Irresponsible Fishery

DEB News


This is Billy Hau

The Virtual School of Biodiversity

Salzburg, Austria-Hong Kong, China: connected





Book Review

Wild Corner

Recent Publications

Free Column

Information for Contributors




A glance at the marine aquarium fish trade in Hong Kong
Restocking - an effective measures to restore the depleted fishery stocks in Hong Kong?
The case of the disappearing croaker, the Chinese bahaba, Bahaba taipingensis
Modification of local wire cage traps to reduce injuries to small mammals
Feral dogs and civet mortality on Kau sai Chau, Sai Kung
The changing bird community of Tai Po Kau
Is the Javan Mongoose native and does it matter?

Feral dogs and civet mortality on Kau Sai Chau, Sai Kung

by Thomas D. Dahmer

Ecosystems Ltd., 2/F Kingsun Computer Bldg., 40 Shek Pai Wan Road, Aberdeen, Hong Kong,


Feral and stray dogs (Canis lupus familiaris) are abundant and widely distributed throughout Hong Kong, but concentrated at the fringes of urban areas (Dahmer, Coman and Robinson 2000). Many are released by their keepers who no longer want them as pets, some may be escapees, and some are wild-reared offspring of stray or feral parents. Between the Agriculture, Fisheries and Conservation Department (AFCD) and the Society for Prevention of Cruelty to Animals (SPCA) more than 20,000 feral/stray dogs (on average) are captured and destroyed each year in Hong Kong. In spite of this effort the population of feral/stray dogs shows no downward trend. Rather, the large number of dogs destroyed each year appears only to compensate for mortality that might otherwise occur due to disease, starvation, or accidental death. While the public sanitation, human health, and personal injury risks of maintaining dense populations of feral/stray dogs are widely acknowledged, the ecological implications are little understood. To take one example, predation by feral/stray dogs upon wildlife may not be widely known because it is difficult and time-consuming to document, and is therefore probably under-reported. The purpose of this note is to document six cases of civet mortality on Kau Sai Chau in Port Shelter, Sai Kung from May 1998 through May 2001, all of which were attributed to attacks by feral/stray dogs.


Due to the apparent scarcity of civets on Kau Sai Chau no attempt has been made to quantify population numbers. However, civet signs have been recorded during other studies on the island. I carried out systematic quarterly bird surveys on Kau Sai Chau beginning in summer 1995. The surveys covered approximately 43 ha of area split equally between the golf course and adjacent shrublands. The areas were surveyed on foot 4 times per quarter, twice beginning before sunrise, and twice during the last hours before sunset. The surveys took place on the northern third of the island (approximately 2.2 km2 in area) where golf facilities were constructed in 1994-6. During the bird surveys, opportunistic observations of mammals and mammal signs were also made, resulting in infrequent records of civet droppings and recovery of one Small Indian Civet (Viverricula indica) skeleton in May 1998.

Greenskeeping personnel work on the island at The Jockey Club Kau Sai Chau Public Golf Course from 0600-1530 hrs Monday through Friday, and 0600-1100 hrs on Saturday. The staff numbers over 50 greenskeepers and maintenance workers who have adopted as standard practice the reporting and collection of mammals found dead on the golf courses. This resulted in recovery of four Small Indian Civet carcasses and one Masked Palm Civet (Paguma larvata) carcass between May 1998 and May 2001.

Results of ecology studies have been documented in periodic reports to the Hong Kong Jockey Club (e.g. Ecosystems Ltd. 2000). Nomenclature used in this report follows Wilson and Reeder (1992).


Over a period of 37 months, six dead civets were recovered from northern Kau Sai Chau. Five were Small Indian Civets (four fresh carcasses and one skeleton), and the sixth was a Masked Palm Civet (fresh carcass) (Table 1). All fresh carcasses were discovered between 0620 and 0730 when the staff spread out over the golf courses to mow grass and carry out other greenskeeping tasks. No flesh or internal organs of any of the five carcasses had been eaten, and none of the body cavities had been opened (except by tooth punctures through skin and underlying flesh). Three of the carcasses were photographed but disposed of before a post-mortem examination could be made. The last two carcasses were examined in detail and body measurements were taken (Table 2)


Material Recovered

Month and Year Recovered

Small Indian Civet

fresh carcass

May 1998

Small Indian Civet


May 1998

Small Indian Civet

fresh carcass

October 1998

Small Indian Civet

fresh carcass

October 1998

Small Indian Civet

fresh carcass

May 2001

Masked Palm Civet

fresh carcass

May 2001

Table 1.Civet fatalities documented at Kau Sai Chau between May 1998 and May 2001.


