Integrative Zoology 2013; 8: 417–426
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
doi: 10.1111/1749-4877.12056
ORIGINAL ARTICLE
Southeast Asian primate communities: the effects of ecology and Pleistocene refuges on species richness Heather Hassel-Finnegan,1,2 Carola Borries,1 Qing Zhao,3 Phaivanh Phiapalath4 and Andreas Koenig1 1
Department of Anthropology, Stony Brook University, Stony Brook, NY, USA, 2Biology Department, Swarthmore College, Swarthmore, PA, USA, 3College of Life Science, Peking University, Beijing, China and 4International Union for Conservation of Nature, Vientiane, Lao PDR
Abstract We examined historical and ecological factors affecting current primate biodiversity in Southeast Asia. In Africa, Madagascar and South America, but not Southeast Asia, primate species richness is positively associated with average rainfall and distance from the equator (latitude). We predicted that Southeast Asia’s non-conformance may be due to the effect of dispersed Pleistocene refuges (locations of constricted tropical forests during glacial maxima which today are at least 305 m in altitude). Based on 45 forested sites (13 on large islands; 32 on the mainland) of at least 100 km2 to minimize recent human impact, we determined correlations between extant primate species richness and rainfall, latitude and supplementary ecological variables, while controlling for refuges and islands. We found that refuge sites had significantly higher primate species richness than nonrefuges (t = –2.76, P < 0.05), and distance from the nearest Pleistocene refuge was negatively correlated with species richness for non-refuge sites (r = –0.51, P < 0.05). There was no difference in species richness between sites on large islands and the mainland (t = –1.4, P = 0.16). The expected positive relationship between rainfall and species richness was not found (r = 0.17, P = 0.28). As predicted, primate species richness was negatively correlated with latitude (r = –0.39, P < 0.05) and positively correlated with mean temperature (r = 0.45, P < 0.05). General linear models indicated that a site’s latitude (F1,38 = 6.18, P < 0.05) and Pleistocene refuge classification (F1,42 = 5.96, P < 0.05) were the best predictors of species richness. Both ecological and historical factors contribute to present day primate species richness in Southeast Asia, making its biodiversity less of an outlier than previously believed. Key words: allopatric speciation, island, latitude, mainland, rainfall, temperature
INTRODUCTION
Correspondence: Carola Borries, Department of Anthropology, Stony Brook University, Stony Brook, NY 11794-4364, USA. Email: [email protected]
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Current patterns of biodiversity are determined by both historical and ecological factors. Historical factors, such as climate change, migrations and continental drift, may influence species distribution and diversification (MacArthur & Wilson 1967; MacArthur 1984;
417
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
H. Hassel-Finnegan et al.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
Davis & Shaw 2001), while ecological factors, such as forest productivity and food availability, may determine whether an area provides a suitable habitat (Rosenzweig & Abramsky 1993). In South America, Africa and Madagascar, primate species richness is positively correlated with rainfall (Reed & Fleagle 1995), which is often considered to be a predictor of forest productivity (Mittlebach et al. 2001). However, this relationship is absent for Asia (Reed & Fleagle 1995; but see Wang et al. 2013). Latitude may also affect forest productivity, as higher levels of solar radiation and less seasonality are expected near the equator. Among nonhuman primates, the expected negative relationship between species richness and distance from the equator is supported in both South America and Africa (Eeley & Lawes 1999; Emmons 1999), but not in Southeast Asia (Emmons 1999). This lack of conformance for Asian areas is potentially related to 2 major confounding factors: island biogeography (Reed & Fleagle 1995; Fleagle 1999) and the distribution of refuges during the Pleistocene (Eudey 1980; BrandonJones 1996; Jablonski et al. 2000). However, tests for the latter have rarely been performed. Based on the theory of island biogeography, it is hypothesized that species richness depends on both the size of a landmass and its distance from a source population (MacArthur & Wilson 1967). Larger landmasses are expected to have greater habitat heterogeneity and more barriers to gene flow; both factors may increase the number of species with time (Williams 1964). Areas closer to source populations are expected to have greater species richness, because of higher rates of immigration than more isolated islands (MacArthur & Wilson 1967). The effects of landmass and isolation on primate species richness are well documented. Primate species– area relationships have been found at a variety of geographic scales. For example, continents with more rainforest area have greater primate species richness than those with less rainforest (Reed & Fleagle 1995). Within both Africa and South America, latitudinal gradients in species richness may be associated with species–area relationships (Eeley & Lawes 1999), as equatorial regions have the greatest landmass (Rosenzweig 1995), as well as the greatest species richness (Eeley & Lawes 1999; Emmons 1999). At a finer scale, larger Southeast Asian islands have greater primate species richness than small islands (Nijman & Meijaard 2008). Isolation may also be an important determinant of primate biodiversity, particularly in Southeast Asia. Isolation affects the rate of immigration, extinction and
418
speciation (MacArthur & Wilson 1967). The land-bridge islands of the Sunda Shelf became isolated due to rises in sea levels approximately 12 000 years ago and many other times in the past 5 million years (Heaney 1986; Lawlor 1986). Even such a recent isolation can have a significant impact on patterns of species richness, because of the potential for faunal collapse due to limited contact with source populations (Heaney 1986). Thus, islands may be predicted to have lower species richness than mainland areas. Pleistocene refuges may have also affected the distribution of Southeast Asian primate species (Eudey 1980; Brandon-Jones 1996; Jablonski et al. 2000). It has been hypothesized that reduced rainfall during the Pleistocene led to temporary separation of previously continuous rainforests, in mid-altitude forest patches surrounded by savannah. Forest taxa later re-expanded their ranges during inter-glacial periods (Mayr 1963). Higher species richness is, indeed, reported within suspected refuges in Southeast Asia for both plants and animals (Rodgers et al. 1982; Pearson & Carroll 2001; Svenning & Flemming 2007). Some suggest that allopatric speciation occurred within refuges (Colyn et al. 1991; Haffer 1997; Abegg & Thierry 2002), although there is controversy as some phylogenies suggest primate speciation predated the Pleistocene epoch (Kay et al. 1997; Collins & Dubach 2000; see also discussion below). During the last glacial maximum, there was a significant increase in Southeast Asian montane forest and savannah, with a corresponding decline in rainforest (Heaney 1991). Pollen and termite-community composition analyses indicate that the pattern of rainforest patches that resulted was quite different from that of other continents, in that refuges were abundant, but geographically dispersed (Heaney 1991; Gathorne-Hardy et al. 2002). In Africa and South America, refuges were rare but more continuous, and concentrated near the equator (Maley 1991; Pennington et al. 2000), where primate species richness is also expected to be high for ecological reasons (Reed & Fleagle 1995; Eeley & Lawes 1999; Emmons 1999). Finally, it has been suggested (Reed & Fleagle 1995; Emmons 1999) that the limited sample size of previous studies and exclusive sampling of insular and peninsular sites affected the results for Southeast Asia. Here, we analyzed a new large dataset, emphasizing mainland sites (all west of the Wallace line), of which many have thus far not been considered. We tested for the relationship between species richness and various ecological variables collected directly at each site. This selection
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Asian primate communities
criterion led to a sample including, besides mainland sites, only sites from the 2 largest islands, Sumatra and Borneo, as well as the Malay Peninsula. We also explored explanations for Southeast Asia’s non-conformance, such as the effect of Pleistocene refuges. We specifically addressed 4 questions related to these topics: 1. Do sites on the largest islands have lower species richness than those on the mainland? 2. Is species richness highest at low latitudes, despite Southeast Asia’s small land area near the equator? 3. Do former refuges have higher species richness than non-refuges? 4. Does species richness decrease with distance from the nearest Pleistocene refuge?
