High Genetic Diversity in a Potentially Vulnerable Tropical Tree Species Despite Extreme Habitat Loss

Annika M. E. Noreen; Edward L. Webb

doi:10.1371/journal.pone.0082632

Abstract

Over the last 150 years, Singapore’s primary forest has been reduced to less than 0.2% of its previous area, resulting in extinctions of native flora and fauna. Remaining species may be threatened by genetic erosion and inbreeding. We surveyed >95% of the remaining primary forest in Singapore and used eight highly polymorphic microsatellite loci to assess genetic diversity indices of 179 adults (>30 cm stem diameter), 193 saplings (>1 yr), and 1,822 seedlings (<1 yr) of the canopy tree Koompassia malaccensis (Fabaceae). We tested hypotheses relevant to the genetic consequences of habitat loss: (1) that the K. malaccensis population in Singapore experienced a genetic bottleneck and a reduction in effective population size, and (2) K. malaccensis recruits would exhibit genetic erosion and inbreeding compared to adults. Contrary to expectations, we detected neither a population bottleneck nor a reduction in effective population size, and high genetic diversity in all age classes. Genetic diversity indices among age classes were not significantly different: we detected overall high expected heterozygosity (H_e = 0.843–0.854), high allelic richness (R = 16.7–19.5), low inbreeding co-efficients (F_IS = 0.013–0.076), and a large proportion (30.1%) of rare alleles (i.e. frequency <1%). However, spatial genetic structure (SGS) analyses showed significant differences between the adults and the recruits. We detected significantly greater SGS intensity, as well as higher relatedness in the 0–10 m distance class, for seedlings and saplings compared to the adults. Demographic factors for this population (i.e. <200 adult trees) are a cause for concern, as rare alleles could be lost due to stochastic factors. The high outcrossing rate (t_m = 0.961), calculated from seedlings, may be instrumental in maintaining genetic diversity and suggests that pollination by highly mobile bee species in the genus Apis may provide resilience to acute habitat loss.

Citation: Noreen AME, Webb EL (2013) High Genetic Diversity in a Potentially Vulnerable Tropical Tree Species Despite Extreme Habitat Loss. PLoS ONE 8(12): e82632. https://doi.org/10.1371/journal.pone.0082632

Editor: Terry Brown, Manchester Institute of Biotechnology, United States of America

Received: July 30, 2013; Accepted: November 4, 2013; Published: December 18, 2013

Copyright: © 2013 Noreen, Webb. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: The study was supported by the Singapore Ministry of Education (grant R154-000-400-133) and the Lee Hiok Kwee Endowment to the Department of Biological Sciences, NUS (grant R-154-000-480-720). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Deforestation and degradation of primary forests are of critical concern for global biodiversity conservation [1]. The island city-state of Singapore (704 km², population 5.18 million) is one of the most extreme examples of urbanization in the world [2], [3], [4]. Until the mid-1800’s the island was predominantly primary lowland rainforest (∼80%) with additional substantial areas of mangrove and freshwater swamp forest [5]. Rapid deforestation beginning ∼150 years ago has reduced overall forest cover to less than 2% of its original extent: 118 ha (0.16%) of primary and 995 ha (1.37%) of old secondary forest [2], [3].

In addition to outright extirpation following habitat reduction, extinctions of local flora and fauna may continue after initial deforestation, as a subset of species within small and isolated habitat remnants may continue to decline in response to stochastic, demographic and genetic factors, i.e. an extinction debt [6], [7]. In line with these predictions, Singapore’s native flora and fauna has been affected by “catastrophic” extinction rates [8], and the continuing loss of native, forest-dependent species remains a concern [3], [9]. Detrimental genetic effects from a reduction in population size can include genetic erosion (i.e. loss of rare alleles and lower genetic diversity), which reduces the genetic variation for selection to act upon [10], [11], and higher inbreeding. These changes may be particularly important for outcrossing species [12], [13], as they can significantly affect fitness, fertility, and offspring viability, as shown by studies of laboratory [14], captive [15], and wild [16] populations.

However, interpreting genetic diversity indices in relation to population viability can depend on phylogenetic history, life history and biogeographic variables. For example, some plant species that are naturally rare and/or endemic have strategies that allow for survival under conditions of self-fertilization or clonality, low genetic diversity, and high inbreeding [17], [18]. However, widespread, abundant, and outcrossing species may lack mechanisms allowing persistence if their populations undergo precipitous declines. Meta-analyses have concluded that previously common tree species that undergo fragmentation and isolation are more susceptible to the negative genetic consequences of deforestation than naturally rare species [19], with “strong, negative and significant effects” on genetic diversity [20]. These consequences may be apparent only after several generations (i.e. >100 years) for some long-lived species [19], [20], [21], and a recent meta-analysis has concluded that woody species are equally prone to loss of genetic diversity compared to other plant species [22]. Indeed, several recent studies have documented a significant decline in genetic diversity in tropical tree species immediately following habitat loss and fragmentation [23], [24], [25]. Additionally, increasing spatial genetic structure (SGS) intensity in younger cohorts has recently been used to document a negative genetic effect of habitat loss and fragmentation on a tropical forest tree despite a lack of genetic diversity reduction [26], although demographic factors relating to age cohorts may also play a significant role in interpreting spatial genetic structure [27].

Immigration and gene flow into small, isolated populations is generally considered to have positive genetic and demographic effects, as gene flow can increase effective population size and genetic diversity, and reduce deleterious inbreeding, e.g. [16], [28]. For Singapore, the next-nearest primary forests are more than 70 km distant (on peninsular Malaysia) and separated by the Singapore Strait as well as urban and agricultural landscapes. Hence, geographic isolation likely corresponds to genetic isolation, as naturally occurring dispersal (of either pollen or seed) from outside populations is improbable. Thus, in the absence of human-mediated migration (e.g. transplants) Singapore’s primary forest species may be fully dependent on recruitment from local individuals for persistence.

Target species

Koompassia malaccensis Maingay ex Benth. (Fabaceae) is a long-lived (up to several hundred years), large (up to 55 m tall), invertebrate-pollinated rainforest tree with wind-dispersed seeds. It is a canopy emergent tree of primary rainforest in Southeast Asia, ranging from Thailand to Borneo. Despite being classified as a species of low extinction risk over its entire range [29], K. malaccensis is locally endangered within Singapore owing to severe habitat loss [9]. As a commercially valuable timber species, K. malaccensis has been widely harvested and therefore tends to be relegated to primary forest patches or occasionally found as isolated remnant adults in secondary forests. In some areas individual trees are left intact as K. malaccensis is frequently utilized by wild honey bees (Apis dorsata in particular) for nesting [30]. Pollination of Koompassia is accomplished by A. dorsata, which is a large, migratory bee species, along with other Apis species (R.T. Corlett, pers. comm.). As a member of the Fabaceae, reproduction is hypothesized to occur yearly [31], although observations suggest that seed set in Singapore might be more sporadic (E.L. Webb and S.K.Y. Lum, pers. obs.). K. malaccensis is fast growing, and reaches reproductive maturity at around 30 cm stem diameter (perhaps 30–40 years old assuming a 1 cm annual diameter increment [32]).

Previous research on K. malaccenis has been conducted on peninsular Malaysia, Singapore’s next-nearest primary forest populations [33], [34], [35]. Across 19 populations throughout peninsular Malaysia, K. malaccensis was found to have high genetic diversity and low differentiation [34]. A Malaysian population also showed low biparental inbreeding and a predominately outcrossing mating system [35], which is in accordance with other tropical tree species. Being the closest populations to Singapore, these Malaysian Koompassia populations can be used as an approximate baseline population, although direct comparisons should be undertaken with caution.

