Back to index of mammal studies

Back to index of 2002 research reports

SPATIAL ORGANISATION, RANGING BEHAVIOUR AND HABITAT USE BY THE MALAY CIVET
VIVERRA TANGALUNGA - A SUMMARY OF WORK COMPLETED IN 2001 - 2002

Andy Jennings & Dr Nigel Dunstone, University of Durham

Introduction 

Three species of civets have been recorded on Sulawesi: the Sulawesi palm civet, Macrogalidia musschenbroekii; the Malay civet, Viverra tangalunga; and the common palm civet, Paradoxurus hermaphroditus (Suyanto et al. 1998). The Sulawesi palm civet is a native species while the Malay and the common palm civets were introduced (Musser 1987). These are the only species of the order Carnivora on Sulawesi and very little is known about their ecology, distribution and conservation status (Colon 2002, Nozaki et al. 1994, Wemmer & Watling 1986, Macdonald & Wise 1979). 

The Sulawesi palm civet is known only from anecdotal information and has never been studied in the wild (Wemmer & Watling 1986). Its status on Sulawesi is uncertain and is listed as vulnerable on the IUCN Red Data list (2002). It occurs in both lowland and montane forests up to about 2,600 meters (Wemmer & Watling 1986). Macrogalidia seems to be dependent on primary forests but also moves onto adjacent grasslands and farms (Wemmer & Watling 1986). It is a skilful climber feeding on small mammals and fruits (Nowak & Paradiso 1999, Wemmer & Watling 1986). All known records and specimens come from central Sulawesi and the Minahassa peninsula, but in former times Sulawesi palm civets were believed to occur over the whole of Sulawesi but were then thought to have vanished from the Southern regions (Veron 2001, Wemmer & Watling 1986). However, a Sulawesi palm civet was very recently camera-trapped in southeast Sulawesi, near Kendari (R. Lee, pers. comm.). Unconfirmed sightings of Macrogalidia have also been reported in north and central Buton Island (Operation Wallacea staff, pers. comm.). 

The common palm civet has only been recorded from Sulawesi’s two major ports: Manado in the north and Makassar in the south. It is believed to have been a recent introduction and the existence of breeding populations is considered questionable (Wemmer & Watling 1986). 

The Malay civet is found on the Malay Peninsula, Sumatra, Borneo, Sulawesi, Maluku Islands, and the Philippines (Kanchanasakha et al. 1998, Suyanto et al. 1998, Corbet & Hill 1992). It occurs in a wide variety of habitats including forests, secondary habitats, cultivated land, and the outskirts of villages (Nowak & Paradiso 1999, Wemmer & Watling 1986). V. tangalunga is one of the sources of civet, a substance that was commonly used commercially in producing perfume. Because of this function, the Malay civet was introduced to several islands throughout Southeast Asia, apparently including Sulawesi (Nowak & Paradiso 1999). Malay civets are generally solitary, feeding on a wide range of food including small mammals, birds, snakes, frogs, insects, eggs, and fruit (Colon 2002, Nowak & Paradiso 1999). They are terrestrial animals, although they can climb trees readily but seldom do so (Nowak & Paradiso 1999). Although the Malay civet is a widespread and common species little is known about its natural history and ecology: only a small number of field projects have been conducted on Borneo (Colon 2002, Heydon & Bulloh 1996, Nozaki et al. 1994, Macdonald & Wise 1979): an island on which 22 species of carnivore have been recorded (Suyanto et al. 1998). 

The effects of habitat loss, fragmentation, and degradation on carnivore populations are largely unknown (Creel 2001) and many species of carnivores are becoming endangered from lack of knowledge of their habitat requirements (Nowell &Jackson 1996, Schreiber et al. 1989). Certain destructive activities in the forest may have negative impacts on carnivores (Colon 2002, Heydon & Bulloh 1996). The factors that promote ecological separation among closely related sympatric carnivores in tropical forests are also poorly understood due to a paucity of field studies (Ray & Sunquist 2001). Inter-specific competition can be a strong limiting factor for carnivore populations and the effects of competition can help explain why some carnivores are prone to extinction (Creel 2001). 