MATERIALS AND METHODS Published and unpublished data from a total of 45 sites (Nisland = 13; Nmainland = 32; details in Fig. S1 and Table S1) were included in this analysis. All sites were protected forested areas of at least 100 km2, chosen to limit the effect of more recent human activity on local primate community composition. We excluded sites over 25° from the equator to limit the influence of extreme seasonality. All data including climate data were collected within the protected areas, rather than from primate distribution maps or wide scale weather databases that may be less reliable. Sites on Java, the Mentawai Islands and Sulawesi could not be included because very few ecological data have thus far been published. The following variables were available for all or most sites: extant primate species richness (N species), distance from the equator (measured as latitude) and annual rainfall (in mm). Because the taxonomy of Asian primates is in flux (Roos et al. 2007; Osterholz et al. 2008; Roos et al. 2008), we followed the species designations of the reference for a given site without questioning it. It has also been shown that taxonomic differences do not necessarily influence zoogeographic analyses of primates in the region (Nijman & Meijaard 2008). We note that if several species were reported to occur in a particular area, they might not always live sympatrically, especially in very large areas, but the resolution of the data did not allow for identifying these cases in our sample, should they exist. Wherever possible, data for the following variables were also included: mean temperature (in °C), mean
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
minimum temperature, mean maximum temperature, tree density (in N/ha) and number of dry months (with an average of < 50 mm precipitation). All variables were tested for normality, using the Kolmogorov–Smirnov test (Sokal & Rohlf 1995). No variables differed significantly from a normal distribution (all “P”s > 0.10). Consequently, correlations were calculated using Pearson’s r for the relationships between all aforementioned variables (Sokal & Rohlf 1995). In addition, assuming that the relationship between rainfall and primate species richness might not be linear, a Lowess localized regression was calculated for mean annual rainfall versus species richness (following Kay et al. 1997). Sites were classified as refuges if they were forested during the last glacial maximum, based on studies of soil, sediments and plant fossils (Adams 1997). These sites are at least 305 m asl today (Maps.com 1999). Refuges are believed to have existed in mid-altitudinal regions where rainfall and temperature may have remained within suitable ranges for sustaining forest primates (Brandon-Jones 1996). For all non-refuges we determined the great circle distance from the center of the site to the center of the nearest refuge. To test for dependency of primate species richness on categorical historical and ecological variables, we used t-tests for independent samples (Sokal & Rohlf 1995) to analyze the individual effects of these variables. To test for multiple dependencies we used 2 general linear models (Grafen & Hails 2002). An overall model testing all variables simultaneously was not possible, because of collinearity between several of the variables (Quinn & Keough 2002; see also Table S2). The first general linear model tested the effect of islands, Pleistocene refuges and rainfall on primate species richness. The second tested the influence of Pleistocene refuges and distance from the equator on primate species richness. All statistical tests were 2-tailed and performed using STATISTICA 6.1 (StatSoft) at an alpha level of 0.05.
RESULTS The distribution of sites ranged from 25°N to 6°S (Fig. S1 and Table S1). Primate species richness ranged from 1 to 12 species overall and was negatively correlated with latitude, decreasing from approximately 8 species at the equator down to approximately 5 species at the most northern sites (Pearson’s r = –0.39, P < 0.05, N = 45; Fig. 1a). At the same time, the number of species was positively correlated with mean temperature (r = 0.45, P < 0.05, N = 20; Fig. 1b). However,
419
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
H. Hassel-Finnegan et al.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
Figure 1 Relationship between primate species richness and (a) latitude (Pearson’s r = –0.39, P < 0.05) as well as (b) mean temperature (r = 0.45, P < 0.05).
Figure 2 Relationships between species richness and mean annual rainfall presented as: (a) linear regression (r = 0.17, P = 0.28) and (b) Lowess localized regression.
species richness did not have a significant relationship with rainfall; species richness was only slightly increasing from approximately 6 to 7 species over the range of rainfall from 1000 to 4500 mm (r = 0.17, P = 0.28, N = 43; Fig. 2a). A Lowess localized regression applied to this dataset did not show a discernible pattern (Fig. 2b). Species richness remains fairly stable except for 2 minima at approximately 1300 and 1700 mm rainfall. Correlations between species richness and all other vari-
ables were not significant: mean maximum temperature, r = 0.42, P = 0.12, N = 15; mean minimum temperature, r = 0.16, P = 0.57, N = 15; tree density r = 0.29, P = 0.54, N = 7; and number of dry months, r = –0.27, P = 0.35, N = 11. With 7.1 primate species for the large island sites, species richness was slightly higher than at mainland sites (mean = 6.0), although the difference was not sta-
420
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Asian primate communities
Figure 4 Relationship between primate species richness and distance from the nearest Pleistocene refuge for non-refuge sites (Pearson’s r = –0.51, P < 0.05).
Figure 3 Mean primate species richness and the influence by: (a) geography (mainland vs large island sites; t = –1.42, P = 0.16) and (b) history (non-refuge vs Pleistocene refuge sites; t = –2.76, P < 0.05). Boxes represent standard errors and whiskers standard deviations.
away it declines to approximately 3 (r = –0.51, P < 0.05; Fig. 4). A first general linear model (Fintercept = 20.13, P < 0.05) found that Pleistocene refuge classification affects species richness (F1,38 = 5.43, P < 0.05), but there was no effect of rainfall (F1,38 = 0.26, P = 0.61) or large island/ mainland classification (F1,38 = 0.73, P = 0.40). A second general linear model (Fintercept = 146.38, P < 0.05) found that both Pleistocene refuge classification (F1,42 = 5.96, P < 0.05) and latitude (F1,42 = 6.18, P < 0.05) had a significant effect on primate species richness. Primate species richness in non-refuge sites decreased from approximately 8 species at the equator down to approximately 4 in the north and south, while primate species in refuge sites showed little change across latitudes (Fig. 5).
DISCUSSION tistically significant (t = –1.42, P = 0.16; Fig. 3a). In contrast, primate species richness was significantly higher in suspected Pleistocene refuges (mean = 7.5) than in non-refuges (mean = 5.6; t = –2.76, P < 0.05; Fig. 3b). For non-refuges, the distance to the nearest refuge was negatively correlated with species richness; close to a refuge the number of species is close to 7, while further
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Our results indicate that distance from the equator, mean temperature and whether a site had been a Pleistocene refuge are the best predictors of the present primate species richness in Southeast Asia. In the current sample, the island effect was absent. Similarly, rainfall was not found to be a reliable predictor of primate species richness in the region.
421
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
H. Hassel-Finnegan et al.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
a
b
Figure 5 Relationship between latitude and history (non-refuge vs Pleistocene refuge) on primate species richness (non-refuges: Pearson’s r = –0.44, P < 0.05; Pleistocene refuges: r = –0.10, P = 0.70).