Small Indian Civet

Masked Palm Civet




age class



body weight (g)



total length (cm)



tail length (cm)



body length (cm)



hind foot length (cm)



ear length (cm)



Table 2. Sex, age class, and morphometrics of two civet carcasses recovered at Kau Sai Chau in May 2001.

The first fresh carcass was found in May 1998, followed in the same month by the only recovered skeleton. The fresh carcass was disposed of before a post-mortem examination could be made. The skeleton was found in tall grass where the civet had apparently slept immediately prior to dying. The posture of the skeleton was typical of a canid or felid curled to sleep. Although all flesh and nearly all skin and hair had been removed or decomposed from the skeleton, it had not been moved after the civet died, and only a few of the larger bones were missing. None of the long bones was broken, but the pelvis was disarticulated, and the right scapula broken and punctured. The puncture appeared to have been made by a single tooth, possibly a canine. The two May 1998 fatalities were found within 0.5 km of one another at the northern periphery of the golf courses.

Similar to the two May 1998 fatalities, those in October 1998 were found within several hundred metres of one another on the golf fairway and rough. The location was at the eastern periphery of the golf courses. Of the two fresh carcasses reported in October 1998, only one could be examined. The civet had suffered a compound fracture of the left femur in addition to numerous bites to the dorsal lumbar region. The bites penetrated the skin and flesh to the spine and pelvis.

The two fresh carcasses recovered in May 2001 were found on the south periphery of the golf courses, but were separated by a distance of approximately 900 m. One carcass was that of a Masked Palm Civet, the first record of the species on Kau Sai Chau. The second carcass was that of a Small Indian Civet. Both civets had suffered bites on the posterior dorsum, upper hind legs, and on the anterior thoracic dorsum. The Masked Palm Civet had numerous bites that penetrated the skin and underlying flesh in both areas, including the upper thorax. The wounds of the Masked Palm Civet caused tissue trauma as evidenced by swelling and infection. The wounds also appeared to have caused blood loss, although rain or irrigation water had washed away most evidence of blood. No physical damage or symptom of disease was detectable in the internal organs of either carcass. Feral dogs were observed in the vicinity of the Masked Palm Civet carcass near the time it was found. The location of recovery of the Small Indian Civet carcass is an area frequented by feral dogs at night. Weights and measurements of the two civets recovered in May 2001 are listed in Table 2.


Fatalities of 6 civets were recorded over a time span of 37 months. Four of the six dead civets bore bite wounds that in some cases caused broken bones and/or blood loss, and penetrated thoracic or abdominal cavities. Although Kau Sai Chau is Hong Kong’s fifth largest island, it is not large (6.67 km2), and civet signs have not been recorded with any regularity since routine, systematic surveys began on the island in summer 1995. This suggests that civets are not abundant on the golf course portion of the island. In a separate study involving 14 nocturnal bird surveys spanning 28 hours and carried out between March 2000 and June 2001, only one Small Indian Civet was seen. This also suggests that civet abundance on the island is low or civets are extremely wary.

Since 1993 I have observed from 1 to 3 dog packs on the northern third of Kau Sai Chau at any given time. Since 1995 more than 60 dogs have been removed from the island by control actions of the golf course or AFCD. As dogs are removed they are replaced by litters born on the island, dogs that swim to the island, or unwanted dogs that are dropped from boats. Total dog numbers may range from zero immediately after a capture operation to over 20.

The feral/stray dogs often receive handouts from golfers, visitors, or golf course staff. As a result, most dogs on the island appear to be in good condition, if not well-fed. The dogs frequently attack or threaten members of other packs or individual dogs. Although injuries to humans have not been recorded, dog fatalities have been attributed to attacks by other dogs or dog packs on the island.

