Simon Hedges and Erik Meijaard. December World Wide Fund for Nature Indonesia (WWF) and Centre for International Forestry Research (CIFOR)

Transcription

1 Reconnaissance Survey for Banteng (Bos javanicus) and Banteng Survey Methods Training Project, Kayan-Mentarang National Park, East Kalimantan, Indonesia Simon Hedges and Erik Meijaard December 1999 World Wide Fund for Nature Indonesia (WWF) and Centre for International Forestry Research (CIFOR)

2 CONTENTS CONTENTS... 2 ACKNOWLEDGEMENTS... 4 EXECUTIVE SUMMARY... 5 JUSTIFICATION... 5 AIMS OF THE PROJECT... 5 SUMMARY OF ACTIVITIES... 5 MAIN FINDINGS... 6 MAIN RECOMMENDATIONS... 7 Conduct further genetic work... 7 Carry out a comprehensive Banteng survey in the upper Bahau area... 7 Use of fire as a habitat management tool for Banteng and other ungulates... 7 Prevention of hunting and management of conflict between Banteng and local people... 7 Proposed resettlement of Saben people to Long Pe - Long Mepun area... 7 Opportunities for eco-tourism... 8 Other... 8 CHAPTER 1: BACKGROUND... 9 BANTENG AND BALI CATTLE... 9 GLOBAL DISTRIBUTION AND STATUS OF BANTENG... 9 Historical distribution... 9 Current distribution Subspecies Global status STATUS AND DISTRIBUTION OF BANTENG ON BORNEO Kalimantan (Indonesian Borneo) Sarawak (Malaysian Borneo) Sabah (Malaysian Borneo) THE BANTENG OF THE UPPER BAHAU GRASSLANDS IN KAYAN-MENTARANG NP, EAST KALIMANTAN The importance of the Bahau grasslands for Banteng CHAPTER 2: AIMS OF THE SURVEY CHAPTER 3: REPORT ON ACTIVITIES CALENDAR OF ACTIVITIES DESCRIPTION OF STUDY AREA Long Tua grasslands Long Kayun grassland Long Pe grasslands REPORT ON TRAINING: TRANSECTS AND PLOTS GENETIC SAMPLING REPORT ON BANTENG HUNTING IN THE UPPER BAHAU AREA AND RESULTS OF THE DOOR-TO-DOOR SURVEY IN APAU PING Records of Banteng hunting in the upper Bahau Door-to-door survey for skulls, horns and other animal parts BANTENG SIGHTINGS (AND OTHER RECORDS) REPORT ON THE GRASSLAND BURNING CARRIED OUT BY LOCAL PEOPLE DURING THE SURVEY OTHER MAMMALS AND BIRDS RECORDED DURING THE SURVEY CHAPTER 4: DESIGN OF FUTURE SURVEYS AND MONITORING METHODS Track-count based methods Dung-count based methods PROTOCOL Survey design

3 Field methods Data analysis NEED FOR FURTHER TRAINING PROTOCOLS FOR GENETIC SURVEY WORK CHAPTER 5: RECOMMENDATIONS FOR 1ST 5-YEAR PHASE OF THE 25-YEAR MANAGEMENT PLAN NEED FOR FURTHER GENETIC SAMPLING NEED FOR FURTHER FIELD SURVEYS FIRE AS A MANAGEMENT TOOL CONTROL OF HUNTING/GUARD POSTS PROPOSED RESETTLEMENT OF SABEN PEOPLE TO LONG PE - LONG MEPUN AREA OPPORTUNITIES FOR ECO-TOURISM BUDGET FOR 1 ST 5-YEAR PHASE OF 25-YEAR MANAGEMENT PLAN POSSIBILITIES FOR FUND RAISING RECOMMENDATIONS FOR CIFOR REFERENCES APPENDIX 1: DATA SHEET FOR COLLECTION OF GENETIC SAMPLES APPENDIX 2: PROTOCOLS FOR COLLECTION OF SOFT TISSUE, BLOOD, OR FAECAL MATTER FOR GENETIC ANALYSIS APPENDIX 3: LIST OF BANTENG SAMPLES COLLECTED FOR GENETIC ANALYSIS APPENDIX 4: BANTENG HEADS FOUND DURING SURVEY APPENDIX 5: RESULTS OF DOOR-TO-DOOR SURVEY FOR HORNS, SKULLS, AND OTHER ANIMALS PARTS IN APAU PING APPENDIX 6: DUNG DECAY RATE MONITORING SHEET FOR BANTENG APPENDIX 7: DATA RECORDING SHEET FOR BANTENG DUNG DENSITY TRANSECTS APPENDIX 8: DUNG DECAY STAGES APPENDIX 9: LIST OF BIRDS AND MAMMALS SEEN APPENDIX 10: DETAILED TIME TABLE OF FIELD WORK

4 ACKNOWLEDGEMENTS We would like to thank the Indonesian Ministry of Forestry s Department of Nature Conservation (PKA) for permission to conduct fieldwork in Kayan Mentarang National Park. We would also like to thank World Wide Fund for Nature - Indonesia (WWF-I) and the Center for International Forestry Research (CIFOR) for financial and logistic support, and WWF-I s Dale Withington and Stephan Wulffraat for support and advice. The Keplala Desa and Kepala Adat Besar in Long Alango made us welcome and provided much useful information for which we are very grateful. Last but not least we would also like to acknowledge the field team, Bpks Swaiki, Daud Lawing, Koling Arang, Taping Usang, Awan, Tului, Irang, and Amban Apui, for their enormous help in gathering the data. 4