In the past, several hypotheses have been proposed for why rainfall does not predict species richness in Southeast Asia, including: extreme seasonal and year-toyear variation in rainfall, the presence of local low-species-diversity forests dominated by Dipterocarpaceae, the possibility that very high amounts of rain limit productivity or that areas with low precipitation were underrepresented in the data (Reed & Fleagle 1995; Kay et al. 1997; Gupta & Chivers 1999; Wang et al. 2013). The first hypothesis predicts no relationship between forest productivity and rainfall due to generally similar monsoonal effects across the Southeast Asian region (Kripalani & Kulkarni 1998). The second hypothesis also predicts no relationship between rainfall and forest productivity, due to wide-scale microhabitat heterogeneity, with dipterocarp areas having uniformly lower primate biomass (Marsh & Wilson 1981). The third hypothesis predicts an inverted-U-shaped or saturated curvilinear relationship between forest productivity and rainfall (Kay et al. 1997), as rainfall >2500 mm may lead to soil leaching (Richter & Babbar 1991) or reduced photosynthesis because of cloud cover (Raich 1989). The final hypothesis predicts a positive linear relationship between rainfall and species richness if sites with low rainfall are included in the analysis (Wang et al. 2013). Such a positive relationship is reported by Wang et al. (2013) for the region in an analysis based on 390 polygons defined by overlaying species distribution maps. These expected species richness values were combined with rainfall values averaged from 1288 climate data
422
points. Thus, in contrast to our present study, the analysis was not based on the actual conditions found at each site, neither in terms of primate species richness nor rainfall, and it remains unclear how realistic this approach was. For example, it is possible that sites with low rainfall and within the distribution range of a species may, in fact, not be inhabited by primates. In contrast, we did not find a relationship between rainfall and primate species richness in our sample, neither in a linear nor in a curvilinear fashion. This makes the first 2 hypotheses more likely; that is, seasonal/year-to-year variation in rainfall or the wide distribution of low diversity dipterocarp areas may explain the lack of a relationship between rainfall and primate species richness. It has also been suggested that the geography of Southeast Asia may have biased past results of primate communities, which exclusively sampled insular and peninsular sites (Reed & Fleagle 1995; Emmons 1999). A species–area relationship was, indeed, found based on 218 small Southeast Asian islands (Nijman & Meijaard 2008). This result could not be reproduced in our analysis likely because the island sites included were from Sumatra, Borneo and the Malay Peninsula. These are very large land masses with a high habitat heterogeneity allowing for niche diversification and limited faunal collapse (Heaney 1984), which may have translated into the absence of an island effect in our sample. Latitudinal gradients in species richness are reported among a wide range of taxonomic groups (Fernandes & Price 1988; Kaufman 1995; Blackburn & Gaston 1996; Emmons 1999). There are 2 general causative hypotheses proposed to explain this phenomenon. One suggests that the latitudinal gradient is due to ecological factors. Near the equator, environmental stability, productivity, physical heterogeneity, solar radiation and temperature are highest, while seasonality and aridity are less extreme. All of these conditions are expected to increase biodiversity (Rohde 1992). The other group of hypotheses suggests that latitudinal gradients are a result of species–area relationships. Most early studies on latitudinal gradients focused on South America and Africa, where the continents are roughly diamond shaped, with the greatest landmass (and species richness) near the equator with land area (and species richness) shrinking towards the poles (Rosenzweig 1995). Our data indicate that there is a strong relationship between distance from the equator and primate species richness in Southeast Asia. In addition, while the data at hand do not allow assessment of factors such as physical heterogeneity and solar radiation, distance to the equator is, indeed,
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Asian primate communities
strongly correlated with seasonality and temperature. This lends support to ecological hypotheses, because the geography of Southeast Asia is not diamond shaped. For a variety of taxa, species richness is high in sites that served as Pleistocene refuges (Rodgers et al. 1982; Pearson & Carroll 2001; Svenning & Flemming 2007) and for non-refuges, decreases according to distance from the nearest refuge (Struhsaker 1981). With respect especially to forest-dependent primates, it is assumed that during glacial maxima refuges were isolated forest patches separated by extended savannahs, which were unsuitable habitats for most forest-dependent species (Morley 2000) even though large areas were connected by land bridges due to the low sea levels. Today these refuges identify areas that have been inhabited for the longest, which should be reflected in high richness of forest-dependent species. It is expected that areas outside of refuges had to be repeatedly re-populated with species from the refuges as the source. Evidence from Southeast Asian forest-dependent termites (Gathorne-Hardy et al. 2002) and mammals (Meijaard 2003) suggests that range expansion to newly reforested areas was often incomplete. This scenario is in line with our finding of species richness reducing with distance to refuges (Fig. 4). Debate remains as to whether the high species richness in refuges is a result of allopatric speciation in isolated forest patches or relic species, which may have been widely distributed before the Pleistocene, but did not re-expand their ranges during post-glacial forest reexpansions because of life history or geographic constraints (Colyn et al. 1991; Fjeldsa & Lovett 1997; Haffer 1997; Collins & Dubach 2000). It seems now likely that already at the beginning of the last glacial maximum between 33.0 and 26.5 ka (Clark et al. 2009) most Southeast Asia primate groups existed: of all Asian macaques, the youngest species occurred approximately 0.8–0.7 Ma (Hayasaka et al. 1996; Disotell & Tosi 2007; Chatterjee et al. 2009); the youngest Asian colobine species around 0.6 Ma (Roos et al. 2008; Meyer et al. 2011; Perelman et al. 2011); and the youngest gibbon species around 0.5–0.3 Ma (Matsudaira & Ishida 2010; Thinh et al. 2010; Perelman et al. 2011). However, at the beginning of the last ice age at around 2.6 Ma, not all species in these groups had evolved. Biogeographic evidence, including analysis of wide-scale species distribution maps, supports the possibility of allopatric speciation within refuges (Eudey 1980; Jablonski et al. 2000). Our results indicate that Pleistocene refuges have higher primate species richness than non-refuges. For non-refug-
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
es, species richness decreases with distance from the nearest refuge. However, it is impossible to decide to what extent this is the result of allopatric speciation or a reduced expansion of relic species, or both. Studies of primate communities, such as this one, may help to determine the ecological and historical factors that typify diversity hotspots. Our results indicate that Southeast Asian sites close to the equator and those that served as Pleistocene refuges have the greatest primate species richness. High species richness near the equator may be due to environmental stability, high levels of primary productivity, high levels of solar radiation and high temperatures. High species richness in and near Pleistocene refuges suggests that there may have been allopatric speciation of primates within these areas, or relic species that never re-expanded their ranges following the last glacial maximum. The differences in species richness patterns between Southeast Asia and Africa/South America may be due to differences in continental shape and the distribution of Pleistocene refuges.
ACKNOWLEDGMENTS We would like to thank Warren Brockelman, Wanlop Chutipong, Tanja Haus, Meyner Nusalawo, Jonathan O’Brien, Rungnapa Phoonjampa and Thomas Ziegler for sharing unpublished data for individual sites. For helpful comments on a previous draft we thank John Fleagle and our anonymous reviewers.
REFERENCES Abegg C, Thierry B (2002). Macaque evolution and dispersal in insular Southeast Asia. Biological Journal of the Linnean Society 75, 555–76. Adams JM (1997). Global Land Environments since the Last Interglacial. [Cited 16 Apr 2013.] Available from URL: http://www.esd.ornl.gov/ern/qen/nerc. html Blackburn TM, Gaston KJ (1996). Spatial patterns in the species richness of birds in the New World. Ecography 19, 369–76. Brandon-Jones D (1996). The Asian Colobinae (Mammalia: Cercopithecidae) as indicators of Quaternary climatic change. Biological Journal of the Linnean Society 59, 327–50. Chatterjee HJ, Ho SYW, Barnes I, Groves C (2009). Estimating the phylogeny and divergence times of pri-
423
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
H. Hassel-Finnegan et al.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
mates using a supermatrix approach. BMC Evolutionary Biology 9, 259.
Grafen A, Hails R (2002). Modern Statistics for the Life Sciences. Oxford University Press, Oxford.
Clark PU, Dyke AS, Shakun JD et al. (2009). The last glacial maximum. Science 325, 710–4.
Gupta AK, Chivers DJ (1999). Biomass and use of resources in South and Southeast Asian primate communities. In: Fleagle JG, Janson CH, Reed KE, eds. Primate Communities. Cambridge University Press, Cambridge, pp. 38–54.
