ISSN 1027-2992 
no 
1 
KH^JSI 
.##/?. 
? ?: 
original contribution 
SHENG Lfr, DAJUN WANG", ZHI LIT, WILLIAM J. MCSHEA2 
Cats living with pandas 
The status of wild felids within giant panda range, China 
From 2002 - 2009, we conducted a camera-trapping survey in 11 nature reserves of 
the five mountain ranges within current giant panda distribution for a total sampling 
effort of 19,151 camera-days. Four felid species were recorded: Panthera pardus, 
Panthera uncia, Catopuma temminckii, and Prionailurus bengalensis. We found no 
significant correlation between the sampling effort and photographic rates of small 
felids, or medium- and large-sized felids, but a positive correlation between the pho- 
tographic rates of medium* and large sized felids and large-sized prey species. A 
well-designed systemic monitoring network using camera-trapping would be the 
best approach to obtain the critical information needed for setting felid conservation 
goals in China. 
Nine felid species have been recorded to oc- 
cur within the current giant panda Ailuropo- 
da melanoleuca range and adjacent areas in 
southwest China. These species include the 
tiger Panthera tigris, leopard, snow leopard, 
clouded leopard Neofelis nebulosa, Eurasi- 
an lynx Lynx lynx, Asiatic golden cat, Pallas's 
cat Otocolobus manul, leopard cat, and Chi- 
nese mountain cat Felis bieti (Hu & Wang 
1984, Wang 1998, Smith & Xie 2008). Giant 
pandas have been long considered a flag- 
ship species in wildlife conservation (Schal- 
ler 1994, Leader-Williams & Dublin 2000, 
Loucks et a!. 2001). Immense financial and 
manpower resources, as well as public at- 
tention, have been devoted to the conserva- 
tion of giant pandas and their habitat, which 
is believed to be beneficial to sympatric 
species, including felids (Cam & O'Doherty 
1999, Lu et al. 2000). From early 1960s to 
2007, 62 nature reserves were established 
with emphasis on giant panda protection 
and covered an area of 29,000 km2 (Wang 
2008). Current giant panda distribution co- 
vers one of the largest remaining forested 
areas in China, comprising five mountains in 
southern Shaanxi, western Sichuan, and a 
southern corner of Gansu province (Loucks 
et al. 2001, SFA 2006). Though a wildlife 
monitoring system based on sign transect 
has been recently established in the giant 
panda reserves (Gu et al. 2004, 2005, SFA 
2007), our knowledge on the status of the 
felid species in this area is still poor due to 
lack of systematic survey and monitoring. 
In 2002, we initiated a large mammal survey 
within the giant panda distribution of Sichu- 
an and Shaanxi provinces using remote-trip 
cameras. We examined this dataset with re- 
spect to felids, and address three major que- 
Fig. 1. Surveyed nature reserves (NR). Label numbers and names referred in Table 1. 
stions: 1) what felid species are currently 
present in the giant panda nature reserves; 
2) what is the spatial distribution pattern of 
these felids; and 3) what is the relative ab- 
undance of known prey for these felids. 
Methods 
Camera-trapping was used to determine the 
presence of large mammal species within 
our study areas. From the five mountain 
ranges, we selected one to five nature re- 
serves in each range as our study area (Fig. 
1). A series of training courses on the ca- 
mera-trapping technique (once a year) were 
held by Smithsonian National Zoological 
Park, Peking University, and Sichuan Fore- 
stry Department in Sichuan province from 
2003 - 2008. At each course, staff from 3 - 5 
reserves were trained to operate the came- 
ra units, identify common mammal species 
based on the photographs, manage the 
camera-trapping data using a database, and 
design the survey plan of their own reser- 
ve within a Geographic Information System 
(GIS). Some camera units and funds were 
provided to each reserve after the training 
to initiate a survey focusing on large mam- 
The surveyed nature reserve was divided 
into 1x1 km2 sampling blocks using GIS. 
