sun bears consumed proportionately more insects than did black bears; insectivory may help sustain the smaller-sized sun bears in the face of competition over fruits with black bears. Also, competition over fruits was reduced by both species cropping a lower proportion of common fruit trees than rarer fruit trees, thereby leaving a potential surplus for the other species. Furthermore, food resources were generally abundant and available year-round: about half the trees in the forest were potential food trees for bears. Bear populations likely were depressed below carrying capacity by previous hunting; as they recover, more competition for resources and greater niche divergence could ensue. Key words: Asiatic black bear, food habits, frugivory, Helarctos malayanus, negative frequency-dependent selection, niche overlap, resource selection, sun bear, Ursus thibetanus Ó 2013 American Society of Mammalogists

DOI: 10.1644/11-MAMM-A-351.1

Asiatic black bears (hereafter black bears; Ursus thibetanus) and sun bears (Helarctos malayanus) co-occur in Myanmar, Thailand, Lao People’s Democratic Republic, Cambodia, Vietnam, and portions of India and China (International Union for the Conservation of Nature 2010). Across this region the 2 species coexist ubiquitously at small spatial scales, including forest blocks as small as 80 km2 (Htun 2006; Vinitpornsawan et al. 2006). Sun bears (40–60 kg) are about half the size of black bears (65–150 kg—Lekagul and McNeely 1988), but otherwise these species are ecologically and behaviorally similar. Both species are opportunistic omnivores with broadly similar diets of fruits and insects (Fredriksson et al. 2006; Hwang et al. 2002; Takahashi et al. 2008; Wong et al. 2004). In this study we ask, given these similarities, how do sun bears and black bears coexist so extensively in this region?

Coexisting species that share limiting resources compete for those resources. Classical ecological theory predicts that over time, the species that is best able to garner the limiting resources, reduce access to the resources by the other species, or express the highest population growth rate will eventually dominate and drive out the competitors. Alternatively, competitors may respond by adapting to the use of alternate resources, thereby diminishing competition through reduced niche overlap, facilitating coexistence (Holt 2001). Despite a plethora of ecological literature on the subject of species coexistence, few studies have actually demonstrated whether or how a species is able to rebound from rarity in the presence of a

www.mammalogy.org 1

2

Vol. 94, No. 1

JOURNAL OF MAMMALOGY

more abundant competitor—this being the strict ecological meaning of coexistence (Siepielski and McPeek 2010). Nevertheless, very long-term co-occurrence of species strongly suggests that they must coexist in this classical ecological sense. Black bears and sun bears have co-occurred in Southeast Asia since the middle Pleistocene (Erdbrink 1953; Tougard 2001). Absence of competitive exclusion over this evolutionary timescale implies that the 2 species maintain divergent niches, whereby they eat different foods, or the same foods in different places (Schoener 1974). The 2 species are often described as having similar diets in Southeast Asia, with fruits and insects as key components (Francis 2008). Lekagul and McNeely (1988) suggested, from mainly anecdotal information, that diets of sun bears and black bears were similar in Thailand, but that sun bears consumed more insects. Later field studies in Borneo showed that sun bears selected fruits over insects (Augeri 2005), but when fruits were rare, they used insects as a fallback food (Fredriksson et al. 2006; Wong et al. 2002). Insects tend to be widespread and abundant in tropical forests (Primack and Corlett 2005), but sun bears apparently find it difficult to subsist on insects over extended periods, because starvation has been documented during lengthy fruiting failures (Fredriksson 2012; Wong et al. 2005). Black bears also regularly eat insects as a fallback food in some areas (Koike 2010), especially during lean seasons with low fruit availability, but recent studies indicate that black bears cannot meet their energy demands on a diet exclusively of insects (Yamazaki et al. 2012). Thus, examination of available data suggests that both black bears and sun bears rely heavily on fruits. Fruit availability in tropical forests is extremely patchy in space and time (Fleming et al. 1987; Jordano 1992); this scarcity generates competition among other frugivores (French and Smith 2005), so we expected it to occur between these 2 species of bears. Sun bears and black bears share most habitats in sympatry (Steinmetz et al. 2011) and have similar temporal activity patterns (Grassman et al. 2006; W. Chutipong, King Monkut’s Institute of Technology Thonburi, pers. comm.). Thus, we hypothesized that the 2 species would exhibit niche differentiation in terms of the fruits and other types of foods that they consumed. We examined 3 hypotheses regarding possible resource partitioning that might explain coexistence of black bears and sun bears. First, the 2 species might rely on fruits to different extents, because one (presumably the sun bear) uses other types of food, notably insects, to a large degree. In terms of species coexistence, this hypothesis would predict that each species was limited, over the long term, by a different type of food. Although this prediction is difficult to test, it suggests that the 2 species should exhibit sizeable differences in the proportion of fruit in their diets. In North America, sympatric populations of brown bears (U. arctos) and American black bears (U. americanus) show some partitioning of foods (Apps et al. 2006; Belant et al. 2006, 2010), with smaller-sized American black bears being better able to subsist on an exclusively fruitbased diet (McLellan 2011). Second, although black bears and

sun bears might both focus on fruits, their preference for fruit taxa might differ. This hypothesis predicts that some fruit taxa would be more commonly eaten by black bears and others more commonly eaten by sun bears. Distinct fruit preferences are an important source of niche separation promoting coexistence of other frugivorous mammals (Stevenson et al. 2000). Third, the 2 bear species might use similar fruits, but from different fruit crop sizes. Smaller-bodied competitors can coexist with larger species by foraging on smaller or lessproductive patches of food (Basset and Angelis 2007). Large brown bears in North America, for example, seek dense patches of fruit where their rate of ingestion is high, whereas smaller American black bears can subsist on more meager patches (Welch et al. 1997). Thus, this hypothesis predicts that larger-bodied black bears should select more abundant crops of fruit from larger trees, whereas smaller sun bears could exploit sparser crops in smaller trees. To assess evidence for resource partitioning we investigated the diets of bears of each species in sympatry, then examined food selection relative to food availability and nutritional value. Thus, our study considers niche overlap of competing species along with the dynamics and characteristics of the available resources. This dual focus provides insights into the competitive processes and underlying mechanisms of coexistence that are difficult to discern by simply comparing species niches (Begon et al. 2006; Rotenberry 1980).

MATERIALS

AND 2

METHODS

Study sites.—The 3,622-km Thung Yai Naresuan Wildlife Sanctuary (hereafter, Thung Yai) is in western Thailand, adjacent to Myanmar. The sanctuary is mountainous, with elevations of 400–1,800 m. Two forest types, semievergreen and mixed deciduous, occur in a mosaic below 1,000 m elevation and comprise 75% of Thung Yai’s forest cover. Montane evergreen forest occurs between 1,000 and 1,800 m and is less extensive (15%). Secondary forest, savanna, and dry dipterocarp forest comprise the remaining vegetation cover (Nakhasathien and Stewart-Cox 1990). Semievergreen forest (hereafter, evergreen) is tall, with a closed canopy at 25–40 m formed predominantly by evergreen tree species (Maxwell 1995). Mixed deciduous forest (hereafter, deciduous) is dominated by deciduous tree species, and tree density and plant species richness is lower than in evergreen forest (Rundel and Boonpragob 1995; van de Bult 2003). Montane evergreen forest (hereafter, montane) has high tree density and richness but a lower canopy than evergreen forest. We defined 2 seasons, dry (November–April) and rainy (May–October). Rainfall is typically ,100 mm/month in the dry season. Mean (6SD) annual rainfall in Thung Yai during our study was 1,731 6 217 mm (Thai Department of Meteorology 2005). Mean (6SD) annual maximum and minimum temperatures were 33.68C 6 0.28C and 20.58C 6 0.78 C, respectively. We established 4 study sites; 3 contained mosaics of evergreen and deciduous forest, and the 4th was montane. Sites were 15–30 km apart. Sign transects and vegetation plots were

February 2013

STEINMETZ ET AL.—FOOD HABITS OF SYMPATRIC TROPICAL BEARS

distributed over an area of 30–100 km2 at each site, encompassing home ranges of multiple bears. Annual home ranges of adult sun bears and black bears have been reported as 4–21 km2 (Fredriksson 2012; Wong et al. 2004) and 30–150 km2 (Garshelis 2004; Hwang et al. 2010; Reid et al. 1991), respectively, and individual home ranges overlap widely. Most fieldwork was conducted from November 2001 to June 2003, including both seasons. Additional vegetation sampling continued until July 2005. Determining bear diets.—Bears leave abundant signs in the forest that are related largely to foraging. We determined diets of bears from observations of foraging signs and scat analysis. We searched for bear signs in 71 straight, 300-m-long strip transects (n ¼ 38 in evergreen, n ¼ 27 in deciduous, and n ¼ 6 in montane; total 31.2 ha). Sampling of transects in evergreen and deciduous forest was distributed across all months except July and October. Each transect was sampled once. Transects were 10 m wide in evergreen and deciduous (0.3 ha) and 20 m wide in montane (0.6 ha) forest, commensurate with differing tree densities. Transect width did not affect results because resource use by bears was expressed in terms of signs per hectare. Transects covered the range of topographical variation within a study site. Within each transect, we looked for bear claw marks by closely examining every tree, and searched the ground for terrestrial insect-foraging signs (holes dug for nests of wasps, ants, and termites; opened termite mounds; and broken logs caused by bears foraging for insects). We assumed that few signs were missed because 3 observers searched exhaustively through each transect at a slow pace (about 100 m/h), and bear foraging signs are generally conspicuous. Bear claw marks are uniquely teardrop-shaped and easily distinguished from those of other animals that climb (e.g., civets) or rake trees (e.g., felids). We considered holes to be dug by bears if accompanied by footprints, or if deeper than 30 cm; other mammals dig shallower holes (R. Steinmetz, pers. obs.). We measured diameter at breast height (DBH) of climbed trees, and identified trees by comparing samples of leaves, fruits, and flowers to reference collections at herbaria in Thailand, and based on Gardner et al. (2000). Multiple climbs of a tree were treated as 1 sign. We also collected scats in the course of sign surveys, and recorded percent frequency of occurrence of different food items as (ni/N) 3 100, where N was the total number of scats and ni the number of samples containing food item i. We did not attempt to differentiate scats to bear species. We differentiated claw-marked trees that appeared to have been climbed for foraging, versus resting or other reasons. We considered climbing to be associated with foraging on fruit (fleshy fruits, pods, or nuts) if fresh claw marks coincided with fruiting in the climbed tree; broken branches occurred in the canopy (bears break branches to reach fruit—Hwang et al. 2002); or multiple ages of claw marks were present, indicating seasonal revisiting; such repeated climbing often is associated with food-producing trees (Fredriksson et al. 2006; Wong et al. 2002). If claw marks led to a raided bee nest, we considered climbing to be associated with foraging on bees. Bears also