During August and September 2000, as part of the Operation Wallacea Biodiversity Survey of Southeast Sulawesi, a live-trapping program to capture civets was undertaken around the village of Labundo Bundo, central Buton Island. Several Malay civets were caught confirming the presence of this species in this area. Although several unconfirmed sightings of the Sulawesi palm civet have been reported near Labundo Bundo and in north Buton attempts to trap this species in these areas have so far been unsuccessful. To investigate the natural history, ecology and competitive interaction of civet species on central Buton Island a radio-tracking project was set-up by Operation Wallacea in June 2001, in the Kakenauwe and Lambasango Forest Reserves. 

Aims and Objectives 

The aim of this radio-tracking project was to collect data concerning the distribution, habitat requirements, behaviour, and interactions of civet species around Labundo Bundo, Buton Island. The objective was to investigate the following ecological parameters of each civet species: home range sizes, spatial relationships, activity patterns, habitat use, rest sites, density estimates, genetic relationships, and diet.

Methods 

Seven wire-box traps were used for capturing civets; 2 wire box-traps made by ‘Longmeadow’ (76cm x 27cm x 25cm) and 5 locally made traps (of the same design and measuring 140cm x 40cm x 40cm). These traps were placed at the base of large trees and along logs and were completely covered by woody debris. Domestic cat food and fruit was used as bait and was placed in small plastic containers tied to the back of the trap. Lures were placed high up on prominent tree branches around the trap site. Traps were left open continuously and checked twice a day (in the morning and afternoon). The bait was replaced when necessary. 

A closed trap was checked to determine what animal had been caught. The surrounding woody debris was removed and an estimation of the animal’s weight was made. The trap was then covered by a dark sheet to keep the animal calm and the processing equipment was set-up on a nearby flat, shaded area. A combination of ketamine hydrochloride and xylazine hydrochloride was used as the anaesthetic drug and an appropriate dosage (based on the animal’s estimated weight) was administered in a 1 or 2ml syringe with a 21G needle. Animals were restrained using metal combs and the drug injected into the animal’s rear leg muscle through the bars of the cage. The trap was then covered and the animal left quietly for 5-10 minutes. If the animal had not gone down sufficiently after 10-15 minutes either half the original dose was administered or the animal was allowed to recover and released unprocessed. 

Once the animal was ataxic it was removed from the trap to the processing area. Eye cream was immediately added to each eye and both eyes were covered with a black cloth. A rectal thermometer was inserted to record body temperature every 5-10 minutes and breathing was monitored by watching chest movements. Civets were fitted with transmitters (MOD-80, Telonics, Mesa, Arizona) on a butyl collar incorporating a whip-antenna (total weight 58g). Each transmitter had an operational life of 6 months and was fitted with activity sensors (S6B) that caused a reduction in signal rate when the animal was inactive. 

The animal was sexed and aged based on body size and condition of the teeth. The age categories were: Juvenile (not full size and still has milk teeth), Young adult (not sexually mature), Adult (full size, showing signs of reproductive activity and with moderate tooth wear), and Old Adult (showing signs of decline in weight and health and with pronounced tooth wear). Reproductive status was determined by checking nipples/lactation in females and testicle size in males. The following body measurements were taken: head and body length, tail length, neck circumference, right ear length, right hind foot length, right fore-footpads (length and width), right canine length (upper and lower). Every animal captured was ear-tagged on both ears with coloured plastic tags and a PIT tag implanted sub-cutaneously in the neck. Hair samples were collected for genetic analysis. A general examination of the body was made and written comments of unusual body features were noted (e.g. injuries, external parasites). The animal was weighed and photographs were taken. The animal was placed back into the trap at the point of capture and was allowed to fully recover before being released. 