The major vegetation types of our sampled 
blocks were deciduous forest, deciduous- 
coniferous mixed forest, and coniferous 
forest. In three reserves (i.e. Wanglang, 
Wolong and Tangjiahe Nature Reserves), 
the sampling area was partly extended to 
alpine shrub and alpine meadow habitat 
above the tree-line. One or two passive 
infrared-triggered camera units were placed 
in each block for a sampling duration of one 
month. Adjacent camera units were set > 
300 m apart. We used multiple camera mo- 
dels over the course of the surveys, inclu- 
ding three passive film models (DeerCairT 
DC-200, CamTrakker? Original 35 mm, and 
one film model manufactured by colleagues 
and based on similar specifications to the 
CamTrakker? Original 35 mm units), and two 
passive digital models (CamTrakker'" Digital 
Ranger, CuddeBack? Digital). Camera units 
were attached to trees along active ani- 
mal trails, 30-50 cm above ground and 3-5 
m from the trail. All the units were set to 
record 24 hours per day with 0.5-2.5 minu- 
te delay between consecutive exposures, 
to film, memory storage or batteries being 
expended on a single detection. Carnivo- 
re scent lure (Montgomery Fur Company, 
CATnews 52 Spring 2010 
cats living with pandas 
Table 1. Photographic rates of wild felids and their prey within surveyed nature reserves. Photographic rate was based on number 
of independent detections per 1000 camera-days. The survey effort is indicated in camera-days. 
Nature reserve Survey #1km! Camera- 
Photographic rate ? 
Mountain Leopard Asiatic Leopard Snow Small Medium Large 
range period blocks days cat golden cat leopard prey prey prey 
Qingling 1 .Changqing 4-12/08 54 4307 5.34 0.70 0.23 16.25 17.88 177.39 
Minshan 2.Tangjiahe 3/02-10/04 
9/08-3/09 
112 4977 0.80 0.20 8.64 62.89 45.21 
3!aohegoua 7/05-4/07 33 1008 0.99 50.60 23.81 43.65 
4.Wanglang 9/04-10/05 
6/08-11/08 
120 4260 3.99 69.48 16.90 12.68 
5.Xiaohegou 9/08-11/08 15 587 170 107.33 25.55 8.52 
6.Xuebaoding 4/08-5/07 23 526 19.01 91.26 89.35 39.92 
Qionglai 7,Wolong 8/05-5/07 
10/08-3/09 
56 2032 7.87 2.46 79.72 83.66 154.04 
8.Anzihe 4/08-1/09 5 506 65.22 41.50 23.72 
Xiangling g.Yele 6/08-12/08 26 547 1.83 731 89.58 14.63 
Liangshan lO.Shenguozhuang 6/08-11/08 12 271 3.69 33.21 3.69 
II.Mamize 4/08-8/08 5 130 69.23 15.38 
"Timber area managed by a local timber company, stopped logging in 1998 and initialized wildlife monitoring in 2004. 
" Include the sample sites that are above tree-line. 
Ogden UT) was placed in front of the camera 
unit to attract the animal in the focal area to 
ensure photographs. 
The camera units were checked at the end 
of one-month sampling duration and moved 
to the next block after changing batteries, 
films, or memory cards if needed. The ani- 
mals in each photograph were identified to 
species by the authors and input to a MS 
Access database to track the date, GPS lo- 
cation, elevation, and habitat information of 
each photographic record. Photographic rate 
was calculated for each felid species, as 
well as potential small, medium and large 
prey species. We defined the photographic 
rate as: 
# independent detection/1000 camera-days 
where the independent detection of a spe- 
cies was defined as one photograph of the 
species taken during a 30-minute period at 
one camera site. Consecutive photographs 
of the same species within 30 minutes were 
considered a single detection. Small prey 
were those prey species with a body weight 
< 5 kg, medium prey were those species 
whose body weight 5-20 kg, and large prey 
were considered those species whose body 
weight > 20 kg. 