Although no pre-deforestation abundance and distribution data for Koompassia or any other species exist for Singapore, it is known that Singapore was predominately lowland tropical rainforest (∼80%) [5]. As the current distribution of K. malaccensis is almost exclusively in primary forest patches, it is fair to assume a high correlation between primary forest habitat loss and population loss. Thus we estimated that owing to habitat loss, K. malaccensis has sustained a population reduction of at least 90% over the last >150 years. It is important to note that whether actual population loss is exactly equivalent to the reduction of primary forest area is somewhat inconsequential given the scale of deforestation [2], [3]. A previous study on the effect of population reduction of K. malaccenis through logging in southern peninsular Malaysia —a disturbance that typically removes a lower proportion of a tree population than deforestation—found fewer alleles per locus for seedling, sapling, and adult K. malaccensis individuals in logged versus unlogged plots, as well as a reduction in the number of polymorphic loci [33]; however, low sample sizes in that study prevented definitive conclusions. Given that previous research over smaller spatial and shorter temporal scales showed genetic consequences of logging on K. malaccensis, a study of the effects of extreme habitat loss in Singapore was well justified.

Objectives

Koompassia malaccensis in Singapore is a putatively previously abundant, widespread and outcrossing species that underwent a severe decline in habitat starting ∼150 years ago and has a low likelihood of immigration and gene flow from outside populations. Hence, we tested two hypotheses relevant to the genetic consequences of habitat loss. First was that the K. malaccensis population in Singapore experienced a population bottleneck and a reduction in effective population size. Second was that recruit cohorts would show genetic erosion (i.e. lower allelic richness and/or heterozygosity) and inbreeding compared to the adult cohort. To test these hypotheses, we assessed the genetic structure of comprehensively sampled adults, and randomly selected samples of saplings and seedlings, using eight highly polymorphic microsatellite loci.

Materials and Methods

Our study area encompassed the central catchment and the Singapore Botanic Gardens (Figure 1). The central catchment, a mosaic of primary and secondary forest patches set within an urban matrix, consists of the Central Catchment Nature Reserve (CCNR) complex (MacRitchie, Lower Pierce, Upper Pierce and Upper Seletar Reservoir forests) and Bukit Timah Nature Reserve (BTNR), and accounts for the largest area of contiguous forest remaining in Singapore: ca. 3,043 ha (primary, old secondary and young secondary forest), of which 2880 ha is in CCNR and 163 ha in BTNR [36] (Figure 1). From November 2010 through November 2011, we conducted fieldwork to locate and survey all K. malaccensis adults with a dbh (stem diameter at 1.4 m above the base) of at least 30 cm in BTNR and MacRitchie Reservoir Park. For BTNR, we were provided access to an existing spatial database of all trees >30 cm dbh, from which we extracted the locations of all adult K. malaccensis. MacRitchie surveys were conducted by referring to existing vegetation maps of the central catchment [2], [5] and taking extensive reconnaissance hikes through primary as well as secondary forest to locate adults. The Botanic Gardens retains five K. malaccensis adults, all of which we sampled. Our data set was collected in >95% of the remaining primary forest area in Singapore, and was therefore expected to account for a roughly equivalent percentage of the remaining K. malaccensis population. However, given that we found a few isolated adults located in secondary forest, it is possible that we captured a somewhat smaller proportion of the remaining population than 95%.

Download:

Figure 1. Map of Singapore’s remaining primary forest fragments within a secondary forest and urban matrix.

Dark green – primary forest; Light green – secondary forest; red dots – adult K. malaccensis individuals.

https://doi.org/10.1371/journal.pone.0082632.g001

Sample collection

We collected bark and/or leaves of 179 adult K. malaccensis (dbh >30 cm) and leaves from 193 saplings (>1 year old, 55 cm to 2+ m height) and 1,822 seedlings (<1 year old, ≤50 cm, often with cotyledons still attached) from beneath 122 adults. Sampling was conducted under Singapore National Parks research permit NP/RP10-078. All necessary permits were obtained for the described study, which complied with all relevant regulations.

Our field observations of isolated adults (i.e. >50 m to the next nearest K. malaccensis) indicated that most recruits occur within 10 m of the parental tree. All recruits sampled had their compass bearing and distance measured to the nearest adult K. malaccensis, which was designated the putative mother. For each recruit, 3 to 5 leaflets were collected, placed in labelled bags and dehydrated with silica gel. We used one or both of the following sampling methods for adults: if no other adult K. malaccensis tree was within 50 m, a freshly fallen leaf was collected directly under the crown. If other adults were within 50 m, a 1 cm×1 cm bark scraping (<4 mm deep) was taken using sterilized equipment, placed in a 1.5 mL Eppendorf tube, and stored at −20°C the same day.

Sample DNA was extracted using a modified CTAB-chloroform extraction [37] using an OMNI bead disruptor for initial tissue lysis. DNA samples are stored in TE at −20°C.

Microsatellite amplification and genotyping

Eight highly robust microsatellite markers [38] (Table 1) were fluorescently labelled (VIC, FAM, NED, or PET: Applied Biosystems, Carlsbad, CA, USA) and grouped into two multiplex reactions of four markers each. PCR conditions for both multiplexes were as follows: 0.25 unit Taq polymerase and 0.5 unit buffer (New England BioSciences, Ipswich, MA, USA), 20 to 80 picomol of each F- and R- primer; 5 mM dNTPs, 2.5 mM MgCl₂ and 1% DMSO final concentration in 5 µl total volume reactions. Cycling conditions: 95°C for 1.5 minutes; 34× 95°C for 30 seconds, 56°C for 30 seconds, 72°C for 30 seconds, and a final 5 minute 72°C extension. For each sample, the two multiplex PCR reactions were pooled for a single fragment analysis (ABI 3100 or 3730x) and run against a 600-LIZ internal size standard (ABI). Samples were scored using GeneMapper v. 4.1. One-third of adults and 0.7% of the recruit cohorts were re-extracted, PCRed and genotyped in their entirety, and 100% of adults and 24.9% of the recruit cohorts were re-PCRed and genotyped at 1 to 8 loci to confirm rare alleles as well as to estimate genotyping errors. Genotyping errors were calculated per locus as the percent of non-matching alleles in a second PCR (Table 1).

Download:

Table 1. Sample size (N) per locus, size range of alleles, total number of alleles detected, observed (H_o) and expected (H_e) heterozygosity, genotyping error rate, percent missing data, percent null alleles, and inbreeding co-efficient values (F_IS).

https://doi.org/10.1371/journal.pone.0082632.t001

Data analysis

Microchecker v. 2.2.3 [39] was used to detect the presence of null alleles as well as possible technical scoring errors. Given the high levels of null alleles detected in locus Km071, this locus was excluded from the calculations of F_IS values (the inbreeding co-efficient, used to detect inbreeding), as the conclusions drawn would not be biologically accurate—i.e. a high F_IS value at this locus would not reflect inbreeding. To evaluate the adequacy of our field sampling in capturing the full range of allelic diversity in Singapore, we produced curves for each locus depicting the accumulation of alleles as a function of the number of samples analyzed.

Allelic richness (R), expected and observed heterozygosity (H_e, H_o) and inbreeding co-efficients (F_IS) were calculated with Fstat v. 2.9.3 [40]. H_e and R among age classes were compared with an Analysis of Variance at the locus level. For mating system calculations, seedlings or saplings were classified as the progeny of the nearest adult K. malaccensis. Mltr v. 3.4 [41] was used to calculate outcrossing rates (t_m) and biparental inbreeding rates (t_m-t_s) under default parameters and with the possibility of null alleles; the putative maternal genotype for each set of putative offspring was included. Four chosen loci (Km011, Km127, Km158a, Km180) were analysed for a random subsample of 9 adults in order to compare the mltr results from the Singapore population to the Malaysia population under more analogous parameters.

M_P_val v. 2 [42] was used to detect a potential reduction in effective population size in the adult as well as recruit cohorts, as theory predicts that a population reduction of 50% is expected to lead to a 20% loss of discrete allele sizes and would correspond to gaps in the allele size distribution [42]. An M-ratio value <0.70 generally indicates that a population has undergone a reduction in effective population size, whereas a population with an M-ratio value >0.80 indicates population stability. The program Bottleneck [43] utilizes the assumption that rare alleles will be lost more quickly than heterozygotes during a bottleneck, and detects a mode-shift under conditions consistent with a recent bottleneck. Depending on the initial population size, the decay in the genetic signature of bottleneck detection under random mating means that bottlenecks can usually be detected no more than 2 to 3 generations prior to the current generation, whereas M_P_val can be used to detect more historical population size reductions. Bottleneck v. 1.2.02 was run for adult, juvenile and seedling populations respectively, using the infinite allele and stepwise mutation models with 10,000 iterations per run.

