Home

I have been studying the behaviour and population ecology of the Malay civet (Viverra tangalunga) on Buton Island since 2003.

The Malay civet is one of 35 species of small to medium sized carnivores in the family Viverridae.  This family, along with the mongooses (Herpestidae) to which they are closely related, consists of mostly omnivorous forest-dwelling animals that are often the most numerous members of mammalian rainforest predator communities in Asia and Africa.  However, despite the potential importance of Viverrids in Old World rainforest ecosystems as predators, frugivores, seed dispersers and pollinators, very little is known about their ecology, largely due to their secretive nature and the inaccessibility of their jungle habitats. 

Identifying the processes that cause civet populations to fluctuate, particularly mechanisms of regulation (processes that stabilise populations), population outbreaks and extinction are of great importance to ecologists, both as a central topic of fundamental ecology and as a way of improving the management of rare or potential pest species in the family.  These processes are likely to be profoundly affected by forest habitat degradation and fragmentation and anomalous climatic events such as El Nino that leave such a strong mark in Southeast Asia.  

There are 13 species and subspecies of Viverrids on the IUCN red list including one species from the same genus, the critically endangered Malabar civet Viverra civettina from the Western Ghats of India (considered as a subspecies of V. megaspila by some authorities).  At the other end of the scale, some very abundant Viverrids such as the masked palm civet Paguma larvata, the principle vector of the SARS corona virus, may be considered as a pest species.  Indeed, the Malay civet itself may be having an adverse effect on rare endemic prey and competitor species on islands such as Sulawesi where it has been introduced.  Almost nothing is known about the population processes of wild Viverrids or indeed any rainforest carnivore.  This study is an attempt to gain insights into the lives of one member of this diverse group of fascinating beautiful carnivores. 

Counting civets

This study focuses on a population of Malay civets inhabiting an area of approximately four square kilometres in the Lambusango Forest.  Since 2003 a grid of traps has been run in the area during the Opwall season (July-September).  The number of traps on the grid increased from seventeen to twenty-four in 2005 in an attempt to boost the sample size (Figure 1).  Between 2003 and 2006 there were 274 capture events involving 66 individual Malay civets.  The total number of Malay civets caught on the trapping grid varied from fifteen in 2005 to twenty-four in 2006 (Figure 2).  However, the number of sexually mature animals varied less between years ranging between 12 and 16 individuals.  Within a trapping season individuals were caught between 1 and 10 times.  In 2003, 2004 and 2006 most individuals were caught more than once during the trapping season (54%, 64% and 60% respectively captured more than once), suggesting that they were ‘resident’ rather than dispersing animals.  In 2005 however, only 29% of individuals were captured more than once.  The 2005 sampling year also stands out in the low numbers of civets caught.  Capture numbers were also relatively low in 2003 but there was less trapping effort that season (the trapping grid was not fully established until half way through the season), so the capture results aren’t strictly comparable with the other years.  There appears to be an inverse relationship between civet abundance and mean civet weight – in years with relatively high numbers of civet captures (2004 and 2006), mean adult body weight is lower than in years when there are relatively few captures (2003 and 2005) (Figure 3).  This difference in body weight is statistically different, and suggests that higher density populations are feeling the effects of intraspecific competition.

Civet kittens

A total of 14 kittens (defined as civets that still have milk teeth) have been caught over the four years of study ranging in weight from 0.30 to 2.62 Kg.  Reproductively active females usually have only one pair of prominent mammae (though a vestigial second pair has been observed in some individuals) suggesting small litters.  Malay civets in captivity in the Singapore zoo usually have litters of two or occasionally three kittens.  Kittens were rarely captured more than once, and show significantly lower recapture rates than adults.  Lower trap rates may reflect lower activity and reduced ranging behaviour in kittens.  Only one kitten was ever captured in a second sampling season – individual M8 marked in 2003 was recaptured in the same area in 2004 as a subadult (adult teeth present, but not sexually mature).  Low recapture rates of cubs between years suggests high mortality and/or emigration rates after weaning, a pattern common in many solitary carnivores.