We tested the correlation between the 
photographic rates and the sampling ef- 
fort (number of camera-days) for all felids, 
and the correlation between the photogra- 
phic rates of felids and their prey. For this 
analysis, we combined the three southern 
nature reserves (Yele, Shenguozhuang and 
Mamize) due to low sampling effort in each 
individual reserve, and excluded the sample 
sites that were above tree-line. All variables 
were examined for normal distribution pri- 
or to analysis and no transformations were 
needed. 
Result 
From 2002-2009, 461 1x1 km2 blocks in 11 
nature reserves were surveyed for a total 
sampling effort of 19,151 camera-days (Fig. 
1, Table!), among which 6 blocks were abo- 
ve tree-line. The 11 surveyed reserves were 
not evenly sampled (sampling efforts varied 
from 130-4977 camera-days) due to logistic 
reasons. Tangjiahe, Changqing, Wanglang, 
and Wolong were the reserves with most 
sampling efforts (>2000 camera-days). 
Four felid species were detected during our 
survey: snow leopard [n = 2 sites, >3 indivi- 
duals, 1 reserve), leopard (n = 1 site, 1 indi- 
vidual, 1 reserve), Asiatic golden cat (n = 4 
sites, 2 reserves), and leopard cat (n = 45 
sites, 9 reserves) (Fig. 2-5). Leopard cat, the 
smallest but most detected species, was re- 
corded at 9 out of the 11 reserves. The ave- 
rage photographic rate of leopard cat among 
these reserves was 5.03 (range 0.80-19,01). 
Large- and medium-sized felids (i.e., leo- 
pard, snow leopard, and Asiatic golden cat) 
were only detected in Changqing, Tangjiahe, 
and Wolong with rather low photographic 
rates. Changqing Nature Reserve detected 
the most felid species (i.e., leopard, Asiatic 
golden cat, and leopard cat), followed by 
Wolong Nature Reserve (i.e., snow leopard 
and leopard cat) and Tangjiahe Nature Re- 
serve (i.e., Asiatic golden cat and leopard 
cat). The sampling effort (number of came- 
ra-days) prior to detection of felid species 
varied broadly among our survey reserves, 
from 50-1250 (average 200) camera-days for 
leopard cats to 600-5000 camera-days for 
the medium- and large-sized felids (Table 1), 
but there was no significant difference bet- 
ween small, medium- and large-sized felids 
(independent-samples t test, t - 1.170, v = 
2.037, p - 0.361). No medium- or large-sized 
felids were detected in the three southern 
reserves {i.e., Yele, Shenguozhuang, and 
Mamize). 
Small prey detected in our survey included 
pheasant and grouse species, pika Ochotona 
spp., wooly hare Lepus oiostolus, small ro- 
dents, and squirrels. Medium prey included 
short-tailed porcupine Hystrix brachyura, 
hog badger Arctonyx collaris, golden monkey 
Hhinopithecus roxeliana, macaque Macaca 
spp., tufted deer Elaphodus cephalophus, 
Reeve's muntjac Muntiacus reevesi, and fo- 
rest musk deer Moschus berezovskii. Large 
21 
CATnews 52 Spring 2010 
Sheng Li et al. 
prey detected were wild boar Sus scrota, 
blue sheep Pseudois nayaur, goral Naemor- 
hedus goral, Chinese serow Capricornis mil- 
needwardsii, and takin Budorcas taxicolor. 
Changqing Nature Reserve had the highest 
photographic rate of large prey, followed 
by Wolong and Tangjiahe Nature Reserves 
(Table 1). In the two southern reserves of 
Liangshan mountains (i.e., Shenguozhuang 
and Mamize), no large prey species were 
detected. 
