Multiscale spatially explicit modelling of livestock depredation by reintroduced tiger (Panthera tigris) to predict conflict risk probability

Multiscale spatially explicit modelling of livestock depredation by reintroduced tiger (Panthera tigris) to predict conflict risk probability | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Multiscale spatially explicit modelling of livestock depredation by reintroduced tiger (Panthera tigris) to predict conflict risk probability Manjari Malviya, Ramesh Krishnamurthy This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-1754621/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Context: Spatial modelling of human-carnivore conflict has recently gained traction, and predictive maps have become a great tool to understand the distribution of present and future conflict risk. However, very few such studies consider scale and use appropriate spatial modelling tools. Objectives: We aimed to understand the ecological predictors of human-tiger ( Panthera tigris ) conflict and predict livestock predation risk by reintroduced tigers in Panna Tiger Reserve, Central India. By modelling livestock kill as a function of various tiger relevant ecological variables at multiple scales employing spatially explicit statistical tools. Methods: We used geostatistical modelling to create raster layers of covariates (prey, cover, human activities), following which we did univariate scaling. We then modelled livestock loss by tiger using spatial Generalized Additive Model (geoGAM), predicted and mapped conflict risk probability. Results: We found that prey and shrub cover, both selected at a fine scale, were key ecological determinants of human-tiger conflict. Prey showed an inverse relationship with livestock predation and shrub nonlinear; livestock predation increasing with an increase in shrub cover but decreasing beyond a certain point. Thus, in habitats where optimum ambush cover is available but prey presence is low at fine-scale, carnivores are more likely to depredate domestic livestock since livestock have lost most of their anti-predator behaviours. Conclusions: Livestock kill by tiger is a culmination of predator choice and foraging tactics, and prey vulnerability and defence mechanism. The spatially explicit predation risk map produced in this study can guide adequate human-tiger conflict prevention measures. human-carnivore conflict ecological predictors prey cover domestic livestock Panna Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Human-wildlife conflict, especially by large carnivores, is among the key drivers of local extinction of several species and is also a major cause for local communities turning hostile toward the conservation agenda (Young and Goldman 1944 ; Seidensticker 1987 ; Clark et al. 2013 ; Babrgir et al. 2017 ; Gross et al. 2021 ). Revealing the ecological reasoning behind large carnivore attacks on domestic livestock is key to designing effective prevention and mitigation strategies. Studies have linked several ecological factors with human-carnivore conflict (HCC) viz. tree cover, forest area and forest/vegetation types (Amirkhiz et al. 2018 ; Zarco-González et al. 2018), distance to forest/protected area/reserve (Treves et al. 2011 ; Broekhuis et al. 2017 ), wild prey abundance and availability (Cavalcanti et al. 2010 ; Davie et al. 2014 ), proximity to predator occupied habitat, predator and cattle density (Silveira et al. 2008 ; Kissling et al. 2009 ; Kaartinen et al. 2009 ), proximity to water (Behdarvand et al. 2014 ; Abade et al 2014 ), distance to settlement and road (Mbiba et al. 2018 ; Amirkhiz et al. 2018 ), temperature and precipitation (Dar et al. 2009 ), topography/terrain, elevation and slope (Naha et al. 2018 ; Chetri et al. 2019 ), season (Mbiba et al. 2018 ), time (Yirga et al. 2012 ; Mazzolli et al. 2002 ), and, predator personality, sex, social status and pack size (Odden et al. 2002 ; Mattisson et al. 2011 ). In order to reveal these ecological predictors and understand the distribution of HCC, spatial modelling of conflict or predictive risk modelling has become one of the important tools (Treves et al. 2004 ; Kissling et al. 2009 ; Marucco and McIntire 2010 ; Edge et al. 2011 ; Zarco-González et al. 2013 ; Mbiba et al. 2018 ). Statistical modelling is used to identify the factors related to depredation events, predict its distribution by extrapolating to similar areas, and predict future conflict risk (Treves et al. 2004 ; Kaartinen et al. 2009 ; Behdarvand et al. 2014 ; Rostro-García et al. 2016 ). The predictive/risk maps so produced help managers identify vulnerable habitats, communities, and species (Treves et al. 2011 ; Mateo-tomas et al. 2012; Davie et al. 2014 ; Soh et al. 2014 ; Broekhuis et al. 2017 ; Amirkhiz et al. 2018 ). However, since all the factors associated with HCC are neither linearly related to kill occurrence nor come into play at the same scale, scale must be considered when modelling habitat correlates of HCC. Most ecological relationships are complex and involve several factors. And because these factors range from macro to microhabitat/environmental covariates, all of them cannot be expected to operate at and influence the relationship at the same scale. Thus, ecological relationships are scale-dependent, such that when examined at different spatio-temporal scales, the relationship and its interpretation are subject to change (Weins et al. 1989; McGarigal et al. 2016 ). Therefore, when identifying the factors related to a process or phenomenon, they need to be examined at multiple scales to identify the meaningful scale and make ecologically sound inferences (McGarigal et al. 2016 ). In the absence of such a multiscale approach, misleading conclusions may be drawn. Even if we are able to identify the causal factors of a problem, we would not know at which level to intervene without understanding the scale at which these causal factors are influencing the problem. In which case, selection of the scale at which the variables are meaningfully correlated with the issue becomes as crucial as the selection of variables themselves (Mateo Sánchez et al. 2013 ). Thus, coupled with variable selection, scale optimisation should be the first step to predictive modelling. Multiscale models have been proven to perform better than single-scale models at identifying and predicting relationships between environmental variables and the phenomenon/process under study (Mateo Sánchez et al. 2013 ; Timm et al. 2016 ). Thus, multiscale modelling has become an important tool for studying a myriad of ecological and biological processes/problems/groupings, including community ecology (Dray et al. 2012 ), ecological niche modelling/niche/resource partitioning (Hearn et al. 2018 ; Khosravi et al. 2019 ), habitat selection (Mateo Sánchez et al. 2013 )/ habitat suitability modelling (Store and Jokimäki 2003 ; Kittle et al. 2018 ; Khosravi et al. 2019 ; Rather et al. 2020 ), predicting indicator species hotspots (Grand et al. 2004 ) and predicting carnivore dispersal (Krishnamurthy et al. 2016 ). Even though HCC often involves different variables and complex interactions, very few studies have tried to examine the factors determining HCC at multiple scales (Wilson et al. 2005 ; Soh et al. 2014 ; Miller et al. 2015 ; Rostro-García et al. 2016 ; Broekhuis et al. 2017 ). These studies have found that scale influences livestock predation risk (Davie et al. 2004), with certain habitat factors influencing livestock depredation at a broad scale and others at a fine scale (Miller et al. 2015 ; Rostro-García et al. 2016 ; Broekhuis et al. 2017 ). Upon comparison of multiscale model with a single-scale model, studies have concluded that scale optimisation improves modelling results of livestock predation risk by large carnivores like tiger ( Panthera tigris ) (Rostro-García et al. 2016 ). However, most studies employ aspatial models to predict predation risk using spatial correlates (Soh et al. 2013; Miller et al. 2015 ). Since most ecological variables exhibit a certain degree of spatial autocorrelation, it is important to account for the spatial nature of the data (Griffith 1992 ; Legendre 1993 ), when modelling predation risk by carnivores. In the absence of which, ‘independence of data points’, a common assumption across most statistical models, is violated, leading to unreliable model outcomes (Legendre 1993 ; Dale and Fortin 2002 ). Moreover, most studies on HCC attempt to only map risk and discuss the causal factors. They rarely address the ecology (like prey-predator dynamics) behind how these factors interact to cause conflict (Wilkinson et al. 2020 ). As there is a dearth of studies examining HCC at multiple scales employing appropriate spatial statistical models; our study aims to identify the ecological determinants of human-tiger conflict (HTC) at suitable scale, predict livestock predation risk by tigers in and around Panna Tiger Reserve, Central India, and reveal the ecology behind livestock depredation by large carnivores in the light of the identified causal factors. For this purpose, we modelled livestock kill as a function of various tiger relevant ecological variables, viz. prey, cover, water, and anthropogenic disturbance (Miquelle et al. 1999 ; Karanth and Sunquist 2000 ; Sunarto et al. 2012 ), at multiple scales employing spatially explicit Generalized Additive Model (GAM). Various statistical tools have been applied to model the relationship between habitat variables and livestock kill, most of the times as presence vs absence (or classification into kill or no kill) using linear parametric models for e.g. discriminant function analysis (Edge et al. 2011 ; Treves et al. 2004 ), binary logistic regression or generalized linear model (GLM) with logit link function and binomial error distribution (Broekhuis et al. 2017 ; Karanth et al. 2013 ; Kissling et al. 2009 ; Michalski et al. 2006 ; Miller et al. 2015 ; Thorn et al. 2012 ); or when modelling the frequency of occurrence of kills (count data), then negative binomial distribution (Penteriani et al. 2016 ) or if kill events are rare, then zero-inflated negative binomial model (Soh et al. 2014 ) or rare event model in a binary logistic regression (Naha et al. 2018 ). However, classical statistical tools like parametric models have several assumptions relating to data distribution and linearity, even though most relationships are not linear in the real world and most of the data does not have a Gaussian distribution (Chambers and Dinsmore 2014 ; Mahmoud 2021 ). Thus, in recent times, machine learning algorithms are being increasingly used to generate accurate predictions without having to worry about the data distributions, a priori (Kuhn and Johnson 2013 ). Several studies have employed machine learning algorithms to model conflict/predation risk (Abade et al. 2014 ; Amirkhiz et al. 2018 ; Mbiba et al. 2018 ; Rostro-García et al., 2016 ). Although the machine learning algorithms may perform better than classical statistical models when it comes to giving more accurate predictions, if the purpose is to draw inferences about the relationship between variables, they are not very interpretable (Stewart 2019 ). GAM, while retaining the interpretability of GLM has the flexibility of machine learning algorithms, because it does not assume a linear relationship between dependent and independent variables (Hastie and Tibshirani 1990 ; Larsen 2015 ). GAM is, as the name suggests, a generalisation of the linear model, in which the linear function of the covariate is replaced with a smooth function (Hastie and Tibshirani 1990 ). Because of their semiparametric nature, GAMs are much more sensitive to unique data distribution than GLM, allowing for the modelling of nonlinear relationships by deriving predictor functions during model building (Härdle and Turlach 1992 ; Larsen 2015 ). At the same time to avoid overfitting, one can control the smoothness or ‘wiggliness’ of the predictor function (Larsen 2015 ). Despite their versatility, GAM has not been explored as much as linear or machine learning models to understand the relationship between livestock depredation by carnivores and the environmental factors (Kaartinen et al. 2009 ; Miller et al. 2015 ; Rostro-García et al. 2016 ; Broekhuis et al. 2017 ; Struebig et al. 2018 ). Therefore, we have carried out multivariate multiscale predictive modelling to identify the ecological factors linked to livestock depredation by tiger, employing geoGAM, and discussed how these factors might be linked to the behavioural ecology of predator and prey. Methodology Study area Panna Tiger Reserve (24°16ʹN to 24°42ʹN and 79°29ʹE to 80°16ʹE), covering an area of 1598.10 km² is situated in the state of Madhya Pradesh in central India. The Critical Tiger Habitat (CTH), or core of the reserve comprises Panna National Park and Gangau Wildlife Sanctuary, covering an area of 576.13 km 2 (Madhya Pradesh Forest Department 2007). The buffer covers an area of about 1,021.97 km 2 (Madhya Pradesh Forest Department 2012). The reserve lies in Vindhyan hills, its altitude ranging between 330 and 540 m a.s.l. (Chawdhry 1996 ; Rodgers et al. 2002 ). It has an average annual humidity of 86%, and temperature ranges from 5 to 45℃. Both monsoon (July-September) and winter (November-February) are short, thus, the climate is mostly hot and dry (Chawdhry 1996 ; Gopal et al. 2010 ). Ken a major tributary of the river Yamuna, cuts through the reserve, flowing from South to North. The major forest type is dry deciduous forest with teak ( Tectona grandis ) as the most dominant flora (Meher-Homji 1990 ). Apart from tigers, the major faunal community comprises of carnivores, viz. leopard ( Panthera pardus ), striped hyena ( Hyaena hyaena ), wild dog ( Cuon alpinus ), golden jackal ( Canis aureus ), Bengal fox ( Vulpes bengalensis ), jungle cat ( Felis chaus ), and sloth bear ( Melursus ursinus ); herbivores, viz. sambar ( Cervus duvauceli ), chital ( Axis axis ), nilgai ( Boselaphus tragocamelus ), chinkara ( Gazella bennetti ), chousingha or four-horned antelope ( Tetraceros quadricornis ) and wild pig ( Sus scrofa ); and primates, viz. hanuman or common langur ( Semnopithecus entellus ) and rhesus macaque ( Macaca mullata ) (Gopal et al. 2010 ). There are four villages within the national park area, however, there are seven villages in sanctuary area and 49 villages in the buffer of the reserve. Many of the communities living in these villages are dependent on the reserve for fuelwood, fodder and NTFPs (Malviya et al. 2022 ). The current study focussed on the CTH and two kilometres of natural buffer around it (Fig. 1 ). The original tiger population in Panna was lost due to poaching and as a population recovery measure, tigers were reintroduced from neighbouring reserves in 2009 (Sarkar et al. 2016 ). Since then, the tiger population in Panna has increased exponentially from seven founders to more than 60 individuals in 2020 (Sharma and Jarande 2020 ). Livestock kill quantification and kill site location In India, forest department of the state compensates livestock depredation by wild carnivores. We collected livestock compensation records from Panna Tiger Reserve management to understand the intensity of HTC within the reserve. But the compensation data was not geo tagged, hence we also obtained the livestock kill data (that has GPS locations), which is collected by tiger monitoring teams in the reserve, for the period 2009–2016. Since in case of Panna, there are several feral cattle, and no distinction is made between feral and domestic cattle in the kill records, we matched kill and compensation data to obtain reliable locations for domestic livestock kill within the reserve. We thus matched 156 locations for the period 2011–2016. For predation risk probability modelling, we treated livestock kill data as presence and generated equal number of random absence points using ‘create random points’ tool in ArcGIS 10.4 (ESRI 2016a). We used livestock kill data for the period 2011–2015 for training the model and 2016 data for testing the model. Measuring ecological variables We considered a total of 27 variables for ecological driver modelling, which can be grouped under the following heads I. Prey : We used distance sampling technique for tiger prey species population estimation (Buckland et al. 2001 ). We surveyed 41 line transects each up to 2 km in length, during winter 2012-13 and 2013-14. All the line transects were walked in a replicate of three. The prey species recorded were sambar, chital, wild pig, nilgai, chinkara, chousingha, hanuman langur, hare, peacock and livestock (cattle and buffalo). We combined the data for both the years and calculated all prey (wild and livestock) encounter rate, wild prey encounter rate, and livestock encounter rate, for each transect. II. Vegetation cover : We quantified vegetation indices, viz. canopy cover and shrub abundance in 15m circular plots laid at every 400 m on the line transects during winter 2012-13 (Jhala et al. 2009 ). We laid a total of 234 circular plots. Within the plots, we made ocular estimations of canopy cover and scored shrub cover according to its abundance (0–4). The same team of two people carried out all the vegetation related estimations to avoid interobserver bias. For further understanding the vegetation cover of the tiger reserve, we downloaded LANDSAT 8 (OLI/TIRS) scenes for the reserve from USGS website for April 2013 (LANDSAT SCENE ID = LC81440432013119LGN01; Download date = 20 April 2015) and November 2013 (LANDSAT SCENE ID = LC81440432013327LGN01; Download date = 15 December 2020). We calculated Normalized Difference Vegetation Index (NDVI) (Rouse et al. 1974 ) using these scenes, employing Raster Calculator in ArcGIS 10.1 (ESRI, 2012) by the