Anthropogenic pressure on large carnivores 265
(7) location within or outside (binary) the proposed Lenya Reserved Forest. The selection of these variables was based on previous studies and on the biological and ecological requirements of the study species (Hossain et al., 2018). We measured distance to road and distance to village
using ArcGIS 10.3 (Esri, Redlands, USA). We used a forest cover map (Connette et al., 2016) to classify land as forest, non-forest and degraded forest within a 1-km radius of each camera-trap location, also using ArcGIS. Forest classi- fications follow Connette et al. (2016), with forest having a canopy cover of.80%and degraded forest having a canopy cover of #80%.
Data analysis
We used logistic regression to analyse the relationships be- tween the detection/non-detection of tigers, leopards and dholes and the potential explanatory covariates. We created separate models for individual prey species and all prey spe- cies. We checked continuous variables for outliers prior to running the models. We then standardized all continuous variables by subtracting the mean and dividing by twice the standard deviation (Gelman, 2008). We treated unequal camera-trap survey effort at each location by using an offset in the model formula. This is equivalent to including survey effort (trap-days) as a regression predictor but with its coef- ficient fixed to 1 (Gelman & Hill, 2007). We assessed spatial autocorrelation using the
spline.correlog function in the ncf package (Ottar, 2018)in R 3.2.2 (R Development Core Team, 2015). This function estimates the spatial dependence of data at discrete distance classes measured using the centred Mantel statistic (Ottar, 2018). We did not include highly correlated (r.0.5)variables in
the same regression model.We compared models using the Akaike information criterion (AIC), and used Akaike model weights (wi)as evidence infavour ofmodel i amongst the models being compared.We assessed themodel classification accuracy of the logistic regression by using the area under the receiver operating characteristic curve (AUC), which varies from 0.5 (models that are no better than random) to 1.0 (high-accuracy models; Franklin, 2009).We calculated AUC thresholds using multiple cut-off points (0.2, 0.4, 0.6, 0.8), where sensitivity is equal to specificity, using the package PresenceAbsence 1.1.9 (Freeman & Moisen, 2008)in R.
Results
During a total of 24,311 trap-days in the 132 locations (20,771 within and 3,540 outside Lenya Reserved Forest), we de- tected 49 mammal and 17 bird species. The total number of detections of prey species were: wild pig, 596 detections in 89 locations (67.4% naïve occupancy); muntjac, 628 de- tections in 93 locations (70.5%); sambar, 11 detections in nine
locations (6.8%); gaur, 70 detections in 27 locations (20.5%); and banteng, six detections in three locations (2.3%: Table 1). The total number of detections of large carnivore species were: leopard, 89 detections in 23 locations (17.4%); tiger, 54 detections in 20 locations (15.2%); and dhole, 49 de- tections in 26 locations (19.7%; Table 2). Several other glo- bally threatened mammal species were also recorded (Supplementary Table 1). We detected tigers only within Lenya Reserved Forest.
Based on the most parsimonious of the 12 models, the pres- ence of tigers was positively associated with the presence of gaur and with increasing distance to the nearest village, with no correlation between gaur and distance to the nearest vil- lage (r = 0.26; Table 3). We detected tigers over altitudes of 69–662 m, and the modelling indicated that tiger presence was not correlated with either altitude or forest type. The model predicted only a 20% probability of tiger presence where camera traps were close to villages (,20 km; Fig. 2a). Tiger presence was correlated with larger prey spe- cies (i.e. gaur, banteng and sambar) and not medium-sized prey. Detections of large prey were mostly of gaur. The probability of tiger presence increased from 0.2 to 0.5 with a doubling (from two to four detections per location) of gaur detections (Fig. 2b). The AUC value was 0.85, indicating ex- cellent discrimination between tiger detections and non- detections (Table 4). Detection of leopards was positively associated with for-
est area (canopy cover.80%) but not with any of the other covariates (Table 3). We detected leopards in eight of the 23 locations (34.8%) outside the Lenya Reserved Forest. The probability of leopard presence was low (,10%) at camera- trap locations with forest areas of,2.23 km2 within a 1-km radius of a camera trap (Fig. 2c), which comprised 21%of forest patches in the study area. Prey detection did not ex- plain the detection of leopards in the study area. The AUC value of 0.68 could be considered acceptable for discrim- ination between leopard detections and non-detections (Table 4)as 0.7 lies within the range of acceptable AUC va- lues (Hosmer & Lemeshow, 2000). We detected dholes in only three locations outside the Re-
served Forest. Of 13 candidatemodels, dholes were positively associated with wild pig detections (Fig. 2d), with total num- ber of prey detections being the second most plausible model (Table 3). Dhole distribution was not associated with forest type or any of the other landscape covariates. The AUC value of 0.70 was considered acceptable for dis- crimination between dhole detections and non-detections (Table 4).
Discussion
Our findings confirm the importance of the study area for biodiversity, with several globally threatened mammal
Oryx, 2023, 57(2), 262–271 © The Author(s), 2022. Published by Cambridge University Press on behalf of Fauna & Flora International doi:10.1017/S0030605321001654
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
