This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright
www.elsevier.de/mambio SHORT COMMUNICATION Food habits of the Tibetan fox (Vulpes ferrilata) in the Kunlun Mountains, Qinghai Province, China Qunxiu Liu a, Richard B. Harris b, Xiaoming Wang a, a College of Life Science, East China Normal University, Shanghai 200062, China b Department of Ecosystem and Conservation Sciences, University of Montana, Missoula, MT 59801, USA Received 10 December 2008; accepted 8 February 2009 Keywords: China; Diet; Kunlun Mountains; Plateau pika; Tibetan fox The Tibetan fox (Vulpes ferrilata) is restricted in geographic distribution to the Tibetan highlands of China and adjacent areas of Bhutan, India, and Nepal (Schaller and Ginsberg 2004; Clark et al. 2008). Studied only during the past few years (Gong and Hu 2003; Wang et al. 2007), it has been assumed to prey principally on plateau pikas, Ochotona curzoniae, which achieve locally high densities on the Tibetan plateau (Smith and Foggin 1999), and with which it is invariably associated (Schaller 1998; Clark et al. 2008; Harris et al. 2008). Although plateau pikas clearly have an important role in facilitating the existence of many other species (Smith and Foggin 1999), they are considered an agricultural pest by Chinese government bureaus, and have been subject to large-scale extermination campaigns using various poisons since the 1950s (Smith et al. 2006). Tibetan foxes have also been reported to consume small rodents (e.g., Alticola, Cricetulus, and Pitymys spp., and plateau zokors [Eospalax fontanierii]), and to scavenge carrion from wild (e.g., Tibetan gazelles [Procapra picticaudata], blue sheep [Pseudois nayaur]) and domestic (yak [Bos mutus], sheep [Ovis aries]) ungulates (Schaller 1998; Clark et al. 2008). Somewhat surprisingly, quantitative documentation of Tibetan fox food habits is limited to that of Zheng (1985), who, unfortunately, lumped lagomorphs (including pikas) together with rodents, and Schaller (1998, p. 187), who Corresponding author. Tel./fax: +1 406 542 6399. E-mail address: xmwang@ecnu.edu.cn (X.M. Wang). noted that rodents were rare within his category pika or small rodent. Thus, clarification of Tibetan foxes potential dependence on plateau pikas has been based largely on anecdotal observations of them preying on pikas, and on the apparent (but not-yet rigorously examined) overlap between their areas of occurrence and areas of high pika density. Here, we report on food habits of Tibetan foxes in the Kunlun Mountains of the northern Tibetan plateau, Qinghai Province, China, based on an analysis of scats. Our study objectives were to quantify Tibetan fox diets, estimate the importance of plateau pikas for foxes, and examine if fluctuating pika numbers affected their importance in fox diets. We used foxes captured and radio-monitored as part of a larger study (Harris 2007; Liu et al. 2007) to help us locate feces, which were later examined. Our study was conducted in Gouli Township, Dulan County, Qinghai Province, China. The study area, centered on approximately 35130 0 N, 98140 0 E, covered approximately 32 km 2 of rolling mountainous terrain in the eastern Kunlun Mountains at elevations of 3800 4500 m. Vegetation formations were alpine steppe (dominated by Stipa purpurea), alpine meadow (dominated by Kobresia spp.), and shrublands (dominated by Salix spp.). The study area formed the winter grazing area for approximately 10 households of Tibetan pastoralists who raised primarily domestic sheep and yaks along with a few domestic goats (Capra hircus) and horses (Equus caballus). 1616-5047/$ - see front matter r 2009 Deutsche Gesellschaft für Säugetierkunde. Published by Elsevier GmbH. All rights reserved. doi:10.1016/j.mambio.2009.02.002 Mamm. biol. 75 (2010) 283 286