3

sometimes mark trees for communication, by biting, rubbing, and clawing (Burst and Pelton 1983), but these signs are conspicuously different from climbing marks that ascend the trunk. Finally, we compared our purported bear foods with bear foods identified from scats and visual observations of bears foraging elsewhere in the region. Both species of bears also eat fruits that have dropped to the forest floor (Fredriksson et al. 2006; Hwang et al. 2002). Such foraging produces no sign. We assumed that any fruits eaten by bears on the ground also were eaten by bears in the trees. For this assumption to bias our inferences about dietary overlap, one of the bear species would have to have fed largely on fallen fruits that the other species did not eat, while neither species climbed the trees that produced those fruits. Such a scenario seems highly improbable, especially given the intensive tree climbing shown by both species, and its concentration in wellknown fruit families (see ‘‘Results’’). Feeding on shrub-borne fruits also leaves little or no sign. Asiatic black bears in temperate environments do eat shrub-borne fruits (Koike 2009), but shrubs are rare in tropical forests (Turner 2001; R. Steinmetz, pers. obs.), and consequently shrub-borne fruits constitute an insignificant portion of the diets of Southeast Asian bears (Fredriksson et al. 2006; Wong et al. 2002). Meateating by sun bears and black bears is uncommon (Fredriksson et al. 2006; Hwang et al. 2002; Wong et al. 2004) so we expected hunting and scavenging by bears to contribute negligible portions of their overall diets. Sign ages and species identification.—Based on experimental work on the degree of bark regrowth in gouges, we distinguished claw marks that were fresh (,3 months), recent (3–10 months), or old (.10 months—Steinmetz and Garshelis 2010). Claw mark ages were used to match fresh climbing with current fruiting in a tree, and to identify seasonal revisitation to a tree. Insect foraging signs such as diggings and opened logs persist for up to a year, roughly coinciding with the time period represented by fresh and recent claw marks. Thus, we could directly compare the proportions of within-year claw marks versus ground signs. Comparing these proportions for each bear species was more difficult, because we relied on footprints to distinguish species for ground signs (see below), and these faded faster than claw marks, thus producing a slight underrepresentation of fresh ground signs. Claw marks from Asiatic black bears tend to be wider than those of sun bears. In previous experimental work we found that the difference in spacing across marks from 3 to 5 claws of a hind foot reliably differentiated 95% of claw marks of these 2 species (Steinmetz and Garshelis 2008). When encountering fresh or recent claw marks on a transect, we pressed a sheet of paper over the tree trunk and punched holes over the center of each gouged mark so that we could later measure the spacing across the marks, and thus distinguish them to species. Small black bears and large sun bears may make similar-sized claw marks; marks in this size range were categorized as indeterminate (Steinmetz and Garshelis 2008) and not used in species-specific analyses. Bear marks that were old and stretched with tree growth, incomplete (,3 claws imprinted on

4

JOURNAL OF MAMMALOGY

the bark), or from front feet, were recorded but not identified to species. Thus, we had 2 samples, 1 of all climbed trees irrespective of bear species, and a subset of that in which the bear species was identified and the marks created mostly within the year (fresh and recent). We used the larger data set, which included many trees climbed 1 year or more in the past, to identify frequently climbed tree taxa targeted by bears, and to examine long-term relationships between tree abundance and use by bears. We used the species-specific tree-climbing data set to assess fruit-tree selection by each bear species. Bear footprints at insect- or vegetation-foraging sites, where present, were classified as black bear if hind pad width was .10 cm and total length .17 cm, sun bear if measurements were below these thresholds, and indeterminate if length and width matched different species (R. Steinmetz, pers. obs., from captive animals). Food availability.—We established vegetation plots to quantify density of bear food-tree species, genera, and families. In deciduous forest we established 60 plots of 20 3 20 m (total 2.4 ha), each .400 m apart, at 2 sites. In evergreen forest we established 1 large plot (100 3 140 m ¼ 1.4 ha), because censusing this species-rich forest type requires many repeat visits to a site. We measured and identified all trees .10 cm DBH. We did not establish a plot in montane forest because of its remote location. For this habitat, we used published data on species composition from a 15-ha plot in northern Thailand (Hara et al. 2002), the only data available on this forest type in Thailand. Thung Yai montane forest has a similar floristic composition to northern Thailand (van de Bult 2003). The montane data set was quantitative only at the family level, but included all montane genera used by bears at our site. We conducted quantitative analyses of trees climbed by bears and in vegetation plots mainly at genus and family levels because not all could be identified to species. This taxonomic resolution might have obscured fruit partitioning if it occurred at the tree species level. However, tree species within most food genera had the same fruit type (e.g., fleshy fruit, pod, nut, etc.) and similar nutritional composition (Herrera 2002), so we reasoned that it was unlikely that the 2 bear species would have different preferences for congeneric tree species. We focused on 26 families that were climbed frequently (each accounting for .2% of all climbed trees within a habitat) and whose constituent species or genera were used by bears for food (as determined by foraging signs and scats). We obtained 2 fruiting phenology data sets. First, we recorded numbers of bear food trees with fruit in sign transects and vegetation plots (n ¼ 197 trees). These data covered 4 rainy and 7 dry season months. We calculated mean density of trees bearing fruit each month (fruiting tree density), then combined months into seasons. Second, we recorded the months that different bear food-tree species contained fruit (n ¼ 1,309 trees), based on observations in sign transects, vegetation plots, and while hiking in the sanctuary. Most months were sampled in multiple (2–3) years, except June, July, and August, which were sampled once, and October, which was not sampled. For months sampled ,2 times, we used published phenology data

Vol. 94, No. 1

from Thung Yai (Maxwell 1995) to augment our observations. We then calculated, for each habitat, numbers of bear food-tree species that fruited per month, and average number of months a tree species had fruit (fruit-crop persistence). Observations were made for 95% of the identified bear tree species (n ¼ 79/ 83 tree species) in deciduous and evergreen forests (¼ 16.6 trees/species). Fieldwork in montane forest occurred in only 1 month so our depiction of phenology was incomplete; fruiting phenology was observed on 13 (59%) of 22 bear food trees in this habitat. Many fruits are high in carbohydrates, whereas lipid content varies greatly and tends to be inversely correlated with carbohydrate content (Jordano 1992). We used percent dry mass of lipids to characterize plant families as either lipid-rich (lipids . 15%) or sugar-rich (lipids , 15%), based on data by Van Steenis (1978), Snow (1981), Leighton and Leighton (1983), Wheelwright et al. (1984), Stiles (1993), Kitamura (2000), and Yasuda et al. (2005). Nutritional composition of fruits is strongly correlated with plant phylogeny at the genus and family levels (Herrera 2002), so these studies, although from other tropical countries, are likely to reflect fruit nutrient characteristics of tree taxa in Thung Yai. We counted termite mounds in 13 sign transects in evergreen forest, 4 in deciduous, 6 in montane, and in 20 deciduous and evergreen vegetation plots. Total area searched was 2.7 ha in deciduous, 4.2 ha in evergreen, and 1.8 ha in montane forest. We assumed that active termite mounds represented available foods for bears. Fruit-tree selection.—We used foraging signs and scats to identify tree taxa that bears used for food; we refer to these taxa as bear food trees. We used densities of climbed trees in these bear food taxa to assess resource use and selection by bears. We compared use of bear food trees to abundance in 2 ways. First, we regressed density of all climbed trees from frequently climbed families (i.e., use) on tree density. Next, using only the data identified to bear species, we compared tree use to abundance using logit-transformed values of proportional use (percent climbed) of trees as the dependent variable in a leastsquares linear regression on tree density (log-transformed). The logit transformation is ln[pi/(1  pi)], where pi is the proportion of available trees climbed in family or genus i. Analyses for each bear species were conducted at family and genera levels in each habitat. We further tested whether selection of tree families and genera was influenced by differential use of fruit nutrient types (lipid- or sugar-rich), or by fruiting phenology relative to other tree families. We assigned families and genera to either high or low periods of fruit availability, depending on whether their constituent species tended to fruit when most other species fruited. We defined high-fruit months as those when at least 50% of species in a forest type were fruiting. We conducted multiple regression with these categorical predictors entered as a block following tree density. We used the change in the multiple coefficient of determination (DR2) and standardized coefficients (b) to assess their contribution to the model. Data

February 2013

5

STEINMETZ ET AL.—FOOD HABITS OF SYMPATRIC TROPICAL BEARS

TABLE 1.—Signs of sun bears and Asiatic black bears recorded in sign transects (n ¼ 71) in Thung Yai Naresuan Wildlife Sanctuary, Thailand, 2001–2003. Fresh and recent signs are within the year. No. signs

Sign category

Likely food

Climbed trees Raided stingless bee nestsa Broken logs; termite mounds opened; holes dug Feeding debris Scat Total

Fruit Stingless bees Insects Vegetation Various

a

Total (%) 623 25 44 5 33 730

(85.3) (3.4) (6.0) (0.7) (4.5) (100)

Bear species identified

Fresh and recent (%) 287 10 44 5 33 379

(75.7) (2.6) (11.6) (1.3) (8.7) (100)

Total

Sun bear

Black bear

n

n (%)

n (%)

279 10 20 4 0 313

152 8 19 3 0 182

(83.5) (4.4) (10.4) (1.6) (0) (100)

127 2 1 1 0 131

(96.9) (1.5) (0.76) (0.76) (0) (100)

Usually also on climbed trees.

points in montane forest were too few for multiple regression, so we assessed the influence of nutrients qualitatively. Niche breadth and overlap.—We calculated trophic niche breadths and diet overlap to compare the selection of food-tree taxa used by each bear species. We examined niche breadth with Levins’ index: ! n X pi 2 ; B¼ i¼1

where n is the number of tree families or genera and p is the proportion of records (i.e., climbed trees) in each family or genus i (Krebs 1999). The index was standardized as Bstandard ¼ (B  1)/(Bmax  1), to produce breadths between 0 and 1. This index measures the uniformity of distribution of individuals (climbed trees) among resource categories (tree families or genera). We estimated overlap in resource use with Pianka’s index (Krebs 1999): sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi n n n X X X pij pik pij 2 pik2 ; Ojk ¼ i

