258 C. Zhang et al.
TABLE 1 Model comparison table using the Akaike information cri- terion adjusted for small sample sizes (AICc) for spatially explicit capture–recapture models, to estimate snow leopard Panthera uncia density in Machang, central and eastern Qilian Mountains, China (Fig. 1), during January–March 2019.
Model1 D(.)g0(.), σ(.) K2 AICc 3 2,675.45 D(.)g0(b), σ(.) 4 2,676.72
ΔAICc3 0.00
2.59
D(.)g0(k), σ(.) 4 2,893.65 16.79 D(.)g0(bk), σ(.) 4 2,894.15 18.42
Model weight 0.65
0.19 0.09 0.06
1D, snow leopard density; g0, detection probability; σ, spatial scale over which the detection probability declines as the distance between the activity
centre of an individual and the camera-trap station increases. 2Number of model parameters. 3Difference in AICc to the top-ranked model.
China. This information will support regional and global snow leopard conservation, research and management. Accurate population and demographic estimates are
critical for snow leopard conservation. Our overall density estimate of 2.26 snow leopards/100 km2 is comparable to other published capture–recapture-based estimates for the species (range 0.12–8.49 snow leopards/100 km2; Jackson et al., 2006, 2009; McCarthy et al., 2008; Alexander et al., 2015; Chen et al., 2017; Kachel et al., 2017; Suryawanshi et al., 2017; Chetri et al., 2019; Nawaz et al., 2021). This variation in density estimates amongst study sites could reflect true differences in abundance or they could be a result of the methods used, or both. At a relatively low mean altitude and with a high human population density, it would be reasonable to expect a lower snow leopard density in our study area compared to the higher-elevation landscapes of the western Qilianshan National Nature Reserve, for which the density estimate is 3.3 snow leop- ards/100 km2 (95%CI 1.4–5.3; Alexander et al., 2015). Earlier estimates of density based on capture‒recapture
models (e.g. McCarthy et al., 2008) as well as spatial den- sity estimates for small study areas might be positively biased. A meta-analysis has demonstrated that surveys in small areas and a bias towards studies in good-quality habitat could have overestimated snow leopard densities by up to five times (Suryawanshi et al., 2019). The spatial coverage of a survey is one of the most important factors affecting the reliability of estimates from spatially explicit capture–recapture models. It has been recommended that a minimum survey area of 573 km2 is required for estimat- ing density reliably (Suryawanshi et al., 2019). Spatially explicit capture–recapture methods overcome the edge effects that are problematic in conventional capture– recapture methods (Otis et al., 1978), and these models performed well if the trapping arrays were similar to or larger than the mean home range of the species of concern (Sollmann et al., 2012). The mean 95% minimum convex polygon home range estimate for the snow leopard is
c. 500 km2 (Johansson et al., 2016). Our study area (625 km2) was larger than both this and the recommended minimum sampling area of 573 km2, and these areas are larger than those used in most previous estimations of snow leopard density (Jackson et al., 2006, 2009; Alexander et al., 2015; Chen et al., 2017; Suryawanshi et al., 2017). An adequate sample size of captured and recaptured
individuals is required for analyses of the precision of spatially explicit capture–recapture estimates (Sun et al., 2014; Alexander et al., 2015). A minimum of 10–20 in- dividuals has been suggested as an adequate sample size (Otis et al., 1978), and a minimum of 10 recaptures is required to draw reliable inferences from spatially explicit capture–recapture analysis (Efford et al., 2004). Many pre- vious snow leopard studies have identified ,10 individuals (Jackson et al., 2006, 2009;Chenetal., 2017;Kacheletal., 2017;Suryawanshi et al., 2017; Nawaz et al., 2021). Usable detections were also limited in our study (10 individuals were only captured once and four were captured twice each), highlighting the importance of optimizing captures for individual identification. To counter the main cause of low-quality images, inadequate night-time illumination (Alexander et al., 2015), we used cameras with flash. However, 60% of the photographs obtained could not be used for individual identification because individual fea- tures were washed out by the flash and/or only a part of the body was visible as the animal moved across the cam- era’s field of view. More sensitive and higher definition cameras and lures such as castor or fish oils could be used to increase photograph quality and capture probability, respectively, in future research. The statistical methods used in spatially explicit capture–
recapture modelling are sensitive to trap configuration (Sun et al., 2014; Nawaz et al., 2021). The recommendation is that trap spacing be at most the maximum distance moved during the research period (the 2σ rule of thumb; Sun et al., 2014). In our study design the 1-km trap spacing was ,2σ (11 km). In addition, individuals are not reliably and accurately identified in many camera-based capture– recapture studies, potentially resulting in systematically inflated population abundance estimates by a third on average (Johansson et al., 2020). The other constraints of density estimation using spat-
ially explicit capture–recapture modelling relate to the method for determining the effective area surveyed. Following Alexander et al. (2015) we excluded areas outside Qilianshan National Nature Reserve identified as unsuit- able habitat for snow leopards, based on our field knowl- edge. Because a snow leopard had been previously ob- served in forest in Qilianshan National Nature Reserve, we included areas below 3,000 m in our study, even though it has been suggested that snow leopards do not use forest (Snow Leopard Network, 2014). We believe there is a negligible probability of snow leopard activity centres
Oryx, 2024, 58(2), 255–260 © The Author(s), 2023. Published by Cambridge University Press on behalf of Fauna & Flora International doi:10.1017/S0030605323000340
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100 |
Page 101 |
Page 102 |
Page 103 |
Page 104 |
Page 105 |
Page 106 |
Page 107 |
Page 108 |
Page 109 |
Page 110 |
Page 111 |
Page 112 |
Page 113 |
Page 114 |
Page 115 |
Page 116 |
Page 117 |
Page 118 |
Page 119 |
Page 120 |
Page 121 |
Page 122 |
Page 123 |
Page 124 |
Page 125 |
Page 126 |
Page 127 |
Page 128 |
Page 129 |
Page 130 |
Page 131 |
Page 132 |
Page 133 |
Page 134 |
Page 135 |
Page 136 |
Page 137 |
Page 138 |
Page 139 |
Page 140