formula: NDVI = (Near Infrared - Red)/ (Near Infrared + Red). III. Water : Using the same scenes as used for calculating NDVI, we calculated Normalized Difference Water Index (NDWI) (McFeeters 1996 ) with Raster Calculator in ArcGIS 10.1 (ESRI 2012), by the formula: NDWI = (Green-Near Infrared)/ (Green + Near Infrared). Additionally, we obtained drainage and water source data from the forest department and created Euclidean distance raster using ‘Euclidean Distance’ tool in the Spatial Analyst toolbox in ArcGIS 10.4 (ESRI 2016a). We also created Euclidean distance rasters for Ken River and its tributaries, and water sources tagged perennial. IV. Topography : We downloaded ASTER Global Digital Elevation Model (DEM) data from USGS Global Visualization Viewer website. We used ‘Slope’ tool in the Spatial Analyst toolbox to calculate slope from DEM layers in ArcGIS 10.1 (ESRI 2012). Additionally, we calculated topographic ruggedness index or terrain ruggedness index (TRI) that measures elevation difference between a cell and mean of its eight neighbouring cells (Riley et al. 1999 ) using raster calculator in ArcGIS 10.1 (ESRI 2012), by the formula (Cooley 2016 ): TRI = SquareRoot (Abs((Square(“3x3max”)-Square (“3x3min”)))). V. Land Use Land Cover and forest contiguity : We procured Land Use Land Cover (LULC) prepared by Forest Survey of India (FSI) for the entire country at 98m resolution for the year 2009 (FSI 2009 ). We studied landscape characteristics and patterns, particularly habitat connectivity, using multiple indices in program FRAGSTATS (ver. 4.2). We ran FRAGSTAT analysis using the FSI LULC and calculated three class-level metrics: Patch Density (PD), Large Patch Index (LPI) (percentage of total landscape area comprised by the largest patch), and Clumpiness Index (CLUMPY) (a measure of fragmentation). LPI is a simple measure of dominance. And CLUMPY that ranges from − 1 (patch type is maximally disaggregated) to 1 (patch type is maximally clumped) provides an index of fragmentation of the focal class that is not affected by changes in class area (McGarigal 2015 ). VI. Disturbance : We also quantified anthropogenic disturbance indices in the circular plots (as discussed earlier under section II). In each plot, we counted all the lopped (only branches were cut) and cut (cut to stump) trees. We deployed camera traps (Cuddeback Attack pairs) in 109 locations in 2x2 km grids within the national park area, accounting for 7459 trap nights, in the winter of 2013-14. We then manually counted the number of tigers, livestock, humans, and vehicles captured in each camera trap and calculated encounter rates (total no. of captures/total trap nights). We also obtained village and road location data from the forest department and calculated Euclidean distance rasters. Additionally, we downloaded human footprint data from Socio Economic Data and Application Centre (SEDAC) website for the year 2009 (Sanderson et al. 2002 ; Venter et al. 2018 ). We also downloaded population census data for the year 2011 from SEDAC (Balk et al. 2020 ). Geostatistical modelling to create rasters : We interpolated canopy cover, prey, human, livestock, and vehicle encounter rates, and cutting and lopping intensity rates, to create rasters using the Geostatistical wizard in ArcGIS 10.4 (Cressie 2015 ; ESRI 2016a). For this purpose, we considered four interpolation tools: Inverse Distance Weighing, Simple Kriging (SK), Ordinary Kriging (OK), and Empirical Bayesian Kriging (EBK). We used statistical measures of correctness (mean prediction error, root-mean-square error, standardized root-mean-square error, average standard error) to compare the kriging algorithms. We selected the model that had the smallest root-mean-squared prediction error (RMSE), standardized mean nearest to zero, the average standard error nearest the root-mean-squared prediction error, and the standardized root-mean-squared prediction error nearest to 1 (ESRI 2016b) (Supplementary Table 1). Scale and variable selection : To construct a multiscale model, we resampled each of the variables (except for human footprint (which was available at ~ 1 km resolution)) at five scales: 30m (the highest resolution available), 50m (mean drag distance for tiger kill (Karanth and Sunquist 2000 )), 100m (midpoint between fine and coarse resolution), 350m (maximum kill drag distance (Karanth and Sunquist 2000 )), 1200m (coarsest resolution used by us and at which most global environmental data is available). LULC (available at 98m) could only be resampled at coarse scale (100-1200m). Thus, in total, we had 118 variables. We then extracted all the variables for each of the presence/absence points using the Spatial Analyst toolbox of ArcGIS 10.4 (ESRI 2016a). For scale and feature selection, firstly, we ran univariate logistic regressions, after performing Box-Tidwell procedure to test for logistic regression’s linearity assumption, i.e., logit transformation of the dependent variable and continuous independent variables have a linear relationship (Shin and Ying 1994 ). Secondly, we ran univariate GAMs to understand how much r square/deviance was explained by each of the predictors at each of the scales. We selected the scales at which the variables were best explaining the response (livestock kill presence/absence). Additionally, we employed Information Value and Weight of evidence for feature selection (Good and Osteyee 1974 ). We studied the results of univariate logistic regression, univariate GAM, and Information value to select the explanatory variables at appropriate scales. We then checked the data for multicollinearity and spatial autocorrelation. To check for multicollinearity among the selected variables, we ran appropriate tests of association (for continuous vs continuous and continuous vs ordinal variables, Kendall’s tau b; for categorical vs continuous variables, logistic regression; and for categorical vs categorical variables, Cramer’s V). Among the correlated variables, we included those variables in the model that better explained the response. For example, all prey (wild prey plus livestock) was highly correlated to wild prey encounter rate (r = 0.812). Between the two, we selected all prey encounter rate since it was explaining higher deviance of the dependent variable. Similarly, NDVI and NDWI were moderately correlated (r = 0.636); we selected NDVI since it was explaining higher deviance. So, after variable selection, the global model consisted of the following 12 variables at these scales: All prey encounter rate (50m), slope (50m), elevation (1200m), slope deviation (100m), NDVI (100m), shrub abundance (50m), canopy cover (30m), distance to water (1200m), livestock encounter rate (ct) (50m), human encounter rate (1200m), distance to road (100m), distance to village (100m) (Fig. 2 ). We checked spatial autocorrelation among these variables using Moran’s I statistic, and it was found that many variables were spatially autocorrelated (Getis 2007 ; Moran 1950 ). Spatial GAM The Box-Tidwell test indicated that many of the variables did not meet the linearity assumption of logistic regression. Therefore, we could not use linear models for our predictive modelling. And since our aim was not to merely get accurate predictions but to identify the drivers of HTC, we did not use machine learning algorithms. Therefore, as discussed within the introduction, we selected GAM to model livestock kill locations as a function of various tiger relevant ecological factors (Wood 2017 ). To account for the spatial autocorrelation in the data, we constructed spatial GAM model using geoGAM package in R ver. 3.6.3. (Nussbaum and Papritz 2017 ). It’s a procedure to build a parsimonious model based on gradient boosting, smoothing splines and a smooth spatial surface to account for the spatial structure. The GAM for spatial data or geoadditive model in its full generality is represented by $$\text{g}\left({\mu }\left(\text{x}\left(\text{s}\right)\right)\right)= {\nu } + \text{f}\left(\text{x}\left(\text{s}\right)\right)=$$ $${\nu } +{\sum }_{u}{f}_{{j}_{u}}\left({x}_{{j}_{u}}\right(\text{s}\left)\right)+ {\sum }_{v}{f}_{{j}_{v}}\left({x}_{{j}_{v}}\right(\text{s}\left)\right) . {f}_{{k}_{v}}\left({x}_{{k}_{v}}\right(\text{s}\left)\right)$$ $$+{\sum }_{w}{f}_{{s}_{w}}\left(\text{s}\right). {f}_{{j}_{w}}\left({x}_{{j}_{w}}\right(\text{s}\left)\right)+ {f}_{s}\left(\text{s}\right)$$ Where, \({f}_{s}\left(\text{s}\right)\) is a smooth function of spatial coordinates, which accounts for residual autocorrelation (Nussbaum and Papritz 2017 ). Since the response variable, in this case, is binary, Bernoulli distribution is assumed, and logit link used $$\text{g}\left({\mu }\left(\text{x}\left(\text{s}\right)\right)\right)= \text{log}\left(\frac{{\mu }\left(\text{x}\left(\text{s}\right)\right)}{1-{\mu }\left(\text{x}\left(\text{s}\right)\right)}\right)$$ where, $${\mu }\left(\text{x}\left(\text{s}\right)\right)= \text{P}\text{r}\text{o}\text{b}\left[\text{Y}\left(\text{s}\right)= 1\right| \text{x}\left(\text{s}\right)]= \frac{\text{e}\text{x}\text{p}({\nu } + \text{f}(\text{x}\left(\text{s}\right)\left)\right)}{1+\text{e}\text{x}\text{p}({\nu } + \text{f}(\text{x}\left(\text{s}\right)\left)\right)}$$ For building parsimonious model geoGAM automatically selects factors, covariates and spatial effects using componentwise gradient boosting, following which model is further reduced using cross validation (Nussbaum and Papritz 2017 ). We ran the final model so selected on test data, to get model performance measures, area under the curve (AUC) and true skill statistic. We performed all the analyses using R Statistical Software (v3.6.3; R Core Team, 2020 ). Risk map We created a raster with 42m cell size (average calculated from mean kill drag distance for tiger as reported by literature (Karanth and Sunquist 2000 ; Miller et al. 2015 )) and masked it to the reserve boundary. We then converted it into points and, for each of these points, extracted the values for all the explanatory variables. We then ran the final selected model on this data to predict predation risk probability for each point. We then converted the points back to raster. Finally, we assigned risk predictions as the value of the raster to create HTC risk map. Results The variables selected to be included in the final geoGAM model were all prey encounter rate (50m), elevation (1200m), NDVI (100m), shrub abundance (50m), and human encounter rate (1200m). Among these smooth terms, all prey encounter rate and shrub abundance were significant at α = 0.05 level, and elevation was significant at α = 0.1 level (Table 1 ). Table 1 Approximate significance of smooth terms of final geoGAM model predicting livestock predation by tiger Smooth terms Effective degree of freedom (edf) Ref. df Chi. sq p-value s(lt_all_50) (prey encounter rate at 50 m) 3.17 3.85 16.65 0.002 s(sa_ebk_50) (shrub abundance at 50 m) 3.23 3.97 10.95 0.025 s(hum_1200) (human encounter rate at 1200 m) 3.26 4.02 6.46 0.170 s(ndvi_100) (NDVI at 100 m) 3.18 3.95 5.72 0.190 s(dem_1200) (elevation at 1200m) 3.41 4.13 9.37 0.089 The effective degree of freedom (edf) is higher than 3 for most of the smooth terms, indicating that the wiggliness is high and relationships nonlinear (Table 1 ). Even more is revealed by examining the partial effect plots of smooth terms, also called rug plots. A partial effect plot shows the effect of an explanatory variable on the response variable after accounting for the effects of all the other variables included in the model. Upon examining the partial effect plot for all prey encounter rate, we found that it has an inverse relationship with log odds of livestock kill i.e., the odds of livestock kill by tiger are higher when prey is low (Figure 3a). In case of shrub abundance, we observed a unique trend, log odds of livestock kill increase with shrub abundance but only till it reaches a certain mark, after which increase in shrub abundance seems to reduce the odds of livestock kill (Figure 3b). NDVI, human encounter rate and elevation, as also indicated by their chi-square p values, do not seem to have a significant relationship with the odds of livestock kill (Figure 3c, d, e). The deviance explained by the model was 44.4%, and AUC of the model was 0.91. When run on the test dataset, the model accuracy was calculated to be 0.65 (Table 2 ), and AUC was found to be 0.70, indicating that the model had fair amount of prediction capability. Table 2 True skill statistic of final geoGAM model predicting livestock predation by tiger Accuracy 0.65 95% CI 0.52–0.77 Kappa 0.30 Sensitivity 0.61 Specificity 0.69 Positive Predicted Value 0.66 Negative Predicted Value 0.65 Prevalence 0.49 Detection Rate 0.30 Detection Prevalence 0.46 Balanced Accuracy 0.65 Discussion Spatial modelling of HCC has enabled conservationists to visualise where the risk of conflict is high and requires mitigation (Kaartinen et al. 2009 ; Treves et al. 2011 ; Zarco-González et al. 2013 ; Amirkhiz et al. 2018 ; Broekhuis et al. 2017 ). Potential habitat/environmental factors identified as drivers of conflict risk can help reduce conflict potential, and design targeted mitigation measures (Behdarvand et al. 2014 ). However, spatial modelling should consider scale/resolution of the data and spatial autocorrelation. In the absence of which model results can be unreliable leading to inaccurate identification of HCC drivers and the resultant conflict risk. Habitat factors that structure the carnivore use of an area are likely to dictate livestock kill by the carnivore and, thereby, HCC. Preferred habitat parameters for tiger have been identified mainly as high prey density, forest contiguity, thick understory, proximity to water, and low human disturbance (Miquelle et al. 1999 ; Karanth and Sunquist 2000 ; Sunarto et al. 2012 ). Among these, past studies have linked HTC with tree cover, elevation/altitude, slope, aspect, proximity to reserve forest, proximity to water, distance to village, distance to road, and density of livestock, settlements, and roads (Li et al. 2009 ; Ahmed et al. 2012 ; Soh et al. 2014 ; Miller et al. 2015 ; Rostro-García et al. 2016 ; Struebig et al. 2018 ; Ramesh et al. 2020 ). Our spatial modelling revealed that the potential ecological drivers of livestock depredation by tigers in Panna Tiger Reserve were prey, and shrub, at a fine scale, i.e., 50m which is the mean drag distance of kill by tigers in tropical landscapes (Karanth and Sunquist 2000 ). Miller et al. ( 2015 ), while studying livestock predation risk by tigers in India, also found that the fine-scale model (20m) performed the best (among the three spatial scales viz. 20m, 100m, and 200m at which they measured vegetation structure). They concluded that fine spatial grain risk models are more accurate in predicting human-carnivore conflict. And although Rostro-García et al. ( 2016 ) while examining livestock depredation by tiger and leopard in Bhutan tested all their variables at five scales, and found that vegetation cover was more influential at a broader scale (2000m), they concluded that scale optimization improves modelling results with multiscale model performing better than single-scale model. Albeit our modelling results also reveal that both the predictors were operating at a fine scale. It should be emphasised that since we tested each variable at multiple scales, our multiscale model is more reliable than the single scale models or models that did not consider scale, employed by past studies on HTC (Li et al. 2009 ; Ahmed et al. 2012 ; Soh et al. 2014 ; Miller et al. 2015 ; Struebig et al. 2018 ; Ramesh et al. 2020 ). Moreover, ours was the only study that accounted for spatial autocorrelation when carrying out spatial ecological modelling of conflict risk. Without which, the earlier studies violated the assumption about the independence of residuals assumed by the statistical techniques employed by them. Our model suggests that when prey encounter is low at fine scale (50m), i.e., tiger encounters less prey, it is more likely to predate upon domestic livestock. Low availability of prey has been linked with livestock depredation by carnivores, including tiger (Fritts et al. 2003 ; Bhattarai and Fischer 2014 ; Burgas et al. 2014 ; Khorozyan et al. 2015 ). Moreover, vulnerability of prey influences predator choice (Greene 1986 ; Onkonburi, and Formanowicz' Jr 1997 ; Provost et al. 2006 ; Cresswell et al. 2010 ). Predators are known to select a kill that is easier to catch (Mueller 1977 ; Lang and Gsödl 2001; Weise et al. 2020 ). Livestock, having lost most of their anti-predator behaviour during the domestication process are vulnerable to becoming easy prey for predators in the absence of human herders (Linnell et al. 1999 ; Laporte et al. 2010 ; Flörcke, and Grandin 2013 ; Weise et al. 2020 ). Thus, in predator-occupied habitats where there is low availability of wild prey if the optimal foraging theory (large prey, high in abundance, easy to catch) is applied (Emlen 1966 ; MacArthur and Pianka 1966 ; Werner and Hall 1974 ), the tiger kills what it can with least effort i.e., livestock. Shrub abundance, the second explanatory variable (also selected at 50m scale), seems to have a unique relationship with livestock kill, resulting in an increase in livestock kill up to a certain point after which increase in shrub abundance decreases the odds of livestock kill. Although it is difficult to explain such a complex relationship, it can be examined in the light of predation technique of tigers. Tiger is an ambush predator therefore, in areas where there is very low cover, it might be very difficult to make a kill, but in areas where cover is high, the chances of success may improve (Greene 1986 ; Murray et al. 1995 ; Karanth and Sunquist 2000 ; Sunquist 2010 ). Studies on tiger and other carnivores have also found that livestock predation risk was higher in habitats with high shrub density because it provides cover for these predators (Davie et al. 2014 ; Miller et al. 2015 ). However, if the cover is too dense, grazers like livestock are also less likely to venture into such patches because they would be devoid of grasses. Thus, making the relationship curve between livestock kill and shrub cover, bell-shaped. Livestock kill by tiger is thus a culmination of predator choice and foraging tactics, and prey vulnerability and defence mechanism. Therefore, from studying the ecological drivers of HTC in Panna Tiger Reserve, we conclude that in a predator-occupied habitat if prey availability is low at fine scale, domestic livestock availability is high, and ambush cover is available, the odds of a predator depredating livestock become high. The risk map produced using spatial modelling shows that domestic livestock predation risk is higher in the south eastern part of the tiger reserve encompassing Panna Range and parts of Gahrighat Range (Fig. 4 ). Preventative measures like fencing, viz. biofencing or electric/solar fencing (Distefano, 2005 ; Sapkota et al., 2014 ), increased protection through livestock entry point monitoring and patrolling (Pettigrew et al. 2012 ), change in livestock husbandry and dependence, or village resettlement (Treves and Karanth 2003 ), education and awareness (Consorte-McCrea et al., 2017 ), should be focussed on the high-risk areas and villages in the proximity of these areas. Summary And Recommendation Ecological drivers of HCC are complex and scale dependent. With the likelihood of conflict being high in a large carnivore habitat that has low prey encounter and an influx of domestic livestock. In case of Panna, we suggest that mitigation efforts should be focussed on the administrative units flagged as high risk by our study. Furthermore, a detailed study should be conducted to understand the lower availability of wild prey and higher availability of livestock in certain parts of the reserve, based on which prey augmentation should be considered where required and deemed feasible. Statements & Declarations Acknowledgments: We are grateful to the National Tiger Conservation Authority (NTCA), Government of India, and Madhya Pradesh Forest Department for funding and providing requisite permissions. We thank the Dean, Director and colleagues (Dr. M. S. Sarkar, Mr. S. K. Roamin, Mr. Naveen, and Mr. Sunil Kumar) at the Wildlife Institute of India and field staff (Mr. R. Mohammad, Mr. M. Kumar, Mr. A. Kondar, Mr. B. Kondar, Mrs. A. Raikwar, and Mr. D. Singh) for providing support and facilitating field data collection. Funding : This study was funded by National Tiger Conservation Authority (NTCA), India [NTCA Letter No1-3/93-PT(Vol.II) dated 05th March 2012] Competing interests: It is stated that there is no conflict of interest among authors, funding agency, or with any other party. The authors have no relevant financial or non-financial interests to disclose. Authors' contributions: All authors contributed to the study conception and design. Funding for the study was secured by Ramesh Krishnamurthy. Material preparation, data collection, and analysis were performed by Manjari Malviya. The first draft of the manuscript was written by Manjari Malviya and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Availability of data and material: The datasets generated and/or analysed during the current study are available on public data repository figshare and can be accessed using the following link: https://figshare.com/s/a9cc82045fd2960f7c9d . Code availability: (software application or custom code): We used ArcGIS 10.1 & 10.4 and R ver. 3.6.3. All R codes are available from the corresponding author on reasonable request. Ethics approval: (include appropriate approvals or waivers): Not applicable. Consent to participate: (include appropriate statements): Not applicable. References Abade L, Macdonald DW, Dickman AJ (2014) Assessing the relative importance of landscape and husbandry factors in determining large carnivore depredation risk in Tanzania’s Ruaha landscape. Biol Conserv 180: 241-248. https://doi.org/10.1016/j.biocon.2014.10.005 Ahmed RA, Prusty K, Jena J, Dave C., Vihar SC (2012) Prevailing Human Carnivore Conflict in Kanha-Achanakmar Corridor, Central India. J Zool 7(2): 158–164. https://doi.org/10.5829/idosi.wjz.2012.7.2.6335 Amirkhiz RG, Frey JK, Cain JW, Breck SW, Bergman DL (2018) Predicting spatial factors associated with cattle depredations by the Mexican wolf ( Canis lupus baileyi ) with recommendations for depredation risk modeling. Biol Conserv 224: 327-335. https://doi.org/10.1016/j.biocon.2018.06.013 Babrgir S, Farhadinia MS, Moqanaki, EM (2017) Socio-economic consequences of cattle predation by the Endangered Persian leopard Panthera pardus saxicolor in a Caucasian conflict hotspot, northern Iran. Oryx 51(1): 124-130. https://doi.org/10.1017/S0030605315000903 Balk D, Montgomery MR, Engin H, Lin N, Major E, Jones B (2020) Spatial Data from the 2011 India Census. Palisades, NY: NASA Socioeconomic Data and Applications Center (SEDAC). https://doi.org/10.7927/gya1-wp91. Accessed 25 August 2020. Behdarvand N, Kaboli M, Ahmadi M, Nourani E, Mahini AS, Aghbolaghi MA (2014) Spatial risk model and mitigation implications for wolf–human conflict in a highly modified agroecosystem in western Iran. Biol Conserv 177:156-164. https://doi.org/10.1016/j.biocon.2014.06.024 Bhattarai BR, Fischer K (2014) Human–tiger Panthera tigris conflict and its perception in Bardia National Park, Nepal. Oryx 48(4): 522-528. https://doi.org/10.1017/S0030605313000483 Broekhuis F, Cushman SA, Elliot NB (2017) Identification of human–carnivore conflict hotspots to prioritize mitigation efforts. Ecol Evol 7(24): 10630-10639. https://doi.org/10.1002/ece3.3565 Buckland ST, Anderson DR, Burnham KP, Laake JL, Borchers DL, Thomas L (2001) Introduction to Distance sampling: Estimating abundance of biological populations. Oxford University Press, Oxford. Burgas A, Amit R, Lopez BC (2014) Do attacks by jaguars Panthera onca and pumas Puma concolor (Carnivora: Felidae) on livestock correlate with species richness and relative abundance of wild prey?. Rev. Biol. Trop. 62(4):1459-67. Cavalcanti SC, Marchini S, Zimmermann A, Gese EM, Macdonald DW (2010) Jaguars, livestock, and people in Brazil: realities and attitudes behind the conflict. In: Macdonald D, Loveridge A (eds) The biology and conservation of wild felids. Oxford University Press, Oxford, pp 383-402. Chambers M, Dinsmore TW (2014) Advanced analytics methodologies: Driving business value with analytics. Pearson Education, USA. Chawdhry PK (1996) Management Plan for Panna National Park. Madhya Pradesh Forest Department, Government of Madhya Pradesh. Chetri M, Odden M, Devineau O, Wegge P (2019) Patterns of livestock depredation by snow leopards and other large carnivores in the Central Himalayas, Nepal. Glob. Ecol Conserv 17: e00536. https://doi.org/10.1016/j.gecco.2019.e00536 Chundawat RS, Gogate N, Johnsingh AJT (1999) Tigers in Panna: preliminary results from an Indian tropical dry forest. In: Seidensticker J, Christie S, Jackson P (eds) Riding the tiger: tiger conservation in human-dominated landscapes. Cambridge University Press, Cambridge, pp 123-129. Clark T, Rutherford M, Casey D (eds) (2013) Coexisting with large carnivores: Lessons from Greater Yellowstone. Island Press, Washington, DC. Cooley SW (2016) Terrain Roughness – 13 ways. http://www.gis4geomorphology.com/ roughness-topographic-position. Accessed January 2017. Consorte-McCrea A, Nigbur D, Bath A (2017) Implications of teenagers' attitudes toward maned wolf conservation in Brazil. Canid Biology & Conservation 20(5): 16-24. https://www.canids.org/CBC/20/Attitudes_toward_maned_wolf_conservation.pdf Cressie N (2015) Statistics for spatial data. John Wiley and Sons, New Jersey, USA. Cresswell W, Lind J, Quinn JL (2010) Predator‐hunting success and prey vulnerability: Quantifying the spatial scale over which lethal and non‐lethal effects of predation occur. J Anim Ecol 79(3): 556-562. https://doi.org/10.1111/j.1365-2656.2010.01671.x Dale MR, Fortin MJ (2002) Spatial autocorrelation and statistical tests in ecology. Ecoscience 9(2): 162-167. https://doi.org/10.1080/11956860.2002.11682702. Dar NI, Minhas RA, Zaman Q, Linkie M (2009) Predicting the patterns, perceptions and causes of human–carnivore conflict in and around Machiara National Park, Pakistan. Biol Conserv 142 (10): 2076–2082. https://doi.org/10.1016/j.biocon.2009.04.003. Davie HS, Murdoch JD, Lhagvasuren A, Reading RP (2014) Measuring and mapping the influence of landscape factors on livestock predation by wolves in Mongolia. J Arid Environ 103: 85-91. https://doi.org/10.1016/j.jaridenv.2014.01.008 Distefano E (2005) Human-Wildlife Conflict worldwide: collection of case studies, analysis of management strategies and good practices. Food and Agricultural Organization of the United Nations (FAO), Sustainable Agriculture and Rural Development Initiative (SARDI), Rome, Italy. http://www. fao. org/documents. Dray S, Pélissier R, Couteron P, Fortin MJ, Legendre P, Peres-Neto PR, Bellier E, Bivand R, Blanchet FG, De Cáceres M, Dufour AB, Heegaard E, Jombart T, Munoz F, Oksanen J, Thioulouse J, Wagner HH (2012) Community ecology in the age of multivariate multiscale spatial analysis. Ecol Monogr 82(3): 257-275. https://doi.org/10.1890/11-1183.1 Edge JL, Beyer DE, Belant JL, Jordan MJ, Roell BJ (2011) Adapting a predictive spatial model for wolf Canis spp. predation on livestock in the Upper Peninsula, Michigan, USA. Wildlife Biol 17(1): 1-10. https://doi.org/10.2981/10-043 Emlen JM (1966) The role of time and energy in food preference. Am Nat 100(916): 611-617. ESRI (Environmental Systems Resource Institute) (2016a) ArcGIS 10.4. ESRI, Redlands, California. ESRI (Environmental Systems Resource Institute) (2016b) Comparing models. Evaluating interpolation results. ArcGIS Desktop. https://desktop.arcgis.com/en/arcmap/10.4/extensions/geostatistical-analyst/comparing-models.htm ESRI (Environmental Systems Resource Institute) (2012) ArcGIS 10.1. ESRI, Redlands, California. Flörcke C, Grandin T (2013) Loss of anti-predator behaviors in cattle and the increased predation losses by wolves in the Northern Rocky Mountains. Open J Anim Sci 3(3): 248. http://dx.doi.org/10.4236/ojas.2013.33037 Fritts SH, Stephensen RO, Hayes RD, Boitani L (2003) Wolves and humans. In Mech LD, Boitani, L. (eds) Wolves: Behavior, Ecology and Conservation. University of Chicago Press, Illinois, pp 289-316. FSI (2009) Indian State of Forest Report (ISFR) 2009. Forest Survey of India. MoEF, GOI, Dehradun, pp 199. Getis A (2007) Reflections on spatial autocorrelation. Reg Sci Urban Econ 37(4): 491-496. Good IJ, Osteyee DB (1974) Information, weight of evidence. The singularity between probability measures and signal detection (Vol. 376). Springer-Verlag, Berlin. Gopal R, Qureshi Q, Bhardwaj M, Singh RJ, Jhala YV (2010) Evaluating the status of the endangered tiger Panthera tigris and its prey in Panna Tiger Reserve, Madhya Pradesh, India. Oryx 44(3): 383-389. https://doi.org/10.1017/S0030605310000529 Grand J, Buonaccorsi J, Cushman SA, Griffin CR, Neel MC (2004) A multiscale landscape approach to predicting bird and moth rarity hotspots in a threatened pitch pine–scrub oak community. Conserv Biol 18(4): 1063-1077. https://doi.org/10.1111/j.1523-1739.2004.00555.x Greene CH (1986) Patterns of prey selection: implications of predator foraging tactics. Am Nat 128(6): 824-839. Griffith DA (1992) What is spatial autocorrelation? Reflections on the past 25 years of spatial statistics. L'Espace géographique 21(3): 265-280. Gross E, Jayasinghe N, Brooks A, Polet G, Wadhwa R, Hilderink-Koopmans F. (2021) A Future for All: The Need for Human-Wildlife Coexistence. WWF, Gland, Switzerland. https://wwfint.awsassets.panda.org/downloads/a_future_for_all___the_need_for_human_willdife_coexistence.pdf. Accessed July 2021. Härdle WK, Turlach BA (1992) Nonparametric Approaches to Generalized Linear Models. In Fahrmeir L, Francis B, Gilchrist R, Tutz G (eds) Advances in GLIM and Statistical Modelling. Lecture Notes in Statistics, vol 78. Springer, New York, pp 213-225. https://doi.org/10.1007/978-1-4612-2952-0_33 Hastie TJ, Tibshirani RJ (1990) Generalized additive models (Vol. 43). Chapman and Hall/CRC press, London. Hearn AJ, Cushman SA, Ross J, Goossens B, Hunter LT, Macdonald DW (2018) Spatio-temporal ecology of sympatric felids on Borneo. Evidence for resource partitioning?. PloS one 13(7): e0200828. https://doi.org/10.1371/journal.pone.0200828 Jhala YV, Qureshi Q, Gopal R, Amin R (2009) Field Guide: Monitoring tigers, co-predators, prey and their habitats, 3 rd ed. Wildlife Institute of India and National Tiger Conservation Authority, Dehradun. https://projecttiger.nic.in/WriteReadData/PublicationFile/FIELD%20GUIDE%202013.pdf Kaartinen S, Luoto M, Kojola I (2009) Carnivore-livestock conflicts: determinants of wolf ( Canis lupus ) depredation on sheep farms in Finland. Biodivers. Conserv. 18(13): 3503–3517. https://doi.org/10.1007/s10531-009-9657-8. Karanth KU, Sunquist ME (2000) Behavioural correlates of predation by tiger ( Panthera tigris ), leopard ( Panthera pardus ) and dhole ( Cuon alpinus ) in Nagarahole, India. J Zool 250(2): 255-265. Karanth, K. K., Gopalaswamy, A. M., Prasad, P. K., & Dasgupta, S. (2013). Patterns of human–wildlife conflicts and compensation: Insights from Western Ghats protected areas. Biological Conservation, 166, 175-185. Khorozyan I, Ghoddousi A, Soofi M, Waltert M (2015) Big cats kill more livestock when wild prey reaches a minimum threshold. Biodivers Conserv 192: 268-275. https://doi.org/10.1016/j.biocon.2015.09.031 Khosravi R, Hemami MR, Cushman SA (2019) Multi-scale niche modeling of three sympatric felids of conservation importance in central Iran. Landsc Ecol 34(10), 2451-2467. https://doi.org/10.1007/s10980-019-00900-0 Kissling DW, Fernández N, Paruelo JM (2009) Spatial risk assessment of livestock exposure to pumas in Patagonia, Argentina. Ecography 32(5): 807-817. https://doi.org/10.1111/j.1600-0587.2009.05781.x Kittle AM, Watson AC, Cushman SA, Macdonald DW (2018) Forest cover and level of protection influence the island-wide distribution of an apex carnivore and umbrella species, the Sri Lankan leopard ( Panthera pardus kotiya ). Biodivers Conserv 27(1): 235-263. https://doi.org/10.1007/s10531-017-1431-8 Krishnamurthy R, Cushman SA, Sarkar MS, Malviya M, Naveen M, Johnson JA, Sen S (2016) Multi-scale prediction of landscape resistance for tiger dispersal in central India. Landsc Ecol 31(6): 1355-1368. https://doi.org/10.1007/s10980-016-0363-0 Kuhn M, Johnson K (2013) Applied predictive modeling (Vol. 26). Springer, New York. Laporte I, Muhly TB, Pitt JA, Alexander M, Musiani M (2010) Effects of wolves on elk and cattle behaviors: implications for livestock production and wolf conservation. PloS one 5(8): e11954. https://doi.org/10.1371/journal.pone.0011954 Larsen K (2015) GAM: the predictive modeling silver bullet. Multithreaded. Stitch Fix 30: 196-223. Legendre P (1993) Spatial autocorrelation: trouble or new paradigm?. Ecology 74(6): 1659-1673. Li B, Zhang E, Liu Z (2009) Livestock depredation by Amur tigers in Hunchun Nature Reserve, Jilin, China. Acta Theriol 29: 231-238. Linnell JDC, Odden J, Smith ME, Aanes R, Swenson JE (1999) Large Carnivores that kill livestock: do “problem individuals” really exist? Wildlife Soc B 27(3): 698–705. MacArthur RH, Pianka ER (1966) On optimal use of a patchy environment. Am Nat 100(916): 603-609. Madhya Pradesh Forest Department (August 9, 2012) The Gazette of Madhya Pradesh Extraordinary No.15-21-2010-X-2. Government of Madhya Pradesh, Bhopal. Madhya Pradesh Forest Department (December 24, 2007) The Gazette of Madhya Pradesh Extraordinary No. 15-31-2007-X2. Government of Madhya Pradesh, Bhopal. Mahmoud, H. F. (2021). Parametric Versus Semi and Nonparametric Regression Models. International Journal of Statistics and Probability 10(2): 1-90. https://doi.org/10.5539/ijsp.v10n2p90 Malviya M, Kalyanasundaram S, Krishnamurthy R. (2022) Paradox of success-mediated conflicts: analyzing attitudes of local communities towards successfully reintroduced tigers in India. Front Conserv Sci 2:783467. Marucco F, McIntire EJB (2010) Predicting spatio‐temporal recolonization of large carnivore populations and livestock depredation risk: wolves in the Italian Alps. J Appl Ecol 47(4): 789-798. https://doi.org/10.1111/j.1365-2664.2010.01831.x Mateo Sánchez MC, Cushman SA, Saura S (2013) Scale dependence in habitat selection: the case of the endangered brown bear ( Ursus arctos ) in the Cantabrian Range (NW Spain). Int J Geogr Inf Sci 28(8): 1531-1546. https://doi.org/10.1080/13658816.2013.776684 Mattisson J, Odden J, Nilsen EB, Linnell JDC, Persson J, Andrén H (2011) Factors affecting Eurasian lynx kill rates on semi-domestic reindeer in northern Scandinavia: Can ecological research contribute to the development of a fair compensation system? Biol Conserv 144(12): 3009–3017. https://doi.org/10.1016/j.biocon.2011.09.004 Mazzolli M, Graipel ME, Dunstone N (2002) Mountain lion depredation in southern Brazil. Biol Conserv 105(1): 43-51. https://doi.org/10.1016/S0006-3207(01)00178-1 Mbiba M, Muvengwi J, Ndaimani H (2018) Environmental correlates of livestock depredation by spotted hyaenas and livestock herding practices in a semi-arid communal landscape. Afr J Ecol 56(4): 984-992. https://doi.org/10.1111/aje.12529 McFeeters SK (1996) The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features. Int J Remote Sens 17(7): 1425-1432. McGarigal K (2015) FRAGSTATS help. University of Massachusetts, Amherst, MA, USA, 182. McGarigal K., Wan HY, Zeller