=

i

i

where O is the mutual overlap between bear species, p is the proportion of food category (tree family or genus) i for bear species j and k, and n is the total number of resources. Values of O range from 0 (no resource use overlap) to 1 (complete overlap). To compare overlap indexes among 3 forest types with disparate ranges of tree densities, we also used Hurlbert’s overlap measure (L), which accounts for resource abundance:  n  X pij pik L¼ ; ai i where p is the proportion of food category (tree family or genus) i for (bear) species j and k, and ai is the proportional abundance of food category i. Hurlbert’s index is 1.0 when both species use each resource in proportion to its abundance, 0 when no resources are shared, and .1.0 when certain resources are used more intensively than others and preferences of the 2 species tend to correspond (Krebs 1999). We used a Fisher exact test to examine whether proportions of insect- versus fruit-foraging signs differed between black bears and sun bears; we applied a 1-tailed test because we

expected sun bears to be more insectivorous. We used 2-tailed Fisher exact tests to examine whether ratios of sun bear : black bear climbing activity among food-tree families and genera differed within habitats. We also calculated selection indexes for frequently climbed tree families and genera used by each bear species, wi ¼ oi/pi , where wi is the selection index for tree family or genus i, oi is the proportion of family or genus i in the climbed tree sample, and pi is the proportion of family or genus i available in the forest (Krebs 1999). We used this index to compare the rank order of tree selection (tested with Spearman’s rank correlation) by each bear species; if the 2 bear species partitioned resources, we expected negative correlations among these indexes. We examined whether bears partitioned fruit crop sizes in 2 ways. First, we compared mean DBH of climbed trees in each habitat, using Mann–Whitney U-tests. Tree DBH tends to correlate strongly with fruit production (Leighton 1993). Second, we categorized fruiting trees with fresh, identifiable bear signs (n ¼ 57; 28 black bear and 29 sun bear) as having low, moderate, or high crops of fruit, based on fruit density and proportion of the canopy with fruit (Bullock and SolisMagallanes 1990). All observations were made by the same observer (RS). We used a chi-square test to assess whether black bears and sun bears climbed trees with different fruit crops. We defined statistical significance at a ¼ 0.05.

RESULTS Diet overview.—We recorded 730 bear signs, of which 313 (42.9%) were identified to bear species (Table 1). Of the 5 main types of bear signs that we observed, claw marks on trees that bears climbed to forage on fruit were most common for both black bears and sun bears (84–97%; Table 1). Less than 15% of signs of each species was related to foraging on insects. Comparing only fresh signs, evidence of insect foraging was more common for sun bears (n ¼ 11, 17.5%) than black bears (n ¼ 2, 4.7%); this difference was statistically significant (P ¼ 0.04 for the 1-tailed test that sun bears consumed proportionately more insects than black bears). Evidence of foraging (climbing associated with fruiting, broken branches, or multiple climbing events) was visible on 70% of all freshly climbed trees (n ¼ 86/123). Also, most scats (79%, n ¼ 26) contained fruit; 70% contained only fruit. Fruits

6

JOURNAL OF MAMMALOGY

from 14 tree families were found in scats; 13 of these also were represented by freshly climbed trees. Oaks (Fagaceae) were the most frequently observed fruit in scats (18% of scats). Insects, especially ants, occurred in 18% of scats. Termites were rarely eaten: none were in scats, and we found just 1 opened mound (Globitermes sulphureus; Termitidae) and 1 dug nest (Termes sp.). Herbaceous vegetation occurred in 15% of scats. We recorded direct or indirect evidence of bear foraging on 93 plant species, in 72 genera and 42 families; most of these taxa had .1 type of evidence of foraging (Appendix I). Most (70%) of these families produce fleshy fruits, typically berries, drupes, or arillate capsules, but a few produce edible nuts (Fagaceae) or pods (Leguminosae; Table 2). Eighty percent (n ¼ 501/623) of all climbed trees (excluding trees climbed to consume bees) were in 26 food-tree families that were frequently climbed and appeared to be targeted by bears; these families included 52 genera and at least 83 species with evidence of foraging (Appendix I). Similar genera or species from most of these same families (81%) have been recorded as producing bear foods in other studies from the region (Table 2), substantiating the indirect evidence we used to infer feeding. Six families (13 species in 10 genera) with foraging evidence were not observed as bear foods in other studies and produce dry capsules or follicles rather than fleshy fruits (Apocynaceae, Bignoniaceae, Juglandaceae, Lythraceae, Sterculiaceae, and Dipterocarpaceae; the last was climbed mostly to raid stinglessbee [Trigona spp.] nests). Of all climbed trees that we observed, 23% in deciduous forest (n ¼ 35 trees) and 29% in evergreen forest (n ¼ 116 trees) had been reclimbed at different times. Only 4% (n ¼ 3) of climbed trees in montane forest had been reclimbed, although 10% of Fagaceae trees had been reclimbed. Food availability.—The density of bear food trees (not counting the 6 uncorroborated families above) was 223 trees/ha in evergreen, 158 trees/ha in deciduous, and 221 trees/ha in montane forest, accounting for 40–53% of tree density in each habitat. The number of bear food-tree taxa that produced fruit was highest during the late dry and early rainy season (April– May) in both evergreen and deciduous forest (Fig. 1). Density of bear food trees that were fruiting in evergreen and deciduous forest, respectively, was 10.9 trees/ha 6 2.8 SD and 13.4 6 12.3 trees/ha each month during the rainy season, and diminished by 20–50% in the dry season (evergreen: 8.0 6 5.6 trees/ha; deciduous: 6.4 6 5.5 trees/ha). Fruiting tree density in montane forest, measured from a single survey in March, was 17.2 trees/ha. Mean persistence of fruits among species was 2.2 months 6 1.2 SD in evergreen and 2.5 6 1.2 months in deciduous forest. Persistence of acorns on individual oak trees tended to be only 1–2 months, but Fagaceae produced staggered acorn crops that spanned 6 months in each habitat. Lauraceae and Meliaceae species also had staggered fruiting that lasted 5–6 months in evergreen forest. Density of termite mounds was highest in deciduous (5.3 mounds/ha 6 10.3 SD), lower in evergreen (1.2 6 1.7 mounds/ha), and zero in montane forest. All but 1 of 15 mounds that we examined was active, and inhabited by G.

Vol. 94, No. 1

sulphureus. However, we found only 1 mound excavated by a bear. Fruit-tree selection.—The density of trees of each family and genus that were climbed was positively and linearly associated with density of each tree family and genus on the landscape (families: R2 ¼ 0.55, P , 0.0001, n ¼ 32; genera: R2 ¼ 0.25, P , 0.0001, n ¼ 65). The slopes of these regressions, however, were ,1 (b . 0.04, SE , 0.007, P , 0.0001), indicating that bears did not visit trees randomly (implied by a slope of 1); rather, bears’ rates of tree climbing declined with increasing abundance of each type of tree. The most frequently climbed tree taxa tended to be the most abundant trees in the forest. In evergreen forest, both black bears and sun bears climbed Lauraceae, Meliaceae, and Fagaceae most (0.5–2.0 climbed trees/ha; Fig. 2). Lauraceae and Meliaceae were very common (each .35 trees/ha; Fig. 2), so only a small proportion were climbed by bears (,5%). In deciduous forest, black bears climbed Leguminosae and Dilleniaceae most (0.3 trees/ha), and sun bears climbed Leguminosae and Labiatae most (0.4 trees/ha). Leguminosae was abundant (25 trees/ha), and ,2% were climbed by bears. The negative relationship between proportional use and abundance of fruit-tree families was significant for both bear species in evergreen forest (R2 . 0.69, P , 0.0005) and deciduous forest (R2 . 0.3, P  0.05; Table 3; Fig. 3). This relationship was even stronger for genera (Table 3). These relationships held excluding the tree taxa that were not corroborated as bear foods in previous studies. No relationship was detected between tree use and abundance in montane forest; however, the power of this analysis was low because of the small number of tree families (n ¼ 5). Again, the 2 most climbed families, Fagaceae (5 trees/ha) and Lauraceae (4.4 trees/ha), also were most common (Fig. 2). These same relationships in each habitat also held with older marks that could not be categorized to bear species (Fig. 2; Bear sp.). Nutrient type, and whether fruit was available mostly in the period of dearth or abundance, explained an additional 22–28% of variation in regression models for both bear species in deciduous forest, although this change was not statistically significant (b ¼ 0.18–0.46, P . 0.11). In evergreen forest, nutrient and season were not influential (DR2 ¼ 0.02 in both cases, b ¼ 0.04–0.15, P . 0.64); tree density alone was the predominant factor influencing tree selection there. Niche breadth and overlap.—Within each habitat, niche breadth of black bears and sun bears was similar (Table 4), indicating that both bear species used a similar array of tree families. Niche overlap was high in every habitat at the family level (O  0.77), especially evergreen forest (Table 4). At both family and genus levels, Hurlbert’s overlap index was about twice as high in evergreen as in deciduous or montane habitat (Table 4); this emphasized that in evergreen forest both bear species made intensive use of the same resources, notably Lauraceae, Meliaceae, Fagaceae, and constituent genera Beilschmiedia, Aglaia, Lithocarpus, and Quercus. In deciduous forest both species shared many families and

February 2013

7

STEINMETZ ET AL.—FOOD HABITS OF SYMPATRIC TROPICAL BEARS

TABLE 2.—Plant families with evidence of foraging by sun bears and black bears in Thung Yai Naresuan Wildlife Sanctuary, Thailand (2001– 2003), cross-referenced with observations from elsewhere in Asia. Evidence from this study (boldface number 15) includes scat, feeding debris, and foraging signs on climbed trees (see Appendix I for details). Families in boldface type were frequently climbed in Thung Yai (.2% of records). For genus- and species-level information, see Appendix I. NA ¼ not applicable.