5 EXECUTIVE SUMMARY Justification Banteng (Bos javanicus) are a wild cattle species listed as Endangered in the 1996 IUCN Red List of Threatened Animals. Their status is a matter for serious concern, since there are very probably only individuals left in the world and the overall population trend is downwards. The Banteng populations in East Kalimantan are little known but potentially very significant given the precarious status of the species on mainland SE Asia, however there have long been concerns about the extent of hybridization between Banteng and domestic and feral livestock on Borneo. The lack of information about the size and genetic status of these populations is therefore a major problem for those charged with formulating conservation strategies for the species, and genetic studies and survey work in East Kalimantan, and especially the upper Bahau area have been identified as top priorities for the species. Aims of the project To conduct a reconnaissance survey in the upper Bahau grasslands area to select the most suitable survey methods for Banteng. To train WWF staff and local people in these survey methods. To collect samples of Banteng blood, skin, muscle, and faeces for subsequent genetic analysis with the aim of determining whether the Banteng in the upper Bahau area have hybridized with domestic or feral livestock. To collect information about the status of the Banteng populations in the upper Bahau area and the threats to those populations. To design a comprehensive survey strategy for Banteng covering: estimation of population densities and if possible population sizes of Banteng groups that regularly use the upper Bahau grasslands; determining other vegetation types that are used and/or needed by Banteng in this area, including the daily and seasonal movements of Banteng between these habitats and the grasslands. Summary of activities An initial reconnaissance of the Long Tua and Long Kayun grasslands and the surrounding forested areas was conducted and the Long Tua grasslands were mapped using a GPS. Once this reconnaissance phase was complete training was provided in dung count-based survey methods. This training included: the laying-out of transects and plots using sighting compasses and topofils (Hipchains); identification of Banteng dung; classification of Banteng dung piles into decay classes and the monitoring of dung decay rates; and the recording and presentation of survey data for analysis. Training was also provided in the methods of collecting tissue and dung samples for genetic analysis. We explained the importance of working with sterile tools and gloves, showed how to take samples, how to store them in buffer tubes, and how to document each sampling event. Banteng horns, skulls, and other trophies were located either by talking to villagers (in Long Alango and Long Berini) or by systematic door-to-door searches (in Apau Ping). All banteng trophies were photographed and, where possible, measured. No recently dead Banteng specimens were found. Skin samples were collected from 11 of the 25 Banteng heads found in villages. No fresh Banteng dung was found but samples were taken from an approximately one-week-old dung pile. The skin and dung specimens will be analyzed in Dr Ettore Randi s laboratory in Italy. A separate report will be written once 5

6 the results of these analyses are known. This second report will also include discussion of Banteng horn morphology (the consultants will compare the data gathered during this survey with that from Banteng specimens held in various museums in Europe, the USA, and Indonesia). To enable WWF staff to collect further samples for genetic analysis of the Banteng populations in the upper Bahau area, we left behind a complete sampling set, including buffer tubes, sterile blades, plastic gloves, sampling protocols, and waterproof markers. This sampling set was taken to the Lalut Birai field station. An experimental burn of a small part of the Long Tua grassland was conducted in order to see how Banteng, Sambar, and muntjac would react to the burning and subsequent re-growth of grass, and to see local rangeland management in practice. The area burnt was mapped using a GPS. All information pertaining to the status of the upper Bahau Banteng populations and threats to those populations gathered from reports and during interviews and village-visits was compiled. Main findings We confirmed that Banteng are still using the Long Tua grassland: a young adult male was seen on 6 November 1999 and what was almost certainly the same animal was seen on a further 2 occasions; and a team member saw a group of 4 Banteng (including females) at the western side of the Long Tua grasslands in the afternoon of 6 November The male was black colored and looked to the authors like a typical Javan Banteng. Evidence of (very probable) Banteng presence was also found at the Long Kayun grassland and the Batu Pahu saltlick (1 recently dropped dung pile and several footprints on 9 November 1999, and recently dropped dung piles and several footprints on 11 November 1999, respectively). Other reports indicated that Banteng are still using the Long Pe area too. When Long Tua village was abandoned in 1979, the villagers cattle were left behind (there were apparently fewer than 10 animals). These cattle were still seen in the Long Tua area until the early 1990s, after that they were not seen again. However, they were seen in mixed groups with Banteng raising the obvious possibility that interbreeding took place, indeed in 1992, Bpk Usang saw an animal on the Long Tua grasslands that he thought was a Banteng x sapi hybrid ( because it had a large dewlap ). It seems likely that similar situations occurred elsewhere in the upper Bahau area. This is obviously a cause for concern and emphasizes the need for continuing genetic study of Banteng in the upper Bahau area to determine the status of these potentially very important populations. Three saltlicks (2 in the Long Tua area and 1 at Batu Pahu) were found and mapped. The Long Kayun grassland and most of Long Tua grassland do not appear to have been burnt for many years, and consequently many areas are now covered in woody scrub. If this forest regeneration continues the grasslands will eventually be lost and with them potentially high quality Banteng habitat. It appears that Banteng are regularly hunted by local people in the upper Bahau area. The main reason seems to be that Banteng eat or otherwise destroy crops, and therefore need to be killed when they approach or enter gardens; although an important secondary reason appears to be trophy hunting. The range of Banteng in the upper Bahau appears to have declined over the last years since we received reports from Long Alango that suggested that Banteng used to be relatively common around this village in the1970s, but now they are never seen. Hunting appears to be one of the main factors contributing to this decline, indeed local people suggested to us that hunting was likely to be the main reason why the species was now rarely seen. Furthermore, this decrease in the species range appears to be continuing because many Banteng had apparently been shot in the Apau Ping and Long Berini areas in the 1990s but local informants told us that the species is now rarely seen near these villages. 6