Spatial genetic structure (SGS) for the adults, saplings, and seedlings was analyzed with GenAlex v. 6.5.3 [44] under the following parameters: 999 permutations and 999 bootstraps, user-defined distance classes (5, 10, 20, 40, 80, 160, 320, 640, 1280, 2560, and 5120 m)—except no 5 m distance class was possible for the adult cohort due to too few individuals within this distance class. Significance of autocorrelogram results (testing both the null hypothesis of no structure for each cohort as well as significant differences between cohorts) was determined using the “Multiple pops” option using the above parameters and the adult cohort distance classes (i.e. 0–10 m for the first distance class). The relatedness metric “r” (Queller & Goodnight) [45] was also obtained from GenAlex; full siblings have r-values of 0.50 and half siblings, 0.25. SpaGeDi v. 1.4 [46] was used to determine spatial genetic structure intensity (S_p) using the formula -b/(1-F₁), where b is the slope of the regression between pairwise kinship coefficients [47] and the logarithm of spatial distance, and F₁ is the mean pairwise kinship coefficient within the first distance class (see [48]). Identical (adult) distances classes were entered for all three cohorts. Approximate 95% confidence intervals for S_p were calculated as±2× the standard error of b [49]. Calculations were performed with 9999 permutations and jackknifing over loci.

Results

Genetic diversity was high and not significantly different between age classes (H_e = 0.843–0.854, p = 0.98; R = 16.72–19.45, p = 0.20) (Table 2). The inbreeding co-efficient values for all three age classes were low (F_IS = 0.013–0.076) and not significant (p = 0.18), although there was an upward trend in younger age classes (Table 2). There were twelve private alleles in the seedling cohort and two private alleles in the adult cohort (Figure 2). Over the whole dataset, the multilocus outcrossing rate was high (t_m = 0.961, SD 0.015) and the biparental inbreeding rate (t_m-t_s) was 0.111 (SD 0.011) (Table 2).

Download:

Figure 2. Allele frequencies for each locus.

Alleles present in very low numbers have the frequency presented above the allele if the bar too small to be easily visualized. All potential allele sizes within the captured range (assuming normal dinucleotide repeat units) are included in the x-axis; blanks indicate that this allele size was not captured. A dot below the allele indicates an unusual size (1 bp difference from nearest allele). ‘a’ indicates a private allele in the adult age class, ‘s’ indicates a private allele in the seedling cohort.

https://doi.org/10.1371/journal.pone.0082632.g002

Download:

Table 2. Comparisons of genetic diversity, reproductive, and spatial genetic structure indices among Singapore K. malaccensis cohorts with reference to Malaysian populations [34], [35].

https://doi.org/10.1371/journal.pone.0082632.t002

We detected neither a genetic bottleneck in any age class (no mode shift) nor a reduction in effective population size (M-ratio = 0.857 and 0.957 for the adult and recruit cohorts, respectively) (Table 2). The large number of rare alleles and the few allele size gaps in most loci support this conclusion (Figure 2). Rare alleles (<1%) were most common in tails of the size distribution for most loci (Figure 2) and comprised 30.1% of the total data set (Figure 3). Allele accumulation curves (Figure 4) indicated that we likely captured the majority of the alleles present at these loci in K. malaccensis within Singapore.

Download:

Figure 3. Frequency of different allele frequency classes.

https://doi.org/10.1371/journal.pone.0082632.g003

Download:

Figure 4. Allelic richness accumulation curves for each locus.

Rarefaction curves were produced in the R statistical environment [69].

https://doi.org/10.1371/journal.pone.0082632.g004

All three cohorts had spatial genetic autocorrelograms that deviated significantly from the null hypothesis, i.e. all had significant genetic structure (p<0.001 for all) (Figure 5). The SGS of the adult cohort was significantly different from the seedling and sapling cohorts (p<0.001), although the seedling and sapling cohorts were not significantly different (p = 0.30). In addition, the overall mean relatedness (r) at the 0–10 m distance class for the adult cohort was approximately half that of the recruit cohorts; this difference was significant between adults and the seedling and sapling cohorts (p = 0.021 and 0.029, respectively) although not between the seedlings and saplings (p = 0.36) (Table 2). S_p was lowest in the adult cohort (0.0044), and increased with younger cohorts (saplings = 0.0131, seedlings = 0.0287) (Table 2). The non-overlapping confidence intervals indicate significant differences in S_p between each of the three age classes.

Download:

Figure 5. Spatial genetic structure autocorrelograms for adults, saplings, and seedlings, respectively.

‘r’ is the relatedness coefficient of Queller & Goodnight [45] as implemented in GenAlex. Error bars are 95% confidence intervals for the value of r. Upper and lower bounds (dashed lines) represent the 95% confidence intervals for the null hypothesis (no spatial structure).

https://doi.org/10.1371/journal.pone.0082632.g005

Discussion

Gene flow and intergenerational studies using exhaustively surveyed landscapes are crucially lacking in the tropical tree literature [50]. In the few recent studies using comprehensive surveys (i.e. >95% of the population likely captured) for formerly widespread tree species, significant reductions in genetic diversity were most often detected in the recruit cohort [22], [23], [24]. Thus, our results stand in contrast to theoretical predictions as well as many experimental studies of the genetic consequences of past deforestation (reviewed in [22]). This study detected neither a genetic bottleneck nor a reduction in effective population size, as well as high genetic diversity and a large proportion of rare alleles in all age classes. Genetic studies of K. malaccensis in Malaysia [33], [34], [35] provide some reference points (Table 2) and indicate that genetic diversity of K. malaccensis in Singapore is generally on par with populations studied in more intact Malaysian forests. While these unexpected results may not be representative of the majority of primary forest species (and demographic factors are still a cause for concern) they demonstrate lower than expected genetic vulnerability for a locally threatened, primary forest tree species in a highly urbanized landscape.

Genetic diversity and population viability

The initial decline in genetic diversity during habitat loss is due to demographic factors, i.e. the removal of individuals. Subsequently, genetic drift and reduced genetic diversity generally results from only a subset of individuals contributing gametes to the next generation, e.g. [25], [51]. However the genetic diversity detected for K. malaccensis in Singapore would be considered high even for common, widespread species in pristine habitats [21] and there were no significant differences in genetic indices between age classes. We captured a large fraction of all alleles based on the accumulation curves (Figure 4): alleles with a frequency of less than 1% constituted 30.1% of the data set, and 84.6% of the total number of alleles had a frequency of less than 10% (Figure 3). In addition, there were few gaps in the size distribution (Figure 2), which is contrary to theoretical expectations as well as experimental studies that have documented a loss of alleles during severe population decline [52], [53].

The majority (∼99%) of alleles present in the adult population were also present in the recruit cohort, in addition to twelve private alleles in the seedling cohort (Figure 2). Given that all private alleles were PCRed at least twice, genotyping errors were unlikely to have contributed substantially to these results. Explanations for these rare, putatively private alleles in the seedlings could include the following: 1. Unsampled adult trees containing these rare alleles within the existing primary forest fragments; 2. Gene flow from isolated trees containing these rare alleles located within smaller forest fragments in the intervening landscape or remnant trees in Singapore’s urbanized matrix; 3. Spontaneous mutations, which could account for some of the private alleles given that ∼2,000 seedlings were genotyped, and experimental mutation rates for dinucleotide repeats in plant species is estimated to be ∼2.4×10⁻⁴ mutations/generation/locus [54].

Low inbreeding and high heterozygosity are positive indicators for short-term population viability; high allelic richness and a large proportion of rare alleles indicate an adequately large effective population size for long-term viability [55]. While only a relatively small number of individuals may be needed in the short term to prevent inbreeding, greater genetic variation is thought to increase the likelihood of survival over much longer time scales [11]. Genetic diversity measured via neutral microsatellite markers is likely correlated to genetic variation at functional genes [56], and adaptation or resilience to novel or changing habitats is most often positively correlated to genetic variation [57], [58]. Thus, management strategies should be implemented that maintain as far as possible the full complement of genetic diversity within Singapore.