Civets at Kau Sai Chau appear to be entirely nocturnal: No civet has been seen during daylight hours. Civet encounters with dogs may go unobserved because such interactions take place at night when observers are few. Although feral/stray dogs have not been observed attacking civets, the injuries suffered by civets on Kau Sai Chau combined with dog presence near civet kill sites suggest that dogs in packs are responsible for inflicting wounds that result in civet fatalities. The wounds on the Small Indian Civet carcass recovered in October 1998 (compound femur fracture) and on the Masked Palm Civet recovered in May 2001 (numerous body cavity and limb punctures, severe tissue trauma from bites to the dorsal pelvic area) suggest attack by dogs. The disarticulated pelvis and punctured and broken scapula of the skeleton found in May 1998 also suggest dog attack.

The only other possible agents of civet mortality are golf course shuttle buses, other civets, and Eagle Owls (Bubo bubo). Buses can be ruled out because there are no roads in areas where the civet carcasses were found. Also all fresh carcasses were found in the early morning and none evidenced Rigor mortis, yet the buses do not operate after 2200 hrs or before 0615 hrs. Civets could kill other civets in mating or territorial disputes, but this would be expected to occur when civet population densities were high and competition for food or mates most intense. If civet density were high, civet signs (faeces at marking stations) would also be abundant. Yet abundant civet signs have never been recorded during routine quarterly bird studies, or by the greenskeeping staff who are on the island 7 days per week throughout the year.

Eagle Owls have been observed on Kau Sai Chau (Ecosystems Ltd. unpubl. data). Eagle Owls would not attack the hind legs or lower back of prey (as would dogs), but would target the mid-back or neck. For that reason an Eagle Owl attack would be unlikely to cause a compound fracture of a civet femur. Eagle Owl attacks could result in breaking of the civet spine, and Eagle Owls would probably open body cavities to feed on internal organs. None of the recovered civets had a broken spine or opened body cavity. There was no evidence of a predator feeding on any of the five recovered civet carcasses. This suggests that the kills were not made by predators in search of food, which also implicates feral/stray dogs rather than Eagle Owls.

Many aspects of civet mortality on Kau Sai Chau remain unexplained. Why were dead civets found only in May and October? Are sub-adult males disproportionately subject to dog attacks? If so, is it because they are being driven from other civet territories or family groups? Why was a Masked Palm Civet fatality not recorded on the island until May 2001? Can civets swim to Kau Sai Chau? Would elimination of feral/stray dogs from the island lead to increased civet numbers? These and other questions will be investigated in future by continued night surveys, detailed examination of all civet carcasses, and reduction of numbers of feral/stray dogs. Readers wishing to contribute information on civet mortality in other locations or to provide carcasses for post-mortem examination are invited to contact the author.


Civet carcasses and other wildlife observations were often reported by Darren Moseley, Cameron Halliday, Rick Hamilton, Paul Yip and other greenskeeping personnel at The Jockey Club Kau Sai Chau Public Golf Course Ltd. Their cooperation and assistance are greatly appreciated. Studies reported here were funded by The Jockey Club Kau Sai Chau Public Golf Course Ltd., which support is gratefully acknowledged. AFCD assisted frequently in capture of feral/stray dogs on Kau Sai Chau. The efforts of former AFCD Dog Control Unit Supervisor Dr. Thomas Sit and his staff are gratefully acknowledged.


Dahmer, T., Coman, B. and Robinson, J. (2000). Ecology, behaviour and persistence of packs of stray/feral dogs with implications and practical recommendations for control. Final report, Dept. Agriculture, Fisheries & Conservation, Government of Hong Kong Special Administrative Region, 10 March 2000, 30pp +app.

Ecosystems Ltd. (2000). Hong Kong Jockey Club Kau Sai Chau Public Golf Course Ecological Monitoring Report, July 1999 to June 2000, July 2000.

Wilson, D. E. and Reeder, D.M. (eds.). (1992). Mammal species of the world: a taxonomic and geographic reference, Smithsonian Institution Press, Washington, D.C. 1206pp.




For more information, contact

Copyright © 2000