770 K. Syxaiyakhamthor et al.


these five variables within the 1-km buffer radius of each listening point using ArcGIS 10.1. Because the proportions of evergreen forest andmixed de-


ciduous forest were highly correlated (r =−0.97), we selected onlymixed deciduous formodelling as this forest type is pre-


sent across the entire Protected Area (Fig. 3a). To estimate le- vels of illegal hunting in the area, we recorded the locations of signs such as direct sightings of poachers, camps and snares, with aGPS. Prior to 2014, ranger patrols were recorded using a database in which details were difficult to access. Ranger pa- trolling started to improve during 2014–2016, but effectiveness was still low. Since 2016, patrols have been conducted from eight ranger stations, with six rangers per station and each pa- trol lasting c. 18 days permonth, resulting inamore systematic monitoring (see Eshoo et al., 2018 for details). To estimate pa- trolling effort, we recorded the geographical coordinates of patrolmovements in the area and number of visits by rangers over the study period (2007–2015). We used the number of visits to represent patrolling effort, because patrolling distance was not recorded prior to 2014. To determine hunting pressurewe created a 1 × 1 km grid


of the study area. For each 1 km2 grid cell, we assigned a value for hunting evidence (number of observed signs of illegal hunting in the grid cell area) and patrolling effort (number of ranger visits to the grid cell from 2014 onwards). We then calculated the hunting pressure index by dividing hunting evidence by patrolling effort for each grid cell. At each survey site the effective listening area may cover mul- tiple cells and thereforemultiple hunting pressure index va- lues; thus, we weighted the overall pressure index of one site by summing up the proportional hunting pressure value (i.e. the cell value multiplied by the proportion of the cell’s area that fell inside the effective listening area; Fig. 3b).


Data analysis


We used an N-mixture model to estimate the abundance of gibbon groups from four replicate counts; i.e. four con- secutive survey days at each listening point (Royle, 2004). This hierarchical method accounts for the imperfect de- tection of gibbon groups using auditory surveys through repeated counts (i.e. multiple visits to the same location). N-mixture models facilitate the investigation of the relation- ship between environmental variables and estimated abun- dance λ while accounting for detection probability p (Royle, 2004; Joseph et al., 2009; Fiske & Chandler, 2011). We con- ducted data analysis using function pcount in the R pack- age unmarked (Fiske & Chandler, 2011; R Core Team, 2017). All landscape variables were standardized before fit- ting the models, and autocorrelated variables (r.0.7) were not incorporated within the same model. We first deter- mined which variables affected gibbon detection probability (sampling covariates) by fitting five models with different weather conditions and incorporating global covariates


(Table 1) for abundance (Adams et al., 2010; Harihar & Pandav, 2012; Kamjing et al., 2017). We then used the wea- ther variables of the best detection models together with an ecologically plausible combination of landscape and anthropogenic variables to predict gibbon abundance. We compared the fitted models using Akaike’s information cri- terion (AIC) by considering a list of candidate models with ΔAIC,2 (Akaike, 1973).Weassessed goodness-of-fit of the best fitted model based on Pearson χ2 P-value by using the function Nmix.gof.test in the R package AICcmodavg with 1,000 simulations (Mazerolle, 2016). We calculated mean density (groups/km2) for the entire


study area by using the mean abundance λ from the best model divided by the site effective listening area (3.14 km2; Chandler et al., 2011; Dawrueng et al., 2017). To compare density estimates between the two forest types, we employed the predict function in the R package unmarked by using the maximum value of mixed deciduous forest (3.14 km2) to predict density in the mixed deciduous forest, and the minimum value of mixed deciduous (0 km2, i.e. evergreen forest) to estimate density in evergreen forest. We estimated total gibbon group abundance over the 40 sites by summing the estimated site-specific abundance along with uncertainty of confidence intervals, after 1,000 replicate bootstrap samplings (Chandler et al., 2011).


Results


During our survey we detected gibbon groups at 24 of 40 listening sites. Duet start times ranged from 05.14 to 09.22, with the majority of calls (71%) starting before 06.00, 16% between 06.00 and 07.00, 7% between 07.00 and 08.00, 4% between 08.00 and 09.00, and 2% after 09.00 (n = 128 calls). Mean call length was 11 min (range 2–30 min). The best fitting detection model incorporated global


abundance covariates and sun as a function of detection (λ(global)p(sun)), with a lowest AIC score of 213.93 and an Akaike weight of 0.63 (the second best fitting model had an AIC score of 216.99; Table 2). The bestmodel, with β = 1.08, suggested that detection probability was positively related to sunny weather. Amongst the landscape variables assessed, area of mixed deciduous forest had the strongest effect on the variability of gibbon group abundance (Fig. 2). Although three of the models had a ΔAIC,2, the model using deciduous only had more support than models with a higher number of variables. Because these variables did not provide any effect in addition to forest type, λ(deciduous)p(sun) was selected as the best fitted model, with abundance positively asso- ciated with area of mixed deciduous forest (log(λ) = 0.23 + 0.56*deciduous) and detection probability (logit(p)


= −1.63 + 1.03*sun). The goodness-of-fit test lends credibil- ity to the selected model with a Pearson χ2 Pof 0.55.


Oryx, 2020, 54(6), 767–775 © 2019 Fauna & Flora International doi:10.1017/S0030605318001515


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  |  Page 141  |  Page 142  |  Page 143  |  Page 144  |  Page 145  |  Page 146  |  Page 147  |  Page 148  |  Page 149  |  Page 150  |  Page 151  |  Page 152  |  Page 153  |  Page 154  |  Page 155  |  Page 156  |  Page 157  |  Page 158  |  Page 159  |  Page 160  |  Page 161  |  Page 162  |  Page 163  |  Page 164