In the present study, the adult cohort had low SGS intensity (S_p = 0.0044) similar to other tropical tree species (see review of S_p values for trees in [59]). This corresponds to recent research showing that long-lived tropical rainforest canopy tree species pollinated by the highly mobile, migratory bee species Apis dorsata exhibited less SGS compared to species pollinated by smaller, less mobile pollinators [60]. Thus, a signature of frequent long-distance pollen dispersal may remain in the adult cohort. In contrast, the recruit cohorts had significantly different SGS autocorrelogram structure compared to the adult cohort and significantly higher mean relatedness in the 0–10 m distance class. Moreover, SGS intensity significantly increased as cohorts got younger (Table 2). While the results provide only a one-time snapshot of the spatial genetic structure, they could point to one of two trends. First, the results could indicate that recruit cohorts of K. malaccensis are showing sensitivity to habitat loss (e.g. [26]). If this was the case, S_p and mean relatedness in the recruit cohorts would remain comparatively higher than the present adult cohort as they mature. Alternatively, given that the majority of seedlings and saplings die before reaching maturity, the present-day recruit cohorts may converge towards adult SGS intensities as their numbers undergo thinning (e.g. [27]). Long-term population monitoring would be necessary to determine which explanation is most fitting.

The overall outcrossing rate (t_m) for the whole dataset as well as the subset was as high as in the Malaysian populations [34]. However, it is possible that our sampling strategy (see Methods) could have led to a slight inflation of outcrossing rate for the Singapore population. In contrast, the biparental inbreeding rate (t_m-t_s) was much higher overall in Singapore than in Malaysia (Table 2), indicating that trees with higher relatedness are mating at a greater frequency compared to the Malaysian population. While this interpretation would not be unexpected given our other results, there could be confounding variables that reduce the validity of this comparison. In particular, the spatial distance between the trees in the Malaysia study was not reported (they were “distant” from each other [34]), and hence the difference in values could simply reflect that a larger number of adjacent adult trees were sampled in Singapore, with the likely consequence of detecting matings between more closely related neighbours. This speculation is supported by the results of a random subsample of 9 Singapore adults (analyzed using only 4 loci, one of which was the same between studies), in which the biparental inbreeding estimates were on par with the Malaysian population (Table 2). Regardless of comparisons to Malaysia, the overall data from Singapore showed a high level of outcrossing, but the moderate level of outcrossing occurring between related adults could ultimately increase the level of inbreeding.

Ecology, demographics, and population viability

Once a population reaches a critically low threshold of individuals, demographics can play an equal—if not more important—role in long-term population persistence compared to genetic factors, although the two are often inexorably linked [57], [61]. A previous study estimating a universal minimum viable population size of ≈5,000 [62] has been criticized for not taking into account important life history or taxonomic variables [63], and plants may be particularly difficult subjects for calculating minimum viable population sizes [64]. Although effective population sizes can range from around one-tenth to one-third as a proportion of census size [65], a population consisting of around 200 adult trees in Singapore is likely demographically and stochastically vulnerable and hence may not be adequate for long-term viability despite the overall positive genetic data [11], [66].

The higher than expected genetic diversity we detected may be explained by demographic as well as ecological factors. In terms of demographics, a larger number of pre-fragmentation trees than expected may remain in the extant adult population. Hence, a high proportion of Singapore’s pre-deforestation genetic diversity may have been retained to the present day. For example, Singapore’s largest and presumably oldest primary forest trees are protected from lightning by conductors, leading to increased longevity. Conductors have been installed on many trees in our study area, and several mature K. malaccensis have had their conductor rods hit by lightning (E.L. Webb, pers. obs.). In addition, the presence of a small number of large, adult K. malaccensis outside of primary forest habitat suggests persistence of pre-deforestation individuals within the degraded secondary forest and urban matrix.

Ecologically, high outcrossing rates also likely contribute to genetic diversity detected. Apis has been identified as a pollinator of Koompassia (cited in [34]) and the distances between adults are generally within the flight limits and foraging distances of native Apis species in Singapore (A. cerana, A. andreniformis, and A. dorsata) [67]. Our results therefore provide strong evidence that pollination may aid tree species’ resilience in a degraded landscape [68], see also [23]. Hence, the ecology of the pollination vector is a critical determinant of whether pollen dispersal mechanisms can be maintained after habitat loss (and/or fragmentation). In the case of K. malaccensis, two key ecological aspects of pollination may be contributing to the high outcrossing rates of K. malaccensis. First is that A. dorsata forms colonies that may nest preferentially in K. malaccensis; secondly, A. dorsata is a highly mobile insect, capable of pollen transfer during foraging bouts or colony migrations of several km (R. Corlett pers. comm.). Other tree species with less mobile pollinators are unlikely to exhibit the same level of resilience to habitat loss as we have found with K. malaccensis.

Conclusions

Our study stands in contrast to predictions of genetic erosion in formerly abundant species that have sustained rapid and dramatic population declines. However, long-lived species such as trees may need centuries for the accumulated effects of habitat loss and reduced population size to reach a critical threshold, and the century-old extinction debt in Singapore is likely ongoing. Stochastic factors (such as tree falls or continued urbanization) could eliminate rare alleles (e.g. six alleles are represented by a single individual, respectively), and demographic factors such as limited primary forest habitat, lack of recruitment in secondary forest, and a high seedling mortality rate may contribute to a potential extinction debt in the future.

Conversely, the continued functionality of insect pollinators implied by the data may help maintain viability of this small population by increasing the likelihood that the majority of alleles present in the adults are passed via sexual reproduction to the next generation (simultaneously maintaining genetic diversity while reducing genetic drift and inbreeding). Hence, the positive genetic results for Koompassia malaccensis in Singapore despite their demographic vulnerability suggest an “Apis rescue effect”: pollination provides a mechanism through which populations may retain resilience in the increasingly degraded and urbanized landscapes of Southeast Asia. However, the higher relatedness at short distances as well as the increasing SGS intensity in the recruit cohorts could indicate that Singaporean K. malaccensis may be undergoing a shift towards higher mating between (related) neighbours and a contracting neighbourhood size. Hence, the direction, frequency, and distance of pollen and seed dispersal within and between Singapore’s remaining primary forest fragments is crucial to determine the trends in population genetic dynamics resulting from habitat fragmentation.

Acknowledgments

The authors thank S.K.Y. Lum and M.S. Khoo for providing data support. M.A. Niissalo and L. Wijedasa provided field assistance. N.T. Abate, W. Eckman, T.H. Ng, A. Seah, C. Zhang, M. Zhang participated in field work. R.T. Corlett and S.K.Y. Lum provided comments on an early version of the manuscript.

Author Contributions

Conceived and designed the experiments: AN EW. Performed the experiments: AN. Analyzed the data: AN. Contributed reagents/materials/analysis tools: EW. Wrote the paper: AN EW.