Plant family

Fruit type

Nutrient typea

Anacardiaceae Annonaceae Apocynaceae Araceae Araliaceae Bignoniaceae Burseraceae Celastraceae Combretaceae Dilleniaceae Dipterocarpaceae Ebenaceae Elaeocarpaceae Euphorbiaceae Fagaceae Flacourtiaceae Guttiferae Icacinaceae Juglandaceae Labiatae Lauraceae Lecythidaceae Leguminosae Lythraceae Magnoliaceae Meliaceae Moraceae Musaceae Myristicaceae Myrtaceae Oleaceae Palmae Pandanaceae Rhamnaceae Rosaceae Rubiaceae Sapindaceae Sapotaceae Sterculiaceae Symplocaceae Theaceae Tiliaceae

Drupe Berry Follicle NA Berry Capsule Drupe Aril Drupe Aril NA Berry Drupe Drupe, berry Nut Berry Berry Drupe Samara Drupe Berry Berry Pod Capsule Aril Aril Berry NA Aril Berry Drupe Drupe Compound Drupe Drupe Berry Aril Berry Follicle Drupe Berry Berry

Lipid Lipid No data No data No data No data Lipid Sugar Sugar Sugar NA Sugar Sugar Sugar Lipid Sugar Sugar Lipid No data Sugar Lipid No data Lipidf No data Lipid Lipid Sugar NA Lipid Sugar Lipid Sugar NA Sugar Sugar Sugar Sugar Sugar Lipid (seeds) Sugar Sugar Sugar

Part eaten (other studies)

Feeding evidenceb (other studies)

Fruit Fruit Fruit Shoot Fruit Unknown Fruit Unknown Unknown Fruit Nesting bees Fruit Fruit Fruit Nuts Fruit Fruit Fruit Unknown Fruit Fruit Flower Fruit Unknown Fruit, flower, bark Fruit Fruit Shoot Fruit Fruit Fruit Fruit, shoot Shoot Fruit Fruit Fruit Fruit Fruit Unknown Fruit Unknown Fruit

Scat, obs. Scat, obs. Scat, obs. Scat, obs. None None Scat, obs. None None Obs. None Scat, obs. Scat Scat, obs. Scat, obs. Scat Scat, obs. Scat, obs. None Scat, obs. Scat, obs. Obs. Scat, obs. None Obs., scat Scat Scat, obs. Obs. Scat, obs. Scat, obs. Branch Obs. Obs. Scat Scat, obs. Scat Scat, obs. Scat None Scat None Scat, obs.

Source of evidencec Sun bear 3, 15 3, 4, 8 3 (liana), 15 15 15 3, 4, 8, 15 15 15 8, 15 3 3, 8, 15 3, 8, 15 3, 15 15, 20 15 3, 8, 15 3 3, 15 15 3, 15 3, 8, 15 3, 4, 8, 15 3, 8, 15 3, 8, 15 3, 15 3, 15 15, 21 3, 8, 15 3

3, 15

Black bear

Bear sp.d

7, 15

16 15 15 15 15 15, 23 6, 12, 26 15 6, 27 1, 2, 6, 7, 11, 12, 15 15 20 15 15, 22 6, 9, 11, 12, 15 15 8, 15, 25 14, 15, 19 15, 19 6, 15

15 15 25 15 18

15 15, 17, 25 5e 15, 17 15, 25 15 15, 25

15, 17, 25 13e, 15

15 15 15

15 24 15

13e, 17 15

16 2, 6, 10, 15

15

6, 15

1 15 15

5e, 15 25 15, 25 15 15, 18, 25 13e, 17

a

Nutrient type indicates whether fruits are lipid-rich or sugar-rich (see text). Obs. ¼ direct observation of a bear feeding; branch ¼ broken branches from feeding. c Sources: (1) Sathyakumar and Viswanath 2003; (2) Manjrekar 1989; (3) Fredriksson et al. 2006; (4) McConkey and Galetti 1999; (5) Sreekumar and Balakrishnan 2002; (6) Hwang et al. 2002; (7) Saberwal 1989; (8) Wong et al. 2002; (9) Kitamura 2000; (10) Huygens and Hayashi 2001; (11) Reid et al. 1991; (12) Schaller et al. 1989; (13) Joshi et al. 1997; (14) Nozaki et al. 1983; (15) this study; (16) Corlett 1998; (17) Kitamura et al. 2002; (18) Kitamura et al. 2005; (19) Takahashi et al. 2008; (20–27) personal communication with, respectively: P. Kinklai, King Monkut’s Institution of Technology, Thonburi, Thailand; W. Sangkametawee, Khon Kaen University, Thailand; L. Nong, Khao Paeng Ma Village, Thailand; M. Tu-U, Tilaipa Village, Thailand; N. Ahktar, National Zoo, Saudi Arabia; D. Ngoprasert, King Monkut’s Institution of Technology, Thonburi, Thailand; R. Phoonjampa, WWF Thailand; and Y. Em-Oad, Kuiburi National Park, Thailand. d Bear sp. ¼ feeding evidence not distinguished to species. e Data for sloth bear (Melursus ursinus), included for comparison. f Seeds are lipid-rich, but pulp of some species (Cassia fistula) is sugary. b

genera but each was used more in proportion to its abundance (i.e., L was close to 1). We recalculated niche breadth and overlap indexes (at the genus level) separately for high fruiting periods (April–June in

deciduous forest and March–June in evergreen forest) and low fruit periods. Pianka’s index of niche overlap was higher, by 29% in deciduous and 64% in evergreen, during the low fruit period (Table 4). When few tree taxa were fruiting in evergreen

8

JOURNAL OF MAMMALOGY

Vol. 94, No. 1

FIG. 1.—Numbers of tree species with fruit used by bears each month in semievergreen forest (SEF) and mixed deciduous forest (MDF), Thung Yai Naresuan Wildlife Sanctuary, Thailand, 2001– 2003.

forest, both bear species concentrated on those that were, notably Beilschmiedia, Magnolia, and Ficus. Signs of black bears and sun bears occurred on trees in almost all major families (96%) and genera (80%), and we detected no differences in the ratio of black bear : sun bear climbing activity among tree families (Fisher exact tests, P . 0.27) or genera (P . 0.44) in any habitat. Post hoc power analyses (with a ¼ 0.2) indicated that probabilities of detecting differences had they existed were reasonably high for these tests (families: power ¼ 0.63–0.76; genera: power ¼ 0.90– 0.98). Thus, there was little evidence that 1 bear species was more associated with use of a particular taxa of fruit. Contrary to the expectation that, if the species partitioned resources, their fruit-tree selection indexes would be negatively correlated, we found either no relationship or positive correlations between species at the family (evergreen: r ¼ 0.9, P , 0.0001, n ¼ 14; deciduous: r ¼ 0.42, P ¼ 0.16, n ¼ 13) and genus level (evergreen: r ¼ 0.04, P ¼ 0.82, n ¼ 27; deciduous: r ¼ 0.16, P ¼ 0.48, n ¼ 22). Power of the 3 statistically insignificant tests to detect differences was low at genus level (,0.47) but high for deciduous forest families (0.79). There were too few data to conduct correlations in montane forest, but black bear signs were predominant there (Fig. 2), accounting for 83% of all identified claw marks. Mean DBH of trees climbed by black bears and sun bears did not differ in any forest type (P . 0.09), although the statistical power of this test was low (0.45). Both species climbed large trees: average black bear climbed trees were 44– 58 cm DBH (range of means for the 3 habitats), and sun bears were 50–66 cm DBH. The 2 species showed no difference in selection for fruit crop sizes (v22 ¼ 1.62, P ¼ 0.45), although the statistical power of this comparison was low (0.43). About 70% of fruiting trees climbed by each bear species had low or moderate fruit crops; the remainder had abundant crops.

DISCUSSION Niche overlap.—Despite the co-occurrence of black bears and sun bears in mainland Southeast Asia since the middle

FIG. 2.—Use (6SE) of frequently climbed tree families by sun bears and black bears in 3 forest types in Thung Yai Naresuan Wildlife Sanctuary, 2001–2003 (arranged by increasing sun bear use). Bears were identified from claw marks on climbed trees. ‘‘Bear sp.’’ represents older marks not identified to species. Inset shows available tree density in the forest, with families arranged in same order as main graph.

Pleistocene, we found little niche partitioning and a remarkable sharing of fruit resources between these species. Our results did not support our initial hypotheses that black bears and sun bears coexisted because of their complementary use of fruit versus insects, use of different kinds of fruit, or partitioning of fruit crop sizes. Sun bears consumed proportionately more insects than did black bears, but both bear species relied primarily on fruits, so if food was limiting, it seemed likely that both species were limited by fruits (counter to our 1st

February 2013

9

STEINMETZ ET AL.—FOOD HABITS OF SYMPATRIC TROPICAL BEARS

TABLE 3.—Results of linear regression analysis of proportion of trees climbed by bears (logit-transformed) on available tree density (logtransformed) in 3 forest types, Thung Yai Naresuan Wildlife Sanctuary, 2001–2003. Data points (n) are numbers of frequently climbed families and genera with claw marks of a species. Oleaceae in semievergreen forest and Euphorbiaceae in mixed deciduous forest were omitted from habitat-specific analyses because the species of bear climbing the trees in these families could not be identified. CI ¼ confidence interval. Forest type, model, and parameter estimates Semievergreen forest n F R2 P Intercept Slope 6 95% CIs Standardized coefficient (b) Mixed deciduous forest n F R2 P Intercept Slope 6 95% CIs Standardized coefficient (b) Montane evergreen forest n F R2 P

Tree families

Tree genera

Black bears

Sun bears

Black bears

Sun bears

14 26.6 0.69 ,0.0005 1.76 0.89 6 0.37 0.83

14 43.96 0.79 ,0.0001 1.30 0.82 6 0.27 0.89

17 71.51 0.83 ,0.0001 1.88 1.0 6 0.25 0.91

24 49.44 0.69 ,0.0001 2.11 0.65 6 0.19 0.83

13 4.75 0.30 0.05 3.07 0.61 6 0.60 0.55

13 6.97 0.38 0.02 2.42 0.83 6 0.69 0.62

18 32.31 0.67 ,0.0001 2.71 0.84 6 0.31 0.82

20 51.1 0.74 ,0.0001 2.52 0.89 6 0.26 0.86

5 0.17 0.05 0.71

5 2.63 0.35 0.25

Not assessed

Not assessed

hypothesis). Notably, dietary similarities between the 2 bear species were highest during periods of diminished fruit availability. Hence, even during crunch times, speciesspecific diets did not diverge. It is clear that interspecific competition for fruit, if operating, presently has little overall influence on fruit selection by bears in Thung Yai, and that species coexistence there is not a result of partitioning the many taxa of fruit-producing trees. The possibility exists that the 2 bear species partitioned resources on a finer scale than we could measure, such as by focusing on different species of fruits within a genus. It seems improbable, though, that the 2 bear species would have vastly different preferences at the species level, but not the genus level, given the similarity of fruits within genera, and that both bears must track the same temporal sequence of fruit availability (see below). Possibly, sun bears, being smaller, foraged in trees (or parts of trees) that black bears were less apt to climb. Although we did not observe partitioning of fruit crop sizes, the power of our tests was inadequate to reject this hypothesis with confidence. Even if such subtle differences existed, the paramount finding of our study was that the 2 species had similar fruit-dominated diets and fed in the same types of fruit trees in the same habitats. Indeed, black bears and sun bears not only selected similar fruit taxa, they also shared preferences for the same families and constituent genera. We considered whether misclassification of species-specific signs could have contributed to the perceived high dietary overlap between these bears. The claw mark classification method we used, although accurate for adult animals (5% misclassification), may misclassify some black bear cubs as

adult sun bears, resulting in inflation of the sun bear sample by as much as 8% (Steinmetz and Garshelis 2008). Although this potential error might have affected the overall relative proportions of black bear and sun bear signs, there was no reason to believe that this would be manifested differently in different tree taxa; hence, the amount of dietary overlap would not be affected. Moreover, in montane forest, the already few sun bear–sized marks, if categorized instead as black bear cubs, would strengthen our conclusion that this habitat was dominated by black bears. More than one-half of the claw marks could not be identified to bear species (Fig. 2) because they were too old. Although this limits the time frame of our investigation, it should not have affected our results, given that the distribution of older marks (for both species combined) across tree taxa was similar to that of contemporary marks (Fig. 2). Moreover, numerous observations from other times and places show that the tendency for these bear species to focus on the same fruits is widespread across the region (Table 2). Black bears and sun bears in Thung Yai were both primarily frugivorous throughout the year; insects were of secondary importance. Insects were probably even less important than indicated by our data, because insect colonies had far less biomass than a tree full of fruit, yet both were counted as a single feeding event (Table 1). Another study of these bear species elsewhere in Thailand documented a similar preponderance of foraging signs on fruit versus insects (Ngoprasert et al. 2011). Likewise, in sites with abundant fruit on Borneo and Sumatra, occupied only by sun bears, the proportion of fruits and insects in the diet (based on both sign and scat composition) was similar