Collins AC, Dubach JM (2000). Biogeographic and ecological forces responsible for speciation in Ateles. International Journal of Primatology 21, 421–44. Colyn M, Gautier-Hion A, Verheyen W (1991). A reappraisal of paleoenvironmental history in Central Africa: evidence for a major fluvial refuge in the Zaire Basin. Journal of Biogeography 18, 403–7. Davis MB, Shaw RG (2001). Range shifts and adaptive response to quaternary climate change. Science 292, 673–9. Disotell TR, Tosi AJ (2007). The monkey’s perspective. Genome Biology 8, 226. Eeley HAC, Lawes MJ (1999). Large-scale patterns of species richness and species range size in anthropoid primates. In: Fleagle JG, Janson CH, Reed KE, eds. Primate Communities. Cambridge University Press, Cambridge, pp. 191–219. Emmons LH (1999). Of mice and monkeys: primates as predictors of mammal community richness. In: Fleagle JG, Janson CH, Reed KE, eds. Primate Communities. Cambridge University Press, Cambridge, pp. 171–88. Eudey AA (1980). Pleistocene glacial phenomena and the evolution of Asian macaques. In: Lindburg DG, ed. The Macaques: Studies in Ecology, Behavior and Evolution. van Nostrand Reinhold, London, pp. 52– 83. Fernandes GW, Price PW (1988). Biogeographical gradients in galling species richness. Oecologia 76, 161– 7. Fjeldsa J, Lovett JC (1997). Geographical patterns of old and young species in African forest biota: the significance of specific montane areas as evolutionary centres. Biodiversity and Conservation 6, 325–46. Fleagle JG (1999). Primate Adaptation and Evolution, 2nd edn. Academic Press, San Diego. Gathorne-Hardy FJ, Syaukani, Davies RG, Eggleton P, Jones DT (2002). Quaternary rainforest refugia in Southeast Asia: using termites (Isoptera) as indicators. Biological Journal of the Linnean Society 75, 453–66.
424
Haffer J (1997). Alternative models of vertebrate speciation in Amazonia: an overview. Biodiversity and Conservation 6, 451–76. Hayasaka K, Fujii K, Horai S (1996). Molecular phylogeny of macaques: implications of nucleotide sequences from an 896-base pair region of the mitochondrial DNA. Molecular Biology and Evolution 13, 1044– 53. Heaney LR (1984). Mammalian species richness on islands on the Sunda Shelf, Southeast Asia. Oecologia 61, 11–7. Heaney LR (1986). Biogeography of mammals in SE Asia: estimates of rates of colonization, extinction and speciation. Biological Journal of the Linnean Society 28, 127–65. Heaney LR (1991). A synopsis of climatic and vegetational change in Southeast Asia. Climate Change 19, 53–61. Jablonski NG, Whitfort MJ, Roberts-Smith N, Xu Q (2000). The influence of life history and diet on the distribution of catarrhine primates during the Pleistocene in eastern Asia. Journal of Human Evolution 39, 131–57. Kaufman DM (1995). Diversity of New World mammals: universality of the latitudinal gradients of species and bauplans. Journal of Mammalogy 76, 322– 34. Kay RF, Madden RH, van Schaik CP, Higdon D (1997). Primate species richness is determined by plant productivity: implications for conservation. PNAS 94, 13023–7. Kripalani RH, Kulkarni A (1998). Rainfall variability over Southeast Asia–connections with Indian monsoon and ENSO extremes: new perspectives. International Journal of Climatology 17, 1155–68. Lawlor TE (1986). Comparative biogeography of mammals on islands. Biological Journal of the Linnean Society 28, 99–125.
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Asian primate communities
MacArthur RH (1984). Geographical Ecology: Patterns in the Distribution of Species. Princeton University Press, Princeton. MacArthur RH, Wilson EO (1967). The Theory of Island Biogeography. Princeton University Press, Princeton. Maley J (1991). The African rain forest vegetation and paleoenvironments during later Quaternary. Climate Change 19, 79–98. Maps.com (1999). Asia Elevation Map. [Cited 13 Apr 2013.] Available from URL: http://www.maps.com/ map.aspx?cid=1425,1503&pid=1309&nav=MS Marsh CW, Wilson WL (1981). Effects of natural habitat differences and disturbance on the abundance of Malaysian primates. Malaysian Applied Biology 10, 227–49.
Pearson DL, Carroll SS (2001). Predicting patterns of tiger beetle (Coleoptera: Cicindelidae) species richness in northwestern South America. Studies on Neotropical Fauna and Environment 36, 125–36. Pennington RT, Prado DE, Pendry CA (2000). Neotropical seasonally dry forests and Quaternary vegetation change. Journal of Biogeography 27, 261–73. Perelman P, Johnson WE, Roos C et al. (2011). A molecular phylogeny of living primates. PLOS Genetics 7, e1001342. Quinn GP, Keough MJ (2002). Experimental Design and Data Analysis for Biologists. Cambridge University Press, Cambridge. Raich JW (1989). Seasonal and spatial variation in the light environment in a tropical dipterocarp forest and gaps. Biotropica 21, 299–302.
Matsudaira K, Ishida T (2010). Phylogenetic relationships and divergence dates of the whole mitochondrial genome sequences among three gibbon genera. Molecular Phylogenetics and Evolution 55, 454–9.
Reed KE, Fleagle JG (1995). Geographic and climatic control of primate diversity. PNAS 92, 7874–6.
Mayr E (1963). Animal Species and Evolution. Harvard University Press, Cambridge.
Rodgers WA, Owen CF, Homewood KM (1982). Biogeography of East African forest mammals. Journal of Biogeography 9, 41–54.
Meijaard E (2003). Mammals of Southeast Asian islands and their Late Pleistocene environments. Journal of Biogeography 30, 1245–57. Meyer D, Rinaldi ID, Ramlee H, Perwitasari-Farajallah D, Hodges JK, Roos C (2011). Mitochondrial phylogeny of leaf monkeys (genus Presbytis, Eschscholtz, 1821) with implications for taxonomy and conservation. Molecular Phylogenetics and Evolution 59, 311–9. Mittlebach GG, Steiner CF, Scheiner SM et al. (2001). What is the observed relationship between species richness and productivity? Ecology 82, 2381–96. Morley RJ (2000). Origin and Evolution of Tropical Rain Forests. Wiley, Chichester. Nijman V, Meijaard E (2008). Zoogeography of primates in insular Southeast Asia: species–area relationships and the effects of taxonomy. Contributions to Zoology 77, 117–26. Osterholz M, Walter L, Roos C (2008). Phylogenetic position of the langur genera Semnopithecus and Trachypithecus among Asian colobines, and genus affiliations of their species groups. BMC Evolutionary Biology 8, 1–12.
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Richter DD, Babbar LI (1991). Soil diversity in the tropics. Advances in Ecological Research 21, 316–89.
Rohde K (1992). Latitudinal gradients in species diversity: the search for primary cause. Oikos 65, 514–27. Roos C, Vu NT, Walter L, Nadler T (2007). Molecular systematics of Indochinese primates. Vietnamese Journal of Primatology 1, 41–53. Roos C, Nadler T, Walter L (2008). Mitochondrial phylogeny, taxonomy and biogeography of the silvered langur species group (Trachypithecus cristatus). Molecular Phylogenetics and Evolution 47, 629–36. Rosenzweig ML (1995). Species Diversity in Space and Time. Cambridge University Press, Cambridge. Rosenzweig ML, Abramsky Z (1993). How are diversity and productivity related? In: Ricklefs RE, Schluter D, eds. Species Diversity in Ecological Communities: Historical and Geographical Perspectives. University of Chicago Press, Chicago, pp. 52–65. Sokal RR, Rohlf FJ (1995). Biometry: The Principles and Practice of Statistics in Biological Research, 3rd edn. W. H. Freeman and Company, New York. Struhsaker TT (1981). Forest and primate conservation in East Africa. African Journal of Ecology 19, 99– 114.
425
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
H. Hassel-Finnegan et al.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
Svenning J, Flemming S (2007). Ice age legacies in the geographical distribution of tree species richness in Europe. Global Ecology and Biogeography 16, 234– 45. Thinh VN, Mootnick AR, Geissmann T et al. (2010). Mitochondrial evidence for multiple radiations in the evolutionary history of small apes. BMC Evolutionary Biology 10, 1–13. Wang Y, Srivathsan A, Feng C, Salim A, Shekelle M (2013). Asian primate species richness correlates with rainfall. PLOS ONE 8, e54995. Williams CB (1964). Patterns in the Balance of Nature. Academic Press, New York.
