Field surveys of the threatened pygmy slow loris Nycticebus pygmaeus using local knowledge in Mondulkiri Province, Cambodia C arly S tarr, K.A.I. Nekaris,Ulrike S treicher and L uke K.-P. Leung Abstract The pygmy slow loris Nycticebus pygmaeus is The pygmy slow (hereafter pygmy) loris Nycticebus pyg- a little-studied primate endemic to Vietnam, Laos, south- maeus is a nocturnal primate endemic to Vietnam, Laos, ern China and eastern Cambodia. Our study aimed to gain southern China and eastern Cambodia (Nisbett & Ciochon, local knowledge on the distribution and ecology of, and 1993; Fooden, 1996; Ratajszczak, 1998; Brandon-Jones et al., threats to, the species by interviewing hunters, traders and 2004). The pygmy loris differs from its congeners in wildlife protection staff, and to verify this information morphology and genetics (Zhang et al., 1993; Wang et al., using a spotlighting survey in three major reserves in Mon- 1996; Groves, 2001, 2004; Roos, 2003, 2004; Roos et al., dulkiri Province, Cambodia. Encounter rates of pygmy 2007) and in seasonal reproductive patterns and litter size loris were assessed along 29 transects (129.5 km), yielding (Izard & Weisenseel, 1989; Fitch-Snyder & Jurke, 1998). observations of 26 individuals. Mean encounter rates were N. pygmaeus is categorized as Vulnerable on the IUCN Red -1 -1 0.40 km in Seima Protection Forest, 0.10 km in Phnom List (Streicher et al., 2008) and all Nycticebus species are on -1 Prich Wildlife Sanctuary and 0.00 km in Mondulkiri Appendix I of CITES because of on increasing and un- Protected Forest. Informants had knowledge of where popu- sustainable demand for slow lorises in international trade lations occurred, their diet, sociality and habitat preferences. (Nekaris & Nijman, 2007). Widespread large population declines were reported and Little is known of the distribution and ecology of slow informants linked this to high hunting pressure, particularly lorises, and most information has come from local inform- in 2001 and 2002. In late 2008 and 2009 we repeated three ants, trade data (Nekaris & Nijman, 2007; Starr et al., 2010) transects with previously high encounter rates in early 2008 and confiscated animals (Streicher, 2003, 2004). The most and failed to detect any lorises. Local informants reported intensive field studies of slow lorises have been conducted high hunting pressure during the previous wet season in two in Malaysia on the greater slow loris Nycticebus coucang of these sites, and a gold mine was underway in the third (Barrett, 1984; Wiens & Zitzmann, 1999, 2003; Wiens, 2002; site. Urgent actions are required to address these population Wiens et al., 2006). Some surveys of slow lorises have been declines and to assess the conservation status of pygmy conducted elsewhere (Duckworth, 1994; Tan, 1994; Fitch- lorises throughout eastern Cambodia. Snyder & Vu, 2002; Radhakrishna et al., 2006; Nekaris Keywords Bunong, Cambodia, hunting, local ecological et al., 2008; Das et al., 2009) as well as a radio-tracking knowledge, Nycticebus pygmaeus, pygmy slow loris study of reintroduced animals in Vietnam (Streicher, 2004). Pygmy lorises are rarely reported in biodiversity assess- ments. The reports available indicate low encounter rates (Nekaris et al., 2008). In regions where pygmy lorises occur Introduction sympatrically with the northern slow loris or the possibility iodiversity loss in Asia is occurring at an unprece- exists that it does so, they are often reported together as Bdented rate, and monitoring the abundance and dis- ‘slow loris’, because of confusion in species identification tribution of species is essential for their conservation and (e.g. Walston, 2005). for assessments for the IUCN Red List (IUCN, 2010). Such Local ecological knowledge is increasingly being used by information is lacking for many Indochinese primates, and ecologists in wildlife monitoring, understanding historical particularly for nocturnal primates because they are often and ongoing threatening processes, and collaborative ac- elusive and are difficult to study during their active periods. tion in conserving wildlife (Steinmetz et al., 2006; Brook & McLachlan, 2008; Jones et al., 2008; Newton et al., 2008; 2008 2009 CARLY STARR (Corresponding author) and LUKE K.-P. LEUNG School of Silvano & Valbo-Jorgensen, ; Anadon et al., ). The Animal Studies, University of Queensland, 4343, Australia. E-mail: knowledge of local people regarding the abundance, ecol- [email protected] ogy and distribution of at least some species of local wildlife K.A.I. NEKARIS Nocturnal Primate Research Group, School of Social Sciences is becoming a useful component of determining a species’ and Law, Oxford Brookes University, Oxford, UK conservation status (Griffon & Griffon, 2000; Steinmetz, ULRIKE STREICHER Wildlife Rescue Programme, Nam Theun 2 Hydropower 2004 2006 2008 Project, Nakai, Laos ; Steinmetz et al., ; Newton et al., ). Large Received 27 August 2009. Revision requested 9 December 2009. numbers of lorises are sold in marketplaces for traditional Accepted 14 January 2010. medicines (Baird, 1993; Walston, 2005; Starr et al., 2010) ª 2010 Fauna & Flora International, Oryx, 0(0), 1–8 doi:10.1017/S0030605310001316 2 C. Starr et al. and hunters may therefore have ecological knowledge of . 