References

1. Gibson L, Lee TM, Koh LP, Brook BW, Gardner TA, et al. (2011) Primary forests are irreplaceable for sustaining tropical biodiversity. Nature 478: 378–381.
- View Article
- Google Scholar
2. Yee ATK, Corlett RT, Liew SC, Tan HTW (2011) The vegetation of Singapore—an updated map. Gardens’ Bulletin Singapore 63: 205–212.
- View Article
- Google Scholar
3. Corlett RT (1992) The ecological transformation of Singapore, 1819–1990. Journal of Biogeography 19: 411–420.
- View Article
- Google Scholar
4. Turner IM, Tan HTW, Wee YC, Bin Ibrahim A, Chew PT, et al. (1994) A study of plant species extinction in Singapore: lessons for the conservation of tropical biodiversity. Conservation Biology 8: 705–712.
- View Article
- Google Scholar
5. Corlett RT (1997) The vegetation in the nature reserves of Singapore. Gardens’ Bulletin Singapore 49: 147–159.
- View Article
- Google Scholar
6. Tilman D, May RM, Lehman CL, Nowak MA (1994) Habitat destruction and the extinction debt. Nature 371: 65.
- View Article
- Google Scholar
7. Gibson L, Lynam AJ, Bradshaw CJA, He F, Bickford DP, et al. (2013) Near-complete extinction of native small mammal fauna 25 years after forest fragmentation. Science 341: 1508–1510.
- View Article
- Google Scholar
8. Brook BW, Sodhi NS, Ng PKL (2003) Catastrophic extinctions follow deforestation in Singapore. Nature 424: 420–423.
- View Article
- Google Scholar
9. Chong KY, Tan HTW, Corlett RT (2009) A checklist of the total vascular plant flora of Singapore: native, naturalized and cultivated species. Raffles Museum of Biodiversity Research, Singapore. http://rmbr.nus.edu.sg/raffles_museum_pub/flora_of_singapore_tc.pdf.
10. van Treuren R, Bijlsma N, Ouborg J, van Delden W (1991) The significance of genetic erosion in the process of extinction. I. Genetic differentiation in Salvia pratensis and Scabiosa columbaria in relation to population size. Heredity 48: 996–1008.
- View Article
- Google Scholar
11. Jamieson IG, Allendorf FW (2012) How does the 50/500 rule apply to MVPs? Trends in Ecology and Evolution 27: 578–584.
- View Article
- Google Scholar
12. Allendorf FW (1986) Genetic drift and the loss of alleles versus heterozygosity. Zoo Biology 5: 181–190.
- View Article
- Google Scholar
13. Keller LF, Waller DM (2002) Inbreeding effects in wild populations. Trends in Ecology and Evolution 17: 230–241.
- View Article
- Google Scholar
14. Bijlsma R, Bundgaard J, Boerema A (2000) Does inbreeding affect the extinction risk of small populations?: predictions from Drosophila. Journal of Experimental Biology 13: 502–514.
- View Article
- Google Scholar
15. McGreevy TJ, Dabek L, Husband TP (2010) Genetic evaluation of the association of zoos and aquariums Matschie's tree kangaroo (Dendrolagus matschiei) captive breeding program. Zoo Biology 30: 636–646.
- View Article
- Google Scholar
16. Palomares F, Godoy JA, López-Bao JV, Rodríguez A, Roques S, et al. (2012) Possible Extinction Vortex for a Population of Iberian Lynx on the Verge of Extirpation. Conservation Biology 26: 689–697.
- View Article
- Google Scholar
17. Oostermeijer JGB, Van Eijck MW, Den Nijs HCM (1994) Offspring fitness in relation to population size and genetic variation in the rare perennial plant species Gentiana pneumonanthe (Gentianaceae). Oecologia 97: 289–296.
- View Article
- Google Scholar
18. Durka W, Michalski SG (2007) High selfing and high inbreeding depression in peripheral populations of Juncus atratus. Molecular Ecology 16: 4715–4727.
- View Article
- Google Scholar
19. Honnay O, Jacquemyn H (2007) Susceptibility of common and rare plant species to the genetic consequences of habitat fragmentation. Conservation Biology 21: 823–831.
- View Article
- Google Scholar
20. Kramer AT, Ison JL, Ashley MV, Howe HF (2008) The paradox of forest fragmentation genetics. Conservation Biology 22: 878–885.
- View Article
- Google Scholar
21. Lowe AJ, Boshier D, Ward M, Bacles CEF, Navarro C (2005) Genetic resource impacts of habitat loss and degradation; reconciling empirical evidence and predicted theory for neotropical trees. Heredity 95: 255–273.
- View Article
- Google Scholar
22. Vranckx G, Jacquemyn H, Muys B, Honnay O (2012) Meta-Analysis of Susceptibility of Woody Plants to Loss of Genetic Diversity through Habitat Fragmentation. Conservation Biology 26: 228–237.
- View Article
- Google Scholar
23. Rosas F, Quesada M, Lobo JA, Sork VL (2011) Effects of habitat fragmentation on pollen flow and genetic diversity of the endangered tropical tree Swietenia humilis (Meliaceae). Biological Conservation 144: 3082–3088.
- View Article
- Google Scholar
24. Sebbenn AM, Carvalho ACM, Freitas MLM, Moraes SMB, Gaino APSC, et al. (2011) Low levels of realized seed and pollen gene flow and strong spatial genetic structure in a small, isolated and fragmented population of the tropical tree Copaifera langsdorffi Desf. Heredity 106: 134–145.
- View Article
- Google Scholar
25. Finger A, Kettle CJ, Kaiser-Bunbury CN, Valentin T, Mougal J, et al. (2012) Forest fragmentation genetics in a formerly widespread island endemic tree: Vateriopsis seychellarum (Dipterocarpaceae). Molecular Ecology 21: 2369–2382.
- View Article
- Google Scholar
26. Wang R, Compton SG, Chen X-Y (2011) Fragmentation can increase spatial genetic structure without decreasing pollen-mediated gene flow in a wind-pollinated tree. Molecular Ecology 20: 4421–4432.
- View Article
- Google Scholar
27. Vieira F de A, Fajardo CG, de Souza AM, Reis CAF, de Carvalho D (2012) Fine-scale genetic dynamics of a dominant neotropical tree in the threatened Brazilian Atlantic Rainforest. Tree Genetics & Genomes 8: 1191–1201.
- View Article
- Google Scholar
28. Newman D, Tallmon DA (2001) Experimental Evidence for Beneficial Fitness Effects of Gene Flow in Recently Isolated Populations. Conservation Biology 15: 1054–1063.
- View Article
- Google Scholar
29. IUCN (2012) Red List of Threatened Species. Version 2012.2. Available at: www.iucnredlist.org. Accessed 10 November 2013.
30. Dutta A (2001) Biodiversity and Ecosystem Conservation, Kalpaz Publications. Pg. 173.
31. Corlett RT (1990) Flora and reproductive phenology of the rain forest at Bukit Timah, Singapore. Journal of Tropical Ecology 6: 55–63.
- View Article
- Google Scholar
32. Shono K, Davies SJ, Chua YK (2007) Performance of 45 native tree species on degraded lands in Singapore. Journal of Tropical Forest Science 19: 25–34.
- View Article
- Google Scholar
33. Lee KM, Wickneswari R, Choong CY (2007) Stand Structure and the Genetic Diversity of Koompassia malaccensis and Dryobalanops aromatica in Unlogged and Logged-over Stands. Sains Malaysiana 36: 233–242.
- View Article
- Google Scholar
34. Lee CT, Lee SL, Faridah QZ, Siraj SS, Ng KKS, et al. (2008) Genetic Diversity Assessment of Koompassia malaccensis.. Pertanika Journal of Tropical Agricultural Science 31: 127–133.
- View Article
- Google Scholar
35. Lee CT, Lee SL, Ng KKS, Faridah QZ, Siraj SS, et al. (2011) Estimation of outcrossing rates in Koompassia malaccensis from an open-pollinated population in peninsular Malaysia using microsatellite markers. Journal of Tropical Forest Science 23: 410–416.
- View Article
- Google Scholar
36. National Parks Board (2012) Available at: www.nparks.gov.sg.Accessed 1 November 2013.
37. Doyle JJ, Doyle JL (1987) A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11–15.
- View Article
- Google Scholar
38. Lee CT, Lee SL, Faridah QZ, Siraj SS, Ng KKS, et al. (2006) Isolation and characterization of microsatellite markers in Koompassia malaccensis (Leguminosae), an important tropical timber species. Molecular Ecology Notes 6: 1198–1201.
- View Article
- Google Scholar
39. van Oosterhout C, Hutchinson WF, Wills DPM, Shirley P (2004) MICRO-CHECKER: software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes 4: 535–538.
- View Article
- Google Scholar
40. Goudet J (2001) FSTAT (version 2.9.3), a program to estimate and test gene diversities and fixation indices.
41. Ritland K (2002) Extensions of models for the estimation of mating systems using independent loci. Heredity 88: 221–228.
- View Article
- Google Scholar
42. Garza JC, Williamson EG (2001) Detection of reduction in population size using data from microsatellite loci. Molecular Ecology 10: 305–318.
- View Article
- Google Scholar
43. Cornuet JM, Luikart G (1997) Description and power analysis of two tests for detecting recent population bottlenecks from allele frequency data. Genetics 144: 2001–2014.
- View Article
- Google Scholar
44. Peakall R, Smouse PE (2012) GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research—an update. Bioinformatics 28: 2537–2539.
- View Article
- Google Scholar
45. Queller CD, Goodnight KF (1989) Estimating relatedness using genetic markers. Evolution 43: 258–275.
- View Article
- Google Scholar
46. Hardy OJ, Vekemans X (2002) SPAGeDi: a versatile computer program to analyse spatial genetic structure at the individual or population levels. Molecular Ecology Notes 2: 618–620.
- View Article
- Google Scholar
47. Loiselle BA, Sork VL, Nason J, Graham C (1995) Spatial genetic structure of a tropical understorey shrub, Psychotria officinalis (Rubiaceae). American Journal of Botany 82: 1420–1425.
- View Article
- Google Scholar
48. Vekemans X, Hardy OJ (2004) New insights from fine-scale spatial genetic structure analysis in plant populations. Molecular Ecology 13: 921–935.
- View Article
- Google Scholar
49. Fenster CB, Vekemans X, Hardy OJ (2003) Quantifying gene flow from spatial genetic structure data in a metapopulation of Chamaecrista fasciculate (Leguminosae). Evolution 57: 995–1007.
- View Article
- Google Scholar
50. Hamrick JL (2012) Tropical breeding systems: one and done? Heredity 109: 330–331.
- View Article
- Google Scholar
51. Fiumera AC, Parker PG, Fuerst PA (2000) Effective Population Size and Maintenance of Genetic Diversity in Captive-Bred Populations of a Lake Victoria Cichlid. Conservation Biology 14: 886–892.
- View Article
- Google Scholar
52. Hauser L, Adcock GJ, Smith PJ, Bernal Ramírez JH, Carvalho GR (2002) Loss of microsatellite diversity and low effective population size in an overexploited population of New Zealand snapper (Pagrus auratus). Proceedings of the National Academy of Science U. S. A. 99: 11742–11747.
- View Article
- Google Scholar
53. Pastor T, Garza JC, Allen P, Amos W, Aguilar A (2004) Low Genetic Variability in the Highly Endangered Mediterranean Monk Seal. Journal of Heredity 95: 291–300.
- View Article
- Google Scholar
54. Thuillet A-C, Bru D, David J, Roumet P, Santoni S, et al. (2002) Direct Estimation of Mutation Rate for 10 Microsatellite Loci in Durum Wheat, Triticum turgidum (L.) Thell. ssp durum desf. Molecular Biology and Evolution 19: 122–125.
- View Article
- Google Scholar
55. Frankham R (2005) Genetics and Extinction. Biological Conservation 126: 131–140.
- View Article
- Google Scholar
56. Väli U, Einarsson A, Waits L, Ellegren H (2008) To what extent do microsatellite markers reflect genome-wide genetic diversity in natural populations? Molecular Ecology 17: 3808–3817.
- View Article
- Google Scholar
57. Willi Y, Hoffmann AA (2009) Demographic factors and genetic variation influence population persistence under environmental change. Journal of Evolutionary Biology 22: 124–133.
- View Article
- Google Scholar
58. Nuismer SL, MacPherson A, Rosenblum EB (2012) Crossing the threshold: gene flow, dominance and the critical level of standing genetic variation required for adaptation to novel environments. Journal of Evolutionary Biology 25: 2665–2671.
- View Article
- Google Scholar
59. Lasso E, Dalling JW, Bermingham E (2011) Strong spatial genetic structure in five tropical Piper species: should the Baker–Fedorov hypothesis be revived for tropical shrubs? Ecology and Evolution 1: 502–516.
- View Article
- Google Scholar
60. Harata T, Nanami S, Yamakura T, Matsuyama S, Chong L, et al. (2012) Fine-scale Spatial Genetic Structure of Ten Dipterocarp Tree Species in a Bornean Rain Forest. Biotropica 44: 586–594.
- View Article
- Google Scholar
61. Lande R (1988) Genetics and Demography in Biological Conservation. Science 241: 1455–1460.
- View Article
- Google Scholar
62. Traill LW, Bradshaw C JA, Brook BW (2007) Minimum viable population size: A meta-analysis of 30 years of published estimates. Biological Conservation 139: 159–166.
- View Article
- Google Scholar
63. Flather CH, Hayward GD, Beissinger SR, Stephens PA (2011) Minimum viable populations: is there a ‘magic number’ for conservation practitioners? Trends in Ecology and Evolution 26: 307–316.
- View Article
- Google Scholar
64. Menges ES (2000) Population viability analyses in plants: challenges and opportunities. Trends in Ecology and Evolution 15: 51–56.
- View Article
- Google Scholar
65. Frankham R (1995) Effective population size/adult population size ratios in wildlife: a review. Genetical Research 66: 95–107.
- View Article
- Google Scholar
66. Traill LW, Brook BW, Frankham R, Bradshaw CJA (2010) Pragmatic population viability targets in a rapidly changing world. Biological Conservation 143: 28–34.
- View Article
- Google Scholar
67. Corlett RT (2011) Honeybees in natural ecosystems. Pp. 215–225 in H.R. Hepburn and S.E. Radloff (eds.) Honeybees of Asia, Springer-Verlag Berlin Heidelberg.
68. Corlett RT (2001) Pollination in a degraded tropical landscape. Journal of Tropical Ecology 17: 155–161.
- View Article
- Google Scholar
69. R Core Team (2012). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available at: http://www.R-project.org/. Accessed 1 November 2013.