10

Vol. 94, No. 1

JOURNAL OF MAMMALOGY

FIG. 3.—Proportional use of frequently climbed tree families by sun bears and black bears in relation to abundance of these trees in the forests of Thung Yai Naresuan Wildlife Sanctuary, 2001–2003. Abbreviations are first 4 letters of family name (see Table 2 for full name). Typical nutrient type of fruits in each family designated as lipid-rich (L) or sugar-rich (S).

to that in Thung Yai (Augeri 2005). The lower proportion of insect foraging signs in all of these studies was not an artifact of poorer detection rates, because where bears foraged mostly on insects, insect foraging signs were readily observed and much more prevalent than climbed trees (Fredriksson et al. 2006;

Wong et al. 2002). The intensive tree climbing observed in Thung Yai might reflect a strategy where bears attempt to exploit high concentrations of fruits before they have been thinned by numerous arboreal competitors such as birds, squirrels, primates, civets, and bats (Augeri 2005; Fleming 1979).

TABLE 4.—Indexes of niche breadth (Levins’ standardized) and overlap for Asiatic black bears and sun bears using frequently climbed fruit-tree families and genera in 3 forest types, Thung Yai Naresuan Wildlife Sanctuary, 2001–2003. Genera indexes were calculated separately for annual high and low fruit availability periods. Niche breadth

Niche overlap

Taxonomic resolution

Forest type

Black bear

Sun bear

Pianka’s index

Hurlbert’s index

Families

Semievergreen Mixed deciduous Montane evergreen Semievergreen High-fruit period Low-fruit period Mixed deciduous High-fruit period Low-fruit period

0.50 0.61 0.59 0.49 0.51 0.58 0.49 0.49 0.74

0.49 0.64 0.64 0.59 0.45 0.71 0.67 0.70 0.50

0.91 0.77 0.78 0.71 0.57 0.94 0.59 0.58 0.75

2.10 1.21 0.79 1.90 1.73 1.72 0.99 0.98 1.12

Genera

February 2013

STEINMETZ ET AL.—FOOD HABITS OF SYMPATRIC TROPICAL BEARS

Continual fruit availability appears to sustain bear frugivory throughout the year in Thung Yai. At least 4 bear food-tree species, and at least 6 trees/ha, bore fruit each month (Fig. 1). Nonetheless, fruit crop size varies from year to year, and some or all trees of a given species may forgo fruiting altogether (Turner 2001; van Schaik et al. 1993). Accordingly, we observed that only about one-fourth of climbed trees had been reclimbed, probably reflecting variable fruit production among individual trees each year, or a general abundance of fruiting trees. Signs of insect foraging in Thung Yai were mostly from sun bears, consistent with previous reports that this species is more insectivorous than black bears (Lekagul and McNeely 1988). Whereas sun bears in Thung Yai ate predominantly fruit, in some parts of Borneo they subsisted predominantly on insects, except during mast fruiting events, when they were almost completely frugivorous (Fredriksson et al. 2007; Wong et al. 2004). These divergent diets might result from underlying differences in climate and tree phenology between Sundaic and mainland Southeast Asia. In aseasonal rain forests in Borneo, most tree species fruit synchronously during masting events, every 3–7 years (Primack and Corlett 2005). In contrast, the seasonal climate that prevails in mainland Southeast Asia promotes asynchronous phenology among tree species, yielding relatively consistent supplies of fruit (Primack and Corlett 2005; Fig. 1). The size and contiguity of forest habitat also may influence bear foraging: sun bears in large unfragmented forests in Borneo and Sumatra fed mostly on fruit, although termites were still a much larger part of their diet than in Thung Yai (Augeri 2005). Patterns of resource use.—Use of trees by bears (climbed trees/ha) increased with abundance (trees/ha) in different families and genera; that is, bears fed more on common foods than on rare foods. Conversely, the proportion of trees of each family and genera that were climbed was inversely related to their abundance, so rarer foods (e.g., Elaeocarpaceae [Elaeocarpus spp.] and Myrtaceae [Syzygium spp.]) were used at a disproportionately high rate (Fig. 3). This relationship, known as negative frequency-dependent selection (Greenwood and Elton 1979), is common among herbivorous animals in both tropical and temperate environments (Dearing and Schall 1992; Edenius et al. 2002; Skarpe et al. 2000), and is thought to occur as a result of animals seeking a mixed, nutritionally balanced diet, or attempting to avoid or counteract dangerous levels of toxic compounds (Begon et al. 2006; Westoby 1978). The mixed diets, broad fruit niches, and disproportionate use of rare fruiting species that we observed also probably relates to temporal availability of fruits in Thung Yai. One-half of the low-density families in deciduous forest (Anacardiaceae and Combretaceae) and evergreen forest (Moraceae, Anacardiaceae, and Icacinaceae) tended to fruit during the low fruiting period, and thus may have attracted higher than expected use. Such taxa may function in a keystone role, providing fruit during periods of general scarcity. The fig family, Moraceae, is a well-known keystone resource for tropical frugivores

11

(Terborgh 1986), including sun bears (Fredriksson et al. 2006; Wong et al. 2005). Niche breadths of bears tended to widen during the annual period of fruit scarcity (Table 4), consistent with predictions from foraging theory (Perry and Pianka 1997). Convergence of bear diets in Thung Yai may be largely a result of phenologic availability: diets of both bear species tracked the same fruiting sequence through time. Resource tracking in seasonally fluctuating environments promotes overlapping diets among other coexisting species (Fleming et al. 1987; Rotenberry 1980). Coexistence.—The broad fruit niches and extensive niche overlap of black bears and sun bears indicate the potential for intense competition, if fruit is a limiting resource. We propose 5 nonexclusive mechanisms that may facilitate this ecological overlap The 1st mechanism involves differential use of some resources. Each bear species used 1 resource nearly exclusively: sun bears were the main consumer of insects, and black bears predominated in fruit-rich montane forest. If different species have some exclusive resources in addition to shared ones, then a certain density of each species can be maintained that is unaffected by competition from the other species, generating conditions for coexistence (Ritchie 2002). This mechanism appears to promote coexistence of other sympatric carnivores with similar diets (Azevedo et al. 2006; Kitchener et al. 1999; Neale and Sacks 2001). Even small niche differences can promote coexistence if species have similar average fitness (Adler et al. 2007). The 2nd mechanism is negative frequency-dependent resource selection. Both bear species cropped a lower proportion of common fruit trees than rarer fruit trees (Fig. 3), thereby leaving a potential surplus for the other species that could lessen competition. For example, sun bears climbed only 4% (2 trees/ha) of available Lauraceae trees (Fig. 3), leaving potentially 45 trees/ha as food for black bears. The 3rd mechanism involves temporal variation in resources. With temporally fluctuating food resources, each species might fare better than the other under certain conditions. That is, instead of partitioning resources, species partition the variation in resource availability (Chesson 1986). Variations over the long term could enable each bear species to periodically flourish more than the other. Each species stores, in a sense, the competitive gains from good years in order to persist in the bad years (hence, this mechanism is called the ‘‘storage effect’’ [Chesson 2008]). The relative abundance of these 2 bear species varies dramatically across Southeast Asia, in part due to variations in habitat (Steinmetz 2011), so it follows that temporal variations within a site would similarly change their relative competitive advantages. Notably, black bears do not range into the most southerly parts of Thailand, where fruiting failures may be too long and too common for them to persist, whereas sun bears live there and farther south, able to survive fruiting failures by feeding on insects. If black bears outcompete sun bears in Thung Yai when fruit is abundant, but suffer more when fruit is sparse, the 2 species could coexist in a stochastic dynamic equilibrium.

12

Vol. 94, No. 1

JOURNAL OF MAMMALOGY

The 4th mechanism involves spatial variation in resources. Fruit availability is distributed in patches of various sizes (from clumps of trees to single trees), and the size and spacing of these patches may influence which bear species benefits more. Wider-ranging black bears might be better able to find widely spaced food patches, enabling them to choose the densest food clumps, and then move on to the next patch. Sun bears, with smaller home ranges, may exploit smaller patches more thoroughly and efficiently. This ‘‘cream-skimmer–crumbpicker’’ trade-off (Brown 1989) provides a means for both species to live off the same resources. The 5th mechanism involves abundant resources. Extensive niche overlap between species may reflect an abundance of food (Sale 1974; Wiens 1993); 2 species may thus share resources without really competing (Smith and Smith 2001). Accordingly, species richness of mammalian frugivores is closely correlated with fruit productivity in tropical environments (Kay et al. 1997). Fruit resources in Thung Yai were diverse and abundant: density of bear food trees in Thung Yai was up to 70% higher (in evergreen forest) than at a site on Borneo inhabited by sun bears (Fredriksson et al. 2007), yet Bornean sun bears had very small home ranges (Fredriksson 2012; Wong et al. 2004), indicative of plentiful food. Moreover, fruit in Thung Yai was consistently available, season to season and year to year, and about half the trees in the forest in each habitat were bear food trees. Nonetheless, an abundance of food would not mollify competition if bear density was correspondingly high, such that individuals significantly constrained foraging opportunities of their neighbors. That is, competition is a function of food availability per individual, which relates both to food abundance and animal abundance. In Thung Yai, not only was food abundant, but also, during the time of our study, bear populations may have been depressed below carrying capacity by previous poaching (Steinmetz et al. 2006); thus food may not have been limiting. It is possible that as bear populations increase, greater competition will induce increased dietary divergence. Future examinations of foraging and habitat use where bear populations were recovering would clarify the importance of niche differences and trade-offs for enabling these species to cooccur at small spatial scales. Should densities of both bear species increase at Thung Yai, we predict sun bears will be less inclined to forage in dense (defensible) fruit patches that attract numerous black bears, and will increase their consumption of less-preferred insects; in contrast, black bears should show little shift in diet or habitat use (Steinmetz et al. 2011). Nevertheless, the outcomes of such competitive interactions are difficult to predict; in North America, brown bears have physically excluded the smaller American black bears from some resources, but in other cases black bears also may have excluded brown bears through exploitation competition (Belant et al. 2010; Mattson et al. 2005; McLellan 2011). The uncertainty of the results of such dynamic interspecific interactions has ramifications for the conservation in Southeast Asia of these sympatric bears, which are already constrained by severe habitat loss.