426
SUPPORTING INFORMATION Additional supporting information may be found in the online version of this article at the publisher’s website. Figure S1 Map of Southeast Asia showing location of the protected areas included in this analysis. Table S1 Complete dataset of sites used in the analysis Table S2 Pearson’s correlations amongst variables Please note: Wiley-Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
doi: 10.1111/1749-4877.12056
ORIGINAL ARTICLE
Southeast Asian primate communities: the effects of ecology and Pleistocene refuges on species richness Heather Hassel-Finnegan,1,2 Carola Borries,1 Qing Zhao,3 Phaivanh Phiapalath4 and Andreas Koenig1 1
Department of Anthropology, Stony Brook University, Stony Brook, NY, USA, 2Biology Department, Swarthmore College, Swarthmore, PA, USA, 3College of Life Science, Peking University, Beijing, China and 4International Union for Conservation of Nature, Vientiane, Lao PDR
Abstract We examined historical and ecological factors affecting current primate biodiversity in Southeast Asia. In Africa, Madagascar and South America, but not Southeast Asia, primate species richness is positively associated with average rainfall and distance from the equator (latitude). We predicted that Southeast Asia’s non-conformance may be due to the effect of dispersed Pleistocene refuges (locations of constricted tropical forests during glacial maxima which today are at least 305 m in altitude). Based on 45 forested sites (13 on large islands; 32 on the mainland) of at least 100 km2 to minimize recent human impact, we determined correlations between extant primate species richness and rainfall, latitude and supplementary ecological variables, while controlling for refuges and islands. We found that refuge sites had significantly higher primate species richness than nonrefuges (t = –2.76, P < 0.05), and distance from the nearest Pleistocene refuge was negatively correlated with species richness for non-refuge sites (r = –0.51, P < 0.05). There was no difference in species richness between sites on large islands and the mainland (t = –1.4, P = 0.16). The expected positive relationship between rainfall and species richness was not found (r = 0.17, P = 0.28). As predicted, primate species richness was negatively correlated with latitude (r = –0.39, P < 0.05) and positively correlated with mean temperature (r = 0.45, P < 0.05). General linear models indicated that a site’s latitude (F1,38 = 6.18, P < 0.05) and Pleistocene refuge classification (F1,42 = 5.96, P < 0.05) were the best predictors of species richness. Both ecological and historical factors contribute to present day primate species richness in Southeast Asia, making its biodiversity less of an outlier than previously believed. Key words: allopatric speciation, island, latitude, mainland, rainfall, temperature
INTRODUCTION
Correspondence: Carola Borries, Department of Anthropology, Stony Brook University, Stony Brook, NY 11794-4364, USA. Email: [email protected]
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Current patterns of biodiversity are determined by both historical and ecological factors. Historical factors, such as climate change, migrations and continental drift, may influence species distribution and diversification (MacArthur & Wilson 1967; MacArthur 1984;
417
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
H. Hassel-Finnegan et al.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
Davis & Shaw 2001), while ecological factors, such as forest productivity and food availability, may determine whether an area provides a suitable habitat (Rosenzweig & Abramsky 1993). In South America, Africa and Madagascar, primate species richness is positively correlated with rainfall (Reed & Fleagle 1995), which is often considered to be a predictor of forest productivity (Mittlebach et al. 2001). However, this relationship is absent for Asia (Reed & Fleagle 1995; but see Wang et al. 2013). Latitude may also affect forest productivity, as higher levels of solar radiation and less seasonality are expected near the equator. Among nonhuman primates, the expected negative relationship between species richness and distance from the equator is supported in both South America and Africa (Eeley & Lawes 1999; Emmons 1999), but not in Southeast Asia (Emmons 1999). This lack of conformance for Asian areas is potentially related to 2 major confounding factors: island biogeography (Reed & Fleagle 1995; Fleagle 1999) and the distribution of refuges during the Pleistocene (Eudey 1980; BrandonJones 1996; Jablonski et al. 2000). However, tests for the latter have rarely been performed. Based on the theory of island biogeography, it is hypothesized that species richness depends on both the size of a landmass and its distance from a source population (MacArthur & Wilson 1967). Larger landmasses are expected to have greater habitat heterogeneity and more barriers to gene flow; both factors may increase the number of species with time (Williams 1964). Areas closer to source populations are expected to have greater species richness, because of higher rates of immigration than more isolated islands (MacArthur & Wilson 1967). The effects of landmass and isolation on primate species richness are well documented. Primate species– area relationships have been found at a variety of geographic scales. For example, continents with more rainforest area have greater primate species richness than those with less rainforest (Reed & Fleagle 1995). Within both Africa and South America, latitudinal gradients in species richness may be associated with species–area relationships (Eeley & Lawes 1999), as equatorial regions have the greatest landmass (Rosenzweig 1995), as well as the greatest species richness (Eeley & Lawes 1999; Emmons 1999). At a finer scale, larger Southeast Asian islands have greater primate species richness than small islands (Nijman & Meijaard 2008). Isolation may also be an important determinant of primate biodiversity, particularly in Southeast Asia. Isolation affects the rate of immigration, extinction and
418
speciation (MacArthur & Wilson 1967). The land-bridge islands of the Sunda Shelf became isolated due to rises in sea levels approximately 12 000 years ago and many other times in the past 5 million years (Heaney 1986; Lawlor 1986). Even such a recent isolation can have a significant impact on patterns of species richness, because of the potential for faunal collapse due to limited contact with source populations (Heaney 1986). Thus, islands may be predicted to have lower species richness than mainland areas. Pleistocene refuges may have also affected the distribution of Southeast Asian primate species (Eudey 1980; Brandon-Jones 1996; Jablonski et al. 2000). It has been hypothesized that reduced rainfall during the Pleistocene led to temporary separation of previously continuous rainforests, in mid-altitude forest patches surrounded by savannah. Forest taxa later re-expanded their ranges during inter-glacial periods (Mayr 1963). Higher species richness is, indeed, reported within suspected refuges in Southeast Asia for both plants and animals (Rodgers et al. 1982; Pearson & Carroll 2001; Svenning & Flemming 2007). Some suggest that allopatric speciation occurred within refuges (Colyn et al. 1991; Haffer 1997; Abegg & Thierry 2002), although there is controversy as some phylogenies suggest primate speciation predated the Pleistocene epoch (Kay et al. 1997; Collins & Dubach 2000; see also discussion below). During the last glacial maximum, there was a significant increase in Southeast Asian montane forest and savannah, with a corresponding decline in rainforest (Heaney 1991). Pollen and termite-community composition analyses indicate that the pattern of rainforest patches that resulted was quite different from that of other continents, in that refuges were abundant, but geographically dispersed (Heaney 1991; Gathorne-Hardy et al. 2002). In Africa and South America, refuges were rare but more continuous, and concentrated near the equator (Maley 1991; Pennington et al. 2000), where primate species richness is also expected to be high for ecological reasons (Reed & Fleagle 1995; Eeley & Lawes 1999; Emmons 1999). Finally, it has been suggested (Reed & Fleagle 1995; Emmons 1999) that the limited sample size of previous studies and exclusive sampling of insular and peninsular sites affected the results for Southeast Asia. Here, we analyzed a new large dataset, emphasizing mainland sites (all west of the Wallace line), of which many have thus far not been considered. We tested for the relationship between species richness and various ecological variables collected directly at each site. This selection
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Asian primate communities
criterion led to a sample including, besides mainland sites, only sites from the 2 largest islands, Sumatra and Borneo, as well as the Malay Peninsula. We also explored explanations for Southeast Asia’s non-conformance, such as the effect of Pleistocene refuges. We specifically addressed 4 questions related to these topics: 1. Do sites on the largest islands have lower species richness than those on the mainland? 2. Is species richness highest at low latitudes, despite Southeast Asia’s small land area near the equator? 3. Do former refuges have higher species richness than non-refuges? 4. Does species richness decrease with distance from the nearest Pleistocene refuge?