3,200 mm in more mountainous parts of the Province the species. (Javier, 1997). Surveyed habitat types were semi-evergreen We therefore aimed to obtain local knowledge of the mixed deciduous and dry dipterocarp forests. Bamboo was distribution and ecology of, and threats to, N. pygmaeus present in all habitat types, with a cover ranging from and verify this information using a spotlight survey. This sparse to continuous. knowledge may be used for developing improved manage- ment, census and conservation strategies for the species. Methods Study area Local ecological knowledge All data were collected in Mondulkiri Province, Cambodia, Ethnic minorities make up 70% of the population in by CRS. Mondulkiri is the largest province within the Mondulkiri; the dominant group is the Bunong people distribution of the pygmy loris in Cambodia (i.e. that part (ADB, 2000; ICC, 2004). A principal livelihood of the of the country east of the Mekong River). It has three major Bunong is the collection of forest products, particularly protected areas, and access to sites was facilitated by liquid tree resin (Evans et al., 2003; McAndrew et al., 2003). conservation organizations. Spotlight surveys and inter- Hunting is a traditional activity for men and is known to views were conducted in the Seima Protection Forest, occur in the area (Richardson, 2003; Drury, 2005). Inform- Phnom Prich Wildlife Sanctuary and Mondulkiri Protected ants were selected from villages in close proximity to Forest (Fig. 1.). Market data were collected opportunisti- surveyed sites. Based on advice from commune leaders or cally from Sen Monorom Township and the neighbouring park staff who came from the area, we selected informants Bousra waterfall (Fig. 1). Mondulkiri has a dry season from on the basis of their reputation of having knowledge and/or November to April and a wet season from May to October, experience with lorises and having lived in the study area with mean annual rainfall of c. 2,000–2,500 mm and for at least 10 years. Semi-structured face-to-face interviews were conducted within an interview framework (Huntington, 2000). In- Thailand Laos terview questions focused on the interviewee’s perceptions of how many types of loris occurred in their area and their Cambodia physical characteristics (as outlined in Streicher & Broadis, 2007). Following this, informants were shown five photo- Vietnam Mondulkiri graphs of each species and were asked to identify if any Protected Forest (and how many) of the pictures resembled the lorises found in the areas described in the interview. Questions about the Phnom Prich ecology of pygmy lorises included dietary observations, Wildlife Sanctuary sociality, time of year young are seen and perceived habitat and vegetation preferences. Questions about threats to local loris populations covered current and historical hunting pressure, where lorises were captured, hunting methods for lorises, and habitats in which they are frequently observed. We used the following definitions of hunting pressures in Sen assessing this during interviews : low (few lorises are hunted Monorom in the area), medium (some hunting of lorises in the area) Seima and high (many lorises are hunted in this area). Protection Forest Given that hunting and trading lorises is illegal we had to rely on a small sample of non-randomly selected ex hunters and traders. The response frequency is reported as Pygmy loris present a y/z, where y 5 number of informants to give the response Pygmy loris not seen and z 5 total number of informants asked the question. 010205km Interviews were conducted with 44 informants in 12 villages near the study sites, immediately before each survey (but only once at sites where we surveyed a transect more FIG. 1 Localities of the 29 spotlight transect surveys in three protected areas in eastern Cambodia. Interviews were conducted than once). All informants were fluent in Khmer. Some at the closest villages to these sites. The shaded rectangle on the questions were open-ended, which allowed the informants inset indicates the location of the main map in eastern to move into other areas of discussion. Interviews lasted Cambodia. 30–120 minutes. ª 2010 Fauna & Flora International, Oryx, 0(0), 1–8 Surveys of Threatened pygmy slow loris 3 Survey method used the linear encounter rate index (number of animals encountered per km; Sutherland, 2002). Slow lorises exhibit During the interviews informants recommended potential numerous characters that enhance detectability: they tend survey sites for lorises. Reconnaissance survey techniques to exploit the middle to lower canopy, they are curious and 2000 (White & Edwards, ) were used for spotlighting sur- do not look away when red lights are used, and they move 2006 2009 29 veys during – . A total of transects were sur- relatively slowly compared to other nocturnal mammals. 4 1 15 veyed during the years (Table ): in Seima Protection Visibility was not noticeably different between sites.