ACKNOWLEDGMENTS We thank the Department of National Parks, Wildlife, and Plant Conservation, and the National Research Council of Thailand, for granting permission for this research. We are grateful to K. Kaewplang, Thiha, and E. Webb for assistance in the field, and to M. van de Bult and J. F. Maxwell for superlative plant identification guidance. RS was supported by a Doctoral Dissertation Fellowship from the Graduate School, and by the Conservation Biology Program, University of Minnesota. We thank D. Smith, E. Cushing, P. Jordan, T. Arnold, and 3 anonymous reviewers for commenting on earlier drafts. We thank the project advisors, S. Maneerat and P. Duangkhae, and previous Superintendents of Thung Yai, P. Sawetmelanon and E. Chirng-saard, for their kind assistance.

LITERATURE CITED ADLER, P. B., J. HILLERISLAMBERS, AND J. M. LEVINE. 2007. A niche for neutrality. Ecology Letters 10:95–104. APPS, C. D., B. N. MCLELLAN, AND J. G. WOODS. 2006. Landscape partitioning and spatial inferences of competition between black and grizzly bears. Ecography 29:561–572. AUGERI, D. M. 2005. On the biogeographic ecology of the Malayan sun bear. Ph.D. dissertation, University of Cambridge, Cambridge, United Kingdom. ` , A. J. AZEVEDO, F. C. C., V. LESTER, W. GORSUCH, S. LARIVIERE WIRSING, AND D. L. MURRAY. 2006. Dietary breadth and overlap among five sympatric prairie carnivores. Journal of Zoology 269:127–135. BASSET, A., AND D. L. ANGELIS. 2007. Body size mediated coexistence of consumers competing for resources in space. Oikos 116:1363– 1377. BEGON, M., C. R. TOWNSHEND, AND J. L. HARPER. 2006. Ecology: from individuals to ecosystems. 4th ed. Blackwell Publishing, Malden, Massachusetts. BELANT, J. L., B. GRIFFITH, Y. ZHANG, E. H. FOLLMANN, AND L. G. ADAMS. 2010. Population-level resource selection by sympatric brown and American black bears in Alaska. Polar Biology 33:31– 40. BELANT, J. L., K. KIELLAND, E. H. FOLLMANN, AND L. G. ADAMS. 2006. Interspecific resource partitioning in sympatric ursids. Ecological Applications 16:2333–2343. BROWN, J. S. 1989. Desert rodent community structure: a test of four mechanisms of coexistence. Ecological Monographs 59:1–20. BULLOCK, S. H., AND J. A. SOLIS-MAGALLANES. 1990. Phenology of canopy trees in a tropical deciduous forest in Mexico. Biotropica 22:22–35. BURST, T. R., AND M. R. PELTON. 1983. Black bear mark trees in the smoky mountains. International Conference on Bear Research and Management 5:45–53. CHESSON, P. L. 1986. Environmental variation and the coexistence of species. Pp. 240–256 in Community ecology (J. Diamond and T. Case, eds.). Harper and Row, New York. CHESSON, P. 2008. Quantifying and testing species coexistence mechanisms. Pp. 119–164 in Unity in diversity: reflections on ecology after the legacy of Ramon Margalef (F. Valladares, et al., eds.). Fundacion BBVA, Bilbao, Spain. CORLETT, R. T. 1998. Frugivory and seed dispersal by vertebrates in the Oriental (Indomalayan) Region. Biological Review 73:413–448. DEARING, M. D., AND J. J. SCHALL. 1992. Testing models of optimal diet assembly by the generalist herbivorous lizard Cnemidophoros murinus. Ecology 73:845–858.

February 2013

STEINMETZ ET AL.—FOOD HABITS OF SYMPATRIC TROPICAL BEARS

EDENIUS, L., G. ERICSSON, AND P. NASLUND. 2002. Selectivity by moose vs the spatial distribution of aspen: a natural experiment. Ecography 25:289–294. ERDBRINK, D. P. 1953. A review of fossil and recent bears of the Old World, with remarks on their phylogeny. Deventer, Netherlands. FLEMING, T. H. 1979. Do tropical frugivores compete for food? American Zoologist 19:1157–1172. FLEMING, T. H., R. BREITWISCH, AND G. H. WHITESIDES. 1987. Patterns of tropical vertebrate frugivore diversity. Annual Review of Ecology and Systematics 18:91–109. FRANCIS, C. M. 2008. A guide to the mammals of Southeast Asia. Princeton University Press, Princeton, New Jersey. FREDRIKSSON, G. M. 2012. Effects of El Ni˜no and large-scale forest fires on the ecology and conservation of Malayan sun bears (Helarctos malayanus) in East Kalimantan, Indonesian Borneo. Ph.D. dissertation, University of Amsterdam, Amsterdam, Netherlands. FREDRIKSSON, G. M., L. S. DANIELSEN, AND J. E. SWENSON. 2007. Impacts of El Ni˜no related drought and forest fires on sun bear fruit resources in lowland dipterocarp forest of East Borneo. Biodiversity and Conservation 16:1823–1838. FREDRIKSSON, G. M., S. A. WICH, AND TRISNO. 2006. Frugivory in sun bears (Helarctos malayanus) is linked to El Nin˜ o–related fluctuations in fruiting phenology, East Kalimantan, Indonesia. Biological Journal of the Linnean Society 89:489–508. FRENCH, A. R., AND T. B. SMITH. 2005. Importance of body size in determining dominance hierarchies among diverse tropical frugivores. Biotropica 37:96–101. GARDNER, S., P. SIDISUNTHORN, AND V. ANUSARNSUNTHORN. 2000. A field guide to forest trees of northern Thailand. Kobfai Publishing Project, Bangkok, Thailand. GARSHELIS, D. L. 2004. Variation in ursid life histories—is there an outlier? Pp. 53–73 in Giant pandas: biology and conservation (D. Lindburg and K. Baragona, eds.). University of California Press, Berkeley. ˇ GRASSMAN, L. I., JR., A. M. HAINES, J. E. JANECKA , AND M. E. TEWES. 2006. Activity periods of photo-captured mammals in north central Thailand. Mammalia 70:306–309. GREENWOOD, J. J. D., AND R. A. ELTON. 1979. Analysing experiments on frequency-dependent selection by predators. Journal of Animal Ecology 48:721–737. HARA, M., ET AL. 2002. The floristic composition of tropical montane forest in Doi Inthanon National Park, northern Thailand, with special reference to its phytogeographical relation with montane forests in tropical Asia. Natural History Research 1:1– 17. HERRERA, C. M. 2002. Seed dispersal by vertebrates. Pp. 185–208 in Plant–animal interactions: an evolutionary approach (C. M. Herrera and O. Pellmyr, eds.). Blackwell Publishing, Malden, Massachusetts. HOLT, R. D. 2001. Species coexistence. Pp. 413–426 in Encyclopedia of biodiversity (S. A. Levin, ed.). Academic Press, Waltham, Massachusetts. HTUN, S. 2006. The status and conservation of bears in Myanmar. Pp. 45–49 in Understanding Asian bears to secure their future (Japan Bear Network, compilers). Japan Bear Network, Ibaraki, Japan. HUYGENS, O. C., AND H. HAYASHI. 2001. Use of stone pine seeds and acorns by Asiatic black bears in central Japan. Ursus 12:47–50. HWANG, M.-S., D. L. GARSHELIS, AND Y. WANG. 2002. Diets of Asiatic black bears in Taiwan, with methodological and geographical comparisons. Ursus 13:111–125.

13

HWANG, M.-S., D. L. GARSHELIS, Y.-H. YU, AND Y. WANG. 2010. Home ranges of Asiatic black bears in the central mountains of Taiwan: gauging whether a reserve is big enough. Ursus 21:81–96. INTERNATIONAL UNION FOR THE CONSERVATION OF NATURE. 2010. IUCN Red list of threatened species. Version 2010.4. www.iucnredlist. org. Accessed 13 November 2010. JORDANO, P. 1992. Fruits and frugivory. Pp. 105–156 in Seeds, the ecology of regeneration on plant communities (M. Fenner, ed.). CAB International, Oxon, United Kingdom. JOSHI, A. R., D. L. GARSHELIS, AND J. L. D. SMITH. 1997. Seasonal and habitat-related diets of sloth bears in Nepal. Journal of Mammalogy 78:584–597. KAY, R. F., R. H. MADDEN, C. VAN SCHAIK, AND D. HIGDON. 1997. Primate species richness is determined by plant productivity: implications for conservation. Proceedings of the National Academy of Sciences 94:13023–13027. KITAMURA, S. 2000. Seed dispersal by hornbills in a tropical rain forest in Khao Yai National Park, Thailand. M.S. thesis, Kyoto University, Kyoto, Japan. KITAMURA, S., ET AL. 2005. A botanical inventory of a tropical seasonal forest in Khao Yai National Park, Thailand: implications for fruit– frugivore interactions. Biodiversity and Conservation 14:1241– 1262. KITAMURA, S., ET AL. 2002. Interactions between fleshy fruits and frugivores in a tropical seasonal forest in Thailand. Oecologia 133:559–572. KITCHENER, A. M., E. M. GESE, AND E. R. SCHAUSTER. 1999. Resource partitioning between coyotes and swift foxes: space, time, and diet. Canadian Journal of Zoology 77:1645–1656. KOIKE, S. 2009. Fruiting phenology and its effect on fruit feeding behavior of Asiatic black bears. Mammal Study 34:47–52. KOIKE, S. 2010. Long-term trends in food habits of Asiatic black bears in the Misaka Mountains on the Pacific coast of central Japan. Mammalian Biology 75:17–28. KREBS, C. 1999. Ecological methodology. HarperCollins, New York. LEIGHTON, M. 1993. Modeling dietary selectivity by Bornean orangutans: evidence for integration of multiple criteria in fruit selection. International Journal of Primatology 14:257–313. LEIGHTON, M., AND D. R. LEIGHTON. 1983. Vertebrate responses to fruiting seasonality within a Bornean rain forest. Pp. 181–196 in Tropical rainforest: ecology and management (S. L. Sutton, T. C. Whitmore, and A. C. Chadwich, eds.). Blackwell Science Publishing, London, United Kingdom. LEKAGUL, B., AND J. A. MCNEELY. 1988. Mammals of Thailand. 2nd ed. Association for the Conservation of Wildlife, Bangkok, Thailand. MANJREKAR, N. 1989. Feeding ecology of the Himalayan black bear (Selenarctos thibetanus Cuvier) in Dachigam National Park. M.c. thesis, Saurashtra University, Rajkot, India. MATTSON, D. J., S. HERRERO, AND T. MERRILL. 2005. Are black bears a factor in the restoration of North American grizzly bear populations? Ursus 16:11–30. MAXWELL, J. F. 1995. Vegetation and vascular flora of the Ban Seneh Pawng area, Lai Wo Subdistrict, Sangklaburi District, Kanchanaburi Province, Thailand. Natural History Bulletin of the Siam Society 43:131–170. MCCONKEY, K., AND M. GALETTI. 1999. Seed dispersal by the sun bear Helarctos malayanus in Central Borneo. Journal of Tropical Ecology 15:237–241. MCLELLAN, B. N. 2011. Implications of a high-energy and low-protein diet on the body composition, fitness, and competitive abilities of