284 Q.X. Liu et al. / Mamm. biol. 75 (2010) 283 286 Blue sheep were common in the higher elevation and/ or steeper slopes of the study area; a few Tibetan gazelles were also year-round residents within home ranges of all of our radio-marked Tibetan foxes. Small bands of argali (Ovis ammon) and red deer (Cervus elaphus) were occasional visitors to the study area. Based on interviews with pastoralists, we believe that both musk deer (Moschus chrysogaster) and wild yaks (Bos grunniens) were originally present in the area, but had long since been extirpated by the time of our study. Other carnivores we observed within the study area were red fox (V. vulpes), wolf (Canis lupus), lynx (Lynx lynx), Pallas cat (Otocolobus manul), and badger (Meles meles). We also observed tracks of brown bear (Ursus arctos) and snow leopard (Uncia uncia; a separate study had obtained a photograph of a free-ranging snow leopard using a remotely-operated camera in 2007, Jiang Zhigang, Institute of Zoology, personal communication, 2007). We collected feces of Tibetan foxes during September 21 October 21, 2006, March 13 May 20, 2007, September 15 October 17, 2007 and March 6 March 21, 2008. To minimize the potential for species mis-identification, scats were considered to originate from Tibetan foxes only when (i) they appeared fresh, (ii) they were found in close proximity to a known, radio-marked Tibetan fox, (iii) they were collected from a known Tibetan fox den or burrow, or (iv) we observed deposition by a radiomarked fox directly. Scats collected were air-dried in the field, and stored in paper bags. We captured 5 Tibetan foxes using Victor Soft-Catch s leg-hold (Woodstream Corporation, Lititz, Pennsylvannia, USA), or Belisle s leg-hold traps and immobilized them with Ketamine hydrochloride (approximately 6 mg/kg) and Medetomidine (approximately 0.05 mg/kg). Foxes were fitted with VHF radio-collars (MOD-225, Telonics, Inc., Mesa, Arizona, USA) weighing approximately 130 g and monitored 1 7 times/week. All capture and handling procedures complied with University of Montana Animal Use Committee Protocols. We also obtained an index of pika density by counting all pikas seen in 30-min time spans within 30 m radius plots placed randomly within each 0.5 km 2 cell of the study area. Plots were surveyed both in early autumn (September October 2007 [n ¼ 54] when it was hypothesized pika density would be high, and late winter (March May 2007 [n ¼ 77], and March May 2008 [n ¼ 61]), when it was hypothesized pika density would be low, following the normal pattern of winter mortality (Wang and Smith 1988). We tested whether the pika density index differed between seasons using a Mann Whitney U-test with a normal approximation. In the laboratory, scats were autoclaved (30 min at 120 1C), then oven-dried (at 80 1C). After being soaked in water for 24 72 h, individual dried scats were filtrated with a 0.5-mm-mesh sieve to separate microscopic from macroscopic fragments and to discard the inconspicuous fragments (Ciucci et al. 1996; Reynolds and Aebischer 1991). Prey determination was performed by microscopic analyses on the basis of hair characteristics, feather and bone. Guard hairs were used to identify mammal prey. Macroscopic and microscopic structures of hairs found in the scats were compared with reference hairs (e.g., Brunner and Coman 1974; Reynolds and Aebischer 1991) obtained in the study area, or with laboratory samples stored at East China Normal University in Shanghai or Northeast Forest University in Harbin. Because of the low presence of rodent remains, and because differentiating among Eospalax, Allactaga, and other rodent species was difficult, we pooled them together as one category termed rodents. We were, however, able to reliably separate Marmota himalayana from other rodents, and pikas from all rodents. Undistinguishable fragments were labeled as unidentified. We expressed diet composition as the percent occurrence (PO), defined simply as the number of scats containing a prey item divided by the total collected ( 100). We tested for difference in FO between our hypothesized low (March May) and high (September - October) pika density periods using a w 2 test of independence. Food niche breadth (B) was calculated from the proportions of the 11 food categories in scats as suggested by Hurlbert (1978): B ¼ 1= X ðp 2 i Þ where p i is the proportion of food items made up by category i. Shannon Weaver diversity (H) and evenness (E) were calculated: H ¼ Xs i¼1 ðp i Þðlog 2 P i Þ E ¼ H=H max (iii) We analyzed 93 scats that included pika, rodent, large mammal, bird, reptile, insect, vegetation, and unidentified animal remains (Table 1). Pika remains were found in 84% of scats, representing the most frequently occurring food item. Rodents, including Himalayan marmots, occurred in 26% of scats (Table 1). All other prey items were found in only a few scats (Table 1). The pika density index was higher in September October (x ¼ 4:15, 95% confidence interval 3.42 4.88) than in March May (x ¼ 2:67, 95% confidence interval 2.31 3.03; Mean Ranks 86.7, 121.5, U ¼ 2376, Z ¼ 3.961, Po0.001). However, we found no evidence of seasonal differences in diets of Tibetan foxes (w 2 ¼ 9.14, 11 df, P ¼ 0.609). The year-round food niche breadth (B) was 1.32, Shannon Weaver diversity and evenness indices were 2.29 and 0.66, respectively. (i) (ii)