KA, Timm BC, Cushman SA (2016) Multi-scale habitat selection modeling: a review and outlook. Landsc Ecol 31(6): 1161-1175. https://doi.org/10.1007/s10980-016-0374-x Meher-Homji VM (1990) Vegetation types of India in relation to environmental conditions. In: Daniel JC, Serrao JS (eds) Conservation in developing countries: problems and prospects. Oxford University Press, Bombay, pp 95–110. Michalski F, Boulhosa RLP, Faria A, Peres CA (2006) Human-wildlife conflicts in a fragmented Amazonian forest landscape: determinants of large felid depredation on livestock. Anim Conserv 9(2): 179–188. https://doi.org/10.1111/j.1469-1795.2006.00025.x Miller JR., Jhala YV, Jena J, Schmitz OJ (2015) Landscape‐scale accessibility of livestock to tigers: implications of spatial grain for modeling predation risk to mitigate human–carnivore conflict. Ecol Evol 5(6): 1354-1367. https://doi.org/10.1002/ece3.1440 Miquelle DG, Smirnov EN, Merrill TW, Myslenkov AE, Quigley HB, Hornocker MG, Schleyer B (1999) Hierarchical spatial analysis of Amur tiger relationships to habitat and prey. In: Seidensticker J, Christie S, Jackson P (eds) Riding the Tiger: tiger conservation in human-dominated landscapes. Cambridge University Press, Cambridge, pp 71-99. Moran PA (1950) Notes on continuous stochastic phenomena. Biometrika 37(1/2): 17-23. Mueller HC (1977) Prey selection in the American kestrel: experiments with two species of prey. Am Nat 111(977): 25-29. Murray DL, Boutin S, O'Donoghue M, Nams VO (1995) Hunting behaviour of a sympatric felid and canid in relation to vegetative cover. Anim Behav 50(5): 1203-1210. Naha D, Sathyakumar S, Rawat GS (2018) Understanding drivers of human-leopard conflicts in the Indian Himalayan region: spatio-temporal patterns of conflicts and perception of local communities towards conserving large carnivores. Plos One 13: e0204528. https://doi.org/10.1371/journal.pone.0204528 Nussbaum M, Papritz A (2017) geoGAM: Select Sparse Geoadditive Models for Spatial Prediction. R package version 0.1-2. Odden J, Linnell JD, Moa PF, Herfindal I, Kvam T, Andersen R (2002) Lynx depredation on domestic sheep in Norway. J Wildlife Manage 66(1): 98-105. Onkonburi J, Formanowicz' Jr DR (1997) Prey choice by predators: effect of prey vulnerability. Ethol Ecol Evol 9(1): 19-25. Penteriani V, del Mar Delgado M, Pinchera F, Naves J, Fernández-Gil A, Kojola I, Härkönen S, Norberg H, Frank J, Fedriani JM, Sahlén V, Støen O, Swenson JE, Wabakken P, Pellegrini M, Herrero S, Sahlén V. (2016) Human behaviour can trigger large carnivore attacks in developed countries. Sci Rep-UK 6: 20552. Pettigrew M, Xie Y, Kang A, Rao M, Goodrich J, Liu T, Berger J (2012) Human–carnivore conflict in China: a review of current approaches with recommendations for improved management. Integr Zool 7(2):210-26. https://doi.org/10.1111/j.1749-4877.2012.00303.x Provost C, Lucas E, Coderre D (2006) Prey preference of Hyaliodes vitripennis as an intraguild predator: Active predator choice or passive selection?. Biol Control 37(2): 148-154. https://doi.org/10.1016/j.biocontrol.2006.01.005 R Core Team (2020) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/. Ramesh T, Kalle R, Milda D, Gayathri V, Thanikodi M, Ashish K, Giordano AJ (2020) Patterns of livestock predation risk by large carnivores in India’s Eastern and Western Ghats. Glob. Ecol. Conserv. 24: e01366. https://doi.org/10.1016/j.gecco.2020.e01366 Rather TA, Kumar S, Khan JA (2020) Multi-scale habitat modelling and predicting change in the distribution of tiger and leopard using random forest algorithm. Sci Rep-UK 10(1): 1-19. https://doi.org/10.1038/s41598-020-68167-z Riley SJ, DeGloria SD, Elliot R (1999) A terrain ruggedness index that quantifies topographic heterogeneity. Intermountain. J. Sci. 5(1-4): 23-27. Rodgers WA, HS Panwar, VB Mathur (2002) Wildlife Protected Area Network in India: A Review. Wildlife Institute of India publication, Dehradun, India, 44pp. Rostro-García S, Tharchen L, Abade L, Astaras C, Cushman SA, Macdonald DW (2016) Scale dependence of felid predation risk: identifying predictors of livestock kills by tiger and leopard in Bhutan. Landsc. Ecol. 31(6): 1277-1298. https://doi.org/10.1007/s10980-015-0335-9 Rouse J Jr, Haas RH, Schell JA, Deering DW (1974) Monitoring vegetation systems in the Great Plains with ERTS. NASA Spec. Publ. 351: 309. Sanderson EW, Jaiteh M, Levy MA, Redford KH, Wannebo AV, Woolmer G (2002) The Human Footprint and the Last of the Wild: The human footprint is a global map of human influence on the land surface, which suggests that human beings are stewards of nature, whether we like it or not. BioScience 52(10): 891–904. https://doi.org/10.1641/0006-3568(2002)052[0891:THFATL]2.0.CO;2 Sapkota S, Aryal A, Baral SR, Hayward MW, Raubenheimer D (2014) Economic analysis of electric fencing for mitigating human-wildlife conflict in Nepal. J. Resour. Ecol. 5(3): 237-243. https://doi.org/10.5814/j.issn.1674-764x.2014.03.006 Sarkar MS, Ramesh K, Johnson JA, Sen S, Nigam P, Gupta SK, Murthy RS, Saha GK (2016) Movement and home range characteristics of reintroduced tiger ( Panthera tigris ) population in Panna Tiger Reserve, Central India. Eur J Wildlife Res 62(5): 537-547. https://doi.org/10.1007/s10344-016-1026-9 Sharma UK, Jarande IR (2020) A brief note on Tiger population dynamics and its future projection in Panna Tiger Reserve. https://www.pannatigerreserve.in/research.htm. Accessed 21 April 2022. Seidensticker J (1987) Bearing witness: Observations on the extinction of Panthera tigris balica and Panthera tigris sondaica . In: Tilson R, Nyhus PJ (eds) Tigers of the world: The Biology, Biopolitics, Management, and Conservation of an Endangered Species. Noyes Publications, New York, pp 416-426. Shin RT, Ying JS (1994) Imposing linear homogeneity on Box–Tidwell flexible functional forms. J Bus Econ Stat 12(2): 261-265. Silveira L, Boulhosa R, Astete S, Jácomo ATA (2008) Management of domestic livestock predation by jaguars in Brazil. Cat News 4: 26-30. Soh YH, Carrasco LR, Miquelle DG, Jiang J, Yang J, Stokes EJ, Tang J, Kang A, Liu P, Rao M (2014) Spatial correlates of livestock depredation by Amur tigers in Hunchun, China: relevance of prey density and implications for protected area management. Biol Conserv 169: 117-127. https://doi.org/10.1016/j.biocon.2013.10.011 Stewart M (2019) The Actual Difference Between Statistics and Machine Learning. Towards data science. https://towardsdatascience.com/the-actual-difference-between-statistics-and-machine-learning-64b49f07ea3 Store R, Jokimäki J (2003) A GIS-based multi-scale approach to habitat suitability modeling. Ecol Model 169(1): 1-15. https://doi.org/10.1016/S0304-3800(03)00203-5 Struebig MJ, Linkie M, Deere NJ, Martyr DJ, Millyanawati B, Faulkner SC, Le Comber SC, Mangunjaya FM, Leader-Williams N, McKay JE, John FAS (2018) Addressing human-tiger conflict using socio-ecological information on tolerance and risk. Nat Commun 9(1): 3455. https://doi.org/10.1038/s41467-018-05983-y Sunarto S, Kelly MJ, Parakkasi K, Klenzendorf S, Septayuda E, Kurniawan H (2012) Tigers need cover: Multi-scale occupancy study of the big cat in Sumatran forest and plantation landscapes. Plos One 7 (1): e30859. https://doi.org/10.1371/journal.pone.0030859 Sunquist M (2010) What is a tiger? Ecology and behavior. In: Tilson R, Nyhus PJ (eds) Tigers of the world: The Science, Politics, and Conservation of Panthera tigris . Academic Press, London, pp 19-33. Thorn M., Green, M., Dalerum, F., Bateman, P. W., & Scott, D. M. (2012). What drives human–carnivore conflict in the North West Province of South Africa? Biological Conservation, 150(1), 23-32. Timm BC, McGarigal K, Cushman SA, Ganey JL (2016) Multi-scale Mexican spotted owl (Strix occidentalis lucida) nest/roost habitat selection in Arizona and a comparison with single-scale modeling results. Landsc Ecol 31(6): 1209-1225. https://doi.org/10.1007/s10980-016-0371-0 Treves A, Karanth KU (2003) Human‐carnivore conflict and perspectives on carnivore management worldwide. Conserv Biol 17(6):1491-9. https://doi.org/10.1111/j.1523-1739.2003.00059.x Treves A, Martin KA, Wydeven AP, Wiedenhoeft JE (2011) Forecasting environmental hazards and the application of risk maps to predator attacks on livestock. BioScience 61(6): 451-458. https://doi.org/10.1525/bio.2011.61.6.7 Treves A, Naughton‐treves L., Harper, E. K., Mladenoff, D. J., Rose, R. A., Sickley, T. A., and Wydeven, A. P. (2004) Predicting human‐carnivore conflict: a spatial model derived from 25 years of data on wolf predation on livestock. Conserv Biol 18(1), 114-125. https://doi.org/10.1111/j.1523-1739.2004.00189.x Venter O, Sanderson EW, Magrach A, Allan JR, Beher J, Jones KR, Watson JE (2018) Last of the wild project, version 3 (LWP-3): 2009 human footprint, 2018 release. NASA Socioeconomic Data and Applications Center (SEDAC), 10, H46T40JQ44, Palisades, NY. Wiens JA (1989) Spatial scaling in ecology. Functional ecology 3(4):385-397. Weise FJ, Tomeletso M, Stein AB, Somers MJ, Hayward MW (2020) Lions Panthera leo prefer killing certain cattle Bos taurus types. Animals 10(4): 692. https://doi.org/10.3390/ani10040692 Werner EE, Hall DJ (1974) Optimal foraging and the size selection of prey by the bluegill sunfish ( Lepomis macrochirus ). Ecology 55(5): 1042-1052. Wilkinson CE, McInturff A, Miller JR, Yovovich V, Gaynor KM, Calhoun K, Karandikar H, Martin JV, Parker‐Shames P, Shawler A, Van Scoyoc A (2020) An ecological framework for contextualizing carnivore–livestock conflict. Conserv Biol 34(4):854-67. Wilson SM, Madel MJ, Mattson DJ, Graham JM, Burchfield JA, Belsky JM (2005) Natural landscape features, human-related attractants, and conflict hotspots: a spatial analysis of human-grizzly bear conflicts. Ursus 16(1): 117-129. https://doi.org/10.2192/1537-6176(2005)016[0117:NLFHAA]2.0.CO;2 Wood SN (2017) Generalized additive models: an introduction with R. CRC press, Florida. Yirga G, De longh HH, Leirs H, Gebrehiwot K, Berhe G, Asmelash T, Gabrehiwot B, Bauer H (2012) The ecology of large carnivores in the highlands of northern Ethiopia. Afr J Ecol 51: 78–86. https://doi.org/10.1111/aje.12008 Young SP, Goldman EA (1944) The wolves of North America. Dover, New York. Zarco-González MM, Monroy-Vilchis O, Alaníz J (2013) Spatial model of livestock predation by jaguar and puma in Mexico: conservation planning. Biol Conserv 159: 80-87. https://doi.org/10.1016/j.biocon.2012.11.007 Additional Declarations No competing interests reported. Supplementary Files ESM1.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-1754621","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":113695401,"identity":"45d58cd1-f4f5-41da-bdf2-dfe324971154","order_by":0,"name":"Manjari Malviya","email":"","orcid":"","institution":"Wildlife Institute of India","correspondingAuthor":false,"prefix":"","firstName":"Manjari","middleName":"","lastName":"Malviya","suffix":""},{"id":113695402,"identity":"d81ed35a-04f2-491b-b192-1b0870daa60a","order_by":1,"name":"Ramesh Krishnamurthy","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAuElEQVRIiWNgGAWjYLACxgYbBjZmCJuHWC1ppGs5TIKbzNtPJz78ueN8Yh87A/OLj20MMuaEtMicyd1szHvmdmIbMwOb5cw2Bh7LBgJaJBhyt0kztt3OBWkx5jnDwGNwgJAW/rfbf/5sO0eKFoncbQy8bQdAWpgf81QQpeXtZmnetuT6NmbGNsYZFRLEOCx348efbXbG8v2HD3/4YGBjT1ALEmBskwAFB0mA+QNp6kfBKBgFo2CkAAA4WTjq/Aq64AAAAABJRU5ErkJggg==","orcid":"","institution":"Wildlife Institute of India","correspondingAuthor":true,"prefix":"","firstName":"Ramesh","middleName":"","lastName":"Krishnamurthy","suffix":""}],"badges":[],"createdAt":"2022-06-13 18:59:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-1754621/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-1754621/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":22791803,"identity":"84ebbe8e-3510-4c93-af2b-5533cf3b011f","added_by":"auto","created_at":"2022-06-17 20:03:57","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":385555,"visible":true,"origin":"","legend":"\u003cp\u003eMap depicting the study site i.e., Core area of Panna Tiger Reserve and 2 km natural buffer around it\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-1754621/v1/b2a417636054f716c32b0e06.png"},{"id":22791805,"identity":"21403ed2-b6f0-416f-9b20-fbf32027746f","added_by":"auto","created_at":"2022-06-17 20:03:57","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":303997,"visible":true,"origin":"","legend":"\u003cp\u003eInterpolated layers of variables included in the global model for livestock depredation by tiger, at selected scales: a) All prey encounter rate (50m), b) shrub abundance (50m), c) canopy cover (30m), d) slope (50m), e) elevation (1200m), f) slope deviation (100m), g) NDVI (100m), h) distance to water hole (1200m), i) distance to road (100m), j) distance to village (100m), k) human encounter rate (1200m), l) livestock encounter rate (ct) (50m)\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-1754621/v1/258c1d50230ce6b32660abe7.png"},{"id":22791804,"identity":"1b212ec3-44df-489a-8e7e-ae65c658728b","added_by":"auto","created_at":"2022-06-17 20:03:57","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":44590,"visible":true,"origin":"","legend":"\u003cp\u003ePartial effect plots for livestock kill by tiger in Panna Tiger Reserve: a) all prey encounter rate (50m), b) shrub abundance (50m), c) human encounter rate (1200m), d) NDVI (100m), e) elevation (1200m). The dashed line represents 95% confidence interval, and the lines on x axis represent the frequency of data.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-1754621/v1/e6fa337fc1d86dca0be81c63.png"},{"id":22791807,"identity":"69ce3358-1849-4131-bbd9-1f1c9b764fbf","added_by":"auto","created_at":"2022-06-17 20:03:57","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":162952,"visible":true,"origin":"","legend":"\u003cp\u003eRisk map depicting probability of livestock kill by tiger in Panna Tiger Reserve predicted employing geoGAM model\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-1754621/v1/781743413447265e8e029d04.png"},{"id":24537039,"identity":"0f9b0b30-2101-4f75-befb-1ef8d6d37d7d","added_by":"auto","created_at":"2022-07-30 00:14:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1234688,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-1754621/v1/0138e3e6-0b0e-4422-a10e-4129e6cf610d.pdf"},{"id":22792095,"identity":"f4e49de3-0303-4caf-b0aa-0569070dda1f","added_by":"auto","created_at":"2022-06-17 20:08:57","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":15287,"visible":true,"origin":"","legend":"","description":"","filename":"ESM1.docx","url":"https://assets-eu.researchsquare.com/files/rs-1754621/v1/77aa56d7ab28cfc8c56c96bb.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Multiscale spatially explicit modelling of livestock depredation by reintroduced tiger (Panthera tigris) to predict conflict risk probability","fulltext":[{"header":"Introduction","content":"\u003cp\u003eHuman-wildlife conflict, especially by large carnivores, is among the key drivers of local extinction of several species and is also a major cause for local communities turning hostile toward the conservation agenda (Young and Goldman \u003cspan citationid=\"CR119\" class=\"CitationRef\"\u003e1944\u003c/span\u003e; Seidensticker \u003cspan citationid=\"CR97\" class=\"CitationRef\"\u003e1987\u003c/span\u003e; Clark et al. \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Babrgir et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Gross et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Revealing the ecological reasoning behind large carnivore attacks on domestic livestock is key to designing effective prevention and mitigation strategies. Studies have linked several ecological factors with human-carnivore conflict (HCC) viz. tree cover, forest area and forest/vegetation types (Amirkhiz et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Zarco-Gonz\u0026aacute;lez et al. 2018), distance to forest/protected area/reserve (Treves et al. \u003cspan citationid=\"CR109\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Broekhuis et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), wild prey abundance and availability (Cavalcanti et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Davie et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), proximity to predator occupied habitat, predator and cattle density (Silveira et al. \u003cspan citationid=\"CR99\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Kissling et al. \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Kaartinen et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2009\u003c/span\u003e), proximity to water (Behdarvand et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Abade et al \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), distance to settlement and road (Mbiba et al. \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Amirkhiz et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), temperature and precipitation (Dar et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2009\u003c/span\u003e), topography/terrain, elevation and slope (Naha et al. \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Chetri et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), season (Mbiba et al. \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2018\u003c/span\u003e ), time (Yirga et al. \u003cspan citationid=\"CR118\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Mazzolli et al. \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2002\u003c/span\u003e), and, predator personality, sex, social status and pack size (Odden et al. \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Mattisson et al. \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). In order to reveal these ecological predictors and understand the distribution of HCC, spatial modelling of conflict or predictive risk modelling has become one of the important tools (Treves et al. \u003cspan citationid=\"CR110\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Kissling et al. \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Marucco and McIntire \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Edge et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Zarco-Gonz\u0026aacute;lez et al. \u003cspan citationid=\"CR120\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Mbiba et al. \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Statistical modelling is used to identify the factors related to depredation events, predict its distribution by extrapolating to similar areas, and predict future conflict risk (Treves et al. \u003cspan citationid=\"CR110\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Kaartinen et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Behdarvand et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Rostro-Garc\u0026iacute;a et al. \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). The predictive/risk maps so produced help managers identify vulnerable habitats, communities, and species (Treves et al. \u003cspan citationid=\"CR109\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Mateo-tomas et al. 2012; Davie et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Soh et al. \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Broekhuis et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Amirkhiz et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). However, since all the factors associated with HCC are neither linearly related to kill occurrence nor come into play at the same scale, scale must be considered when modelling habitat correlates of HCC.