Civet survival

There was a high rate of population turnover of civets in the study site.  Only five animals are known to have remained on the study site for all four years of the study.  The majority of animals tagged in any year (>60%) were never recaptured in subsequent years (Figure 4).  Only 7% (1 individual) of kittens tagged were ever recaptured.  Subadults and young adults (sexually mature but very little tooth wear) showed similar recapture rates (37% and 35% respectively), whilst adults had the highest recapture rate at 43%.  Recapture is determined not only by survival but by capture probability, so these capture rates do not necessarily reflect survival, especially if there are big differences in emigration or trap shyness.  However, the pattern of recapture rates with age classes is consistent with a pattern of mortality common in a carnivore of this size – high first year mortality as weaned offspring disperse to search for territories, and reduced mortality as adults when a higher proportion of individuals will be territory holders.

Civets face a number of dangers.  They sometimes interact aggressively with one another and civets were frequently heard fighting in the vicinity of camp Lapago.  Many civets have scars, which may be a result of aggressive interactions.  Intraspecific competition is likely to be the most important source of mortality in civets.  Civets have relatively few natural enemies in the Lambusango Forest, but they are doubtless occasionally preyed upon by reticulated pythons and young civets may be vulnerable to attack by Salvator’s monitor lizards. Two adult civets had injuries caused by snares set for catching jungle fowl – one was completely missing a forefoot, and another individual was captured with a snare wire embedded in one of the hind feet.  In the vicinity of villages there are dogs, which have been reported to kill civets.  There were no signs of disease in captured civets, and external parasite loads were low, usually just a few ticks were observed.  Finally, some civets may become victims of traffic accidents; though considering the limited road network and traffic this is not likely to be an important source of mortality.

A zoo specimen of a related species, Viverra zibetha was recorded to have lived to 20 years.  Civets in wild populations are likely to live for much less.  One female individual (F2) that was caught in Lapago as an adult in 2003 was recaptured in 2006, making her at least five years old.

Social behaviour

Malay civets are considered to be solitary animals, with adults only getting together during mating.  In fact very little is known about the ranging, social and mating behaviour of Viverrids.  To shed light on the behaviour of these animals we have so far radio-tracked fifteen individuals (7 females, 8 males) in the Lambusango Forest between 2001 and 2006.  The Malay civets in our study site occupy distinct home ranges which vary in size from 6 ha (individual F17) to 189 ha (male tracked in 2001).  Males had significantly larger territories than females.  The amount of intrasexual range overlap (i.e. the percentage overlap between ranges of neighbouring animals of the same sex) varied considerably.  In 2002 there was zero range overlap between neighbouring females (Figure 5).  However, in 2006 there was considerable range overlap in neighbouring females – indeed, the range of female F17 was observed to lie almost entirely within the range of female F09 (Figure 6).  Female F09 was first caught as a young adult in 2003, and F17 was first caught as a subadult in the following year.  It may be that F17 is F09’s offspring and is remaining in her mother’s territory as an adult.  The ranges of the two neighbouring females F09 and F07 also overlapped each other considerably – thirty-three percent of F07’s range lay within F09’s range.  The 2006 results suggest that females tolerate one another’s presence.  In fact, simultaneous radio-tracking studies revealed that these pairs of females frequently came into close proximity to one another at night when they were active, implying that these animals do not actively avoid one another and may be more social than originally thought.  Male ranges also seem to overlap to a certain degree, averaging around 8% overlap in the 2001 study and up to 30% in the 2006 study.  One hypothesis to explain variation in intrasexual spatial organisation in relatively solitary species is the resource dispersion hypothesis, which puts forward that variable and patchy resources (e.g. food) will promote intrasexual territory overlap because most of the time the territory will be able to support additional animals at almost no extra cost to the primary territory holder.  However, this idea is difficult to test not least because the composition of this civet population’s diet remains largely a mystery.

Photo: Dr Dave Bird ©.

What do civets eat?