Q.X. Liu et al. / Mamm. biol. 75 (2010) 283 286 285 Table 1. Number and percent occurrence (PO) of Tibetan fox scats (n ¼ 93) containing various food items, collected in Gouli Township, Dulan County, Qinghai Province, China, September 15 October 17, 2007, March 13 May 20, 2007, March 6 March 21, 2008. Items September October (high pika density, n ¼ 52) March May (low pika density, n ¼ 41) n PO n PO Plateau pika Ochotona curzoniae 45 86.5 33 80.5 Undifferentiated rodents 10 19.2 6 14.6 Marmot Marmota himalayana 4 7.7 4 9.8 Blue sheep Pseudois nayaur 0 0 4 9.8 Tibetan gazelle Procapra picticaudata 1 1.9 2 4.9 Domestic sheep Ovis aries 1 1.9 2 4.9 Domestic yak Bos mutus 3 5.8 0 0 Undifferentiated bird 1 1.9 1 2.4 Undifferentiated reptile 0 0 1 2.4 Undifferentiated insect 7 13.5 2 4.9 Vegetation 1 1.9 2 4.9 Unidentified animal 13 25.0 10 24.4 Our results confirm previous suggestions that Tibetan foxes prey principally on plateau pikas, and secondarily on various rodents (Schaller and Ginsberg 2004, Clark et al. 2008). Remains of wild ungulates found in fox feces probably represent scavenging on carrion (Schaller and Ginsberg 2004); we believe it unlikely that Tibetan foxes, at only 4 5.5 kg (Harris et al. 2008) are capable of killing adult ungulates (although newly-born ungulates may be vulnerable to fox predation for a few days). Tibetan foxes are probably capable of killing young domestic sheep, but in informal interviews, pastoralists never reported that foxes were a depredation concern, even during lambing season. Marmot remains were detected in fox scats collected in March, April and September. Himalayan marmots in our study area were rarely above ground before May or after August; thus, we speculate that, in addition to direct predation, foxes may have scavenged the carcasses of marmots killed at other times of year, and possibly by other predators. In collecting feces in the field, we endeavored to ensure that they came from Tibetan foxes rather than other sympatric carnivores; however, we cannot rule out the possibility of some mis-identification. Based on unquantified visual observations, Tibetan foxes appeared to be far more common in the study area than any other mammalian carnivore. The most likely misidentification would have come from red foxes. However, we only made 3 red fox observations during the 2 years of field work, whereas we observed Tibetan foxes on well over 100 occasions even without the aid of telemetry. We captured only a single red fox in 4200 capture nights, compared with 7 captures of Tibetan foxes. This, coupled with our field procedures, suggested that we erroneously included very few red fox feces within our sample. An additional shortcoming of basing diet estimates on raw frequency of occurrence in feces is that analysis of different size prey species may result in overestimating the proportion of smaller animals because of their higher surface: volume ratios (Floyd et al. 1978; Geraldine 1978; Corbett 1989). Our research may therefore have overestimated the importance of plateau pikas and smaller rodents relative to larger-bodied prey. Regression equations, such have been developed for other carnivores, would be useful to assess this shortcoming. Unfortunately, we are unaware of any Tibetan foxes in captivity that could be used for such a controlled experiment. Notwithstanding these uncertainties, our results confirm the central role of plateau pikas in the life-history of Tibetan foxes. Acknowledgements Funding for this project came from the Robert M. Lee Foundation, the Research Fellowship Project (RFP) of the Wildlife Conservation Society (WCS), the Denver Zoological Foundation, and the doctoral training program at East China Normal University. We thank Zhou Jiake for assistance in the field, and Da Shenglin for administrative support. References Brunner, H., Coman, B.J., 1974. The Identification of Mammalian Hair. Inkata Press, Melbourne. Clark Jr., H.O., Newman, D.P., Murdoch, J.D., Tseng, J., Wang, Z.H., Harris, R.B., 2008. Vulpes ferrilata. Mammalian Species 821, 1 6.
