Introduction
Decapod crustaceans perform an important role in ecosystem function in tropical streams, serving as detritivores and omnivorous consumers (Mantel & Dudgeon, Reference Mantel and Dudgeon2004; De Souza & Moulton, Reference De Souza and Moulton2005) and as an important food source for predators (Hinojosa-Garro & Zambrano, Reference Hinojosa–Garro and Zambrano2004; Nyström et al., Reference Nyström, Stenroth, Holmqvist, Berglund, Larsson and Granelí2006). However, relatively little is known about the ecology of tropical stream decapods, with most studies focused on their zoogeography and taxonomy (Dobson et al., Reference Dobson, Magana, Lancaster and Mathooko2007).
Primary freshwater crabs of the superfamily Potamoidea are arguably the least well studied freshwater decapods. These crabs are adapted to freshwater environments and are independent of the marine environment for their entire life cycle (Yeo et al., Reference Yeo, Ng, Cumberlidge, Magalhães, Daniels and Campos2008a). Many primary freshwater crabs have a restricted geographical distribution, show a high degree of endemism, and display direct development, producing small broods of large, yolky eggs, which hatch directly into juvenile crabs (Hartnoll, Reference Hartnoll, Burggren and McMahon1988; Ng, Reference Ng1988; Anger, Reference Anger1995; Yeo et al., Reference Yeo, Ng, Cumberlidge, Magalhães, Daniels and Campos2008a; Cumberlidge et al., Reference Cumberlidge, Ng, Yeo, Magalhaes, Campos and Alvarez2009).
Anthropogenic habitat disturbance is threatening many species of freshwater crabs, with potamids and pseudothelphusids facing the highest risk of extinction (Cumberlidge et al., Reference Cumberlidge, Ng, Yeo, Magalhaes, Campos and Alvarez2009). Species under threat are often semi-terrestrial, with restricted distributions (Cumberlidge et al., Reference Cumberlidge, Ng, Yeo, Magalhaes, Campos and Alvarez2009). Stenotopic species that are endemic to islands are vulnerable to human-induced disturbances such as development and pollution (Bahir et al., Reference Bahir, Ng, Crandall and Pethiyagoda2005; Ng & Yeo, Reference Ng, Yeo and Chua2007; Yeo et al., Reference Yeo, Ng, Cumberlidge, Magalhães, Daniels and Campos2008a; Cumberlidge et al., Reference Cumberlidge, Ng, Yeo, Magalhaes, Campos and Alvarez2009). Conserving these species is challenging, as they are often found in unprotected areas (Cumberlidge et al., Reference Cumberlidge, Ng, Yeo, Magalhaes, Campos and Alvarez2009) and therefore their decline may go unnoticed because of lack of monitoring (Ng & Yeo, Reference Ng, Yeo and Chua2007).
The freshwater crab fauna of Singapore consists of six species in three families: Geosesarma peraccae and Geosesarma nemesis (Sesarmidae), Parathelphusa maculata, Parathelphusa reticulata and Irmengardia johnsoni (Gecarcinucidae), and Johora singaporensis (Potamidae). The latter three are endemic (Ng, Reference Ng, Chou and Ng1990, Reference Ng1997). With the exception of P. maculata, the species are mostly confined to forest streams (Ng, Reference Ng1997).