Malay civets are omnivorous and are known to eat small vertebrates, arthropods, roots and fruit.  In the villages around the Lambusango Forest, Malay civets have a reputation as a chicken raider.  In camp Lapago in the Lambusango Forest young Malay civets have been observed hunting frogs along a stream at night.  Surprisingly, it has been very difficult to locate any scats.  The one or two scats that have been recovered have yielded large amounts of animal fur.  An examination of hair scale patterns reveals that much of the fur in the two scats found comes from an endemic murid rodent Bunomys andrewsii, the most abundant small mammal on the forest floor (Figure 7). 

Island effects and competition with other species

The geographical range of the Malay civet originally covered the Malay Peninsula, Sumatra, Borneo and some smaller satellite islands, but has been extended by human agency to many parts of the Philippines and the Wallacea region, including Sulawesi, Buton and the Moluccas.  It is not known whether it was first introduced to Sulawesi within the last few centuries (perhaps in response to demand for civet musk from Europe) or within the last few millennia (by prehistoric man).  The Malay civet is the only mammalian carnivore confirmed to inhabit the forests of Buton, though there have been unverified sightings of the rare Sulawesi palm civet Macrogalidia musschenbroekii.  In contrast, throughout their original range, the Malay civet is sympatric with a large guild of mammalian carnivores.  On Borneo, for example, the Malay civet shares its rainforest habitat with 22 other species of mammalian carnivore including eight species of civet.

A reduction in interspecific competition and confinement to an island has been shown to have important effects on behaviour and morphology due to competitive release and island effects.  The Malay civets from the Lambusango Forest appear to have significantly shorter body lengths than those in Sabah in Malaysian Borneo, but whether this is a product of character release due to decreased competition, or something else (e.g. founder effect) is not know.

Future Research

There are several avenues of future research.  Firstly, data on civet abundance and survival shall continue to be collected to generate a long-term data set unique amongst studies of rainforest carnivores.  Secondly, the amount of behavioural data from radio-tracking will increase to shed more light on social behaviour and maybe help explain what happens to individuals that are captured only once.  We plan to take samples for microsatellite analysis in the future so that we can determine pedigrees and mating success of individuals within the population.  Finally, we may open a similar trapping grid on the small island of Kaledupa some six hours boat ride from Buton, where there is said to be an isolated population of Malay civets.  This would allow us to explore how confinement to a small island affects morphology and behaviour.

Relevant Publications

Jennings, A.P., Seymour, A.S. and Dunstone N. (2006) Ranging behaviour, spatial organisation and activity of Malay civets Viverra tangalunga on Buton island, Sulawesi. Journal of Zoology, 268: 63-71.

Malay civet in Lapago Camp, Buton Island                           (Photo: Dr Dave Bird ©)
 

Figure 1. Civet trapping grid

Figure 2. Number of Malay civets from each age class captured per year.

Figure 3. Mean weight (± s.d.) of adult Malay civets caught on the Lambusango forest trapping grid. 
 

Figure 4. Recapture of civet cohorts first captured in 2003, 2004 and 2005.

Figure 5.  Home ranges (95% MCP) for civets radio-tracked around the Kakenauwe study grid (just to the north of the civet trapping grid) in 2001, 2002 and 2003.  Blue polygons represent males ranges, and pink polygons represent female ranges.  The white area is cultivated land including citrus orchard, root crops and rice paddy. a

rea is mostly secondary

Figure 6.  95% MCP home ranges for Malay civets radio-tracked by the Lapago study grid in the Lambusango Forest in 2005 and 2006.  Blue polygons represent male ranges and red polygons represent female ranges.  Home range for M28 an underestimate – this individual was frequently out of range of the radio receivers.

(b)                                         (a)

Figure 7.  Hair scale patterns under microscope (a) ventral hair from Bunomys andrewsii and (b) hair found in a Malay civet scat.

Radio-tracking Malay civets on Buton Island 2005.

Malay civets on Buton are very curious! Photo: Dr Dave Bird ©