Radio-tracking was done using TR-4 receivers and two-element RA-14 antennas (Telonics, Mesa, AZ). Signal bearings were taken using a Silva Ranger sighting compass (Type 15). An attempt to locate each tagged individual was made every day by triangulating from known points along the grid/trail network. Tracking teams consisted of 2-4 people working 8-hour shifts. From marked positions, compass bearings were taken on tagged animals. For each bearing, the Civet ID, date and current location were recorded. The peak signal position was found by rotating the antenna through 360°. The gain on the receiver was then turned down until the signal could only just be heard. The antenna was slowly swung to the side until the signal disappeared and then swung back until it could just be heard again. A landmark along the line of the antenna boom was visually noted. This procedure was then repeated on the other side. The signal bearing was then the line that bisected the angle between these two directions. The time and signal bearing were recorded and plotted onto field maps. At least 3 bearings were obtained, ideally with a maximum of 5 minutes between successive bearings. Bearings that appeared erroneous due to signal bounce were eliminated from the data set. 

At the same time that signal bearings were taken an animal’s activity was also recorded by listening to the radio signal. Each radio-collar was equipped with an activity sensor that indicated whether an animal was active or inactive. An animal was recorded as ‘active’ if the signal rate was 75 pulses per minute (ppm) and there was fluctuation in the signal strength. An animal was recorded as ‘inactive’ if the signal rate was 45 ppm and the signal strength was steady. Continuous monitoring of each animal’s activity over 8 to 24-hour periods were also undertaken whenever possible. During this period, an animal’s activity was recorded every 15 minutes. 

Bearing data was entered into the computer program LOAS (Version 2.06, Ecological Software Solutions, www.ecostats.com) to generate location fixes. Estimated locations with an error ellipse greater than the acceptable minimum (1ha) were rejected. Location fixes were exported into RANGES V software (Kenward & Hodder 1996) for further analysis. All statistical analysis was carried out with SPSS for Windows (Version 11.0). 

A few walk-ins were attempted on inactive animals to investigate rest site selection. The location of a resting animal was determined by taking several compass bearings of the radio signal from different positions on the road/trail network. These were plotted onto a field map to give an approximate fix. The walk-in was then begun from the closest point on the road/trail network to the calculated position of the resting animal. The signal from the animal’s radio-collar was followed through the forest. At the point when a box signal could be heard (the signal from the receiver without the antenna and cable) the animal was less than 10m away. The surrounding forest was then scanned for the most likely resting locations. The actual rest site was confirmed by making a complete 360º rotation while checking on the direction of the radio signal. A description of the rest site was made and its GPS location was recorded. 

Results 

During the 2001 and 2002 field seasons (June to September), 9 Malay civets (5 males, 4 females) were captured. A total of 14 captures (including recaptures) were made in 432 trap- nights, resulting in an average capture rate of 1 animal per 31 trap-nights. 

From June to September 2001, 4 Malay civets (1 male, 3 females) were radio-collared in Kakenauwe forest. From June to September 2002, 3 Malay civets (3 males) were radio-collared in Kakenauwe forest and 2 Malay civets (1 male, 1 female) were radio-collared in Lambasango forest. 

(i) Home range sizes 

Home range estimates (95% minimum convex polygons) were calculated for all Malay civets that were radio-tracked in Kakenauwe Forest Reserve in 2001 and 2002 (Table 1). The mean home range size was 78 ha for adult males (range = 33 - 176 ha, n = 4) and 37 ha for adult females (range = 20 - 53 ha, n = 2). There was no significant difference between the home ranges of adult males and adult females (t = 0.806, d.f. = 4, p = 0.465). 

Table 1. Home ranges (95% MCP) of Malay civets radio-tracked in Kakenauwe Forest Reserve in 2001 and 2002.

There was no significant difference between the home range sizes of adult males in Kakenauwe forest and adult males in unlogged (t = 0.416, d.f. = 6, p = 0.692) and logged (t = 1.215, d.f. = 5, p = 0.279) forests in Sabah, Malaysia (Colon 2002). There was also no significant difference between the home range sizes of adult females in Kakenauwe forest and adult females in unlogged forests in Sabah, Malaysia (t = 2.595, d.f. = 2, p = 0.122) (Colon 2002). However, there was a significant difference between the home range sizes of adult females in Kakenauwe forest and adult females in logged forests in Sabah, Malaysia (t = 3.552, d.f. = 2, p = 0.038) (Colon 2002).

(ii) Spatial relationships - home range overlaps 

Figure 1 shows the home ranges (100% MCP) of all civets radio-tracked in Kakenauwe Forest Reserve in 2001 and 2002. There was no overlap of home ranges between females (F01, F02, and F03) (Table 2). The mean overlap of home ranges among males was 4.5% (Table 2).