\u003c/p\u003e \u003cp\u003eMost ecological relationships are complex and involve several factors. And because these factors range from macro to microhabitat/environmental covariates, all of them cannot be expected to operate at and influence the relationship at the same scale. Thus, ecological relationships are scale-dependent, such that when examined at different spatio-temporal scales, the relationship and its interpretation are subject to change (Weins et al. 1989; McGarigal et al. \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Therefore, when identifying the factors related to a process or phenomenon, they need to be examined at multiple scales to identify the meaningful scale and make ecologically sound inferences (McGarigal et al. \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). In the absence of such a multiscale approach, misleading conclusions may be drawn. Even if we are able to identify the causal factors of a problem, we would not know at which level to intervene without understanding the scale at which these causal factors are influencing the problem. In which case, selection of the scale at which the variables are meaningfully correlated with the issue becomes as crucial as the selection of variables themselves (Mateo S\u0026aacute;nchez et al. \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Thus, coupled with variable selection, scale optimisation should be the first step to predictive modelling.\u003c/p\u003e \u003cp\u003eMultiscale models have been proven to perform better than single-scale models at identifying and predicting relationships between environmental variables and the phenomenon/process under study (Mateo S\u0026aacute;nchez et al. \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Timm et al. \u003cspan citationid=\"CR107\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Thus, multiscale modelling has become an important tool for studying a myriad of ecological and biological processes/problems/groupings, including community ecology (Dray et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), ecological niche modelling/niche/resource partitioning (Hearn et al. \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Khosravi et al. \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), habitat selection (Mateo S\u0026aacute;nchez et al. \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2013\u003c/span\u003e)/ habitat suitability modelling (Store and Jokim\u0026auml;ki \u003cspan citationid=\"CR102\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Kittle et al. \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Khosravi et al. \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Rather et al. \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), predicting indicator species hotspots (Grand et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2004\u003c/span\u003e) and predicting carnivore dispersal (Krishnamurthy et al. \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Even though HCC often involves different variables and complex interactions, very few studies have tried to examine the factors determining HCC at multiple scales (Wilson et al. \u003cspan citationid=\"CR116\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Soh et al. \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Miller et al. \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Rostro-Garc\u0026iacute;a et al. \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Broekhuis et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). These studies have found that scale influences livestock predation risk (Davie et al. 2004), with certain habitat factors influencing livestock depredation at a broad scale and others at a fine scale (Miller et al. \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Rostro-Garc\u0026iacute;a et al. \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Broekhuis et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Upon comparison of multiscale model with a single-scale model, studies have concluded that scale optimisation improves modelling results of livestock predation risk by large carnivores like tiger (\u003cem\u003ePanthera tigris\u003c/em\u003e) (Rostro-Garc\u0026iacute;a et al. \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). However, most studies employ aspatial models to predict predation risk using spatial correlates (Soh et al. 2013; Miller et al. \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Since most ecological variables exhibit a certain degree of spatial autocorrelation, it is important to account for the spatial nature of the data (Griffith \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e1992\u003c/span\u003e; Legendre \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e1993\u003c/span\u003e), when modelling predation risk by carnivores. In the absence of which, \u0026lsquo;independence of data points\u0026rsquo;, a common assumption across most statistical models, is violated, leading to unreliable model outcomes (Legendre \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e1993\u003c/span\u003e; Dale and Fortin \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2002\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eMoreover, most studies on HCC attempt to only map risk and discuss the causal factors. They rarely address the ecology (like prey-predator dynamics) behind how these factors interact to cause conflict (Wilkinson et al. \u003cspan citationid=\"CR115\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). As there is a dearth of studies examining HCC at multiple scales employing appropriate spatial statistical models; our study aims to identify the ecological determinants of human-tiger conflict (HTC) at suitable scale, predict livestock predation risk by tigers in and around Panna Tiger Reserve, Central India, and reveal the ecology behind livestock depredation by large carnivores in the light of the identified causal factors. For this purpose, we modelled livestock kill as a function of various tiger relevant ecological variables, viz. prey, cover, water, and anthropogenic disturbance (Miquelle et al. \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e1999\u003c/span\u003e; Karanth and Sunquist \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Sunarto et al. \u003cspan citationid=\"CR104\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), at multiple scales employing spatially explicit Generalized Additive Model (GAM).\u003c/p\u003e \u003cp\u003eVarious statistical tools have been applied to model the relationship between habitat variables and livestock kill, most of the times as presence vs absence (or classification into kill or no kill) using linear parametric models for e.g. discriminant function analysis (Edge et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Treves et al. \u003cspan citationid=\"CR110\" class=\"CitationRef\"\u003e2004\u003c/span\u003e), binary logistic regression or generalized linear model (GLM) with logit link function and binomial error distribution (Broekhuis et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Karanth et al. \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Kissling et al. \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Michalski et al. \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Miller et al. \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Thorn et al. \u003cspan citationid=\"CR106\" class=\"CitationRef\"\u003e2012\u003c/span\u003e); or when modelling the frequency of occurrence of kills (count data), then negative binomial distribution (Penteriani et al. \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) or if kill events are rare, then zero-inflated negative binomial model (Soh et al. \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) or rare event model in a binary logistic regression (Naha et al. \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). However, classical statistical tools like parametric models have several assumptions relating to data distribution and linearity, even though most relationships are not linear in the real world and most of the data does not have a Gaussian distribution (Chambers and Dinsmore \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Mahmoud \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Thus, in recent times, machine learning algorithms are being increasingly used to generate accurate predictions without having to worry about the data distributions, a priori (Kuhn and Johnson \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Several studies have employed machine learning algorithms to model conflict/predation risk (Abade et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Amirkhiz et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Mbiba et al. \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Rostro-Garc\u0026iacute;a et al., \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Although the machine learning algorithms may perform better than classical statistical models when it comes to giving more accurate predictions, if the purpose is to draw inferences about the relationship between variables, they are not very interpretable (Stewart \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eGAM, while retaining the interpretability of GLM has the flexibility of machine learning algorithms, because it does not assume a linear relationship between dependent and independent variables (Hastie and Tibshirani \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e1990\u003c/span\u003e; Larsen \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). GAM is, as the name suggests, a generalisation of the linear model, in which the linear function of the covariate is replaced with a smooth function (Hastie and Tibshirani \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e1990\u003c/span\u003e). Because of their semiparametric nature, GAMs are much more sensitive to unique data distribution than GLM, allowing for the modelling of nonlinear relationships by deriving predictor functions during model building (H\u0026auml;rdle and Turlach \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e1992\u003c/span\u003e; Larsen \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). At the same time to avoid overfitting, one can control the smoothness or \u0026lsquo;wiggliness\u0026rsquo; of the predictor function (Larsen \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Despite their versatility, GAM has not been explored as much as linear or machine learning models to understand the relationship between livestock depredation by carnivores and the environmental factors (Kaartinen et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Miller et al. \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Rostro-Garc\u0026iacute;a et al. \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Broekhuis et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Struebig et al. \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Therefore, we have carried out multivariate multiscale predictive modelling to identify the ecological factors linked to livestock depredation by tiger, employing geoGAM, and discussed how these factors might be linked to the behavioural ecology of predator and prey.\u003c/p\u003e"},{"header":"Methodology","content":"\u003cp\u003e\u003cstrong\u003eStudy area\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePanna Tiger Reserve (24\u0026deg;16ʹN to 24\u0026deg;42ʹN and 79\u0026deg;29ʹE to 80\u0026deg;16ʹE), covering an area of 1598.10 km\u0026sup2; is situated in the state of Madhya Pradesh in central India. The Critical Tiger Habitat (CTH), or core of the reserve comprises Panna National Park and Gangau Wildlife Sanctuary, covering an area of 576.13 km\u003csup\u003e2\u003c/sup\u003e (Madhya Pradesh Forest Department 2007). The buffer covers an area of about 1,021.97 km\u003csup\u003e2\u003c/sup\u003e (Madhya Pradesh Forest Department 2012). The reserve lies in Vindhyan hills, its altitude ranging between 330 and 540 m a.s.l. (Chawdhry \u003cspan class=\"CitationRef\"\u003e1996\u003c/span\u003e; Rodgers et al. \u003cspan class=\"CitationRef\"\u003e2002\u003c/span\u003e). It has an average annual humidity of 86%, and temperature ranges from 5 to 45℃. Both monsoon (July-September) and winter (November-February) are short, thus, the climate is mostly hot and dry (Chawdhry \u003cspan class=\"CitationRef\"\u003e1996\u003c/span\u003e; Gopal et al. \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e). Ken a major tributary of the river Yamuna, cuts through the reserve, flowing from South to North. The major forest type is dry deciduous forest with teak (\u003cem\u003eTectona grandis\u003c/em\u003e) as the most dominant flora (Meher-Homji \u003cspan class=\"CitationRef\"\u003e1990\u003c/span\u003e). Apart from tigers, the major faunal community comprises of carnivores, viz. leopard (\u003cem\u003ePanthera pardus\u003c/em\u003e), striped hyena (\u003cem\u003eHyaena hyaena\u003c/em\u003e), wild dog (\u003cem\u003eCuon alpinus\u003c/em\u003e), golden jackal (\u003cem\u003eCanis aureus\u003c/em\u003e), Bengal fox (\u003cem\u003eVulpes bengalensis\u003c/em\u003e), jungle cat (\u003cem\u003eFelis chaus\u003c/em\u003e), and sloth bear (\u003cem\u003eMelursus ursinus\u003c/em\u003e); herbivores, viz. sambar (\u003cem\u003eCervus duvauceli\u003c/em\u003e), chital (\u003cem\u003eAxis axis\u003c/em\u003e), nilgai (\u003cem\u003eBoselaphus tragocamelus\u003c/em\u003e), chinkara (\u003cem\u003eGazella bennetti\u003c/em\u003e), chousingha or four-horned antelope (\u003cem\u003eTetraceros quadricornis\u003c/em\u003e) and wild pig (\u003cem\u003eSus scrofa\u003c/em\u003e); and primates, viz. hanuman or common langur (\u003cem\u003eSemnopithecus entellus\u003c/em\u003e) and rhesus macaque (\u003cem\u003eMacaca mullata\u003c/em\u003e) (Gopal et al. \u003cspan class=\"CitationRef\"\u003e2010\u003c/span\u003e). There are four villages within the national park area, however, there are seven villages in sanctuary area and 49 villages in the buffer of the reserve. Many of the communities living in these villages are dependent on the reserve for fuelwood, fodder and NTFPs (Malviya et al. \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e). The current study focussed on the CTH and two kilometres of natural buffer around it (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eThe original tiger population in Panna was lost due to poaching and as a population recovery measure, tigers were reintroduced from neighbouring reserves in 2009 (Sarkar et al. \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e). Since then, the tiger population in Panna has increased exponentially from seven founders to more than 60 individuals in 2020 (Sharma and Jarande \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLivestock kill quantification and kill site location\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn India, forest department of the state compensates livestock depredation by wild carnivores. We collected livestock compensation records from Panna Tiger Reserve management to understand the intensity of HTC within the reserve. But the compensation data was not geo tagged, hence we also obtained the livestock kill data (that has GPS locations), which is collected by tiger monitoring teams in the reserve, for the period 2009\u0026ndash;2016. Since in case of Panna, there are several feral cattle, and no distinction is made between feral and domestic cattle in the kill records, we matched kill and compensation data to obtain reliable locations for domestic livestock kill within the reserve. We thus matched 156 locations for the period 2011\u0026ndash;2016. For predation risk probability modelling, we treated livestock kill data as presence and generated equal number of random absence points using \u0026lsquo;create random points\u0026rsquo; tool in ArcGIS 10.4 (ESRI 2016a). We used livestock kill data for the period 2011\u0026ndash;2015 for training the model and 2016 data for testing the model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMeasuring ecological variables\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe considered a total of 27 variables for ecological driver modelling, which can be grouped under the following heads\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eI. Prey\u003c/em\u003e: We used distance sampling technique for tiger prey species population estimation (Buckland et al. \u003cspan class=\"CitationRef\"\u003e2001\u003c/span\u003e). We surveyed 41 line transects each up to 2 km in length, during winter 2012-13 and 2013-14. All the line transects were walked in a replicate of three. The prey species recorded were sambar, chital, wild pig, nilgai, chinkara, chousingha, hanuman langur, hare, peacock and livestock (cattle and buffalo). We combined the data for both the years and calculated all prey (wild and livestock) encounter rate, wild prey encounter rate, and livestock encounter rate, for each transect.