7 We found 25 Banteng heads in Long Alango, Long Berini, and Apau Ping. These heads usually comprised the horns, parietal, frontal and sometimes the nasal bones, and associated dry skin and hair; in a few cases complete skulls were also found. We were told of a proposed resettlement by up to 1000 Saben people now mostly living in Malaysia in the Long Pe Long Mepun area within what would become an enclave within Kayan-Mentarang National Park. The proposed resettlement seems to be mostly driven by exploitation and development opportunities, and particularly the possibility of establishing an oil palm plantation within the enclave. Needless to say this would present serious problems for Banteng and other wildlife in the Long Pe area, as well as setting a very bad precedent. Main recommendations Conduct further genetic work Continue collecting samples from Banteng faeces on Long Tua and on the other grasslands in the upper Bahau area. Continue collecting samples of blood and other tissues from any Banteng carcasses found during fieldwork. If the samples from Banteng trophies collected during this survey prove tractable (i.e. if they yield useful amounts of Banteng DNA), further trophies should be sampled in upper Bahau villages (and elsewhere in Kayan Mentarang NP). Carry out a comprehensive Banteng survey in the upper Bahau area Purchase recent Landsat TM imagery or other remotely sensed data for the area between Apau Ping and Long Pe. Once this remotely sensed data is available design an extensive dung-count based survey covering the whole upper Bahau area using the guidelines given in Chapter 4. Use of fire as a habitat management tool for Banteng and other ungulates Hire a rangeland management specialist to assess whether a regular burning strategy is needed, and if so how it should be managed for the Long Tua, Long Pe, and Long Kayun grasslands (and other grasslands if these are found to be important for Banteng during the future surveys). Implement the burning strategy by seeking funds for and employing a local supervisor for the Long Tua and Long Pe grasslands (if recommended by the rangeland specialist). Prevention of hunting and management of conflict between Banteng and local people Inform local communities that hunting Banteng is illegal and discuss proper law-enforcement strategies with the conservation authorities. Develop a compensation scheme for damage caused by Banteng to agricultural fields and gardens Establish guard posts at Long Tua and Long Pe. Proposed resettlement of Saben people to Long Pe - Long Mepun area We strongly recommend that WWF and PHPA oppose the resettlement within the proposed Long Pe - Long Mepun enclave in Kayan-Mentarang NP. 7

8 Opportunities for eco-tourism Conduct a feasibility study. Ensure that proper management of the grasslands is in place before promoting eco-tourism to the areas. Investigate how eco-tourism could contribute to the financing of the grassland management project. Other Apply for funding for the Banteng work from the WWF Species Action Fund. 8

9 CHAPTER 1: BACKGROUND Banteng and Bali cattle Banteng (Bos javanicus) are a wild cattle species listed as Endangered in the 1996 IUCN Red List of Threatened Animals (IUCN, 1996). Their status is a matter for serious concern, since there are very probably only individuals left in the world and the overall population trend is downwards (Duckworth and Hedges, 1998; Hedges in prep.; Hedges et al., in prep.; see below). Banteng have been domesticated and the domestic form, which occur widely in Indonesia, are often referred to as Bali cattle. That Bali cattle are in fact a domestic form of Banteng is supported by studies of milk proteins (Bell et al a & b); multivariate analysis of cranial measurements (Hayashi, Otsuka and Nishida 1988); and hybridization studies (Jellinek et al. 1980). Bali cattle and Banteng are fully interfertile, while F1 males from crosses between Bali cattle and Bos taurus (i.e. domestic cattle of both the European and Zebu types) are sterile due to cessation of sperm development at the secondary spermatocyte stage. Infertility has also been noted in 1/4 and 3/4 Bali bulls and the fertility of 1/4 and 3/4 Bali heifers appears to be lower than that of F1 heifers (National Research Council 1983; Moran 1987). Although very similar in appearance to wild Banteng Bali cattle do differ in a number of ways from their wild ancestor: they are smaller and have less well developed withers; their skulls are narrower and lighter than those from wild Banteng; sexual dimorphism is less pronounced; their horns are less developed; sexual maturity is attained earlier and the gestation period is shorter. In addition, some domestic bulls tend to remain rather reddish-brown when mature, rather than attaining the black coloration of most wild male Banteng (of the Javan and Bornean races). Typical mature liveweights under village management are 400 kg for bulls and 300 kg for females, but under feedlot conditions males can grow to 500 kg (National Research Council 1983; Moran 1987; Hayashi, Otsuka and Nishida 1988). An average-sized male wild Banteng of the Javanese or mainland subspecies stands 160 cm at the shoulder and weighs about 635 kg. Adult females are smaller and considerably more slender than adult males. Average-sized cows of the Javan and mainland races stand 140 cm at the shoulder and weigh 400 kg. However, Bornean Banteng are reported to be smaller in stature than the Javan or mainland forms (National Research Council 1983). Global distribution and status of Banteng Historical distribution Historically the species occurred from the hill tracts of eastern Bangladesh, north-eastern India (Manipur), and Myanmar, in the west, to Vietnam in the east and northern peninsular Malaysia in the south; as well as on Java, Borneo, and probably Bali (see below) (Corbet and Hill, 1992; Hedges et al., in prep.). No evidence supports Flower and Lydekker s (1891), Dammerman s (1929), or Pfeffer s (1965) suggestions that Banteng also occurred on Lombok. No fossil evidence is known from Bali, and Wilson and Reeder (1993) state that Banteng were introduced to the island; Corbet and Hill (1992) do not include Bali in the range of wild Banteng, and Van Strien (1986) simply states that they occur as domestic animals. Nevertheless, it does not seem unreasonable to suggest that Banteng should be included among the native fauna of Bali, since several Indo-Malayan mammal species are known to have penetrated eastwards as far as Flores and Timor during times of lowered sea-levels in the Pleistocene (Hooijer, 1975). Ashby and Santiapillai (1988) treat the small number of free-living Banteng which occur in West Bali as wild animals as does Watling (1991); although Watling notes that interbreeding with Bali cattle (domestic Banteng) is a problem and consequently the population is unlikely to consist solely of pure-bred animals. Wind and Amir (1977) had earlier raised similar fears. It is probable that Banteng also occurred on Sumatra. Prehistoric (sub-fossil) remains have been found in the Padang highlands of West Sumatra, and although they are somewhat larger in size than modern Banteng (as were the remains found in the Niah caves in Sarawak) the presence of Banteng on Sumatra would not be surprising considering the land bridges which have existed between it and Java in the past (Hooijer, 1975; Whitten et al., 1987). Wild Banteng are no longer present on Sumatra, however. 9