We found no significant correlation bet- 
ween the sampling effort and photographic 
rates of small felid (leopard cat, Pearson 
test, r - 0.409, p = 0.667, n = 9), or medium- 
and large-sized felids (e.g. leopard, snow 
leopard and Asiatic golden cat combined, 
Pearson test, r - 0.763, p = 0.100, n= 9). We 
also found no significant correlation bet- 
ween the photographic rates of small felid 
(leopard cat) and small prey (Pearson test, 
r = 0.767, p = 0.319, n = 9), but a positive 
correlation among the photographic rates of 
medium- and large-sized felids and large- 
sized prey [Pearson test, r - 0.821, p = 0.047, 
n = 9). 
nn       Discussion 
       Our surveys provide current information on 
the status of felids and their prey within 
the study area. Our record of snow leopard 
in Wolong Nature Reserve (Fig. 2) was the 
first photographic evidence of this endan- 
gered species at the southeast edge of its 
distribution (Nowell & Jackson 1996, IUCN 
2009). The photograph of an Asiatic golden 
cat at Tangjiahe Nature Reserve showed a 
rare coat pattern with rosettes and spots, 
which is the first field photo of this pheno- 
type in China and only occasionally reported 
elsewhere (Nowell & Jackson 1996, Wang 
2007, Wilson & Mittermeier 2009, Fig. 3). 
These preliminary results indicate extending 
the sampling effort of these surveys would 
benefit felid research and conservation in 
China. 
There are three potential reasons under- 
lining the spatial pattern in our survey, in 
which more felids were detected in northern 
than in southern reserves: 1) sampling effort; 
2) prey abundance; and 3) poaching pressu- 
re on these felids. We found no significant 
correlation between sample effort and car- 
nivore abundance but our sample effort was 
lowest in the south. Camera-trapping is con- 
sidered the most effective methodology to 
detect the presence of wild felids (Cutler & 
Swann 1999, Karanth & Nichols 1998), but 
nondetection cannot be directly considered 
absence, especially when the sampling ef- 
fort is low. Previous studies show that the 
number of camera-days required for de- 
tection of rare felid species varies broadly 
(e.g., 230-2945 camera-days for tiger and 
16-4418 camera-days for clouded leopard 
in southeast Asia; Datta et al. 2008), so its 
possible our sampling effort in southern 
reserves (130-547 camera-days) was insuf- 
ficient to detect rare large felids. However, 
this sampling effort was sufficient to detect 
large mammals in the northern reserves and 
in other regional studies conducted by this 
research team (Wang et al. 2006, Li et al., 
in press), but detected no large prey spe- 
cies in these southern reserves, a finding 
in agreement with informal interviews with 
local people. Large mammals are probably 
missing from these reserves because most 
of the nature reserves in Xiangliang and Li- 
angshan mountains were established from 
1993 to 2002 after a long history of commer- 
cial timber logging, and no control over har- 
vest of both the felids and their prey. Of our 
three probable reasons for lack of detection 
in these southern reserves, we consider low 
survey effort the least likely. 
All the surveyed nature reserves have an 
official mammal species list generated from 
baseline investigation and historical litera- 
ture. Two to seven felid species are listed 
in each nature reserve's formal species list 
(S0M T1), though there may be no evidence 
for the presence of some species within the 
reserve in recent decades. For example, six 
felid species are on the baseline species list 
of Tangjiahe (Hu 2005), but no photographs, 
sightings or dead/captured animal have 
been reported for three species (i.e., clou- 
ded leopard, Eurasian lynx, and Pallass cat) 
since the reserve was established in 1978 
(Liu X., pers. comm.). Our large sampling 
effort (approximate 5000 camera-days), 
covering all possible habitat types within 
this 300 km2 reserve, indicates a conspicu- 
ous gap between the felid species detected 
and the official list maintained by the reser- 
ve. Similar discrepancies exist in all other 
surveyed reserves. During the 3rd National 
Giant Panda Survey from 1999-2003, only 
three felid species (i.e., leopard, Asiatic gol- 
den cat and leopard cat) were detected with 
an extensive sampling effort of 11,174 sign 
transects throughout the giant panda di- 
stribution (SPA 2006). Among the nine felid 
species historically recorded within current 
giant panda range (SOM Tl), the tiger has 
been probably extinct since the late 20th 
century (IUCN 2009), and the record of Chi- 
nese mountain cat remains doubtful (He et 
al. 2004). Though our camera-trapping effort 
didn't cover all possible felid habitat types 
in some reserves (e.g. above treeline), our 
results do highlight the need for detailed 
survey of felids in the major reserves rather 
than reliance on the reserve species lists to 
determine carnivore status and distribution. 