\u003c/p\u003e\n\u003cp\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eII. Vegetation cover\u003c/em\u003e: We quantified vegetation indices, viz. canopy cover and shrub abundance in 15m circular plots laid at every 400 m on the line transects during winter 2012-13 (Jhala et al. \u003cspan class=\"CitationRef\"\u003e2009\u003c/span\u003e). We laid a total of 234 circular plots. Within the plots, we made ocular estimations of canopy cover and scored shrub cover according to its abundance (0\u0026ndash;4). The same team of two people carried out all the vegetation related estimations to avoid interobserver bias.\u003c/p\u003e\n\u003cp\u003e\u003c/p\u003e\n\u003cp\u003eFor further understanding the vegetation cover of the tiger reserve, we downloaded LANDSAT 8 (OLI/TIRS) scenes for the reserve from USGS website for April 2013 (LANDSAT SCENE ID\u0026thinsp;=\u0026thinsp;LC81440432013119LGN01; Download date\u0026thinsp;=\u0026thinsp;20 April 2015) and November 2013 (LANDSAT SCENE ID\u0026thinsp;=\u0026thinsp;LC81440432013327LGN01; Download date\u0026thinsp;=\u0026thinsp;15 December 2020). We calculated Normalized Difference Vegetation Index (NDVI) (Rouse et al. \u003cspan class=\"CitationRef\"\u003e1974\u003c/span\u003e) using these scenes, employing Raster Calculator in ArcGIS 10.1 (ESRI, 2012) by the formula:\u003c/p\u003e\n\u003cp\u003eNDVI = (Near Infrared - Red)/ (Near Infrared\u0026thinsp;+\u0026thinsp;Red).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eIII. Water\u003c/em\u003e: Using the same scenes as used for calculating NDVI, we calculated Normalized Difference Water Index (NDWI) (McFeeters \u003cspan class=\"CitationRef\"\u003e1996\u003c/span\u003e) with Raster Calculator in ArcGIS 10.1 (ESRI 2012), by the formula:\u003c/p\u003e\n\u003cp\u003eNDWI = (Green-Near Infrared)/ (Green\u0026thinsp;+\u0026thinsp;Near Infrared).\u003c/p\u003e\n\u003cp\u003eAdditionally, we obtained drainage and water source data from the forest department and created Euclidean distance raster using \u0026lsquo;Euclidean Distance\u0026rsquo; tool in the Spatial Analyst toolbox in ArcGIS 10.4 (ESRI 2016a). We also created Euclidean distance rasters for Ken River and its tributaries, and water sources tagged perennial.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eIV. Topography\u003c/em\u003e: We downloaded ASTER Global Digital Elevation Model (DEM) data from USGS Global Visualization Viewer website. We used \u0026lsquo;Slope\u0026rsquo; tool in the Spatial Analyst toolbox to calculate slope from DEM layers in ArcGIS 10.1 (ESRI 2012). Additionally, we calculated topographic ruggedness index or terrain ruggedness index (TRI) that measures elevation difference between a cell and mean of its eight neighbouring cells (Riley et al. \u003cspan class=\"CitationRef\"\u003e1999\u003c/span\u003e) using raster calculator in ArcGIS 10.1 (ESRI 2012), by the formula (Cooley \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e):\u003c/p\u003e\n\u003cp\u003eTRI\u0026thinsp;=\u0026thinsp;SquareRoot (Abs((Square(\u0026ldquo;3x3max\u0026rdquo;)-Square (\u0026ldquo;3x3min\u0026rdquo;)))).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eV. Land Use Land Cover and forest contiguity\u003c/em\u003e: We procured Land Use Land Cover (LULC) prepared by Forest Survey of India (FSI) for the entire country at 98m resolution for the year 2009 (FSI \u003cspan class=\"CitationRef\"\u003e2009\u003c/span\u003e). We studied landscape characteristics and patterns, particularly habitat connectivity, using multiple indices in program FRAGSTATS (ver. 4.2). We ran FRAGSTAT analysis using the FSI LULC and calculated three class-level metrics: Patch Density (PD), Large Patch Index (LPI) (percentage of total landscape area comprised by the largest patch), and Clumpiness Index (CLUMPY) (a measure of fragmentation). LPI is a simple measure of dominance. And CLUMPY that ranges from \u0026minus;\u0026thinsp;1 (patch type is maximally disaggregated) to 1 (patch type is maximally clumped) provides an index of fragmentation of the focal class that is not affected by changes in class area (McGarigal \u003cspan class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eVI. Disturbance\u003c/em\u003e: We also quantified anthropogenic disturbance indices in the circular plots (as discussed earlier under section II). In each plot, we counted all the lopped (only branches were cut) and cut (cut to stump) trees. We deployed camera traps (Cuddeback Attack pairs) in 109 locations in 2x2 km grids within the national park area, accounting for 7459 trap nights, in the winter of 2013-14. We then manually counted the number of tigers, livestock, humans, and vehicles captured in each camera trap and calculated encounter rates (total no. of captures/total trap nights).\u003c/p\u003e\n\u003cp\u003e\u003c/p\u003e\n\u003cp\u003eWe also obtained village and road location data from the forest department and calculated Euclidean distance rasters. Additionally, we downloaded human footprint data from Socio Economic Data and Application Centre (SEDAC) website for the year 2009 (Sanderson et al. \u003cspan class=\"CitationRef\"\u003e2002\u003c/span\u003e; Venter et al. \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e). We also downloaded population census data for the year 2011 from SEDAC (Balk et al. \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cspan class=\"BoldItalic\" name=\"Emphasis\" type=\"BoldItalic\"\u003eGeostatistical modelling to create rasters\u003c/span\u003e: We interpolated canopy cover, prey, human, livestock, and vehicle encounter rates, and cutting and lopping intensity rates, to create rasters using the Geostatistical wizard in ArcGIS 10.4 (Cressie \u003cspan class=\"CitationRef\"\u003e2015\u003c/span\u003e; ESRI 2016a). For this purpose, we considered four interpolation tools: Inverse Distance Weighing, Simple Kriging (SK), Ordinary Kriging (OK), and Empirical Bayesian Kriging (EBK). We used statistical measures of correctness (mean prediction error, root-mean-square error, standardized root-mean-square error, average standard error) to compare the kriging algorithms. We selected the model that had the smallest root-mean-squared prediction error (RMSE), standardized mean nearest to zero, the average standard error nearest the root-mean-squared prediction error, and the standardized root-mean-squared prediction error nearest to 1 (ESRI 2016b) (Supplementary Table\u0026nbsp;1).\u003c/p\u003e\n\u003cp\u003e\u003cspan class=\"BoldItalic\" name=\"Emphasis\" type=\"BoldItalic\"\u003eScale and variable selection\u003c/span\u003e: To construct a multiscale model, we resampled each of the variables (except for human footprint (which was available at ~\u0026thinsp;1 km resolution)) at five scales: 30m (the highest resolution available), 50m (mean drag distance for tiger kill (Karanth and Sunquist \u003cspan class=\"CitationRef\"\u003e2000\u003c/span\u003e)), 100m (midpoint between fine and coarse resolution), 350m (maximum kill drag distance (Karanth and Sunquist \u003cspan class=\"CitationRef\"\u003e2000\u003c/span\u003e)), 1200m (coarsest resolution used by us and at which most global environmental data is available). LULC (available at 98m) could only be resampled at coarse scale (100-1200m). Thus, in total, we had 118 variables. We then extracted all the variables for each of the presence/absence points using the Spatial Analyst toolbox of ArcGIS 10.4 (ESRI 2016a).\u003c/p\u003e\n\u003cp\u003eFor scale and feature selection, firstly, we ran univariate logistic regressions, after performing Box-Tidwell procedure to test for logistic regression\u0026rsquo;s linearity assumption, i.e., logit transformation of the dependent variable and continuous independent variables have a linear relationship (Shin and Ying \u003cspan class=\"CitationRef\"\u003e1994\u003c/span\u003e). Secondly, we ran univariate GAMs to understand how much r square/deviance was explained by each of the predictors at each of the scales. We selected the scales at which the variables were best explaining the response (livestock kill presence/absence). Additionally, we employed Information Value and Weight of evidence for feature selection (Good and Osteyee \u003cspan class=\"CitationRef\"\u003e1974\u003c/span\u003e). We studied the results of univariate logistic regression, univariate GAM, and Information value to select the explanatory variables at appropriate scales. We then checked the data for multicollinearity and spatial autocorrelation. To check for multicollinearity among the selected variables, we ran appropriate tests of association (for continuous vs continuous and continuous vs ordinal variables, Kendall\u0026rsquo;s tau b; for categorical vs continuous variables, logistic regression; and for categorical vs categorical variables, Cramer\u0026rsquo;s V). Among the correlated variables, we included those variables in the model that better explained the response. For example, all prey (wild prey plus livestock) was highly correlated to wild prey encounter rate (r\u0026thinsp;=\u0026thinsp;0.812). Between the two, we selected all prey encounter rate since it was explaining higher deviance of the dependent variable. Similarly, NDVI and NDWI were moderately correlated (r\u0026thinsp;=\u0026thinsp;0.636); we selected NDVI since it was explaining higher deviance.\u003c/p\u003e\n\u003cp\u003eSo, after variable selection, the global model consisted of the following 12 variables at these scales: All prey encounter rate (50m), slope (50m), elevation (1200m), slope deviation (100m), NDVI (100m), shrub abundance (50m), canopy cover (30m), distance to water (1200m), livestock encounter rate (ct) (50m), human encounter rate (1200m), distance to road (100m), distance to village (100m) (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eWe checked spatial autocorrelation among these variables using Moran\u0026rsquo;s I statistic, and it was found that many variables were spatially autocorrelated (Getis \u003cspan class=\"CitationRef\"\u003e2007\u003c/span\u003e; Moran \u003cspan class=\"CitationRef\"\u003e1950\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSpatial GAM\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe Box-Tidwell test indicated that many of the variables did not meet the linearity assumption of logistic regression. Therefore, we could not use linear models for our predictive modelling. And since our aim was not to merely get accurate predictions but to identify the drivers of HTC, we did not use machine learning algorithms. Therefore, as discussed within the introduction, we selected GAM to model livestock kill locations as a function of various tiger relevant ecological factors (Wood \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e). To account for the spatial autocorrelation in the data, we constructed spatial GAM model using geoGAM package in R ver. 3.6.3. (Nussbaum and Papritz \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e). It\u0026rsquo;s a procedure to build a parsimonious model based on gradient boosting, smoothing splines and a smooth spatial surface to account for the spatial structure. The GAM for spatial data or geoadditive model in its full generality is represented by\u003c/p\u003e\n\u003cdiv class=\"Equation\" id=\"Equa\"\u003e\n \u003cdiv class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e$$\\text{g}\\left({\\mu }\\left(\\text{x}\\left(\\text{s}\\right)\\right)\\right)= {\\nu } + \\text{f}\\left(\\text{x}\\left(\\text{s}\\right)\\right)=$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv class=\"Equation\" id=\"Equb\"\u003e\n \u003cdiv class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e$${\\nu } +{\\sum }_{u}{f}_{{j}_{u}}\\left({x}_{{j}_{u}}\\right(\\text{s}\\left)\\right)+ {\\sum }_{v}{f}_{{j}_{v}}\\left({x}_{{j}_{v}}\\right(\\text{s}\\left)\\right) . {f}_{{k}_{v}}\\left({x}_{{k}_{v}}\\right(\\text{s}\\left)\\right)$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv class=\"Equation\" id=\"Equc\"\u003e\n \u003cdiv class=\"mathdisplay\" id=\"FileID_Equc\" name=\"EquationSource\"\u003e$$+{\\sum }_{w}{f}_{{s}_{w}}\\left(\\text{s}\\right). {f}_{{j}_{w}}\\left({x}_{{j}_{w}}\\right(\\text{s}\\left)\\right)+ {f}_{s}\\left(\\text{s}\\right)$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cp\u003eWhere, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({f}_{s}\\left(\\text{s}\\right)\\)\u003c/span\u003e\u003c/span\u003e is a smooth function of spatial coordinates, which accounts for residual autocorrelation (Nussbaum and Papritz \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e\n\u003cp\u003eSince the response variable, in this case, is binary, Bernoulli distribution is assumed, and logit link used\u003c/p\u003e\n\u003cdiv class=\"Equation\" id=\"Equd\"\u003e\n \u003cdiv class=\"mathdisplay\" id=\"FileID_Equd\" name=\"EquationSource\"\u003e$$\\text{g}\\left({\\mu }\\left(\\text{x}\\left(\\text{s}\\right)\\right)\\right)= \\text{log}\\left(\\frac{{\\mu }\\left(\\text{x}\\left(\\text{s}\\right)\\right)}{1-{\\mu }\\left(\\text{x}\\left(\\text{s}\\right)\\right)}\\right)$$\u003c/div\u003e\u003c/div\u003e\u003cp\u003ewhere,\u003c/p\u003e\u003cdiv class=\"Equation\" id=\"Eque\"\u003e\u003cdiv class=\"mathdisplay\" id=\"FileID_Eque\" name=\"EquationSource\"\u003e$${\\mu }\\left(\\text{x}\\left(\\text{s}\\right)\\right)= \\text{P}\\text{r}\\text{o}\\text{b}\\left[\\text{Y}\\left(\\text{s}\\right)= 1\\right| \\text{x}\\left(\\text{s}\\right)]= \\frac{\\text{e}\\text{x}\\text{p}({\\nu } + \\text{f}(\\text{x}\\left(\\text{s}\\right)\\left)\\right)}{1+\\text{e}\\text{x}\\text{p}({\\nu } + \\text{f}(\\text{x}\\left(\\text{s}\\right)\\left)\\right)}$$\u003c/div\u003e\u003c/div\u003e\u003cp\u003eFor building parsimonious model geoGAM automatically selects factors, covariates and spatial effects using componentwise gradient boosting, following which model is further reduced using cross validation (Nussbaum and Papritz \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eWe ran the final model so selected on test data, to get model performance measures, area under the curve (AUC) and true skill statistic. We performed all the analyses using R Statistical Software (v3.6.3; R Core Team, \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eRisk map\u003c/strong\u003e\u003c/p\u003e\u003cp\u003eWe created a raster with 42m cell size (average calculated from mean kill drag distance for tiger as reported by literature (Karanth and Sunquist \u003cspan class=\"CitationRef\"\u003e2000\u003c/span\u003e; Miller et al. \u003cspan class=\"CitationRef\"\u003e2015\u003c/span\u003e)) and masked it to the reserve boundary. We then converted it into points and, for each of these points, extracted the values for all the explanatory variables. We then ran the final selected model on this data to predict predation risk probability for each point. We then converted the points back to raster. Finally, we assigned risk predictions as the value of the raster to create HTC risk map.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eThe variables selected to be included in the final geoGAM model were all prey encounter rate (50m), elevation (1200m), NDVI (100m), shrub abundance (50m), and human encounter rate (1200m). Among these smooth terms, all prey encounter rate and shrub abundance were significant at \u0026alpha;\u0026thinsp;=\u0026thinsp;0.05 level, and elevation was significant at \u0026alpha;\u0026thinsp;=\u0026thinsp;0.1 level (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\u0026nbsp;\u003ctable border=\"1\" id=\"Tab1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eApproximate significance of smooth terms of final geoGAM model predicting livestock predation by tiger\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"5\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eSmooth terms\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eEffective degree of freedom (edf)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eRef. df\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eChi. sq\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ep-value\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003es(lt_all_50) (prey encounter rate at 50 m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e16.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003es(sa_ebk_50) (shrub abundance at 50 m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e10.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.025\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003es(hum_1200) (human encounter rate at 1200 m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e6.46\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.170\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003es(ndvi_100) (NDVI at 100 m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e5.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.190\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003es(dem_1200) (elevation at 1200m)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e3.