10 Banteng were first reported for China by Wang Yingxiang (1987), although he accidentally listed Xishuangbanna, in southern Yunnan Province. The true site is Zulin, also in southern Yunnan, which remains the only record. Nevertheless, pending further confirmation of the presence of Banteng, Bleisch (cited in Duckworth and Hedges, 1998) feels that the species should be considered extirpated from China, while Wang Sung argues that until there is a publicly accessible specimen with unambiguous locality data, Banteng should not be considered as having occurred in China (Duckworth and Hedges, 1998). Banteng are extinct in India (Prater, 1971; IUCN, 1978) and Bangladesh (Gittins and Akonda, 1982). On peninsular Malaysia, Banteng have probably been extinct since at least the 1950s (Hislop, 1961); and although Medway (1978) thought that they may still occur in the Sungai Muda area of Kedah near the border with Thailand, possibly also extending into parts of Kelantan, there have been no recent reports. They are also thought to be extinct in Brunei (Payne et al., 1985). Current distribution Wild Banteng currently occur in Myanmar, Thailand, Laos, Cambodia, and Vietnam; and on Java (Indonesia), Bali (Indonesia; although the Banteng there may well be hybrids), and Borneo [Kalimantan (Indonesian Borneo), Sabah (Malaysian Borneo), and possibly Sarawak (Malaysian Borneo)] (Duckworth and Hedges 1998; Hedges et al., in prep.). Domestic Banteng occur on, or have been introduced to, many of the islands of Indonesia including Bali, Java, Lombok, Sulawesi, Sumbawa, Sumba, Sumatra, Timor, and Indonesian Borneo (Kalimantan) (Van der Maarel, 1932; Rollinson, 1984; Corbet and Hill 1992), and introduced to Australia (where there are now large feral herds in the Northern Territory), West Malaysia, New Guinea, the Philippines, and the USA (Kirby, 1979; Moran, 1987; Bowman, 1993). Feral Banteng also occur in Kalimantan (National Research Council, 1983; R. Sözer pers. commun. to E. Meijaard), and introduced Banteng (probably feral animals) occur on the Indonesian islands of Enggano (off Sumatra) and Sangihe (off Sulawesi) (Corbet and Hill 1992; Wilson and Reeder 1993). Subspecies Originally Lydekker (1898) distinguished three races of Banteng: the Javan Bos sondaicus (= javanicus) typicus, the Burmese B. s. birmanicus, and the Manipur B. s. var. The last two were founded on differences in skin coloration. He also provisionally identified a fourth race from Borneo but he was unsure whether the differences in horn shape, which he had noticed in his early specimens were a constant feature. By 1912 he had been able to examine further specimens and it seemed that the differences were indeed diagnostic: the Bornean form apparently has horns that are relatively stouter, are less curved, and have a more upright direction, giving them a smaller maximum span; the Bornean Banteng is also characterized by the flatness of the forehead and the straight intercornual ridge. In Lydekker s opinion these differences fully justified the right to racial distinction and he gave it the name B. s. lowi (Lydekker, 1898; 1912; Van der Maarel, 1932). Lydekker (1905) also described a subspecies from Perak, peninsular Malaysia, B. s. butleri (although this may have been a domesticated form); and another, B. s. porteri, from central Thailand whose chief claim to racial distinction seems to have been a flecked skin (Lydekker, 1909; Van der Maarel, 1932; Corbet and Hill, 1992). Lekagul disputed the validity of B. s. porteri on the grounds that only a few old bulls in the Me Wong area of central Thailand, where the type specimen was shot, showed the white flecks; that the amount of flecking varied from individual to individual; and because the flecked animals lived in herds with Banteng of normal colour (Lekagul and McNeely, 1977). Of these six subspecies, three have come to be generally recognized: (1) Bos javanicus javanicus on Java and Bali, (2) B. j. lowi on Borneo, and (3) B. j. birmanicus on the Asian mainland (Lekagul and McNeely, 1977; National Research Council, 1983). However, Groves and his colleagues recently examined Banteng skulls and horns from Java, Borneo, and the Asian mainland and they concluded that while the Bornean form was subspecifically separable as B. j. lowi, the Javan and mainland forms could not be clearly distinguished; and since javanicus has priority over birmanicus the Javan and mainland forms should both be referred to B. j. javanicus (C. Groves, pers. commun.). 10