The endemic Singapore freshwater crab J. singaporensis Ng, Reference Ng1986 is one of only three freshwater invertebrates in the top 100 most threatened species (Baillie & Butcher, Reference Baillie and Butcher2012). Although most species of Johora are confined to relatively fast-flowing streams in montane regions (Yeo et al., Reference Yeo, Shih, Meier and Ng2007), J. singaporensis is found at relatively low altitudes, although restricted to hilly parts (Ng, Reference Ng1988). First discovered at Jungle Fall Valley stream in Bukit Timah Nature Reserve in 1986, the species was only known from this type locality until subsequent surveys of other parts of the island revealed two other populations in the early 1990s. Historically, the type locality supported the largest population, although numbers there appeared to have fallen (Ng, Reference Ng, Chin, Corlett, Wee and Geh1995, Reference Ng1997), with the last confirmed sighting in the early 1990s. Prior to this study J. singaporensis had only been reported from the two other locations: a stream within a small patch of disturbed secondary forest close to Bukit Batok Nature Park, and a stream at Ministry of Defence land at Bukit Gombak (Ng, Reference Ng, Chou and Ng1990, Reference Ng, Chin, Corlett, Wee and Geh1995, Reference Ng1997; Yeo et al., Reference Yeo, Tan, Ng, Davidson, Ng and Ho2008b). Bukit Timah Nature Reserve is one of only four gazetted nature reserves with the highest level of protection in Singapore; the other two known locations of the species are not within a formal nature reserve and are therefore unprotected. The ecological information available for this species is based on ad-hoc observations and no systematic ecological study has been carried out (Ng, Reference Ng1988; Ng & Yeo, Reference Ng and Yeo2013). The species was initially categorized as Endangered on the IUCN Red List (Yeo et al., Reference Yeo, Tan, Ng, Davidson, Ng and Ho2008b), based on its restricted range, but following a reappraisal it was categorized as Critically Endangered (Cumberlidge et al., Reference Cumberlidge, Ng, Yeo, Magalhaes, Campos and Alvarez2009). J. singaporensis is part of a basal lineage of Johora (Yeo et al., Reference Yeo, Shih, Meier and Ng2007) and represents the southern limit for the distribution of the genus (Ng, Reference Ng1988), which makes its conservation an even greater priority. Our objectives were to determine the status of J. singaporensis through a survey of its known distribution range, including historical collection sites, and to assess the challenges for its conservation.
Study area
Singapore (c. 700 km2) is an island in South-east Asia, at the southern tip of the Malay Peninsula. It is a highly urbanized lowland island, with only remnants of primary forest remaining (Corlett, Reference Corlett1997). The temperature is relatively consistent, with an annual mean of 27 °C, mean annual relative humidity is 84.2%, and rain falls throughout the year (2,342 mm annual total).
Singapore has few hill streams. These are usually shallow (< 1 m), fast flowing, well shaded by the forest canopy but sparse in riparian vegetation, and less disturbed than lowland streams. The substratum consists mainly of rocks and/or sand, with accumulations of leaf litter (Ng, Reference Ng1991, Reference Ng, Wee and Ng1994; Yeo & Lim, Reference Yeo, Lim, Ng, Corlett and Tan2011).
Based on Ng (Reference Ng, Chou and Ng1990, Reference Ng, Chin, Corlett, Wee and Geh1995, Reference Ng1997), museum records in the Raffles Museum of Biodiversity Research at the National University of Singapore, our own unpublished data (PKLN and DCJY), and discussions with carcinologists who have worked and sampled in freshwater environments in Singapore (C.D. Schubart, S.H. Tan & N. Tohru), the known distribution of J. singaporensis comprised three sites: a stream at Jungle Fall Valley in Bukit Timah Nature Reserve (S1), a stream in Bukit Gombak (S2) and a stream in Bukit Batok (S3). The exact locations of these sites are not shown (Fig. 1), to prevent poaching.
The study area, with built-up areas and major roads (exact localities of the Singapore freshwater crab Johora singaporensis populations have been omitted at the request of the National Parks Board). The rectangle on the inset indicates the location of the main map in Singapore.
Methods
We surveyed all three sites during October 2007–February 2008 to determine the status of their J. singaporensis populations. Care was taken to minimize disturbance to the species.
We conducted preliminary surveys (1 day and 1 night), including sieving coarse woody debris and leaf litter, to determine whether the species was present at each of the three sites. We detected J. singaporensis at S2 and S3 but not at S1.
Based on the preliminary surveys we set different objectives for each site. At S1 the objective was to conduct a thorough investigation to determine if J. singaporensis was present. On eight occasions two experienced field workers searched the entire stream by hand during 19.30–21.30. They wore headlamps and combed through coarse debris, turning over small rocks. Nocturnal surveys were conducted because the results of the preliminary surveys suggested that the species was more active and more easily observed at night. In addition, two 3-hour surveys were conducted in the daytime, during 09.30–12.30, using a hand-held push net (0.25 m2, mesh size 0.5 mm). To minimize disturbance to the habitat, care was taken to restore sifted or overturned leaf litter and stones to their original location at the end of each session.