[ref1] 1. Gibson L, Lee TM, Koh LP, Brook BW, Gardner TA, et al. (2011) Primary forests are irreplaceable for sustaining tropical biodiversity. Nature 478: 378–381.
View Article
Google Scholar

[2] View Article

[3] Google Scholar

[ref2] 2. Yee ATK, Corlett RT, Liew SC, Tan HTW (2011) The vegetation of Singapore—an updated map. Gardens’ Bulletin Singapore 63: 205–212.
View Article
Google Scholar

[5] View Article

[6] Google Scholar

[ref3] 3. Corlett RT (1992) The ecological transformation of Singapore, 1819–1990. Journal of Biogeography 19: 411–420.
View Article
Google Scholar

[8] View Article

[9] Google Scholar

[ref4] 4. Turner IM, Tan HTW, Wee YC, Bin Ibrahim A, Chew PT, et al. (1994) A study of plant species extinction in Singapore: lessons for the conservation of tropical biodiversity. Conservation Biology 8: 705–712.
View Article
Google Scholar

[11] View Article

[12] Google Scholar

[ref5] 5. Corlett RT (1997) The vegetation in the nature reserves of Singapore. Gardens’ Bulletin Singapore 49: 147–159.
View Article
Google Scholar

[14] View Article

[15] Google Scholar

[ref6] 6. Tilman D, May RM, Lehman CL, Nowak MA (1994) Habitat destruction and the extinction debt. Nature 371: 65.
View Article
Google Scholar

[17] View Article

[18] Google Scholar

[ref7] 7. Gibson L, Lynam AJ, Bradshaw CJA, He F, Bickford DP, et al. (2013) Near-complete extinction of native small mammal fauna 25 years after forest fragmentation. Science 341: 1508–1510.
View Article
Google Scholar

[20] View Article

[21] Google Scholar

[ref8] 8. Brook BW, Sodhi NS, Ng PKL (2003) Catastrophic extinctions follow deforestation in Singapore. Nature 424: 420–423.
View Article
Google Scholar

[23] View Article

[24] Google Scholar

[ref9] 9. Chong KY, Tan HTW, Corlett RT (2009) A checklist of the total vascular plant flora of Singapore: native, naturalized and cultivated species. Raffles Museum of Biodiversity Research, Singapore. http://rmbr.nus.edu.sg/raffles_museum_pub/flora_of_singapore_tc.pdf.