MATERIALS AND METHODS Published and unpublished data from a total of 45 sites (Nisland = 13; Nmainland = 32; details in Fig. S1 and Table S1) were included in this analysis. All sites were protected forested areas of at least 100 km2, chosen to limit the effect of more recent human activity on local primate community composition. We excluded sites over 25° from the equator to limit the influence of extreme seasonality. All data including climate data were collected within the protected areas, rather than from primate distribution maps or wide scale weather databases that may be less reliable. Sites on Java, the Mentawai Islands and Sulawesi could not be included because very few ecological data have thus far been published. The following variables were available for all or most sites: extant primate species richness (N species), distance from the equator (measured as latitude) and annual rainfall (in mm). Because the taxonomy of Asian primates is in flux (Roos et al. 2007; Osterholz et al. 2008; Roos et al. 2008), we followed the species designations of the reference for a given site without questioning it. It has also been shown that taxonomic differences do not necessarily influence zoogeographic analyses of primates in the region (Nijman & Meijaard 2008). We note that if several species were reported to occur in a particular area, they might not always live sympatrically, especially in very large areas, but the resolution of the data did not allow for identifying these cases in our sample, should they exist. Wherever possible, data for the following variables were also included: mean temperature (in °C), mean
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
minimum temperature, mean maximum temperature, tree density (in N/ha) and number of dry months (with an average of < 50 mm precipitation). All variables were tested for normality, using the Kolmogorov–Smirnov test (Sokal & Rohlf 1995). No variables differed significantly from a normal distribution (all “P”s > 0.10). Consequently, correlations were calculated using Pearson’s r for the relationships between all aforementioned variables (Sokal & Rohlf 1995). In addition, assuming that the relationship between rainfall and primate species richness might not be linear, a Lowess localized regression was calculated for mean annual rainfall versus species richness (following Kay et al. 1997). Sites were classified as refuges if they were forested during the last glacial maximum, based on studies of soil, sediments and plant fossils (Adams 1997). These sites are at least 305 m asl today (Maps.com 1999). Refuges are believed to have existed in mid-altitudinal regions where rainfall and temperature may have remained within suitable ranges for sustaining forest primates (Brandon-Jones 1996). For all non-refuges we determined the great circle distance from the center of the site to the center of the nearest refuge. To test for dependency of primate species richness on categorical historical and ecological variables, we used t-tests for independent samples (Sokal & Rohlf 1995) to analyze the individual effects of these variables. To test for multiple dependencies we used 2 general linear models (Grafen & Hails 2002). An overall model testing all variables simultaneously was not possible, because of collinearity between several of the variables (Quinn & Keough 2002; see also Table S2). The first general linear model tested the effect of islands, Pleistocene refuges and rainfall on primate species richness. The second tested the influence of Pleistocene refuges and distance from the equator on primate species richness. All statistical tests were 2-tailed and performed using STATISTICA 6.1 (StatSoft) at an alpha level of 0.05.
RESULTS The distribution of sites ranged from 25°N to 6°S (Fig. S1 and Table S1). Primate species richness ranged from 1 to 12 species overall and was negatively correlated with latitude, decreasing from approximately 8 species at the equator down to approximately 5 species at the most northern sites (Pearson’s r = –0.39, P < 0.05, N = 45; Fig. 1a). At the same time, the number of species was positively correlated with mean temperature (r = 0.45, P < 0.05, N = 20; Fig. 1b). However,
419
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
H. Hassel-Finnegan et al.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
Figure 1 Relationship between primate species richness and (a) latitude (Pearson’s r = –0.39, P < 0.05) as well as (b) mean temperature (r = 0.45, P < 0.05).
Figure 2 Relationships between species richness and mean annual rainfall presented as: (a) linear regression (r = 0.17, P = 0.28) and (b) Lowess localized regression.
species richness did not have a significant relationship with rainfall; species richness was only slightly increasing from approximately 6 to 7 species over the range of rainfall from 1000 to 4500 mm (r = 0.17, P = 0.28, N = 43; Fig. 2a). A Lowess localized regression applied to this dataset did not show a discernible pattern (Fig. 2b). Species richness remains fairly stable except for 2 minima at approximately 1300 and 1700 mm rainfall. Correlations between species richness and all other vari-
ables were not significant: mean maximum temperature, r = 0.42, P = 0.12, N = 15; mean minimum temperature, r = 0.16, P = 0.57, N = 15; tree density r = 0.29, P = 0.54, N = 7; and number of dry months, r = –0.27, P = 0.35, N = 11. With 7.1 primate species for the large island sites, species richness was slightly higher than at mainland sites (mean = 6.0), although the difference was not sta-
420
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Asian primate communities
Figure 4 Relationship between primate species richness and distance from the nearest Pleistocene refuge for non-refuge sites (Pearson’s r = –0.51, P < 0.05).
Figure 3 Mean primate species richness and the influence by: (a) geography (mainland vs large island sites; t = –1.42, P = 0.16) and (b) history (non-refuge vs Pleistocene refuge sites; t = –2.76, P < 0.05). Boxes represent standard errors and whiskers standard deviations.
away it declines to approximately 3 (r = –0.51, P < 0.05; Fig. 4). A first general linear model (Fintercept = 20.13, P < 0.05) found that Pleistocene refuge classification affects species richness (F1,38 = 5.43, P < 0.05), but there was no effect of rainfall (F1,38 = 0.26, P = 0.61) or large island/ mainland classification (F1,38 = 0.73, P = 0.40). A second general linear model (Fintercept = 146.38, P < 0.05) found that both Pleistocene refuge classification (F1,42 = 5.96, P < 0.05) and latitude (F1,42 = 6.18, P < 0.05) had a significant effect on primate species richness. Primate species richness in non-refuge sites decreased from approximately 8 species at the equator down to approximately 4 in the north and south, while primate species in refuge sites showed little change across latitudes (Fig. 5).
DISCUSSION tistically significant (t = –1.42, P = 0.16; Fig. 3a). In contrast, primate species richness was significantly higher in suspected Pleistocene refuges (mean = 7.5) than in non-refuges (mean = 5.6; t = –2.76, P < 0.05; Fig. 3b). For non-refuges, the distance to the nearest refuge was negatively correlated with species richness; close to a refuge the number of species is close to 7, while further
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Our results indicate that distance from the equator, mean temperature and whether a site had been a Pleistocene refuge are the best predictors of the present primate species richness in Southeast Asia. In the current sample, the island effect was absent. Similarly, rainfall was not found to be a reliable predictor of primate species richness in the region.
421
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
H. Hassel-Finnegan et al.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
a
b
Figure 5 Relationship between latitude and history (non-refuge vs Pleistocene refuge) on primate species richness (non-refuges: Pearson’s r = –0.44, P < 0.05; Pleistocene refuges: r = –0.10, P = 0.70).