11 Unfortunately, the subspecies question is clouded by the uncertain history of Banteng on Borneo. For example, t Hoen (1921 cited by Van der Maarel, 1932) mentions the occurrence of feral Bali cattle on the island of Mojo (north of Sumbawa), which he considered to be probably identical with the form on Borneo and there appears, therefore, to exist still a great amount of doubt as to the origin of the Bornean Banteng (Van der Maarel, 1932:55). Unless of course the Mojo cattle had actually been introduced from Borneo. Nevertheless, Van der Maarel s doubts, together with the long history of reports mentioning interbreeding between wild Banteng and domestic and feral livestock on Borneo (discussed below) call the validity of B. j. lowi into question. Indeed, Grzimek (1990) argues against separating any of the local forms of Banteng into true subspecies because of the effects of interbreeding with domestic or feral cattle, which have apparently affected a number of the remaining populations of wild Banteng and Corbet and Hill (1992) do not recognize any of the putative subspecies of Banteng as valid taxa. Global status The current status of Banteng is not well known but from the limited information that is available it would seem that the total number of Banteng in the world in 1999 is unlikely to be more than 8000 and is quite possibly fewer than 5000 animals. No populations of more than 500 Banteng are known (Ujung Kulon NP in West Java is often reported to contain 800 or more Banteng but such figures are not derived from well designed surveys). Only seven or eight populations of more than 50 Banteng have been reported in the last ten years (six on Java and one or two in Thailand); and even if we assume that several remain in Myanmar and on Borneo it is very unlikely that the total number of such populations exceeds twenty. These figures are informed guesses based on a consideration of aerial and terrestrial survey data (from parts of Thailand, eastern Java, and eastern Cambodia); field visits and interviews with local people, including hunters; and information about levels of trade in Banteng parts, particularly horns. Banteng population trend is clearly downwards on the Asian mainland, apparently relatively stable on Java, and unknown on Borneo (Duckworth and Hedges, 1998; Hedges, in prep.; Hedges et al., in prep). Status and distribution of Banteng on Borneo Doubts have also been expressed about whether Banteng are indigenous to Borneo [see for example Van der Maarel (1932)]. Nevertheless Bos javanicus fossils dating from the later Upper Pleistocene have been recovered from Niah cave in Sarawak (Medway 1972); and Payne et al. (1985), Van Strien (1986), Corbet and Hill (1992), and Wilson and Reeder (1993) all consider Banteng to be part of the island s native fauna. During the last glacial maximum, open vegetation types such as woodland and grassland may have been more widespread that at present. Drier and colder conditions probably created areas of monsoon forest, wooded grassland, and grassland, although the area covered by these vegetation types is still unknown (Adams and Faure, 1997; P. Kershaw, pers. commun., 1999). Thus these conditions may have provided significant areas of suitable habitat for large herbivores such as Banteng. When climatic conditions became warmed and wetter, the re-growth of rainforest must have started to restrict Banteng to sites where open vegetation was maintained as a result of human activities or other natural processes. Early modern humans arriving in the Indonesian archipelago between 5,000 4,000 years ago started to create arable lands, grasslands, savannas, and secondary forests. These vegetation types would have provided Banteng with high quality habitat patches surrounded by rainforest, which would have represented much lower quality habitat (Hoogerwerf, 1938 in Brookfield, 1997). Brookfield (1997) suggests that the human population decline that occurred in Central Borneo in the 19 th century mostly because of disease (Knapen, 1997) reduced the extent of this human-created Banteng habitat, leading to partial or total extinction. Another factor of possible importance for Banteng were the outbreaks of rinderpest in 1871 and 1878, which killed a large proportion of the cattle population of Southeast Borneo, before spreading to the wild pig population, which it affected in areas as far afield as the upper Kahayan and Kapuas Rivers (Knapen, 1997). It is unknown whether these epidemics also affected wild Banteng populations. 11

12 Kalimantan (Indonesian Borneo) Westermann (1939), using his own sources and referring to Witkamp (1932) and Zondag (1931) reported Banteng to be reasonably common in the following areas of south and east Borneo (although they had already started to disappear from some areas): the mountainous areas of the Tidoengsche Landen, the upper Sekatak area (small herds), along the Elor Beroesoe, the Bengara, and the upper reaches of the Malinau River, the open areas near Muara Wahau, the upper S. Boh, Sangkulirang, the Sekerat Mountain, Sangatta, and other areas in east Kutai, the lower Mahakam area (where small herds occurred), and the area between the Mahakam and the Bay of Balikpapan. He also reported Banteng from the Pa Oemoem and Pa Melada areas in the Krayan District, where apparently the species was rarely hunted because of a lack of firearms. Westermann continues to say that Banteng herds move through the Bulungan area, while in Berau they are found in the higher areas between the swamps and in the mountainous forest, especially near Talisayan, Domaring, Madang, Menoembar, Mankalanorang, and Mempaka. In the Apo Kayan area Banteng occurred between Long Heban and Kiham Awon, on the slopes of the watershed in the upper reaches of the Iwan and Maroeng Rivers, and along the lower Kayaniut River. In southern Borneo Westermann reported that Banteng were more or less common in the following areas: the slopes of the Meratus Range (especially on the east side), on Pulau Laut, in the Pasir District, around Gunung Luang, in the eastern and upper Barito lands, and in the Kotawaringin area (from the coast to the Schwaner Mountains!). He also mentions a population that was thought to occur in the Maluka swamps south of Banjarmasin, although this population may have been feral. The current status of Banteng in Kalimantan is poorly known, but they appear to be widespread in East Kalimantan since there have been recent (mid- and late-1990s) reports from many areas including Kayan Mentarang NP, and within or close to Hutan Kapur Sangkulirang (an unprotected area that has been proposed for protection since the early 1980s) (Hedges, in prep.; Hedges et al., in prep.). However, the genetic status of these animals is perhaps a matter for concern since Hoogerwerf (1970) refers to several reports dating from the 1930s and 1940s which mention that many groups of Banteng in Kalimantan and particularly East Kalimantan were no longer pure-bred having interbred with stray domestic cattle. And Schneeberger (1945) mentions domestic cattle that were left behind by villagers in the Kerayan-Kalabit highlands of northern East Kalimantan. Furthermore, similar reports of interbreeding are still being made in the 1990s (R. Sözer, pers. comm; this survey, see Chapter 3). In addition, the Banteng which have in recent times lived in and around the grasslands of south-eastern Kalimantan may be the descendants of feral Bali cattle (domestic Banteng) because in the early 20 th century they were described as cattle (they were kept under a system of free-range husbandry and only rounded up when needed for meat but they were not wild Banteng) (Brookfield, 1997). Sarawak (Malaysian Borneo) Banteng are thought to have followed shifting cultivators into the interior between years ago; prior to this they were probably largely restricted to the alluvial plains of the coastal belt (Wharton, 1968). Low (1848, cited by Wharton) reported that the Banteng in Sarawak lived chiefly in the bamboo forests along the Sangow and Baram rivers, where according to Banks (1931) domestic Banteng were kept by the Kalabit people. These people obtained cattle from Dutch Borneo. It remains unclear, however, whether these imported cattle were of wild or domestic stock, thus raising the possibility that the Banteng populations in Sarawak and neighboring parts of Kalimantan may be partly feral in origin, or contain hybrid animals. Beccari (1904) indicated that the species was not scarce but kept mainly to areas of secondary growth in the interior. Banks (1931) reported that Banteng did not occur south of the Baleh River (a tributary of the Rejang River) in Sarawak, and were found in the headwaters of most rivers to the north of this. He further mentions the Niah district where Banteng were relatively common, Merapok in the Lawas district, Ulu Trusan, Limbang, various places in the Baram, above Tubau in the Ulu Bintulu, at Belaga and down to the head of the Pelagus rapids, but not into the neighboring Mukah and Oya Rivers. In the 1950s, Banteng were still common around the Niah Caves according to Harrisson (1961) but this is one of the few places where it is so in Borneo today, and by 1967 few Banteng were thought to be left in Sarawak (Anderson pers. commun. to Wharton, 1968). In the early-1980s, Banteng apparently persisted in 12