[ref10] 10. van Treuren R, Bijlsma N, Ouborg J, van Delden W (1991) The significance of genetic erosion in the process of extinction. I. Genetic differentiation in Salvia pratensis and Scabiosa columbaria in relation to population size. Heredity 48: 996–1008.
View Article
Google Scholar

[27] View Article

[28] Google Scholar

[ref11] 11. Jamieson IG, Allendorf FW (2012) How does the 50/500 rule apply to MVPs? Trends in Ecology and Evolution 27: 578–584.
View Article
Google Scholar

[30] View Article

[31] Google Scholar

[ref12] 12. Allendorf FW (1986) Genetic drift and the loss of alleles versus heterozygosity. Zoo Biology 5: 181–190.
View Article
Google Scholar

[33] View Article

[34] Google Scholar

[ref13] 13. Keller LF, Waller DM (2002) Inbreeding effects in wild populations. Trends in Ecology and Evolution 17: 230–241.
View Article
Google Scholar

[36] View Article

[37] Google Scholar

[ref14] 14. Bijlsma R, Bundgaard J, Boerema A (2000) Does inbreeding affect the extinction risk of small populations?: predictions from Drosophila. Journal of Experimental Biology 13: 502–514.
View Article
Google Scholar

[39] View Article

[40] Google Scholar

[ref15] 15. McGreevy TJ, Dabek L, Husband TP (2010) Genetic evaluation of the association of zoos and aquariums Matschie's tree kangaroo (Dendrolagus matschiei) captive breeding program. Zoo Biology 30: 636–646.
View Article
Google Scholar

[42] View Article

[43] Google Scholar

[ref16] 16. Palomares F, Godoy JA, López-Bao JV, Rodríguez A, Roques S, et al. (2012) Possible Extinction Vortex for a Population of Iberian Lynx on the Verge of Extirpation. Conservation Biology 26: 689–697.
View Article
Google Scholar

[45] View Article

[46] Google Scholar

[ref17] 17. Oostermeijer JGB, Van Eijck MW, Den Nijs HCM (1994) Offspring fitness in relation to population size and genetic variation in the rare perennial plant species Gentiana pneumonanthe (Gentianaceae). Oecologia 97: 289–296.
View Article
Google Scholar

[48] View Article

[49] Google Scholar

[ref18] 18. Durka W, Michalski SG (2007) High selfing and high inbreeding depression in peripheral populations of Juncus atratus. Molecular Ecology 16: 4715–4727.
View Article
Google Scholar

[51] View Article

[52] Google Scholar

[ref19] 19. Honnay O, Jacquemyn H (2007) Susceptibility of common and rare plant species to the genetic consequences of habitat fragmentation. Conservation Biology 21: 823–831.
View Article
Google Scholar

[54] View Article

[55] Google Scholar

[ref20] 20. Kramer AT, Ison JL, Ashley MV, Howe HF (2008) The paradox of forest fragmentation genetics. Conservation Biology 22: 878–885.
View Article
Google Scholar

[57] View Article

[58] Google Scholar

[ref21] 21. Lowe AJ, Boshier D, Ward M, Bacles CEF, Navarro C (2005) Genetic resource impacts of habitat loss and degradation; reconciling empirical evidence and predicted theory for neotropical trees. Heredity 95: 255–273.
View Article
Google Scholar

[60] View Article

[61] Google Scholar

[ref22] 22. Vranckx G, Jacquemyn H, Muys B, Honnay O (2012) Meta-Analysis of Susceptibility of Woody Plants to Loss of Genetic Diversity through Habitat Fragmentation. Conservation Biology 26: 228–237.
View Article
Google Scholar

[63] View Article

[64] Google Scholar

[ref23] 23. Rosas F, Quesada M, Lobo JA, Sork VL (2011) Effects of habitat fragmentation on pollen flow and genetic diversity of the endangered tropical tree Swietenia humilis (Meliaceae). Biological Conservation 144: 3082–3088.
View Article
Google Scholar

[66] View Article

[67] Google Scholar

[ref24] 24. Sebbenn AM, Carvalho ACM, Freitas MLM, Moraes SMB, Gaino APSC, et al. (2011) Low levels of realized seed and pollen gene flow and strong spatial genetic structure in a small, isolated and fragmented population of the tropical tree Copaifera langsdorffi Desf. Heredity 106: 134–145.
View Article
Google Scholar

[69] View Article

[70] Google Scholar

[ref25] 25. Finger A, Kettle CJ, Kaiser-Bunbury CN, Valentin T, Mougal J, et al. (2012) Forest fragmentation genetics in a formerly widespread island endemic tree: Vateriopsis seychellarum (Dipterocarpaceae). Molecular Ecology 21: 2369–2382.
View Article
Google Scholar

[72] View Article

[73] Google Scholar

[ref26] 26. Wang R, Compton SG, Chen X-Y (2011) Fragmentation can increase spatial genetic structure without decreasing pollen-mediated gene flow in a wind-pollinated tree. Molecular Ecology 20: 4421–4432.
View Article
Google Scholar

[75] View Article

[76] Google Scholar

[ref27] 27. Vieira F de A, Fajardo CG, de Souza AM, Reis CAF, de Carvalho D (2012) Fine-scale genetic dynamics of a dominant neotropical tree in the threatened Brazilian Atlantic Rainforest. Tree Genetics & Genomes 8: 1191–1201.
View Article
Google Scholar

[78] View Article

[79] Google Scholar

[ref28] 28. Newman D, Tallmon DA (2001) Experimental Evidence for Beneficial Fitness Effects of Gene Flow in Recently Isolated Populations. Conservation Biology 15: 1054–1063.
View Article
Google Scholar

[81] View Article

[82] Google Scholar

[ref29] 29. IUCN (2012) Red List of Threatened Species. Version 2012.2. Available at: www.iucnredlist.org. Accessed 10 November 2013.

[ref30] 30. Dutta A (2001) Biodiversity and Ecosystem Conservation, Kalpaz Publications. Pg. 173.

[ref31] 31. Corlett RT (1990) Flora and reproductive phenology of the rain forest at Bukit Timah, Singapore. Journal of Tropical Ecology 6: 55–63.
View Article
Google Scholar

[86] View Article

[87] Google Scholar

[ref32] 32. Shono K, Davies SJ, Chua YK (2007) Performance of 45 native tree species on degraded lands in Singapore. Journal of Tropical Forest Science 19: 25–34.
View Article
Google Scholar

[89] View Article

[90] Google Scholar

[ref33] 33. Lee KM, Wickneswari R, Choong CY (2007) Stand Structure and the Genetic Diversity of Koompassia malaccensis and Dryobalanops aromatica in Unlogged and Logged-over Stands. Sains Malaysiana 36: 233–242.
View Article
Google Scholar

[92] View Article

[93] Google Scholar

[ref34] 34. Lee CT, Lee SL, Faridah QZ, Siraj SS, Ng KKS, et al. (2008) Genetic Diversity Assessment of Koompassia malaccensis.. Pertanika Journal of Tropical Agricultural Science 31: 127–133.
View Article
Google Scholar

[95] View Article

[96] Google Scholar

[ref35] 35. Lee CT, Lee SL, Ng KKS, Faridah QZ, Siraj SS, et al. (2011) Estimation of outcrossing rates in Koompassia malaccensis from an open-pollinated population in peninsular Malaysia using microsatellite markers. Journal of Tropical Forest Science 23: 410–416.
View Article
Google Scholar

[98] View Article

[99] Google Scholar

[ref36] 36. National Parks Board (2012) Available at: www.nparks.gov.sg.Accessed 1 November 2013.

[ref37] 37. Doyle JJ, Doyle JL (1987) A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11–15.
View Article
Google Scholar

[102] View Article

[103] Google Scholar

[ref38] 38. Lee CT, Lee SL, Faridah QZ, Siraj SS, Ng KKS, et al. (2006) Isolation and characterization of microsatellite markers in Koompassia malaccensis (Leguminosae), an important tropical timber species. Molecular Ecology Notes 6: 1198–1201.
View Article
Google Scholar

[105] View Article

[106] Google Scholar

[ref39] 39. van Oosterhout C, Hutchinson WF, Wills DPM, Shirley P (2004) MICRO-CHECKER: software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes 4: 535–538.
View Article
Google Scholar

[108] View Article

[109] Google Scholar

[ref40] 40. Goudet J (2001) FSTAT (version 2.9.3), a program to estimate and test gene diversities and fixation indices.