Figure 1. Home range polygons (100% MCP) of all Malay civets radio-tracked in Kakenauwe Forest Reserve in 2001 and 2002.

Table 2. Overlap of individual Malay civet home ranges as determined from 95% MCP analysis. The overlap matrix shows the percentage of areas in rows overlapped by those in columns.

 (iii) Activity patterns

A total of 1,701 activity readings were recorded. Figure 2 depicts mean activity of all civets (n = 9) divided into 24 hour blocks.

Figure 2. Mean percent activity levels divided into 24 hour blocks of 9 Malay civets radio-tracked in Kakenauwe and Lambasango Forest Reserves in 2001 and 2002.

There was no significant difference in activity levels between males (n = 5) and females (n = 4)(t = 0.441, d.f. = 6, p = 0.675). Mean activity level for all civets (n = 9) was significantly lower from 0600 to 1800 hrs (mean = 60.4%, range = 44.8 – 91.6%) than from 1800 to 0600 hrs (mean = 93.5%, range = 70.0 – 100%) (t = 7.043, d.f. = 22, p < 0.001). 

(iv) Rest Sites 

A total of 6 rest sites were found for 3 individual civets (Table 3). All were situated at ground level and were associated with some form of cover such as logs, dense brush pile or thick herbaceous vegetation. 

Table 3. Location and description of Malay civet rest sites found in Kakenauwe and Lambasango Forest Reserves in 2001 and 2002.

(v) Habitat use 

GIS information for Buton Island has just been obtained. This will be analysed to determine the habitat use and preference of Malay civets within Kakenauwe Forest Reserve.

(vi) Genetics  

During the 2000 to 2002 field seasons, DNA samples (mostly hair) have been collected from 13 Malay civets (8 males, 5 females) within Kakenauwe and Lambasango Forest Reserves. Dr Geraldine Veron (Paris Natural History Museum) is conducting research on the phylogenetics of viverrids within Southeast Asia.  She has been contacted and is interested in a collaborative project on Malay civet genetics.

(vii) Diet 

Malay civets are known to use latrine sites to deposit scats. Three sites were found in Lambasango forest and were collected for dietary analysis. Nicola Grimwood has compiled a reference collection of Sulawesi small mammal hair that will be used to identify the small mammal prey items in the scat samples. Samples of whiskers were collected from 3 civets (2 males, 1 female). Isotope analysis of these samples should also reveal information on dietary preferences. 

Discussion

When three or four animals have adjacent ranges and a mean overlap of less than 10% there is a strong indication of exclusive home ranges (Sandell 1989). In this study, there was no overlap of home ranges among females and a mean overlap of 4.5% between the home ranges of males, suggesting that male and female Malay civets were maintaining exclusive intra-sexual home ranges. However, the trapping program conducted within Kakenauwe forest indicated that not all the civets present in the study area had been caught and radio-tagged in any one field season, possibly due to the low number of traps that were used over the study site. However, if we assume that all the tagged civets were present and maintained similar home ranges during both seasons, and there were no other civets within the study area, the pattern of social organisation that is revealed is intriguing. Colon (2002) found considerable home range overlap of both sexes and concluded that the Malay civet was not territorial.

Temporal and spatial variation in food availability should lead to a system of overlapping ranges (Sandell 1989). Exclusive home ranges are expected when food resources are stable and evenly distributed (Sandell 1989), suggesting that this may be the case in this study. However, Colon (2002) expected food abundance and distribution in rainforests to vary in space and time and that this accounted for the overlapping home ranges of Malay civets in her study area. A similar pattern was observed among common palm civets in Nepal (Joshi et al. 1995) and Thailand (Rabinowitz 1991), and in kodkods (Oncifelis guigna) in South America (Dunstone et al. 2002). However, these predictions have not been rigorously investigated and further studies are required. 