For S2 and S3 our objective was to determine the population densities of J. singaporensis at these sites. Ten 50 × 50 cm quadrats were chosen randomly and searched at night (after 19.30) by two experienced field workers, using headlamps. Each site was sampled on eight occasions. The maximum carapace length of captured crabs (widest part across the carapace) was measured in the field, to the nearest 0.01 mm, using a dial calliper. Gender of adults was determined on the basis of abdomen morphology, with adult males characterized by a triangular-shaped abdomen and adult females by a broad oval-shaped abdomen. The abdomens of smaller individuals (maximum carapace length < 13 mm) are morphologically indistinct and thus could not be differentiated in the field. As such, smaller individuals were considered to be juveniles. Similar care was taken to minimize disturbance to the habitat at sites S2 and S3. Water pH was measured at the start of each survey.
One-way ANOVA was used for inter-site comparison of the number of J. singaporensis observed per survey, population density and water pH. The Student–Newman–Keuls post-hoc test was used to detect significant differences in the pH between sites. Significance levels for all tests were set at P < 0.05. Population density (per m2) was calculated as the number of crabs per quadrat divided by the quadrat area. To calculate the probability of species absence the formula (1 − P)n was used, where n is the number of surveys and P is the mean probability of detection for these three historical sites (McArdle, Reference McArdle1990).
Results
J. singaporensis was absent from S1 (confidence level > 99.9%) but present at S2 and S3 (Table 1). The number of crabs observed per survey and the population density of J. singaporensis were significantly greater at S3 compared to S2 (Table 1). The water pH at S1 was significantly lower than at the other two sites (Table 1). A follow-up survey also discovered a small population of J. singaporensis in a stream in another part of Bukit Timah Nature Reserve, where the pH was higher compared to S1 (Ng, Reference Ng2008).
Results of the surveys for the Singapore freshwater crab Johora Singaporensis, from three sites in Singapore (Fig. 1), with mean number of individuals detected per survey, density, and water pH. One-way ANOVA was used to compare numbers detected per survey, density and pH between sites. The Student–Newman–Keuls test was used to assess the difference in pH between sites.
1 Significant difference between S2 and S3
2 Significant difference between S1 and S2 and between S1 and S3
Over the sampling period of 4 months, 75 individuals of J. singaporensis, comprising 11 males, 12 females and 52 juveniles, were observed at S2. An equal sampling effort at S3 yielded a total of 222 individuals: 26 males, 17 females and 179 juveniles (Fig. 2). No ovigerous or brooding crabs were observed.
Size–frequency distributions of J. singaporensis at S2 and S3.
Discussion
Population status
The results indicate a significantly higher population density of J. singaporensis at S3 compared to S2 (Table 1). A possible reason for this is the presence at S2 of a second species of freshwater crab, the gecarcinucid P. maculata, which is absent from S3. Adults of this more aggressive sympatric species are larger (maximum carapace length > 50 mm in P. maculata, c. 20 mm in J. singaporensis) and may be competing for limited resources such as food and shelter as well as being a predator of J. singaporensis (Ng, Reference Ng1989; authors, pers. obs.). Another consideration is that the stream at S2 runs through an ornamental plant garden, whereas the stream at S3 flows through secondary forest. Other factors such as the quality of the habitat, the number of predators and the amount of food available could potentially influence the population (Donnay & Beissinger, Reference Donnay and Beissinger1993).
A total of 10 surveys (two diurnal and eight nocturnal) failed to detect J. singaporensis in the type locality in Bukit Timah Nature Reserve (S1), indicating at the very least a significant population decline, with the possibility that the species may have been extirpated where it was originally discovered. This stream is located in a natural area with the highest level of formal protection in Singapore, and the population of J. singaporensis there was assumed to be secure and generally free from anthropogenic disturbances. Perhaps for this reason, as well as to minimize disturbance to the species and the locality, this site was rarely surveyed. After the two other populations were discovered, a few subsequent scientific collections of specimens were all conducted at these unprotected sites, mainly for systematic work.
Population structure
Estimating freshwater crab populations is challenging, with different sampling methods predisposed to detecting different size classes (Kino, Reference Kino1990; Dobson et al., Reference Dobson, Magana, Lancaster and Mathooko2007). Dobson et al. (2007) reported that baited traps selectively captured larger crabs, whereas Surber samplers tended to collect smaller crabs; they also reported that only a small number of crabs belonging to the largest size class were captured during an active search by hand, which is consistent with the results of our intensive physical searches.