In the past, several hypotheses have been proposed for why rainfall does not predict species richness in Southeast Asia, including: extreme seasonal and year-toyear variation in rainfall, the presence of local low-species-diversity forests dominated by Dipterocarpaceae, the possibility that very high amounts of rain limit productivity or that areas with low precipitation were underrepresented in the data (Reed & Fleagle 1995; Kay et al. 1997; Gupta & Chivers 1999; Wang et al. 2013). The first hypothesis predicts no relationship between forest productivity and rainfall due to generally similar monsoonal effects across the Southeast Asian region (Kripalani & Kulkarni 1998). The second hypothesis also predicts no relationship between rainfall and forest productivity, due to wide-scale microhabitat heterogeneity, with dipterocarp areas having uniformly lower primate biomass (Marsh & Wilson 1981). The third hypothesis predicts an inverted-U-shaped or saturated curvilinear relationship between forest productivity and rainfall (Kay et al. 1997), as rainfall >2500 mm may lead to soil leaching (Richter & Babbar 1991) or reduced photosynthesis because of cloud cover (Raich 1989). The final hypothesis predicts a positive linear relationship between rainfall and species richness if sites with low rainfall are included in the analysis (Wang et al. 2013). Such a positive relationship is reported by Wang et al. (2013) for the region in an analysis based on 390 polygons defined by overlaying species distribution maps. These expected species richness values were combined with rainfall values averaged from 1288 climate data
422
points. Thus, in contrast to our present study, the analysis was not based on the actual conditions found at each site, neither in terms of primate species richness nor rainfall, and it remains unclear how realistic this approach was. For example, it is possible that sites with low rainfall and within the distribution range of a species may, in fact, not be inhabited by primates. In contrast, we did not find a relationship between rainfall and primate species richness in our sample, neither in a linear nor in a curvilinear fashion. This makes the first 2 hypotheses more likely; that is, seasonal/year-to-year variation in rainfall or the wide distribution of low diversity dipterocarp areas may explain the lack of a relationship between rainfall and primate species richness. It has also been suggested that the geography of Southeast Asia may have biased past results of primate communities, which exclusively sampled insular and peninsular sites (Reed & Fleagle 1995; Emmons 1999). A species–area relationship was, indeed, found based on 218 small Southeast Asian islands (Nijman & Meijaard 2008). This result could not be reproduced in our analysis likely because the island sites included were from Sumatra, Borneo and the Malay Peninsula. These are very large land masses with a high habitat heterogeneity allowing for niche diversification and limited faunal collapse (Heaney 1984), which may have translated into the absence of an island effect in our sample. Latitudinal gradients in species richness are reported among a wide range of taxonomic groups (Fernandes & Price 1988; Kaufman 1995; Blackburn & Gaston 1996; Emmons 1999). There are 2 general causative hypotheses proposed to explain this phenomenon. One suggests that the latitudinal gradient is due to ecological factors. Near the equator, environmental stability, productivity, physical heterogeneity, solar radiation and temperature are highest, while seasonality and aridity are less extreme. All of these conditions are expected to increase biodiversity (Rohde 1992). The other group of hypotheses suggests that latitudinal gradients are a result of species–area relationships. Most early studies on latitudinal gradients focused on South America and Africa, where the continents are roughly diamond shaped, with the greatest landmass (and species richness) near the equator with land area (and species richness) shrinking towards the poles (Rosenzweig 1995). Our data indicate that there is a strong relationship between distance from the equator and primate species richness in Southeast Asia. In addition, while the data at hand do not allow assessment of factors such as physical heterogeneity and solar radiation, distance to the equator is, indeed,
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Asian primate communities
strongly correlated with seasonality and temperature. This lends support to ecological hypotheses, because the geography of Southeast Asia is not diamond shaped. For a variety of taxa, species richness is high in sites that served as Pleistocene refuges (Rodgers et al. 1982; Pearson & Carroll 2001; Svenning & Flemming 2007) and for non-refuges, decreases according to distance from the nearest refuge (Struhsaker 1981). With respect especially to forest-dependent primates, it is assumed that during glacial maxima refuges were isolated forest patches separated by extended savannahs, which were unsuitable habitats for most forest-dependent species (Morley 2000) even though large areas were connected by land bridges due to the low sea levels. Today these refuges identify areas that have been inhabited for the longest, which should be reflected in high richness of forest-dependent species. It is expected that areas outside of refuges had to be repeatedly re-populated with species from the refuges as the source. Evidence from Southeast Asian forest-dependent termites (Gathorne-Hardy et al. 2002) and mammals (Meijaard 2003) suggests that range expansion to newly reforested areas was often incomplete. This scenario is in line with our finding of species richness reducing with distance to refuges (Fig. 4). Debate remains as to whether the high species richness in refuges is a result of allopatric speciation in isolated forest patches or relic species, which may have been widely distributed before the Pleistocene, but did not re-expand their ranges during post-glacial forest reexpansions because of life history or geographic constraints (Colyn et al. 1991; Fjeldsa & Lovett 1997; Haffer 1997; Collins & Dubach 2000). It seems now likely that already at the beginning of the last glacial maximum between 33.0 and 26.5 ka (Clark et al. 2009) most Southeast Asia primate groups existed: of all Asian macaques, the youngest species occurred approximately 0.8–0.7 Ma (Hayasaka et al. 1996; Disotell & Tosi 2007; Chatterjee et al. 2009); the youngest Asian colobine species around 0.6 Ma (Roos et al. 2008; Meyer et al. 2011; Perelman et al. 2011); and the youngest gibbon species around 0.5–0.3 Ma (Matsudaira & Ishida 2010; Thinh et al. 2010; Perelman et al. 2011). However, at the beginning of the last ice age at around 2.6 Ma, not all species in these groups had evolved. Biogeographic evidence, including analysis of wide-scale species distribution maps, supports the possibility of allopatric speciation within refuges (Eudey 1980; Jablonski et al. 2000). Our results indicate that Pleistocene refuges have higher primate species richness than non-refuges. For non-refug-
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
es, species richness decreases with distance from the nearest refuge. However, it is impossible to decide to what extent this is the result of allopatric speciation or a reduced expansion of relic species, or both. Studies of primate communities, such as this one, may help to determine the ecological and historical factors that typify diversity hotspots. Our results indicate that Southeast Asian sites close to the equator and those that served as Pleistocene refuges have the greatest primate species richness. High species richness near the equator may be due to environmental stability, high levels of primary productivity, high levels of solar radiation and high temperatures. High species richness in and near Pleistocene refuges suggests that there may have been allopatric speciation of primates within these areas, or relic species that never re-expanded their ranges following the last glacial maximum. The differences in species richness patterns between Southeast Asia and Africa/South America may be due to differences in continental shape and the distribution of Pleistocene refuges.
ACKNOWLEDGMENTS We would like to thank Warren Brockelman, Wanlop Chutipong, Tanja Haus, Meyner Nusalawo, Jonathan O’Brien, Rungnapa Phoonjampa and Thomas Ziegler for sharing unpublished data for individual sites. For helpful comments on a previous draft we thank John Fleagle and our anonymous reviewers.
REFERENCES Abegg C, Thierry B (2002). Macaque evolution and dispersal in insular Southeast Asia. Biological Journal of the Linnean Society 75, 555–76. Adams JM (1997). Global Land Environments since the Last Interglacial. [Cited 16 Apr 2013.] Available from URL: http://www.esd.ornl.gov/ern/qen/nerc. html Blackburn TM, Gaston KJ (1996). Spatial patterns in the species richness of birds in the New World. Ecography 19, 369–76. Brandon-Jones D (1996). The Asian Colobinae (Mammalia: Cercopithecidae) as indicators of Quaternary climatic change. Biological Journal of the Linnean Society 59, 327–50. Chatterjee HJ, Ho SYW, Barnes I, Groves C (2009). Estimating the phylogeny and divergence times of pri-
423
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
H. Hassel-Finnegan et al.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
mates using a supermatrix approach. BMC Evolutionary Biology 9, 259.
Grafen A, Hails R (2002). Modern Statistics for the Life Sciences. Oxford University Press, Oxford.
Clark PU, Dyke AS, Shakun JD et al. (2009). The last glacial maximum. Science 325, 710–4.
Gupta AK, Chivers DJ (1999). Biomass and use of resources in South and Southeast Asian primate communities. In: Fleagle JG, Janson CH, Reed KE, eds. Primate Communities. Cambridge University Press, Cambridge, pp. 38–54.
Collins AC, Dubach JM (2000). Biogeographic and ecological forces responsible for speciation in Ateles. International Journal of Primatology 21, 421–44. Colyn M, Gautier-Hion A, Verheyen W (1991). A reappraisal of paleoenvironmental history in Central Africa: evidence for a major fluvial refuge in the Zaire Basin. Journal of Biogeography 18, 403–7. Davis MB, Shaw RG (2001). Range shifts and adaptive response to quaternary climate change. Science 292, 673–9. Disotell TR, Tosi AJ (2007). The monkey’s perspective. Genome Biology 8, 226. Eeley HAC, Lawes MJ (1999). Large-scale patterns of species richness and species range size in anthropoid primates. In: Fleagle JG, Janson CH, Reed KE, eds. Primate Communities. Cambridge University Press, Cambridge, pp. 191–219. Emmons LH (1999). Of mice and monkeys: primates as predictors of mammal community richness. In: Fleagle JG, Janson CH, Reed KE, eds. Primate Communities. Cambridge University Press, Cambridge, pp. 171–88. Eudey AA (1980). Pleistocene glacial phenomena and the evolution of Asian macaques. In: Lindburg DG, ed. The Macaques: Studies in Ecology, Behavior and Evolution. van Nostrand Reinhold, London, pp. 52– 83. Fernandes GW, Price PW (1988). Biogeographical gradients in galling species richness. Oecologia 76, 161– 7. Fjeldsa J, Lovett JC (1997). Geographical patterns of old and young species in African forest biota: the significance of specific montane areas as evolutionary centres. Biodiversity and Conservation 6, 325–46. Fleagle JG (1999). Primate Adaptation and Evolution, 2nd edn. Academic Press, San Diego. Gathorne-Hardy FJ, Syaukani, Davies RG, Eggleton P, Jones DT (2002). Quaternary rainforest refugia in Southeast Asia: using termites (Isoptera) as indicators. Biological Journal of the Linnean Society 75, 453–66.