14

JOURNAL OF MAMMALOGY

black (Ursus americanus) and grizzly (Ursus arctos) bears. Canadian Journal of Zoology 89:546–558. NAKHASATHIEN, S., AND B. STEWART-COX. 1990. Nomination of Thung Yai-Huai Kha KhaengWildlife Sanctuary to be a UNESCOWorld Heritage Site. Royal Forest Department, Bangkok, Thailand. NEALE, J. C. C., AND B. N. SACKS. 2001. Resource utilization and interspecific relations of sympatric bobcats and coyotes. Oikos 94:236–249. NGOPRASERT, D., R. STEINMETZ, D. H. REED, T. SAVINI, AND G. A. GALE. 2011. Influence of fruit on habitat selection of Asian bears in a tropical forest. Journal of Wildlife Management 75:588–595. NOZAKI, E., S. AZUMA, T. AOI, H. TORII, T. ITO, AND K. MAEDA. 1983. Food habits of Japanese black bear. International Conference on Bear Research and Management 5:106–109. PERRY, G., AND E. R. PIANKA. 1997. Animal foraging: past, present and future. Trends in Ecology & Evolution 12:360–364. PRIMACK, R., AND R. CORLETT. 2005. Tropical rain forests: an ecological and biogeographical comparison. Blackwell Publishing, Malden, Massachusetts. REID, D., Q. JIANG, Z. TENG, Z. QIN, AND J. HU. 1991. Ecology of the Asiatic black bear (Ursus thibetanus) in Sichuan, China. Mammalia 55:221–237. RITCHIE, M. 2002. Competition and coexistence in mobile animals. Pp. 109–131 in Competition and coexistence (U. Sommer and B. Worm, eds.). Springer-Verlag, Berlin, Germany. ROTENBERRY, J. T. 1980. Dietary relationships among shrubsteppe passerine birds: competition or opportunism in a variable environment? Ecological Monographs 50:93–110. RUNDEL, P. W., AND K. BOONPRAGOB. 1995. Dry forest ecosystems of Thailand. Pp. 93–119 in Seasonally dry tropical forests (S. H. Bullock, H. A. Mooney, and E. Medina, eds.). Cambridge University Press, Cambridge, United Kingdom. SABERWAL, V. 1989. Distribution and movement patterns of the Himalayan black bear (Selenarctos thibetanus Cuvier) in Dachigam National Park, Kashmir. M.S. thesis, Saurashtra University, Rajkot, India. SALE, P. F. 1974. Overlap in resource use, and interspecific competition. Oecologia 17:245–256. SATHYAKUMAR, S., AND S. VISWANATH. 2003. Observations on food habits of Asiatic black bear in Kedarnath Wildlife Sanctuary, India: preliminary evidence on their role in seed germination and dispersal. Ursus 14:99–103. SCHALLER, G. B., Q. T. TENG, K. G. JOHNSON, X. M. WANG, H. M. SHEN, AND J. C. HU. 1989. The feeding ecology of giant pandas and Asiatic black bears in the Tangjiahe Reserve, China. Pp. 212–241 in Carnivore behavior, ecology and evolution (J. L. Gittleman, ed.). Cornell University Press, New York. SCHOENER, T. W. 1974. Resource partitioning in ecological communities. Science 185:27–39. SIEPIELSKI, A. M., AND M. A. MCPEEK. 2010. On the evidence for species coexistence: a critique of the coexistence program. Ecology 91:3153–3164. SKARPE, C., R. BERGSTROM, A. L. BRATAN, AND K. DANELL. 2000. Browsing in a heterogeneous savanna. Ecography 23:632–640. SMITH, R. L., AND T. M. SMITH. 2001. Ecology and field biology. Benjamin Cummings, San Francisco, California. SNOW, D. W. 1981. Tropical frugivorous birds and their food plants: a world survey. Biotropica 13:1–14. SREEKUMAR, P. G., AND M. BALAKRISHNAN. 2002. Seed dispersal by the sloth bear (Melursus ursinus) in South India. Biotropica 34:474– 477.

Vol. 94, No. 1

STEINMETZ, R. 2011. Ecology and distribution of sympatric Asiatic black bears and sun bears in the tropical dry forest ecosystem of Southeast Asia. Pp. 249–273 in Dry forests of Asia: conservation and ecology (W. McShea, S. Davies, and N. Bhumpakphan, eds.). Smithsonian Institution Press, Washington, D.C. STEINMETZ, R., W. CHUTIPONG, AND N. SEUATURIEN. 2006. Collaborating to conserve large mammals in Southeast Asia. Conservation Biology 20:1391–1401. STEINMETZ, R., AND D. L. GARSHELIS. 2008. Distinguishing Asiatic black bears and sun bears by claw marks on climbed trees. Journal of Wildlife Management 72:814–821. STEINMETZ, R., AND D. L. GARSHELIS. 2010. Estimating ages of bear claw marks in Southeast Asian tropical forests as an aid to population monitoring. Ursus 21:143–153. STEINMETZ, R., D. L. GARSHELIS, W. CHUTIPONG, AND N. SEUATURIEN. 2011. The shared preference niche of sympatric Asiatic black bears and sun bears in a tropical forest mosaic. PLoS ONE 6:e14509. ˜ , AND J. A. AHUMADA. 2000. Influence STEVENSON, P. R., M. J. QUINONES of fruit availability on ecological overlap among four Neotropical primates at Tinigua National Park, Colombia. Biotropica 32:533–544. STILES, E. W. 1993. The influence of pulp lipids on fruit preference by birds. Pp. 227–235 in Frugivory and seed dispersal: ecological and evolutionary aspects (T. H. Fleming and A. Estrada, eds.). Kluwer Academic Publishers, Dordrecht, Netherlands. TAKAHASHI, K., T. SHIOTA, H. TAMATANI, M. KOYAMA, AND I. WASHITANI. 2008. Seasonal variation in fleshy fruit use and seed dispersal by the Japanese black bear (Ursus thibetanus japonicus). Ecological Research 23:471–478. TERBORGH, J. 1986. Keystone plant resources in the tropical forest. Pp. 330–344 in Conservation biology: the science of scarcity and diversity (M. E. Soul´e, ed.). Sinauer Associates, Inc., Publishers, Sunderland, Massachusetts. THAI DEPARTMENT OF METEOROLOGY. 2005. Weather. Bangkok, Thailand. 2005. http://www.tmd.go.th/en/. Accessed 15 August 2005. TOUGARD, C. 2001. Biogeography and migration routes of large mammal faunas in South-East Asia during the Late Middle Pleistocene: focus on the fossil and extant faunas from Thailand. Palaeogeography, Palaeoclimatology, Palaeoecology 168:337–358. TURNER, I. M. 2001. The ecology of trees in the tropical rain forest. Cambridge University Press, Cambridge, United Kingdom. VAN DE BULT, M. 2003. The vegetation and flora of the Western Forest Complex. Western Forest Complex Ecosystem Management Project. National Park, Wildlife and Plant Conservation Department, Bangkok, Thailand. VAN SCHAIK, C. P., J. W. TERBORGH, AND S. J. WRIGHT. 1993. The phenology of tropical forests: adaptive significance and consequences for primary consumers. Annual Review of Ecology and Systematics 24:353–377. VAN STEENIS, C. G. G. J. 1978. Flora Malesiana. Series 1. Vol. 8. P. Noordhoff, Leyden, Netherlands. VINITPORNSAWAN, S., R. STEINMETZ, AND B. KANCHANASAKHA. 2006. The status of bears in Thailand. Pp. 50–56 in Understanding Asian bears to secure their future (Japan Bear Network, compilers). Japan Bear Network, Ibaraki, Japan. WELCH, C. A., J. KEAY, K. C. KENDALL, AND C. T. ROBBINS. 1997. Constraints on frugivory by bears. Ecology 78:1105–1119. WESTOBY, M. 1978. What are the biological bases of varied diets? American Naturalist 112:627–631. WHEELWRIGHT, N. T., W. A. HABER, K. G. MURRAY, AND C. GUINDON. 1984. Tropical fruit-eating birds and their food plants: a survey of Costa Rican lowland moist forest. Biotropica 16:173–192.

February 2013

15

STEINMETZ ET AL.—FOOD HABITS OF SYMPATRIC TROPICAL BEARS

WIENS, J. A. 1993. Fat times, lean times, and competition among predators. Trends in Ecology & Evolution 8:348–349. WONG, S. T., C. W. SERVHEEN, AND L. AMBU. 2002. Food habits of Malayan sun bears in lowland tropical forests of Borneo. Ursus 13:127–136. WONG, S. T., C. W. SERVHEEN, AND L. AMBU. 2004. Home range, movement and activity patterns, and bedding sites of Malayan sun bears Helarctos malayanus in the rainforest of Borneo. Biological Conservation 119:169–181. WONG, S. T., C. W. SERVHEEN, L. AMBU, AND A. NORHAYATI. 2005. Impacts of fruit production cycles on Malaysian sun bears and bearded pigs in lowland tropical forest of Sabah, Malaysian Borneo. Journal of Tropical Ecology 21:627–639.