The absence of brooding or ovigerous females suggests possible cryptic behaviour. Females may have been hiding in water-filled excavations located further away from the stream, to protect their broods from predators. This behaviour has been suggested for the freshwater crabs Geosesarma notophorum (Ng & Tan, Reference Ng and Tan1995) and Trichodactylus fluviatilis (Alarcón et al., Reference Alarcón, Arruda Leme, Cobo, Escobar-Briones and Alvarez2002). There is also the possible influence of seasonality (we did not survey throughout the year). This possibility is being investigated in an ongoing collaborative project between the National Parks Board of Singapore and the National University of Singapore, which will involve sampling over 1 year.
Stream acidification—a possible cause of decline?
Acidification is one possible reason for the apparent disappearance of J. singaporensis at stream S1. It appears that this acidification has occurred relatively recently, in the 1990s; the pH at S1 was 5–6 in the early 1990s (Ong, Reference Ong1965; P.K.L. Ng, unpubl. data). The persistence of J. singaporensis in the other streams, where the pH is significantly higher (Table 1), seems to corroborate its absence in S1. The decrease in pH at S1 may be attributable to acid precipitation (mean pH of precipitation in Singapore is 4.1–4.5; Balasubramanian et al., Reference Balasubramanian, Victor and Chun2001; Zhong et al., Reference Zhong, Victor and Balasubramanian2001; Hu et al., Reference Hu, Balasubramanian and Wu2003), with the difference in pH between S1 and streams S2 and S3 possibly maintained by different buffering capabilities of the bedrock of the streams (Hornung et al., Reference Hornung, Bull, Cresser, Ullyett, Hall and Langan1995). However, this is difficult to verify because of the lack of long-term data on the pH of precipitation and of the streams. We cannot rule out the possibility of other causes of stream acidification, such as subterranean exposure of sulphur or sulphide deposits or decomposition of vegetation (Kwok, Reference Kwok2010).
Stream acidification can lead to a reduction in the diversity of freshwater organisms (Feldman & Connor, Reference Feldman and Connor1992; Courtney & Clements, Reference Courtney and Clements1998). In particular, it is known to have caused declines, or even extirpation, of freshwater crayfish (Malley, Reference Malley1980; Berrill et al., Reference Berrill, Hollett, Margosian and Hudson1985; France & Collins, Reference France and Collins1993). Species may be intolerant of the altered chemical composition of acidified water (Sutcliffe & Hildrew, Reference Sutcliffe, Hildrew, Morris, Taylor, Brown and Brown1989; Rosemond et al., Reference Rosemond, Reice, Elwood and Mulholland1992) or they may be affected indirectly through microhabitat modification and food reduction. However, there are no equivalent studies for any species of freshwater crab. It is possible that in the case of J. singaporensis, juveniles, eggs and individuals that are undergoing ecdysis may be more vulnerable to acidic waters than adults (Y.K. Ip, pers. comm.).
J. singaporensis is the first tropical freshwater invertebrate for which a significant decline has been documented. There are two other endemic crab species in Singapore, I. johnsoni and P. reticulata, which are found only in certain lowland forest streams and freshwater swamp habitats in nature reserves. However, they are believed to be less at risk from acidification because the pH of their preferred habitats is naturally lower (pH 5; Ng & Lim, Reference Ng and Lim1992). I. johnsoni is still relatively common at S1, where J. singaporensis has disappeared. However, it is not known whether the drop in pH affected population levels of I. johnsoni, because there is a lack of baseline data for this species.
Acid precipitation can have far-reaching consequences for all taxa in an ecosystem. It is a growing problem especially in developing countries with rapidly rising emissions of sulphur and nitrogen oxides but, despite its ecological effects, it remains relatively poorly studied in the tropics (Galloway, Reference Galloway1995). Based on soil type, land cover and soil moisture, regions that are particularly sensitive to acidic deposition include southern China, parts of South-east Asia, Amazonia, and Central Africa (Kuylenstierna et al., Reference Kuylenstierna, Cambridge, Cinderby and Chadwick1995, Reference Kuylenstierna, Rodhe, Cinderby and Hicks2001). As these areas are rich in biodiversity and high in endemism (Myers et al., Reference Myers, Mittermeier, Mittermeier, da Fonseca and Kent2000), it is projected that many aquatic species with stenotopic habitat preferences will be threatened by acid rain in the future.