13 the more remote parts of north and east Sarawak (Aken and Kravanagh, 1982) but Payne et al. (1985) stated that there had been no recent reports, although Labang (1987 cited by Caldecott, 1988) reportedly found evidence of their continued presence. However, few, if any Banteng persist in Sarawak today, although they may cross into Sarawak from neighboring Kalimantan where they are known to occur in Kayan Mentarang NP (see above). Sabah (Malaysian Borneo) Prior to the 1940s Banteng were reported to be common in riverain areas in eastern Sabah and in many areas of shifting cultivation in the west and north, even in the hilly interior. However, the subsequent widespread use of firearms led to their rapid extermination from most areas. In 1985, they were reported to be locally common in logged forest on flat terrain, but were again under threat as their habitat was converted into permanent agricultural land (Payne et al., 1985). The current status of Banteng in Sabah is poorly known. They are still present in the eastern lowlands and in the Danum Valley area, and in 1990 Banteng still occurred south of Gunung Lumaku (in the upper reaches of the Padas River) but they had been extirpated from all other parts of south-western Sabah according to local people (Payne, 1990). Recent (mid-1990s) surveys have revealed signs of Banteng in the Ulu Segama, Malua, Kuamut, and Sungai Pinangah Forest Reserves, Danum Valley Protection Forest Reserve, Malubuk Virgin Jungle Reserve, and the Lower Segama region (including the Tabin and Kulamba Wildlife Reserves). However, the survey results suggested that Banteng were living at very low densities in these areas, although a significant population still occurs in the western part of Tabin Wildlife Reserve (Boonratana, no date). Signs of Banteng were also seen in the Lower Kinabatangan area (Boonratana, 1993). Boonratana s surveys suggest that the species status had deteriorated since the beginning of the 1980s because Banteng are presently rare in areas in which they were previously reported as common by Davies and Payne (1982). The Banteng of the upper Bahau grasslands in Kayan-Mentarang NP, East Kalimantan Banteng are found throughout much of the upper Bahau area, but they seem to occur at very low densities (Foead, 1993; Puri, pers. commun. 1995; Foead, 1997; this survey). They appear to be concentrated around the grasslands and forest gardens and rice fields. In the last two, they cause considerable damage. Not only do they feed on rice plants and other crops, they also damage large numbers of crop plants when they climb through fields that are laid out on the steep and slippery hillsides. Banteng are reported to occur in groups of ten or more animals in the upper Bahau area (Foead, 1997; this survey) and once they enter a rice field or forest garden they can destroy a significant percentage of a crop in one night. Foead considered Banteng to be the worst agricultural pest in the upper Bahau area; that is worse than other mammals such as Sambar (Cervus unicolor), Indian Muntjac (Muntiacus muntjak), mouse deer (Tragulus spp.), Sunda Pangolin (Manis javanica), and Pig-tailed macaque (Macaca nemestrina). The importance of the Bahau grasslands for Banteng Foead (1995; 1997) found that forest succession was well underway on the Long Tua grassland. It remained unclear whether the succession proceeded from the grassland seed banks or from seedlings becoming established at the forest edge, but the result was that forest was slowly replacing the alangalang (Imperata cylindrica) grassland. This, and similar processes at the other grasslands including Long Pe, could have important implication for Banteng in the area. The grasslands appear to serve as sinks for Banteng drawing them away from the agricultural areas (Foead 1997). The disappearance of the grasslands could limit Banteng food supplies and increase conflict with local people leading to more poaching of Banteng. Foead therefore recommended maintaining the grasslands by burning. The flipside of this is the possibility that maintaining the grasslands could allow hunters to target the Banteng more effectively (see recommendations about guard posts, etc. in Chapter 5 below). 13

14 CHAPTER 2: AIMS OF THE SURVEY Our trip to the upper Bahau grasslands area of Kayan-Mentarang National Park had the following aims: 1. to conduct a reconnaissance survey in the upper Bahau grasslands area to select the most suitable survey methods for Banteng; 2. to train WWF staff and local people in these survey methods; 3. to collect samples of Banteng blood, skin, muscle, and faeces for subsequent genetic analysis with the aim of determining whether the Banteng in the upper Bahau area have hybridized with domestic or feral livestock; 4. to collect information about the status of the Banteng populations in the upper Bahau area and the threats to those populations; and 5. to design a comprehensive survey strategy for Banteng covering (a) estimation of population densities and if possible population sizes of Banteng groups that regularly use the upper Bahau grasslands (b) determining other vegetation types that are used and/or needed by Banteng in this area, including the daily and seasonal movements of Banteng between these habitats and the grasslands 14