[ref41] 41. Ritland K (2002) Extensions of models for the estimation of mating systems using independent loci. Heredity 88: 221–228.
View Article
Google Scholar

[112] View Article

[113] Google Scholar

[ref42] 42. Garza JC, Williamson EG (2001) Detection of reduction in population size using data from microsatellite loci. Molecular Ecology 10: 305–318.
View Article
Google Scholar

[115] View Article

[116] Google Scholar

[ref43] 43. Cornuet JM, Luikart G (1997) Description and power analysis of two tests for detecting recent population bottlenecks from allele frequency data. Genetics 144: 2001–2014.
View Article
Google Scholar

[118] View Article

[119] Google Scholar

[ref44] 44. Peakall R, Smouse PE (2012) GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research—an update. Bioinformatics 28: 2537–2539.
View Article
Google Scholar

[121] View Article

[122] Google Scholar

[ref45] 45. Queller CD, Goodnight KF (1989) Estimating relatedness using genetic markers. Evolution 43: 258–275.
View Article
Google Scholar

[124] View Article

[125] Google Scholar

[ref46] 46. Hardy OJ, Vekemans X (2002) SPAGeDi: a versatile computer program to analyse spatial genetic structure at the individual or population levels. Molecular Ecology Notes 2: 618–620.
View Article
Google Scholar

[127] View Article

[128] Google Scholar

[ref47] 47. Loiselle BA, Sork VL, Nason J, Graham C (1995) Spatial genetic structure of a tropical understorey shrub, Psychotria officinalis (Rubiaceae). American Journal of Botany 82: 1420–1425.
View Article
Google Scholar

[130] View Article

[131] Google Scholar

[ref48] 48. Vekemans X, Hardy OJ (2004) New insights from fine-scale spatial genetic structure analysis in plant populations. Molecular Ecology 13: 921–935.
View Article
Google Scholar

[133] View Article

[134] Google Scholar

[ref49] 49. Fenster CB, Vekemans X, Hardy OJ (2003) Quantifying gene flow from spatial genetic structure data in a metapopulation of Chamaecrista fasciculate (Leguminosae). Evolution 57: 995–1007.
View Article
Google Scholar

[136] View Article

[137] Google Scholar

[ref50] 50. Hamrick JL (2012) Tropical breeding systems: one and done? Heredity 109: 330–331.
View Article
Google Scholar

[139] View Article

[140] Google Scholar

[ref51] 51. Fiumera AC, Parker PG, Fuerst PA (2000) Effective Population Size and Maintenance of Genetic Diversity in Captive-Bred Populations of a Lake Victoria Cichlid. Conservation Biology 14: 886–892.
View Article
Google Scholar

[142] View Article

[143] Google Scholar

[ref52] 52. Hauser L, Adcock GJ, Smith PJ, Bernal Ramírez JH, Carvalho GR (2002) Loss of microsatellite diversity and low effective population size in an overexploited population of New Zealand snapper (Pagrus auratus). Proceedings of the National Academy of Science U. S. A. 99: 11742–11747.
View Article
Google Scholar

[145] View Article

[146] Google Scholar

[ref53] 53. Pastor T, Garza JC, Allen P, Amos W, Aguilar A (2004) Low Genetic Variability in the Highly Endangered Mediterranean Monk Seal. Journal of Heredity 95: 291–300.
View Article
Google Scholar

[148] View Article

[149] Google Scholar

[ref54] 54. Thuillet A-C, Bru D, David J, Roumet P, Santoni S, et al. (2002) Direct Estimation of Mutation Rate for 10 Microsatellite Loci in Durum Wheat, Triticum turgidum (L.) Thell. ssp durum desf. Molecular Biology and Evolution 19: 122–125.
View Article
Google Scholar

[151] View Article

[152] Google Scholar

[ref55] 55. Frankham R (2005) Genetics and Extinction. Biological Conservation 126: 131–140.
View Article
Google Scholar

[154] View Article

[155] Google Scholar

[ref56] 56. Väli U, Einarsson A, Waits L, Ellegren H (2008) To what extent do microsatellite markers reflect genome-wide genetic diversity in natural populations? Molecular Ecology 17: 3808–3817.
View Article
Google Scholar

[157] View Article

[158] Google Scholar

[ref57] 57. Willi Y, Hoffmann AA (2009) Demographic factors and genetic variation influence population persistence under environmental change. Journal of Evolutionary Biology 22: 124–133.
View Article
Google Scholar

[160] View Article

[161] Google Scholar

[ref58] 58. Nuismer SL, MacPherson A, Rosenblum EB (2012) Crossing the threshold: gene flow, dominance and the critical level of standing genetic variation required for adaptation to novel environments. Journal of Evolutionary Biology 25: 2665–2671.
View Article
Google Scholar

[163] View Article

[164] Google Scholar

[ref59] 59. Lasso E, Dalling JW, Bermingham E (2011) Strong spatial genetic structure in five tropical Piper species: should the Baker–Fedorov hypothesis be revived for tropical shrubs? Ecology and Evolution 1: 502–516.
View Article
Google Scholar

[166] View Article

[167] Google Scholar

[ref60] 60. Harata T, Nanami S, Yamakura T, Matsuyama S, Chong L, et al. (2012) Fine-scale Spatial Genetic Structure of Ten Dipterocarp Tree Species in a Bornean Rain Forest. Biotropica 44: 586–594.
View Article
Google Scholar

[169] View Article

[170] Google Scholar

[ref61] 61. Lande R (1988) Genetics and Demography in Biological Conservation. Science 241: 1455–1460.
View Article
Google Scholar

[172] View Article

[173] Google Scholar

[ref62] 62. Traill LW, Bradshaw C JA, Brook BW (2007) Minimum viable population size: A meta-analysis of 30 years of published estimates. Biological Conservation 139: 159–166.
View Article
Google Scholar

[175] View Article

[176] Google Scholar

[ref63] 63. Flather CH, Hayward GD, Beissinger SR, Stephens PA (2011) Minimum viable populations: is there a ‘magic number’ for conservation practitioners? Trends in Ecology and Evolution 26: 307–316.
View Article
Google Scholar

[178] View Article

[179] Google Scholar

[ref64] 64. Menges ES (2000) Population viability analyses in plants: challenges and opportunities. Trends in Ecology and Evolution 15: 51–56.
View Article
Google Scholar

[181] View Article

[182] Google Scholar

[ref65] 65. Frankham R (1995) Effective population size/adult population size ratios in wildlife: a review. Genetical Research 66: 95–107.
View Article
Google Scholar

[184] View Article

[185] Google Scholar

[ref66] 66. Traill LW, Brook BW, Frankham R, Bradshaw CJA (2010) Pragmatic population viability targets in a rapidly changing world. Biological Conservation 143: 28–34.
View Article
Google Scholar

[187] View Article

[188] Google Scholar

[ref67] 67. Corlett RT (2011) Honeybees in natural ecosystems. Pp. 215–225 in H.R. Hepburn and S.E. Radloff (eds.) Honeybees of Asia, Springer-Verlag Berlin Heidelberg.

[ref68] 68. Corlett RT (2001) Pollination in a degraded tropical landscape. Journal of Tropical Ecology 17: 155–161.
View Article
Google Scholar

[191] View Article

[192] Google Scholar

[ref69] 69. R Core Team (2012). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available at: http://www.R-project.org/. Accessed 1 November 2013.

Figures

Abstract

Introduction

Target species

Objectives

Materials and Methods

Sample collection

Microsatellite amplification and genotyping

Data analysis

Results

Discussion

Genetic diversity and population viability

Ecology, demographics, and population viability

Conclusions

Acknowledgments

Author Contributions

References