286 Q.X. Liu et al. / Mamm. biol. 75 (2010) 283 286 Ciucci, P.L., Boitani, E.R., Pelliccioni, M.R., Guy, I., 1996. A comparison of scat-analysis methods to assess the diet of the wolf (Canis lupus). Wildlife Biology 2, 37 48. Corbett, J.K., 1989. Assessing the diets of dingos from feces: a comparison of 3 methods. Journal of Wildlife Management 53, 343 346. Floyd, T.J., Mech Jr., L.D., Jordan, P.A., 1978. Relating wolf scat content to prey consumed. Journal of Wildlife Management 42, 528 532. Geraldine, M.B., 1978. The use of scat samples in primate diet analysis. Primates 19, 215 221. Gong, M.H., Hu, J.C., 2003. The summer microhabitat selection of Tibetan fox in the northwest plateau of Sichuan. Acta Theriologica Sinica 23, 266 269 (in Chinese). Harris, R.B., 2007. Wildlife Conservation in China: Preserving the Habitat of China s Wild West. M.E. Sharpe, Inc., Armonk, NY. Harris, R.B., Wang, Z.H., Zhou, J.K., Liu, Q.X., 2008. Notes on biology of the Tibetan fox (Vulpes ferrilata). Canid News 11, 1 7 (http://www.canids.org/canidnews/11/biology_ of_tibetan_fox.pdf). Hurlbert, S.H., 1978. The measurement of niche overlap and some relatives. Ecology 59, 67 77. Liu, Q.X., Harris, R.B., Wang, X.M., Wang, Z.H., 2007. Home range size and overlap of Tibetan foxes (Vulpes ferrilata) in Dulan County, Qinghai Province. Acta Theriologica Sinica 27, 370 375 (in Chinese). Reynolds, J.C., Aebischer, N.J., 1991. Comparison and quantification of carnivore diet by faecal analysis: a critique, with recommendations, based on a study of the red fox Vulpes vulpes. Mammal Review 21 (3), 97 122. Schaller, G.B., Ginsberg, J.R., 2004. Tibetan fox Vulpes ferrilata. In: Sillero-Zubiri, C., Hoffmann, M., Macdonald, D.W. (Eds.), Canids: Foxes, Wolves, Jackals and Dogs. Status Survey and Conservation Action Plan. International Union for Conservation of Nature and Natural Resources/ Species Survival Commission Canid Specialist Group. The World Conservation Union, pp. 148 152. Schaller, G.B., 1998. Wildlife of the Tibetan Steppe. University of Chicago Press, Chicago. Smith, A.T., Foggin, J.M., 1999. The plateau pika (Ochotona curzoniae) is a keystone species for biodiversity on the Tibetan plateau. Animal Conservation 2, 235 240. Smith, A.T., Zahler, P., Hinds, L.A., 2006. Poisoning of native small mammals in central Asia is an undesirable and unsustainable activity. In: McNeely, J., McCarthy, T.M., Smith, A.T., Olsvig-Whittaker, J., Wikramanayake, E.D. (Eds.), Biodiversity Conservation in Asia. Society for Conservation Biology, Asian Section and Resources Himalaya Foundation, Kathmandu, Nepal, pp. 285 293. Wang, X.G., Smith, A.T., 1988. On the natural winter mortality of the plateau pika (Ochotona curzoniae). Acta Theriologica Sinica 8, 152 156 (in Chinese). Wang, Z.H., Wang, X.M., Lu, Q.B., 2007. Selection of landcover by the Tibetan fox Vulpes ferrilata on the eastern Tibetan Plateau, western Sichuan Province, China. Acta Theriologica 52, 215 223. Zheng, S.W., 1985. Data on the foods of Tibetan sand fox. Acta Theriologica Sinica 5, 222 223 (in Chinese).