YAMAZAKI, K., C. KOZAKAI, S. KOIKE, H. MORIMOTO, Y. GOTO, AND K. FURUBAYASHI. 2012. Myrmecophagy of Japanese black bears in the grasslands of the Ashio area, Nikko National Park, Japan. Ursus 23:52–64. YASUDA, M., S. MIURA, N. ISHII, T. OKUDA, AND N. A. HUSSEIN. 2005. Fallen fruits and terrestrial vertebrate frugivores: a case study in a lowland tropical rainforest in peninsular Malaysia. Pp. 151–173 in Seed fate: predation, dispersal, and seedling establishment (P. M. Forget, J. E. Lambert, P. E. Hulme, and S. B. Venderwall, eds.). CAB International, Wallingford, United Kingdom.

Submitted 15 October 2011. Accepted 16 August 2012. Associate Editor was Roger A. Powell.

APPENDIX I Plant taxa used (i.e., climbed or eaten) by sun bears and black bears, as determined from bear signs in 3 forest types in Thung Yai Naresuan Wildlife Sanctuary, 2001–2003. SEF: semievergreen forest; MDF: mixed deciduous forest; MEF: montane evergreen forest. Values in parentheses are percentages of all climbed trees in a forest type from that family (values do not apply to genera or species, or to nontree families). No value indicates that no trees from that family were used in that forest type. Families in boldface type were frequently climbed (accounting for .2% of climbed trees in a habitat). Bear signs Climbed trees Forest type in which taxa used (x) Family

Genera

Species

SEF

Aceraceae Alangiaceae Anacardiaceae

Acer Alangium Lannea Mangifera Rhus Semecarpus Spondias Cyathocalyx Polyalthia Polyalthia Polyalthia Pseuduvaria Unidentified Unidentified Alstonia Alocasia Schefflera Ilex Fernandoa Markhamia Stereospermum Stereospermum Bombax Canarium Garuga Viburnum Euonymus Euonymus Lophopetalum Terminalia Terminalia Mastixia Tetrameles Dillenia Dillenia

oblongum kurzii coromandelica caloneura chinensis cochinchinensis pinnata cf. martabanicus simiarum viridis sp. sp. sp. 1 sp. 2 rostrata sp. pueckleri sp. adenophylla stipulata colais neuranthum anceps subulatum pinnata punctatum cochinchinensis colonoides wallichii bellerica chebula cf. euonymoides nudiflora ovata parviflora

x (0.3) x (0.5) (2.4) x

Annonaceae

Apocynaceae Araceaea Araliaceae Aquifoliaceae Bignoniaceae

Bombacaceae Burseraceae Caprifoliaceae Celastraceae

Combretaceae Cornaceae Datiscaceae Dilleniaceae

x x x (4.2) x x x x x x x (1.0)

MDF

MEF

Broken branch

Multiple climbing events

Fresh marks on fruiting tree

Other Claw marks only

Scat, feeding debris

x x x

x (4.5) x x x x

x x x

x

x

x

x x x x x x x x

x (0.5)

x x (1.5)

x

x (7.6)

x

x x x (0.6) x (3.8) x

x x

(0.6)

x x x

x

x (0.3)

x (0.8) x x (0.3) x (4.7) x x x (0.8)

x

x x (5.1) x

x x x x

x

x x x x

x (0.3) x (0.5)

x x x (8.9)

x (0.5)

x x

x x

16

Vol. 94, No. 1

JOURNAL OF MAMMALOGY

APPENDIX I.—Continued. Bear signs Climbed trees Forest type in which taxa used (x) Family

Genera

Species

Dipterocarpaceae

Shorea Shorea Diospyros Elaeocarpus Elaeocarpus Elaeocarpus Elaeocarpus Elaeocarpus Elaeocarpus Elaeocarpus Elaeocarpus Antidesma Aporosa Baccaurea Balakata Bischofia Bridelia Bridelia Bridelia Macaranga Mallotus Mallotus Trewia Unidentified Castanopsis Castanopsis Castanopsis Lithocarpus Lithocarpus Lithocarpus Lithocarpus Lithocarpus Quercus Quercus Quercus Casearia Casearia Calophyllum Garcinia Altingia Apodytes Gomphandra Gomphandra Platea Juglandaceae Callicarpa Gmelina Premna Premna Vitex Vitex Alseodaphne Beilschmiedia Cinnamomum Cinnamomum Cinnamomum Cinnamomum Cinnamomum

obtusa siamensis ehretioides or glandulosa cf. griffithii prunifolius cf. rugosa stipularis sp. 1 sp. 2 sp. 3 sp. 4 bunius dioica ramiflora baccata javanica anceps glauca sp. denticulata paniculatus sp. nudiflora sp. tribuloides sp. 1 sp. 2 aggregatus elegans fenestratus polystachyus truncatus brandisiana kerrii semiserrata grewiifolia mekongii sp. sp. excelsa dimidiata quadrifida tetrandra latifolia Engelhardia arborea arborea latifolia villosa peduncularis quinata nigrescens sp. caudatum iners porrectum sp. 1 sp. 2

Ebenaceae Elaeocarpaceae

Euphorbiaceae

Fagaceae

Flacourtiaceae Guttiferae Hammamelidaceae Icacinaceae

Labiatae

Lauraceae

SEF

(2.1) x x x x x x x (3.7)

MDF x (3.8) x x (0.6)

MEF

Broken branch

Multiple climbing events

Fresh marks on fruiting tree

Claw marks only

Scat, feeding debris

x x x x (1.5)

x x x x x x x

x (2.5) x

x x

x x

x x x

x

x

x x

x (6.4)

x x x x x x x x x x x

x x

x x x x (7.9) x x

Other

(29.2)

x

x x x

x x x x x

x x

x

x x x

x

x

x x x x

x x x

x x (1.0)

x

x x x

x x

x x

x (3.1) x x (0.3) x (4.7) x

x x x

(1.5)

x x x x

x x spicata (0.8)

x x (21.5) x x x x

x (0.3) x (11.5) x x x x x (3.2)

x x

x x x

x x

(32.3)

x x x

x

x x

x x x x x

x x x

x

February 2013

17

STEINMETZ ET AL.—FOOD HABITS OF SYMPATRIC TROPICAL BEARS

APPENDIX I.—Continued. Bear signs Climbed trees Forest type in which taxa used (x) Family

Lecythidaceae Leguminosae

Lythraceae Magnoliaceae

Meliaceae

Moraceae

Myristicaceae

Myrtaceae

Oleaceae Palmaea

Genera

Species

SEF

Cryptocarya Dehaasia Litsea Litsea Litsea Nothaphoebe Phoebe Phoebe Phoebe Phoebe Unidentified Unidentified Unidentified Barringtonia Albizia Callerya Cassia Dalbergia Dalbergia Erythrina Erythrina Millettia Pterocarpus Unidentified Lagerstroemia Magnolia Magnolia Michelia Michelia Aglaia Chisocheton Chukrasia Dysoxylum Dysoxylum Sandoricum Unidentified Artocarpus Artocarpus Ficus Ficus Ficus Ficus Ficus Ficus Musaceaea Horsefieldia Horsefieldia Knema Syzygium Syzygium Syzygium Syzygium Syzygium Syzygium Syzygium Syzygium Chionanthes (Rattan)

sp. cuneata garrettii cf. salicifolia sp. umbelliflora cathia lanceolata cf. paniculata sp. sp. 1 sp. 2 sp. 3 acutangula chinensis atropurpurea fistula cf. cana sp. subumbrans sp. brandisiana macrocarpus sp. tomentosa henryi liliifera baillonii champaca spectabilus sp. tabularis excelsum sp. koetjape sp. lakoocha sp. benjamina cf. lacor variegata virens sp. 1 sp. 2 Musa glabra sp. sp. albiflora cumini megacarpa sp. 1 sp. 2 sp. 3 sp. 4 sp.5 ramiflorus sp.

x x x x x x x x

MDF

MEF

Broken branch

Fresh marks on fruiting tree

x

x x

x

x x x

x x x x x

x

x x x

x

x (12.2)

x x x

x x

x x x

x x

x

x x

x x x

x (1.9)

x x

x

x x x

x (1.3)

(1.5) x

(0.6)

x

x x x x x

x x x x

x (1.9) x x x x x

x x x x

x x x x x

(4.5) x

(9.2)

x x

x

x

x x x

x x x x x x (2.1)

xb

x x x

x sp. x (3.4) x x x (1.6)

xb

x

x x x (2.6) x x

Scat, feeding debris

x x

x x x

x x (0.8) x x x (5.0) x x (8.1) x x x

Claw marks only x x

x x x x (0.8) (1.3) x

Multiple climbing events

Other

x

x

x

x x

x x (0.6)

x x

18

Vol. 94, No. 1

JOURNAL OF MAMMALOGY

APPENDIX I.—Continued. Bear signs Climbed trees Forest type in which taxa used (x) Family Pandanaceaea Polyosmaceae Rhamnaceae Rosaceae Rubiaceae

Sabiaceae Sapindaceae

Sapotaceae Sonneratiaceae Sterculiaceae

Symplocaceae

Theaceae

Tiliaceae

Ulmaceae Unidentified Total climbed trees a b

Genera

Species

Pandanus Polyosma Ziziphus Eriobotrya Prunus Adina Canthium Wendlandia Meliosma Dimocarpus Lepisanthes Litchi Mischocarpus Nephelium Schleichera Unidentified Duabanga Firmiana Pterocymbium Pterospermum Sterculia Sterculia Sterculia Symplocos Symplocos Symplocos Camellia Pyrenaria Ternstroemia Unidentified Berrya Berrya Colona Grewia Gironniera

sp. elongata rugosa bengalensis cerasoides diversifolia sp. tinctoria simplicifolia longan sp. chinensis pentopetalus hypoleucum oleosa sp. grandiflora colorata macranthum cinnamonemum pexa urena sp. macrophylla sumnita sp. connata garrettiana gymnanthera sp. cf. cordifolia mollis cf. winitii eriocarpa nervosa

Not a tree family. Seeds in scat identified only to genus or family.

SEF

MDF

MEF

Broken branch

Multiple climbing events

Fresh marks on fruiting tree

Other Claw marks only

Scat, feeding debris x

x (1.5) x (1.8) x (0.3) x

x (0.6) (0.6) x (0.6)

x x x

x

x x

x x (0.3) x (5.2) x

x

x

x x

x (7.6)

x

x

x

x

x

x

x

x x

x x

x

x x

x x

xb x (1.0) x (2.4) x x x x x (1.0) x x (0.8) x

x (0.6)

x x

x (0.3) (6.5) 412

x x x x

(0.6) x

x (1.6)

x x x x x

x (18.5) x x x

(7.0) x x x (3.8) 168

x

x x x x x

x x x (1.5) (1.5) 68

x x x

x x x x