The home ranges of Malay civets in this study were generally smaller than those reported by Colon (2002) in Malaysia, although only female home ranges were significantly different. The lack of inter-specific interference and/or competition for food from other mammalian carnivores may result in an increased prey base available for Malay civets on Buton, allowing individuals to meet their daily energy requirements within smaller home ranges.  However, there are several bird, reptile, and other mammalian species present on Buton that are potential competitors for food. Further investigation is required to discern the relationship between Malay civets and their prey base on Buton Island.  

The activity pattern in this study clearly shows that the Malay civet is predominately nocturnal. Colon (2002) found a similar pattern although the activity levels of civets on her study sites were lower during the day and night. Malay civets on Buton may have different activity pattern in response to the availability of food resources and inter-specific competitors.

References 

Colon, C.P. 2002. Ranging behaviour and activity of the Malay civet (Viverra tangalunga) in a logged and an unlogged forest in Danum Valley, East Malaysia. J. Zoology 257: 473- 485 

Corbet, G.B., & Hill, J.E. 1992. The mammals of the Indomalayan region: a systematic review. Oxford University Press, Oxford, UK. 488pp. 

Creel, S. 2001. Four factors modifying the effect of competition on carnivore population dynamics as illustrated by African wild dogs. Conservation Biology 15(1): 271-274 

Dunstone, N., Durbin, L., Wyllie, I., Freer, R., Jamett, G.A., Mazzolli, M. & Rose, S. 2002. Spatial organization, ranging behaviour and habitat use of the kodkod (Oncifelis guigna) in southern Chile. J. Zool. 257: 1-11. 

Heydon, M.J. & Bulloh, P. 1996. The impact of selective logging upon sympatric civet species (Viverridae) in Borneo. Oryx 30(1): 31-36. 

Joshi, A.R., David Smith, J.L., and F.J. Cuthbert. 1995. Influence of food distribution and predation pressure on spacing behavior in palm civets. J.Mammal. 76(4): 1205-1212 

Kanchanasakha, B., Simcharoen, S. & Than, U.T. 1998. Carnivores of Mainland South East Asia. Siam Tong Kit Printing Co. Ltd., Bankok. 236 pp. 

Kenward, R.E. & Hodder, K.H. 1996. RangesV: an analysis system for biological location data. U.K.: Natural Environment Research Council. 

Macdonald, D.W., & Wise, M.J. 1979. Notes on the behavior of the Malay civet Viverrra tangalunga Gray. Sarawak Museum Journal 48: 295-299. 

Musser, G.G. 1987. The Mammals of Sulawesi. In: Biogeographical Evolution of the Malay Archipelago. Ed. By T.C. Whitmore, pp 74-95. Clarendon Press, Oxford. 

Nowak, R.M. &. Paradiso, L. (1999) Walker's Mammals of the World. John Hopkins, Baltimore. 

Nowell, K. & Jackson, P. 1996. Wild Cats: Status Survey and Conservation Plan. Gland, Switzerland: IUCN.

Nozaki, E. et al. 1994. Home range of the Malay Civet (Viverra tangalunga) in Teluk Kaba area, Kutai National Park, East Kalimantan, Indonesia. Kyoto Univ. Overseas Rep. Spec. Numb. 85-94 

Rabinowitz, A.R. 1991. Behaviour and movement of sympatric civet species in Huai Kha Khaeng Wildlife Sanctuary, Thailand. J. Zool. (Lond.) 223: 281-298. 

Ray, J.C. & Sunquist, M.E. 2001. Trophic relations in a community of African rainforest carnivores. Oecologia (Berlin) 127(3): 395-408. 

Sandell, M. 1989. The mating tactics and spacing patterns of solitary carnivores. In Carnivore behaviour, ecology, and evolution: 164-182. Gittleman, J. L. (ed.). London: Chapman and Hall.

Suyanto, A., Yoneda, M., Maryanto, I.,  Maharadatunkamsi & Sugardjito, J. 1998. Checklist of The Mammals of Indonesia. Published by LIPI & JICA, Indonesia. 

Veron, G. 2001. The palm civets of Sulawesi. Small Carnivore Conservation. 24: 13-14. 

Wemmer, C. & Watling, D. 1986. Ecology and Status of the Sulawesi Palm Civet Macrogalidia musschenbroekii Schelgel. Biological Conservation, 35: 1-17.