15 CHAPTER 3: REPORT ON ACTIVITIES Calendar of activities This report describes the results of fieldwork and interviews conducted in the upper Bahau area in East Kalimantan, Indonesia, 3 16 November For a detailed calendar of activities, see Appendix 10. Description of study area The Apau Ping grasslands are located along the upper reaches of the Bahau River in the north of the Kayan-Mentarang National Park (1,400,000 ha). There are several discrete areas of grassland in this region, including Long Mepun, Long Pe, Long Kayun, Long Tua, Long Nan, and Long Lian (see Fig. 1). All these were once settlements of various Kenyah Dayak subgroups according to Foead (1997). Little is known about these areas. Although in April 1993, and again between April and June 1994, Foead explored the area (Foead, 1993; Foead, 1995); and in June 1994, an expedition to the Apau Ping grasslands was conducted by entomologists of the Royal Ontario Museum and the Zoological Museum in Bogor. It is strongly suspected that the grasslands are man-made, because according to local oral history the areas were once ladang belonging to people that have since moved to Long Bangak, Sarawak, and the Krayan area (information gathered during this survey). However, Momberg et al. (1994) report other oral history sources that say that there were no former ladang sites in the grasslands of the upper Bahau. They suggest that other ecological or climatic conditions must be responsible for the grasslands. Nevertheless, however the grasslands were formed it seems certain that they are now maintained as a result of local people burning them. The grasslands are relatively frequently visited by local people because they are located along well-used paths between Indonesia and Malaysia and between the Bahau and Krayan areas. These travellers are said to burn the grasslands in a somewhat opportunistic manner (Foead, 1995). Long Tua grasslands The Long Tua grasslands were the focus for this reconnaissance survey and training exercise. They are located at latitude 3º10 48 N and longitude 115º46 48 E, around the lower reaches of the S. Tua, just upstream from where the S. Berau branches off from the S. Bahau. This site used to be part of the Long Tua village area until the inhabitants left in the late-1970s. Most Long Tua people moved to the Apau Ping area, where they established the Long Tua desa. The grassland area as found today appears to have been covered in grass for at least several decades, as none of the informants remembers there ever being other vegetation, such as ladang or forest. In the 1960s and 1970s the area was more densely inhabited. In addition to the Long Tua village, in the1960s there used to be a village in the upper Tua River area named Long Turan. These people left and moved to Long Tua and then to the now also abandoned village of Mangau. A footpath that used to connect the villages of Long Tua, Long Turan, and Mangau still runs along the Tua River. Further footpaths run from Long Tua to villages in the S. Mepun and S. Mangen areas. These last two are still used by people travelling between the upper Bahau area and Malaysia. The Long Tua grasslands are located on the slopes on the west and east sides of the S. Tua valley. Figure 2 shows the location of the Long Tua grasslands in a two-dimensional perspective (no reliable altitudinal data are yet available). Their total area is estimated to be between 175 and 225 hectares (ha), on slopes varying between approximately 30 and 50 degrees. On the east bank there are two blocks of grasslands, hereafter named the Akoi and Sangan blocks, the former approximately 60 ha, the latter some 15 ha. On the south-west bank there are two grassland areas, named the Pangkat and Bahau blocks, of respectively 60 and 25 ha. Finally, in the north-west, there are one large and one very small areas of grassland, together named the north-west blocks and totaling some 55 ha in area. Exact size estimates are difficult because the areas were measured using a GPS, which only provided a two dimensional picture of the area. 15

16

17

18

19 The Long Tua grasslands are largely covered by Imperata cylindrica. However, many other, yet unidentified, herbaceous and woody plant species were also found. Ferns were particularly common in the areas nearest to the forest edge. The age of the grasses varied, with some patches in the Akoi block having been burnt recently, whilst others had apparently not been burnt for a decade or more, resulting in very tall and dense stands of grass. We found two saltlicks in the S. Tua area, one at the southern end of the Akoi block and one on the other side of the S. Bahau (see Fig. 2). A third saltlick, supposedly located north of the north-west blocks in the S. Aloh area (ca. N 3º ; E 115º ) was not found and may have dried up. Long Kayun grassland The Long Kayun grassland is located some 3.5 kilometers upstream from the mouth of the S. Tua on the S. Bahau. The grassland consists of one hectare area on the slope of a hill, and is covered in very tall and dense grass. It has apparently not been burnt for many years. Long Pe grasslands Although not visited during this survey, several informants told us that this grassland area was considerably larger than the Long Tua grasslands, and the area should be surveyed as soon as possible (see Chapter 4). Report on training: transects and plots During the reconnaissance phase of the fieldwork we assessed the suitability of dung counts and footprint counts for Banteng survey and monitoring purposes in the Long Tua and Long Kayun grasslands and surrounding forested areas. This was accomplished through trial transects to determine the visibility of dung piles and dung pile encounter rates; and by searching for fresh dung to determine the feasibility of assessing dung decay rates in the different vegetation types (an essential component of dung-based survey methods; see Chapter 4). For the dung counts we employed standard line transects during which the perpendicular distances from the straight line transects to all dung piles seen was measured (see Chapter 4 for further details of protocols). We also tried actively searching for dung in 30m² circular plots placed at 50m intervals along a number of transects (because poor visibility in tall dense grass and scrubby forest reduced the encounter dung rate along the line transects; see Chapter 4 for more details). All dung piles encountered were classified into 4 classes (A, B, C, and D) based on their state of decomposition (see Chapter 4 and Appendix 8 for further details). The trial transect s were also used to provide hands-on training in dung count-based survey methods for WWF staff and local villagers (because they are likely to be employed as field assistants in future Banteng surveys). This training included the laying-out of transects and plots using sighting compasses and topofils (Hipchains); identification of Banteng dung; classification of Banteng dung piles into decay classes and the monitoring of dung decay rates; and the recording and presentation of survey data for analysis. The results of the trial transect are presented in the table below. Following the completion of the reconnaissance surveys and trial transects we concluded that footprintcount based methods were not suitable for the upper Bahau area because it appeared that Banteng footprints would be found too infrequently as a result of thick leaf litter, tall, dense grass and very hard substrates in some areas, and frequent heavy rainfall. The trial transects suggested that Banteng dung pile density was too low in the forested areas to justify the time that would be required to conduct an extensive dung-count based survey throughout the large forested areas of the upper Bahau. Instead, we recommend regular monitoring in the grasslands and along river edges using dung-count line transects together with active searches within 30m² plots spaced every 50m along the same transects (to help overcome low visibility in areas of tall dense grass and scrub). Please see Chapter 4 for fuller discussion of survey methods and protocols. 19