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e9.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.089\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eThe effective degree of freedom (edf) is higher than 3 for most of the smooth terms, indicating that the wiggliness is high and relationships nonlinear (Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). Even more is revealed by examining the partial effect plots of smooth terms, also called rug plots. A partial effect plot shows the effect of an explanatory variable on the response variable after accounting for the effects of all the other variables included in the model. Upon examining the partial effect plot for all prey encounter rate, we found that it has an inverse relationship with log odds of livestock kill i.e., the odds of livestock kill by tiger are higher when prey is low (Figure 3a). In case of shrub abundance, we observed a unique trend, log odds of livestock kill increase with shrub abundance but only till it reaches a certain mark, after which increase in shrub abundance seems to reduce the odds of livestock kill (Figure 3b). NDVI, human encounter rate and elevation, as also indicated by their chi-square p values, do not seem to have a significant relationship with the odds of livestock kill (Figure 3c, d, e).\u003c/p\u003e\n\u003cp\u003eThe deviance explained by the model was 44.4%, and AUC of the model was 0.91. When run on the test dataset, the model accuracy was calculated to be 0.65 (Table \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e), and AUC was found to be 0.70, indicating that the model had fair amount of prediction capability.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\u0026nbsp;\u003ctable border=\"1\" id=\"Tab2\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eTrue skill statistic of final geoGAM model predicting livestock predation by tiger\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"2\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAccuracy\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003e0.65\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e95% CI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.52\u0026ndash;0.77\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eKappa\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.30\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSensitivity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.61\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSpecificity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.69\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePositive Predicted Value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.66\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNegative Predicted Value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.65\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePrevalence\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDetection Rate\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.30\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDetection Prevalence\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.46\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBalanced Accuracy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.65\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eSpatial modelling of HCC has enabled conservationists to visualise where the risk of conflict is high and requires mitigation (Kaartinen et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Treves et al. \u003cspan citationid=\"CR109\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Zarco-Gonz\u0026aacute;lez et al. \u003cspan citationid=\"CR120\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Amirkhiz et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Broekhuis et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Potential habitat/environmental factors identified as drivers of conflict risk can help reduce conflict potential, and design targeted mitigation measures (Behdarvand et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). However, spatial modelling should consider scale/resolution of the data and spatial autocorrelation. In the absence of which model results can be unreliable leading to inaccurate identification of HCC drivers and the resultant conflict risk.\u003c/p\u003e \u003cp\u003eHabitat factors that structure the carnivore use of an area are likely to dictate livestock kill by the carnivore and, thereby, HCC. Preferred habitat parameters for tiger have been identified mainly as high prey density, forest contiguity, thick understory, proximity to water, and low human disturbance (Miquelle et al. \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e1999\u003c/span\u003e; Karanth and Sunquist \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Sunarto et al. \u003cspan citationid=\"CR104\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Among these, past studies have linked HTC with tree cover, elevation/altitude, slope, aspect, proximity to reserve forest, proximity to water, distance to village, distance to road, and density of livestock, settlements, and roads (Li et al. \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Ahmed et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Soh et al. \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Miller et al. \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Rostro-Garc\u0026iacute;a et al. \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Struebig et al. \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Ramesh et al. \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Our spatial modelling revealed that the potential ecological drivers of livestock depredation by tigers in Panna Tiger Reserve were prey, and shrub, at a fine scale, i.e., 50m which is the mean drag distance of kill by tigers in tropical landscapes (Karanth and Sunquist \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). Miller et al. (\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), while studying livestock predation risk by tigers in India, also found that the fine-scale model (20m) performed the best (among the three spatial scales viz. 20m, 100m, and 200m at which they measured vegetation structure). They concluded that fine spatial grain risk models are more accurate in predicting human-carnivore conflict. And although Rostro-Garc\u0026iacute;a et al. (\u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) while examining livestock depredation by tiger and leopard in Bhutan tested all their variables at five scales, and found that vegetation cover was more influential at a broader scale (2000m), they concluded that scale optimization improves modelling results with multiscale model performing better than single-scale model. Albeit our modelling results also reveal that both the predictors were operating at a fine scale. It should be emphasised that since we tested each variable at multiple scales, our multiscale model is more reliable than the single scale models or models that did not consider scale, employed by past studies on HTC (Li et al. \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Ahmed et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Soh et al. \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Miller et al. \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Struebig et al. \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Ramesh et al. \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Moreover, ours was the only study that accounted for spatial autocorrelation when carrying out spatial ecological modelling of conflict risk. Without which, the earlier studies violated the assumption about the independence of residuals assumed by the statistical techniques employed by them.\u003c/p\u003e \u003cp\u003eOur model suggests that when prey encounter is low at fine scale (50m), i.e., tiger encounters less prey, it is more likely to predate upon domestic livestock. Low availability of prey has been linked with livestock depredation by carnivores, including tiger (Fritts et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Bhattarai and Fischer \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Burgas et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Khorozyan et al. \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Moreover, vulnerability of prey influences predator choice (Greene \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1986\u003c/span\u003e; Onkonburi, and Formanowicz' Jr \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e1997\u003c/span\u003e; Provost et al. \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Cresswell et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). Predators are known to select a kill that is easier to catch (Mueller \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e1977\u003c/span\u003e; Lang and Gs\u0026ouml;dl 2001; Weise et al. \u003cspan citationid=\"CR113\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Livestock, having lost most of their anti-predator behaviour during the domestication process are vulnerable to becoming easy prey for predators in the absence of human herders (Linnell et al. \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e1999\u003c/span\u003e; Laporte et al. \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Fl\u0026ouml;rcke, and Grandin \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Weise et al. \u003cspan citationid=\"CR113\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Thus, in predator-occupied habitats where there is low availability of wild prey if the optimal foraging theory (large prey, high in abundance, easy to catch) is applied (Emlen \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e1966\u003c/span\u003e; MacArthur and Pianka \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e1966\u003c/span\u003e; Werner and Hall \u003cspan citationid=\"CR114\" class=\"CitationRef\"\u003e1974\u003c/span\u003e), the tiger kills what it can with least effort i.e., livestock.\u003c/p\u003e \u003cp\u003eShrub abundance, the second explanatory variable (also selected at 50m scale), seems to have a unique relationship with livestock kill, resulting in an increase in livestock kill up to a certain point after which increase in shrub abundance decreases the odds of livestock kill. Although it is difficult to explain such a complex relationship, it can be examined in the light of predation technique of tigers. Tiger is an ambush predator therefore, in areas where there is very low cover, it might be very difficult to make a kill, but in areas where cover is high, the chances of success may improve (Greene \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1986\u003c/span\u003e; Murray et al. \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e1995\u003c/span\u003e; Karanth and Sunquist \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Sunquist \u003cspan citationid=\"CR105\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). Studies on tiger and other carnivores have also found that livestock predation risk was higher in habitats with high shrub density because it provides cover for these predators (Davie et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Miller et al. \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). However, if the cover is too dense, grazers like livestock are also less likely to venture into such patches because they would be devoid of grasses. Thus, making the relationship curve between livestock kill and shrub cover, bell-shaped. Livestock kill by tiger is thus a culmination of predator choice and foraging tactics, and prey vulnerability and defence mechanism.\u003c/p\u003e \u003cp\u003eTherefore, from studying the ecological drivers of HTC in Panna Tiger Reserve, we conclude that in a predator-occupied habitat if prey availability is low at fine scale, domestic livestock availability is high, and ambush cover is available, the odds of a predator depredating livestock become high.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe risk map produced using spatial modelling shows that domestic livestock predation risk is higher in the south eastern part of the tiger reserve encompassing Panna Range and parts of Gahrighat Range (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Preventative measures like fencing, viz. biofencing or electric/solar fencing (Distefano, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Sapkota et al., \u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), increased protection through livestock entry point monitoring and patrolling (Pettigrew et al. \u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), change in livestock husbandry and dependence, or village resettlement (Treves and Karanth \u003cspan citationid=\"CR108\" class=\"CitationRef\"\u003e2003\u003c/span\u003e), education and awareness (Consorte-McCrea et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), should be focussed on the high-risk areas and villages in the proximity of these areas.\u003c/p\u003e\n\u003ch2\u003eSummary And Recommendation\u003c/h2\u003e\n\u003cp\u003eEcological drivers of HCC are complex and scale dependent. With the likelihood of conflict being high in a large carnivore habitat that has low prey encounter and an influx of domestic livestock. In case of Panna, we suggest that mitigation efforts should be focussed on the administrative units flagged as high risk by our study. Furthermore, a detailed study should be conducted to understand the lower availability of wild prey and higher availability of livestock in certain parts of the reserve, based on which prey augmentation should be considered where required and deemed feasible.\u003c/p\u003e"},{"header":"Statements \u0026 Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments:\u003c/strong\u003e We are grateful to the National Tiger Conservation Authority (NTCA), Government of India, and Madhya Pradesh Forest Department for funding and providing requisite permissions. We thank the Dean, Director and colleagues (Dr. M. S. Sarkar, Mr. S. K. Roamin, Mr. Naveen, and Mr. Sunil Kumar) at the Wildlife Institute of India and field staff (Mr. R. Mohammad, Mr. M. Kumar, Mr. A. Kondar, Mr. B. Kondar, Mrs. A. Raikwar, and Mr. D. Singh) for providing support and facilitating field data collection.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e: This study was funded by National Tiger Conservation Authority (NTCA), India [NTCA Letter No1-3/93-PT(Vol.II) dated 05th March 2012]\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests:\u003c/strong\u003e It is stated that there is no conflict of interest among authors, funding agency, or with any other party.\u0026nbsp;The authors have no relevant financial or non-financial interests to disclose.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions:\u003c/strong\u003e All authors contributed to the study conception and design. Funding for the study was secured by Ramesh Krishnamurthy. Material preparation, data collection, and analysis were performed by Manjari Malviya. The first draft of the manuscript was written by Manjari Malviya and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and material:\u0026nbsp;\u003c/strong\u003eThe datasets generated and/or analysed during the current study are available on public data repository figshare and can be accessed using the following\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cbr\u003e\u003cstrong\u003elink:\u0026nbsp;\u003c/strong\u003ehttps://figshare.com/s/a9cc82045fd2960f7c9d\u003cstrong\u003e.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode availability:\u003c/strong\u003e (software application or custom code): We used ArcGIS 10.1 \u0026amp; 10.4 and R ver. 3.6.3. All R codes are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval:\u003c/strong\u003e (include appropriate approvals or waivers): Not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to participate:\u0026nbsp;\u003c/strong\u003e(include appropriate statements): Not applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAbade L, Macdonald DW, Dickman AJ (2014) Assessing the relative importance of landscape and husbandry factors in determining large carnivore depredation risk in Tanzania\u0026rsquo;s Ruaha landscape. Biol Conserv 180: 241-248. https://doi.org/10.1016/j.biocon.2014.10.005\u003c/li\u003e\n \u003cli\u003eAhmed RA, Prusty K, Jena J, Dave C., Vihar SC (2012) Prevailing Human Carnivore Conflict in Kanha-Achanakmar Corridor, Central India. J Zool\u003cem\u003e\u0026nbsp;\u003c/em\u003e7(2): 158\u0026ndash;164. https://doi.org/10.5829/idosi.wjz.2012.7.2.6335\u003c/li\u003e\n \u003cli\u003eAmirkhiz RG, Frey JK, Cain JW, Breck SW, Bergman DL (2018) Predicting spatial factors associated with cattle depredations by the Mexican wolf (\u003cem\u003eCanis lupus baileyi\u003c/em\u003e) with recommendations for depredation risk modeling. Biol Conserv 224: 327-335. https://doi.org/10.1016/j.biocon.2018.06.013\u003c/li\u003e\n \u003cli\u003eBabrgir S, Farhadinia MS, Moqanaki, EM (2017) Socio-economic consequences of cattle predation by the Endangered Persian leopard Panthera pardus saxicolor in a Caucasian conflict hotspot, northern Iran. Oryx 51(1): 124-130. https://doi.org/10.1017/S0030605315000903\u003c/li\u003e\n \u003cli\u003eBalk D, Montgomery MR, Engin H, Lin N, Major E, Jones B (2020) Spatial Data from the 2011 India Census. Palisades, NY: NASA Socioeconomic Data and Applications Center (SEDAC). https://doi.org/10.7927/gya1-wp91. Accessed 25 August 2020.