Studies on the distribution range of each in- 
dividual felid species need to be conducted 
prior to setting felid conservation goals in 
China. We think a well-designed systemic 
monitoring network using camera-trapping 
would be the best approach to obtain such 
information (Li et al., in press). 
Although none of our surveyed nature reser- 
ves were established for the conservation of 
felids, the giant panda may serve as an um- 
brella species for the sympatric felids and 
their prey (Schaller 1994, Lu et al. 2000). Ne- 
vertheless, there may be size limitations for 
reserves established for giant pandas that 
render the reserves of insufficient area for 
large carnivore species. Large felids normal- 
ly maintain low densities due to social inter- 
actions and large home ranges for foraging 
requirement. For example, the home range 
of individual snow leopard (McCarthy et al. 
2005) and leopard (Norton & Lawson 1985) 
can exceed 500 km2, whereas the average 
size of giant panda reserve is 544 km2 (ran- 
ge: 83-4069 km2, SPA 2006). The area nee- 
ded to support a viable population for these 
large felids may require several adjacent 
nature reserves. Isolated reserves, such as 
those in the southern mountains, may be in- 
sufficient in size for large felid conservation. 
For an effective conservation of felid spe- 
cies, adjacent giant panda reserves should 
be considered an integrated management 
unit with connectivity between reserves of 
primary concern. 
Acknowledgements 
This study was financially supported by Smithso- 
nian National Zoological Park, Peking University, 
Shanshui Conservation Center, Conservation In- 
ternational and World Wildlife Fund. We thank 
Sichuan Forestry Department, Shaanxi Forestry 
Department, and local forestry bureaus for permis- 
sions and logistical support. We thank the staff of 
all involved nature reserves for their contribution 
on project management, field work, data collection 
and logistics. We cannot collect these high quality 
data without their hard work and effort. We also 
thank James Sanderson, Shu-Jin Luo, Li Zhang, 
CATnews 52 Spring 2010 
cats living with pandas 
Eva Jutzeler, Urs Breitenmoser and an anonymous 
reviewer for reviewing our manuscript and provi- 
ding valuable comments. 
References 
Cam T. M. & O'Doherty G. 1999. On the use of sur- 
rogate species in conservation biology. Conser- 
vation Biology 13,805-814. 
CutlerT. L. & Swann D. E. 1999. Using remote pho- 
tography in wildlife ecology: a review. Wildlife 
Society Bulletin 27,571-581. 
Datta A., Anand M. 0. & Naniwadekar R. 2008. 
Empty forests: large carnivore and prey abun- 
dance in Namdapha National Park, north-east 
India. Biological Conservation 141,1429-1435. 
Gu X. D., Wang H. J. & Liu F. W. 2005. Biodiversity 
monitoring in giant panda nature reserves in 
Minshan Mountain Region (In Chinese). Sichu- 
an Journal of Zoology 24,168-170. 
Gu X. D? Yang Z. S. & Wang G. 2004. Status ana- 
lysis of eco-monitoring in giant panda nature 
reserves, Sichuan (In Chinese). Sichuan Jour- 
nal of Zoology 23,146-148. 
He L, Garcia-Perea ft, Li M. & Wei F. W. 2004. 
Distribution and conservation status of the 
endemic Chinese mountain cat Fells bieti. Oryx 
38,55-61. 
Hu J. C. 2005. A report of the comprehensive sur- 
vey on Tangjiahe Nature Reserve in Sichuan, 
China (In Chinese). Sichuan Science and Tech- 
nical Press, Chengdu. 
Hu J. C. & Wang Y. Z. 1984. The resource animals 
in Sichuan (Volume 2: Mammals) (In Chinese). 