20 Transect location Length of transect No. of class A, B, or C dung piles along transect No. of class A, B, or C dung piles in 30m² circular plots Grassland, Long Tua 765m 56 n/a Grassland, Long Tua 372m 8 n/a Grassland, Long Tua 634m 32 n/a Grassland, Long Tua 753m 83 n/a TOTAL FOR GRASSLAND Remarks 2524m 179 n/a A significant number of the dung piles found along the transect would also have been recorded in 30m² circular plots had they been [placed at 50m intervals along these transects Grass/scrub/forest mosaic, Long Tua 1000m plots, 1 pile Transect cut across headwaters of S. Tua, most of the dung piles (12 out of 14) were in the first 320m of the transect, which was in a mosaic of grass and scrub, rather than in the forested 680m. Forest, Batu Pahu 1150m 4 24 plots, 0 piles Transect ran perpendicular to river from the saltlick at the river s edge. The 4 dung piles were within 77m of the river (S. Bahau), no piles found further into forest Forest, ca. 700m downriver from Long Tua 1000m 0 21 plots, 0 piles Transect ran perpendicular to river; no dung piles found. TOTAL FOR FOREST 2830m 6 59 plots, 0 piles Includes 680m of forest from the Long Tua mosaic transect (above). Genetic sampling The genetic sampling was an important part of this survey because it aims to indicate to what extent the upper Bahau Banteng have hybridized with domestic cattle. Unfortunately we were unable to obtain fresh Banteng samples, either tissue or faeces, and therefore we had to use older material. We collected skin samples from 11 of the 25 Banteng heads that we found during the survey (see below); samples were also taken from an approximately one-week-old dung pile (see Appendix 3 for a detailed list of all samples taken for genetic analysis). On several occasions, we provided training in genetic sampling methods for our WWF colleague. On 13 November, we demonstrated the sampling protocol for Banteng dung. As there was no fresh dung available, we demonstrated the technique on some slightly older dung piles. We explained the importance of working with sterile tools and gloves, showed how to take samples, how to store them in buffer tubes, and how to document each sampling event. In addition, on several occasions our local field assistants hunted or snared mammals, which they brought back to our camp for consumption. These included Thick-spined Porcupine (Hystrix crassispinis), Common Palm Civet (Paradoxurus hermaphroditus), Sambar (Cervus unicolor), and Bearded Pig (Sus barbatus). As part of the training, it was shown how fresh muscle or liver tissue samples should be taken for genetic analysis. To enable WWF staff to collect further samples for genetic analysis of the Banteng populations in the upper Bahau area, we left behind a complete sampling set, including buffer tubes, sterile blades, plastic gloves, sampling protocols, and waterproof markers. This sampling set was taken to the Lalut Birai field station. 20

21

22 Report on Banteng hunting in the upper Bahau area and results of the door-to-door survey in Apau Ping Based on our interviews, door-to-door surveys in Apau Ping, and visits to houses reported to hold Banteng trophies in Long Berini and Long Alango, it would appear that Banteng are being regularly hunted by local people in the upper Bahau area. The main reason seems to be that Banteng eat or otherwise destroy crops, and therefore need to be killed when they approach or enter gardens; although an important secondary reason appears to be trophy hunting. In most cases, only part of the animal killed was eaten and the rest of the carcass was left to rot. This was either because the animals are too heavy to transport, or because the meat is not considered to be particularly tasty by some local people. Records of Banteng hunting in the upper Bahau We found 25 Banteng heads in Long Alango, Long Berini, and Apau Ping. These heads usually comprised the horns, parietal, frontal and sometimes the nasal bones, and associated dry skin and hair; in a few cases complete skulls were also found. The 25 specimens comprised of the following animals (for a complete list, including horn measurements see Appendix 4): Juvenile male Young adult male Adult male Subadult male Subadult female Young adult female Adult female Old adult female Further records of Banteng hunting and trade in Banteng parts are listed below: Unaged Female A couple of years ago Banteng were hunted by helicopter from Malaysia. This apparently took place on the Long Pe grasslands. In Long Berini Banteng heads have been sold to high government officials in Tanjung Selor for Rp 300,000/head. One man told us that he had recently sold 5 heads out of a collection of 12 (the other 7 were examined and measured, see Appendix 4). Banteng heads are reportedly often sold to Malaysians for Ringgit/head (information from Bpk Kolin). Bpk Awang in Apau Ping recently sold 3 Banteng heads to Malaysian villagers for 100 Ringgit/head. These people, who were described as ordinary farmers, wanted to use the heads for house decoration, which indicates that they may not have had any intention to resell the trophies. The Kepala Adat Besar in Long Alango reported that Banteng on the Long Pe grasslands had been darted (from helicopters) then captured, crated, and taken to a fenced-off area in Sarawak for cattle breeding. At least 2 Banteng heads were reportedly sold by villagers from Apau Ping (reported during the door-to-door survey in Apau Ping during which 11 Banteng heads were found, see below). Door-to-door survey for skulls, horns and other animal parts On Sunday 13 November 1999, we conducted a door-to-door survey in Apau Ping village to find out which species were hunted in the area and to examine any Banteng trophies (horns and skull) that had been kept by villagers. Appendix 5 shows the complete data set for the 38 houses visited, but the table below provides a summary of what we discovered. 22