424
Haffer J (1997). Alternative models of vertebrate speciation in Amazonia: an overview. Biodiversity and Conservation 6, 451–76. Hayasaka K, Fujii K, Horai S (1996). Molecular phylogeny of macaques: implications of nucleotide sequences from an 896-base pair region of the mitochondrial DNA. Molecular Biology and Evolution 13, 1044– 53. Heaney LR (1984). Mammalian species richness on islands on the Sunda Shelf, Southeast Asia. Oecologia 61, 11–7. Heaney LR (1986). Biogeography of mammals in SE Asia: estimates of rates of colonization, extinction and speciation. Biological Journal of the Linnean Society 28, 127–65. Heaney LR (1991). A synopsis of climatic and vegetational change in Southeast Asia. Climate Change 19, 53–61. Jablonski NG, Whitfort MJ, Roberts-Smith N, Xu Q (2000). The influence of life history and diet on the distribution of catarrhine primates during the Pleistocene in eastern Asia. Journal of Human Evolution 39, 131–57. Kaufman DM (1995). Diversity of New World mammals: universality of the latitudinal gradients of species and bauplans. Journal of Mammalogy 76, 322– 34. Kay RF, Madden RH, van Schaik CP, Higdon D (1997). Primate species richness is determined by plant productivity: implications for conservation. PNAS 94, 13023–7. Kripalani RH, Kulkarni A (1998). Rainfall variability over Southeast Asia–connections with Indian monsoon and ENSO extremes: new perspectives. International Journal of Climatology 17, 1155–68. Lawlor TE (1986). Comparative biogeography of mammals on islands. Biological Journal of the Linnean Society 28, 99–125.
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Asian primate communities
MacArthur RH (1984). Geographical Ecology: Patterns in the Distribution of Species. Princeton University Press, Princeton. MacArthur RH, Wilson EO (1967). The Theory of Island Biogeography. Princeton University Press, Princeton. Maley J (1991). The African rain forest vegetation and paleoenvironments during later Quaternary. Climate Change 19, 79–98. Maps.com (1999). Asia Elevation Map. [Cited 13 Apr 2013.] Available from URL: http://www.maps.com/ map.aspx?cid=1425,1503&pid=1309&nav=MS Marsh CW, Wilson WL (1981). Effects of natural habitat differences and disturbance on the abundance of Malaysian primates. Malaysian Applied Biology 10, 227–49.
Pearson DL, Carroll SS (2001). Predicting patterns of tiger beetle (Coleoptera: Cicindelidae) species richness in northwestern South America. Studies on Neotropical Fauna and Environment 36, 125–36. Pennington RT, Prado DE, Pendry CA (2000). Neotropical seasonally dry forests and Quaternary vegetation change. Journal of Biogeography 27, 261–73. Perelman P, Johnson WE, Roos C et al. (2011). A molecular phylogeny of living primates. PLOS Genetics 7, e1001342. Quinn GP, Keough MJ (2002). Experimental Design and Data Analysis for Biologists. Cambridge University Press, Cambridge. Raich JW (1989). Seasonal and spatial variation in the light environment in a tropical dipterocarp forest and gaps. Biotropica 21, 299–302.
Matsudaira K, Ishida T (2010). Phylogenetic relationships and divergence dates of the whole mitochondrial genome sequences among three gibbon genera. Molecular Phylogenetics and Evolution 55, 454–9.
Reed KE, Fleagle JG (1995). Geographic and climatic control of primate diversity. PNAS 92, 7874–6.
Mayr E (1963). Animal Species and Evolution. Harvard University Press, Cambridge.
Rodgers WA, Owen CF, Homewood KM (1982). Biogeography of East African forest mammals. Journal of Biogeography 9, 41–54.
Meijaard E (2003). Mammals of Southeast Asian islands and their Late Pleistocene environments. Journal of Biogeography 30, 1245–57. Meyer D, Rinaldi ID, Ramlee H, Perwitasari-Farajallah D, Hodges JK, Roos C (2011). Mitochondrial phylogeny of leaf monkeys (genus Presbytis, Eschscholtz, 1821) with implications for taxonomy and conservation. Molecular Phylogenetics and Evolution 59, 311–9. Mittlebach GG, Steiner CF, Scheiner SM et al. (2001). What is the observed relationship between species richness and productivity? Ecology 82, 2381–96. Morley RJ (2000). Origin and Evolution of Tropical Rain Forests. Wiley, Chichester. Nijman V, Meijaard E (2008). Zoogeography of primates in insular Southeast Asia: species–area relationships and the effects of taxonomy. Contributions to Zoology 77, 117–26. Osterholz M, Walter L, Roos C (2008). Phylogenetic position of the langur genera Semnopithecus and Trachypithecus among Asian colobines, and genus affiliations of their species groups. BMC Evolutionary Biology 8, 1–12.
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
Richter DD, Babbar LI (1991). Soil diversity in the tropics. Advances in Ecological Research 21, 316–89.
Rohde K (1992). Latitudinal gradients in species diversity: the search for primary cause. Oikos 65, 514–27. Roos C, Vu NT, Walter L, Nadler T (2007). Molecular systematics of Indochinese primates. Vietnamese Journal of Primatology 1, 41–53. Roos C, Nadler T, Walter L (2008). Mitochondrial phylogeny, taxonomy and biogeography of the silvered langur species group (Trachypithecus cristatus). Molecular Phylogenetics and Evolution 47, 629–36. Rosenzweig ML (1995). Species Diversity in Space and Time. Cambridge University Press, Cambridge. Rosenzweig ML, Abramsky Z (1993). How are diversity and productivity related? In: Ricklefs RE, Schluter D, eds. Species Diversity in Ecological Communities: Historical and Geographical Perspectives. University of Chicago Press, Chicago, pp. 52–65. Sokal RR, Rohlf FJ (1995). Biometry: The Principles and Practice of Statistics in Biological Research, 3rd edn. W. H. Freeman and Company, New York. Struhsaker TT (1981). Forest and primate conservation in East Africa. African Journal of Ecology 19, 99– 114.
425
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
H. Hassel-Finnegan et al.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
Svenning J, Flemming S (2007). Ice age legacies in the geographical distribution of tree species richness in Europe. Global Ecology and Biogeography 16, 234– 45. Thinh VN, Mootnick AR, Geissmann T et al. (2010). Mitochondrial evidence for multiple radiations in the evolutionary history of small apes. BMC Evolutionary Biology 10, 1–13. Wang Y, Srivathsan A, Feng C, Salim A, Shekelle M (2013). Asian primate species richness correlates with rainfall. PLOS ONE 8, e54995. Williams CB (1964). Patterns in the Balance of Nature. Academic Press, New York.
426
SUPPORTING INFORMATION Additional supporting information may be found in the online version of this article at the publisher’s website. Figure S1 Map of Southeast Asia showing location of the protected areas included in this analysis. Table S1 Complete dataset of sites used in the analysis Table S2 Pearson’s correlations amongst variables Please note: Wiley-Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
© 2013 Wiley Publishing Asia Pty Ltd, ISZS and IOZ/CAS