Sichuan Science and Technical Press, Cheng- 
du. 
IUCN. 2009. IUCN Red List of Threatened Species. 
Http://www.red I ist.org [accessed 15 May 
2009]. 
Karanth K. U. & Nichols J. 0.1998. Estimation of 
tiger densities in India using photographic cap- 
tures and recaptures. Ecology 79,2852-2862. 
Leader-Williams N. & Dublin H. T. 2000. Cha- 
rismatic megafauna as 'flagship species'. In 
Priorities for the conservation of mammalian 
diversity. Entwistle A. & Dunstone N. (Eds). 
Cambridge University Press, Cambridge, pp. 
53-81. 
Li S? Wang D. J., Gu X. D. & McShea M. J. 2010. 
Beyond pandas, the need for a standardized 
monitoring protocol for large mammals in Chi- 
nese nature reserves. Biodiversity and Conser- 
vation, in press. 
Loucks C.J., Lu Z., Dinerstein E? Wang H., Olson 0. 
M? Zhu C. Q. & Wang D. J. 2001. Giant pandas 
in a changing landscape. Science 294,1465. 
Lu Z? Pan W. S? Zhu X. J., Wang D. J. & Wang 
H. 2000. What has the panda taught us? In 
Priorities for the conservation of mammalian 
diversity. Entwistle A. & Dunstone N. (Eds). 
Cambridge University Press, Cambridge, pp. 
325-334. 
McCarthy T. M., Fuller T. K. & Munkhtsog B. 2005. 
Movements and activities of snow leopards in 
Southwestern Mongolia. Biological Conserva- 
tion 124, 527-537. 
Norton P. M. & Lawson A. B. 1985. Radio tracking 
of leopards and caracals in the Stellenbosch 
area. Cape Province. South African Journal of 
Wildlife Research 15,17-24. 
Nowell K. & Jackson P. 1996. Wild cats: status sur- 
vey and conservation action plan. Switzerland: 
IUCN/SSC Cat Specialist Group. 
Schaller G. B. 1994. The last panda. University of 
Chicago Press, Chicago. 
[SFAj State Forestry Administration. 2006. The 3rd 
national survey report on giant panda in China 
(In Chinese). Science Press, Beijing. 
[SFA] State Forestry Administration. 2007. State 
monitoring protocol on giant panda and its ha- 
bitat (In Chinese). Center for Wetland & Fauna 
and Flora Monitoring, SFA, Beijing, China. 
Smith A. & Xie Y. 2008. A Guide to the Mammals 
of China. Princeton University Press, Princeton. 
Wang D. J., Li S, McShea W. J. & Li M. F. 2006. 
Use of remote-trip cameras for wildlife surveys 
and evaluating the effectiveness of conserva- 
tion activities at a nature reserve in Sichuan 
Province, China. Environmental Management 
38,942-951. 
Wang S. 1998. China red data book of endangered 
animals: Mammalia. Science Press, Beijing. 
Wang S. W, 2007. A rare morph of the Asiatic gol- 
den cat in Bhutan's Jigme Singye Wangchuk 
National Park. Cat News 47,27-28. 
Wang X. Z. 2008. Influences of human activities 
on the giant panda's habitat utilization in the 
Minshan mountain range [In Chinese). Ph. D. 
Dissertation, Chinese Academy of Sciences. 
Wilson D. E. & Mittermeier R. A. 2009. Handbook 
of the mammals of the world, Vol. 1: Carni- 
vores. Lynx Ediciones, Barcelona. 
Supporting Online Material SOM 
Table Tl at www.catsg.org 
College of Life Sciences, Peking University, 
Beijing, China;' corresponding author 
<lis@si.edu> 
Smithsonian Conservation Biology Institute, 
National Zoological Park, USA 
Shanshui Conservation Center, Beijing, China 
23 
Fig. 4. Asiatic golden cat (spotted moiph), 
Tangjiahe NR, 2008 
Fig. 5. Leopard cat, Wanglang NR, 2004 
CATnews 52 Spring 2010