Conservation challenges and action
Anthropogenic disturbance continues to threaten the survival of J. singaporensis at the two unprotected sites where the species is extant. At S2 a heavy rainfall event in December 2006 caused a landslide along a portion of the hill catchment upstream, and drainage modification measures were taken in early 2008 to address this problem. These measures diverted subterranean water away from the stream and caused a drop in the water table and a reduction in water flow (authors, pers. obs.). The relevant authorities were informed promptly and re-diversion of the water source restored the water flow (authors, pers. obs.) but siltation has increased. The land at S3 has been designated for residential use and such development could threaten the survival of the largest known population of J. singaporensis (Ng et al., Reference Ng, Lim, Tan, Ng, Corlett and Tan2011).
Conserving the natural environment while maintaining economic growth and development poses a challenge in Singapore, where land is scarce. The remaining forested areas are fragmented and under pressure to be converted for other use (Fig. 1). A balance between development and habitat conservation is required, with conservation priority preferably being given to areas with endemic species such as J. singaporensis. Efforts are being made to improve connectivity between the fragmented forested/vegetated patches in Singapore, with the construction of Eco-Link, a bridge planted with vegetation, spanning a six-lane highway, to connect the Bukit Timah Nature Reserve and the larger Central Catchment Nature Reserve (NParks, 2011), and the Park Connector Network, which links green areas (e.g. urban parks) and forested nature reserves throughout Singapore.
Following the completion of this study in 2008 the National Parks Board of Singapore was informed about the possible extirpation of J. singaporensis from its type locality in Bukit Timah Nature Reserve and embarked on an action plan to conserve the species. Immediate action was taken to survey other hill streams in Bukit Timah, Bukit Batok and Bukit Gombak to determine whether the species could be present at additional sites. This led to the discovery of a population of J. singaporensis in a second locality in Bukit Timah, in part of a different drainage system (Ng, Reference Ng2008). However, it appears that the population density at this new locality is still lower than that of the other remaining populations at S2 and S3 (Ng, Reference Ng2008). This could, in part, be a result of opportunistic predation by the fish Betta pugnax and competition pressure from the prawn Macrobrachium malayanum, which are absent from S2 and S3 (Ng, Reference Ng2008). Another possible reason for the low abundance of crabs here could be the relatively low pH (lower than S2 and S3 but not as low as S1). Hydrological and biodiversity studies on hill streams were subsequently conducted to examine the distribution and abundance of J. singaporensis and to monitor the conditions of these streams.
The sites of the two unprotected populations are now monitored by the National Parks Board, with the assistance and support of other government agencies, which have taken steps to prevent these two biologically diverse and important sites from being further developed or disturbed.
The status of J. singaporensis remains unchanged; its survival is threatened, as it occurs only at three sites and is absent from its type locality. The biology and ecology of the species remain poorly known. However, ongoing research is evaluating in situ and ex situ conservation strategies for the species and studying its biology and ecology. This project will also investigate the optimal physico-chemical and biological conditions for the species, evaluate the current conditions of the sites and recommend remedial actions to optimize them for the species, recommend a method for long-term monitoring, and identify potentially suitable sites in Singapore for population enhancement.
Acknowledgements
We are grateful to the late Dr Ivan Polunin and his family for generously allowing us to sample the stream on their property, without which this study would not have been possible. We thank Christoph Schubart, David Laurence Higgitt, Alex Ip Yuen Kwong, Tan Swee Hee and Tohru Naruse for their advice, the National Parks Board for granting the survey permit and for providing information included here in the section on conservation action, and Neil Cumberlidge and an anonymous reviewer for their helpful comments and suggestions. We thank the NParks staff Koa Tian Leng and Mishak, who provided field assistance during the follow-up survey. This research was conducted under the permit NP/RP743 and was partially supported by a grant from the National Parks Board (WBS: R167-000-001-490).
Biographical sketches
Daniel Ng's research interests include freshwater stream ecology and conservation, herpetology, and climate change and other anthropogenic impacts. Darren Yeo studies freshwater ecology and biodiversity, aquatic invasions, and systematics and conservation of freshwater decapod crustaceans. Sivasothi N. is interested in the biodiversity, ecology and conservation of mangroves, decapod crustaceans and otters. Peter Ng's research interests lie in the diversity and biology of marine and freshwater crabs in the Indo-West Pacific as well as crustacean and fish systematics.