\u003c/li\u003e\n \u003cli\u003eBehdarvand N, Kaboli M, Ahmadi M, Nourani E, Mahini AS, Aghbolaghi MA (2014) Spatial risk model and mitigation implications for wolf\u0026ndash;human conflict in a highly modified agroecosystem in western Iran. Biol Conserv\u003cem\u003e\u0026nbsp;\u003c/em\u003e177:156-164. https://doi.org/10.1016/j.biocon.2014.06.024\u003c/li\u003e\n \u003cli\u003eBhattarai BR, Fischer K (2014) Human\u0026ndash;tiger \u003cem\u003ePanthera tigris\u003c/em\u003e conflict and its perception in Bardia National Park, Nepal. Oryx 48(4): 522-528. https://doi.org/10.1017/S0030605313000483\u003c/li\u003e\n \u003cli\u003eBroekhuis F, Cushman SA, Elliot NB (2017) Identification of human\u0026ndash;carnivore conflict hotspots to prioritize mitigation efforts. Ecol Evol 7(24): 10630-10639. https://doi.org/10.1002/ece3.3565\u003c/li\u003e\n \u003cli\u003eBuckland ST, Anderson DR, Burnham KP, Laake JL, Borchers DL, Thomas L (2001) Introduction to Distance sampling: Estimating abundance of biological populations. Oxford University Press, Oxford.\u003c/li\u003e\n \u003cli\u003eBurgas A, Amit R, Lopez BC (2014) Do attacks by jaguars \u003cem\u003ePanthera onca\u003c/em\u003e and pumas \u003cem\u003ePuma concolor\u003c/em\u003e (Carnivora: Felidae) on livestock correlate with species richness and relative abundance of wild prey?. Rev. Biol. Trop. 62(4):1459-67.\u003c/li\u003e\n \u003cli\u003eCavalcanti SC, Marchini S, Zimmermann A, Gese EM, Macdonald DW (2010) Jaguars, livestock, and people in Brazil: realities and attitudes behind the conflict. In: Macdonald D, Loveridge A (eds) The biology and conservation of wild felids. Oxford University Press, Oxford, pp 383-402.\u003c/li\u003e\n \u003cli\u003eChambers M, Dinsmore TW (2014) Advanced analytics methodologies: Driving business value with analytics. Pearson Education, USA.\u003c/li\u003e\n \u003cli\u003eChawdhry PK (1996) Management Plan for Panna National Park. Madhya Pradesh Forest Department, Government of Madhya Pradesh.\u003c/li\u003e\n \u003cli\u003eChetri M, Odden M, Devineau O, Wegge P (2019) Patterns of livestock depredation by snow leopards and other large carnivores in the Central Himalayas, Nepal. Glob. Ecol Conserv 17: e00536. https://doi.org/10.1016/j.gecco.2019.e00536\u003c/li\u003e\n \u003cli\u003eChundawat RS, Gogate N, Johnsingh AJT (1999) Tigers in Panna: preliminary results from an Indian tropical dry forest. In: Seidensticker J, Christie S, Jackson P (eds)\u003cem\u003e\u0026nbsp;\u003c/em\u003eRiding the tiger: tiger conservation in human-dominated landscapes. Cambridge University Press, Cambridge, pp 123-129.\u003c/li\u003e\n \u003cli\u003eClark T, Rutherford M, Casey D (eds) (2013) Coexisting with large carnivores: Lessons from Greater Yellowstone. Island Press, Washington, DC.\u003c/li\u003e\n \u003cli\u003eCooley SW (2016) Terrain Roughness \u0026ndash; 13 ways. http://www.gis4geomorphology.com/ roughness-topographic-position. Accessed January 2017.\u003c/li\u003e\n \u003cli\u003eConsorte-McCrea A, Nigbur D, Bath A (2017) Implications of teenagers\u0026apos; attitudes toward maned wolf conservation in Brazil. Canid Biology \u0026amp; Conservation 20(5): 16-24. https://www.canids.org/CBC/20/Attitudes_toward_maned_wolf_conservation.pdf\u003c/li\u003e\n \u003cli\u003eCressie N (2015) Statistics for spatial data. John Wiley and Sons, New Jersey, USA.\u003c/li\u003e\n \u003cli\u003eCresswell W, Lind J, Quinn JL (2010) Predator‐hunting success and prey vulnerability: Quantifying the spatial scale over which lethal and non‐lethal effects of predation occur. J Anim Ecol 79(3): 556-562. https://doi.org/10.1111/j.1365-2656.2010.01671.x\u003c/li\u003e\n \u003cli\u003eDale MR, Fortin MJ (2002) Spatial autocorrelation and statistical tests in ecology. Ecoscience 9(2): 162-167. https://doi.org/10.1080/11956860.2002.11682702.\u003c/li\u003e\n \u003cli\u003eDar NI, Minhas RA, Zaman Q, Linkie M (2009) Predicting the patterns, perceptions and causes of human\u0026ndash;carnivore conflict in and around Machiara National Park, Pakistan. Biol Conserv 142 (10): 2076\u0026ndash;2082. https://doi.org/10.1016/j.biocon.2009.04.003.\u003c/li\u003e\n \u003cli\u003eDavie HS, Murdoch JD, Lhagvasuren A, Reading RP (2014) Measuring and mapping the influence of landscape factors on livestock predation by wolves in Mongolia. J Arid Environ 103: 85-91. https://doi.org/10.1016/j.jaridenv.2014.01.008\u003c/li\u003e\n \u003cli\u003eDistefano E (2005) Human-Wildlife Conflict worldwide: collection of case studies, analysis of management strategies and good practices. Food and Agricultural Organization of the United Nations (FAO), Sustainable Agriculture and Rural Development Initiative (SARDI), Rome, Italy. http://www. fao. org/documents.\u003c/li\u003e\n \u003cli\u003eDray S, P\u0026eacute;lissier R, Couteron P, Fortin MJ, Legendre P, Peres-Neto PR, Bellier E, Bivand R, Blanchet FG, De C\u0026aacute;ceres M, Dufour AB, Heegaard E, Jombart T, Munoz F, Oksanen J, Thioulouse J, Wagner HH (2012) Community ecology in the age of multivariate multiscale spatial analysis. Ecol Monogr 82(3): 257-275. https://doi.org/10.1890/11-1183.1\u003c/li\u003e\n \u003cli\u003eEdge JL, Beyer DE, Belant JL, Jordan MJ, Roell BJ (2011) Adapting a predictive spatial model for wolf Canis spp. predation on livestock in the Upper Peninsula, Michigan, USA. Wildlife Biol 17(1): 1-10. https://doi.org/10.2981/10-043\u003c/li\u003e\n \u003cli\u003eEmlen JM (1966) The role of time and energy in food preference. Am Nat 100(916): 611-617.\u003c/li\u003e\n \u003cli\u003eESRI (Environmental Systems Resource Institute) (2016a) ArcGIS 10.4. ESRI, Redlands, California.\u003c/li\u003e\n \u003cli\u003eESRI (Environmental Systems Resource Institute) (2016b) Comparing models. Evaluating interpolation results. ArcGIS Desktop. https://desktop.arcgis.com/en/arcmap/10.4/extensions/geostatistical-analyst/comparing-models.htm\u003c/li\u003e\n \u003cli\u003eESRI (Environmental Systems Resource Institute) (2012) ArcGIS 10.1. ESRI, Redlands, California.\u003c/li\u003e\n \u003cli\u003eFl\u0026ouml;rcke C, Grandin T (2013) Loss of anti-predator behaviors in cattle and the increased predation losses by wolves in the Northern Rocky Mountains. Open J Anim Sci 3(3): 248. http://dx.doi.org/10.4236/ojas.2013.33037\u003c/li\u003e\n \u003cli\u003eFritts SH, Stephensen RO, Hayes RD, Boitani L (2003) Wolves and humans. In Mech LD, Boitani, L. (eds) Wolves: Behavior, Ecology and Conservation. University of Chicago Press, Illinois, pp 289-316.\u003c/li\u003e\n \u003cli\u003eFSI (2009) Indian State of Forest Report (ISFR) 2009. Forest Survey of India. MoEF, GOI, Dehradun, pp 199.\u003c/li\u003e\n \u003cli\u003eGetis A (2007) Reflections on spatial autocorrelation. Reg Sci Urban Econ 37(4): 491-496.\u003c/li\u003e\n \u003cli\u003eGood IJ, Osteyee DB (1974) Information, weight of evidence. The singularity between probability measures and signal detection (Vol. 376). Springer-Verlag, Berlin.\u003c/li\u003e\n \u003cli\u003eGopal R, Qureshi Q, Bhardwaj M, Singh RJ, Jhala YV (2010) Evaluating the status of the endangered tiger \u003cem\u003ePanthera tigris\u003c/em\u003e and its prey in Panna Tiger Reserve, Madhya Pradesh, India. Oryx 44(3): 383-389. https://doi.org/10.1017/S0030605310000529\u003c/li\u003e\n \u003cli\u003eGrand J, Buonaccorsi J, Cushman SA, Griffin CR, Neel MC (2004) A multiscale landscape approach to predicting bird and moth rarity hotspots in a threatened pitch pine\u0026ndash;scrub oak community. Conserv Biol 18(4): 1063-1077. https://doi.org/10.1111/j.1523-1739.2004.00555.x\u003c/li\u003e\n \u003cli\u003eGreene CH (1986) Patterns of prey selection: implications of predator foraging tactics. Am Nat 128(6): 824-839.\u003c/li\u003e\n \u003cli\u003eGriffith DA (1992) What is spatial autocorrelation? Reflections on the past 25 years of spatial statistics. L\u0026apos;Espace g\u0026eacute;ographique 21(3): 265-280.\u003c/li\u003e\n \u003cli\u003eGross E, Jayasinghe N, Brooks A, Polet G, Wadhwa R, Hilderink-Koopmans F. (2021) A Future for All: The Need for Human-Wildlife Coexistence. WWF, Gland, Switzerland. https://wwfint.awsassets.panda.org/downloads/a_future_for_all___the_need_for_human_willdife_coexistence.pdf. Accessed July 2021.\u003c/li\u003e\n \u003cli\u003eH\u0026auml;rdle WK, Turlach BA (1992) Nonparametric Approaches to Generalized Linear Models. In Fahrmeir L, Francis B, Gilchrist R, Tutz G (eds) Advances in GLIM and Statistical Modelling. Lecture Notes in Statistics, vol 78. Springer, New York, pp 213-225. https://doi.org/10.1007/978-1-4612-2952-0_33\u003c/li\u003e\n \u003cli\u003eHastie TJ, Tibshirani RJ (1990) Generalized additive models (Vol. 43). Chapman and Hall/CRC press, London.\u003c/li\u003e\n \u003cli\u003eHearn AJ, Cushman SA, Ross J, Goossens B, Hunter LT, Macdonald DW (2018) Spatio-temporal ecology of sympatric felids on Borneo. Evidence for resource partitioning?. PloS one 13(7): e0200828. https://doi.org/10.1371/journal.pone.0200828\u003c/li\u003e\n \u003cli\u003eJhala YV, Qureshi Q, Gopal R, Amin R (2009) Field Guide: Monitoring tigers, co-predators, prey and their habitats, 3\u003csup\u003erd\u003c/sup\u003e ed. Wildlife Institute of India and National Tiger Conservation Authority, Dehradun. https://projecttiger.nic.in/WriteReadData/PublicationFile/FIELD%20GUIDE%202013.pdf\u003c/li\u003e\n \u003cli\u003eKaartinen S, Luoto M, Kojola I (2009) Carnivore-livestock conflicts: determinants of wolf (\u003cem\u003eCanis lupus\u003c/em\u003e) depredation on sheep farms in Finland. Biodivers. Conserv. 18(13): 3503\u0026ndash;3517. https://doi.org/10.1007/s10531-009-9657-8.\u003c/li\u003e\n \u003cli\u003eKaranth KU, Sunquist ME (2000) Behavioural correlates of predation by tiger (\u003cem\u003ePanthera tigris\u003c/em\u003e), leopard (\u003cem\u003ePanthera pardus\u003c/em\u003e) and dhole (\u003cem\u003eCuon alpinus\u003c/em\u003e) in Nagarahole, India. J Zool 250(2): 255-265.\u003c/li\u003e\n \u003cli\u003eKaranth, K. K., Gopalaswamy, A. M., Prasad, P. K., \u0026amp; Dasgupta, S. (2013). Patterns of human\u0026ndash;wildlife conflicts and compensation: Insights from Western Ghats protected areas. Biological Conservation, 166, 175-185.\u003c/li\u003e\n \u003cli\u003eKhorozyan I, Ghoddousi A, Soofi M, Waltert M (2015) Big cats kill more livestock when wild prey reaches a minimum threshold. Biodivers Conserv 192: 268-275. https://doi.org/10.1016/j.biocon.2015.09.031\u003c/li\u003e\n \u003cli\u003eKhosravi R, Hemami MR, Cushman SA (2019) Multi-scale niche modeling of three sympatric felids of conservation importance in central Iran. Landsc Ecol 34(10), 2451-2467. https://doi.org/10.1007/s10980-019-00900-0\u003c/li\u003e\n \u003cli\u003eKissling DW, Fern\u0026aacute;ndez N, Paruelo JM (2009) Spatial risk assessment of livestock exposure to pumas in Patagonia, Argentina. Ecography 32(5): 807-817. https://doi.org/10.1111/j.1600-0587.2009.05781.x\u003c/li\u003e\n \u003cli\u003eKittle AM, Watson AC, Cushman SA, Macdonald DW (2018) Forest cover and level of protection influence the island-wide distribution of an apex carnivore and umbrella species, the Sri Lankan leopard (\u003cem\u003ePanthera pardus kotiya\u003c/em\u003e). Biodivers Conserv 27(1): 235-263. https://doi.org/10.1007/s10531-017-1431-8\u003c/li\u003e\n \u003cli\u003eKrishnamurthy R, Cushman SA, Sarkar MS, Malviya M, Naveen M, Johnson JA, Sen S (2016) Multi-scale prediction of landscape resistance for tiger dispersal in central India. Landsc Ecol 31(6): 1355-1368. https://doi.org/10.1007/s10980-016-0363-0\u003c/li\u003e\n \u003cli\u003eKuhn M, Johnson K (2013) Applied predictive modeling (Vol. 26). Springer, New York.\u003c/li\u003e\n \u003cli\u003eLaporte I, Muhly TB, Pitt JA, Alexander M, Musiani M (2010) Effects of wolves on elk and cattle behaviors: implications for livestock production and wolf conservation. PloS one 5(8): e11954. https://doi.org/10.1371/journal.pone.0011954\u003c/li\u003e\n \u003cli\u003eLarsen K (2015) GAM: the predictive modeling silver bullet. Multithreaded. Stitch Fix 30: 196-223.\u003c/li\u003e\n \u003cli\u003eLegendre P (1993) Spatial autocorrelation: trouble or new paradigm?. Ecology 74(6): 1659-1673.\u003c/li\u003e\n \u003cli\u003eLi B, Zhang E, Liu Z (2009) Livestock depredation by Amur tigers in Hunchun Nature Reserve, Jilin, China. Acta Theriol 29: 231-238.\u003c/li\u003e\n \u003cli\u003eLinnell JDC, Odden J, Smith ME, Aanes R, Swenson JE (1999) Large Carnivores that kill livestock: do \u0026ldquo;problem individuals\u0026rdquo; really exist? Wildlife Soc B 27(3): 698\u0026ndash;705.\u003c/li\u003e\n \u003cli\u003eMacArthur RH, Pianka ER (1966) On optimal use of a patchy environment. Am Nat 100(916): 603-609.\u003c/li\u003e\n \u003cli\u003eMadhya Pradesh Forest Department (August 9, 2012) The Gazette of Madhya Pradesh Extraordinary No.15-21-2010-X-2. Government of Madhya Pradesh, Bhopal.\u003c/li\u003e\n \u003cli\u003eMadhya Pradesh Forest Department (December 24, 2007) The Gazette of Madhya Pradesh Extraordinary No. 15-31-2007-X2. Government of Madhya Pradesh, Bhopal.\u003c/li\u003e\n \u003cli\u003eMahmoud, H. F. (2021). Parametric Versus Semi and Nonparametric Regression Models. International Journal of Statistics and Probability 10(2): 1-90. https://doi.org/10.5539/ijsp.v10n2p90\u003c/li\u003e\n \u003cli\u003eMalviya M, Kalyanasundaram S, Krishnamurthy R. (2022) Paradox of success-mediated conflicts: analyzing attitudes of local communities towards successfully reintroduced tigers in India. Front Conserv Sci 2:783467.\u003c/li\u003e\n \u003cli\u003eMarucco F, McIntire EJB (2010) Predicting spatio‐temporal recolonization of large carnivore populations and livestock depredation risk: wolves in the Italian Alps. J Appl Ecol 47(4): 789-798. https://doi.org/10.1111/j.1365-2664.2010.01831.x\u003c/li\u003e\n \u003cli\u003eMateo S\u0026aacute;nchez MC, Cushman SA, Saura S (2013) Scale dependence in habitat selection: the case of the endangered brown bear (\u003cem\u003eUrsus arctos\u003c/em\u003e) in the Cantabrian Range (NW Spain). Int J Geogr Inf Sci\u003cem\u003e\u0026nbsp;\u003c/em\u003e28(8): 1531-1546. https://doi.org/10.1080/13658816.2013.776684\u003c/li\u003e\n \u003cli\u003eMattisson J, Odden J, Nilsen EB, Linnell JDC, Persson J, Andr\u0026eacute;n H (2011) Factors affecting Eurasian lynx kill rates on semi-domestic reindeer in northern Scandinavia: Can ecological research contribute to the development of a fair compensation system? Biol Conserv 144(12): 3009\u0026ndash;3017. https://doi.org/10.1016/j.biocon.2011.09.004\u003c/li\u003e\n \u003cli\u003eMazzolli M, Graipel ME, Dunstone N (2002) Mountain lion depredation in southern Brazil. Biol Conserv 105(1): 43-51. https://doi.org/10.1016/S0006-3207(01)00178-1\u003c/li\u003e\n \u003cli\u003eMbiba M, Muvengwi J, Ndaimani H (2018) Environmental correlates of livestock depredation by spotted hyaenas and livestock herding practices in a semi-arid communal landscape. Afr J Ecol 56(4): 984-992. https://doi.org/10.1111/aje.12529\u003c/li\u003e\n \u003cli\u003eMcFeeters SK (1996) The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features. Int J Remote Sens 17(7): 1425-1432.\u003c/li\u003e\n \u003cli\u003eMcGarigal K (2015) FRAGSTATS help. University of Massachusetts, Amherst, MA, USA, 182.\u003c/li\u003e\n \u003cli\u003eMcGarigal K., Wan HY, Zeller KA, Timm BC, Cushman SA (2016) Multi-scale habitat selection modeling: a review and outlook. Landsc Ecol 31(6): 1161-1175. https://doi.org/10.1007/s10980-016-0374-x\u003c/li\u003e\n \u003cli\u003eMeher-Homji VM (1990) Vegetation types of India in relation to environmental conditions. In: Daniel JC, Serrao JS (eds) Conservation in developing countries: problems and prospects. Oxford University Press, Bombay, pp 95\u0026ndash;110.\u003c/li\u003e\n \u003cli\u003eMichalski F, Boulhosa RLP, Faria A, Peres CA (2006) Human-wildlife conflicts in a fragmented Amazonian forest landscape: determinants of large felid depredation on livestock. Anim Conserv 9(2): 179\u0026ndash;188. https://doi.org/10.1111/j.1469-1795.2006.00025.x\u003c/li\u003e\n \u003cli\u003eMiller JR., Jhala YV, Jena J, Schmitz OJ (2015) Landscape‐scale accessibility of livestock to tigers: implications of spatial grain for modeling predation risk to mitigate human\u0026ndash;carnivore conflict. Ecol Evol 5(6): 1354-1367. https://doi.org/10.1002/ece3.1440\u003c/li\u003e\n \u003cli\u003eMiquelle DG, Smirnov EN, Merrill TW, Myslenkov AE, Quigley HB, Hornocker MG, Schleyer B (1999) Hierarchical spatial analysis of Amur tiger relationships to habitat and prey. In: Seidensticker J, Christie S, Jackson P (eds) Riding the Tiger: tiger conservation in human-dominated landscapes. Cambridge University Press, Cambridge, pp 71-99.\u003c/li\u003e\n \u003cli\u003eMoran PA (1950) Notes on continuous stochastic phenomena. Biometrika 37(1/2): 17-23.\u003c/li\u003e\n \u003cli\u003eMueller HC (1977) Prey selection in the American kestrel: experiments with two species of prey. Am Nat 111(977): 25-29.\u003c/li\u003e\n \u003cli\u003eMurray DL, Boutin S, O\u0026apos;Donoghue M, Nams VO (1995) Hunting behaviour of a sympatric felid and canid in relation to vegetative cover. Anim Behav 50(5): 1203-1210.\u003c/li\u003e\n \u003cli\u003eNaha D, Sathyakumar S, Rawat GS (2018) Understanding drivers of human-leopard conflicts in the Indian Himalayan region: spatio-temporal patterns of conflicts and perception of local communities towards conserving large carnivores. Plos One 13: e0204528. https://doi.org/10.1371/journal.pone.0204528\u003c/li\u003e\n \u003cli\u003eNussbaum M, Papritz A (2017) geoGAM: Select Sparse Geoadditive Models for Spatial Prediction. R package version 0.1-2.\u003c/li\u003e\n \u003cli\u003eOdden J, Linnell JD, Moa PF, Herfindal I, Kvam T, Andersen R (2002) Lynx depredation on domestic sheep in Norway. J Wildlife Manage 66(1): 98-105.\u003c/li\u003e\n \u003cli\u003eOnkonburi J, Formanowicz\u0026apos; Jr DR (1997) Prey choice by predators: effect of prey vulnerability. Ethol Ecol Evol 9(1): 19-25.\u003c/li\u003e\n \u003cli\u003ePenteriani V, del Mar Delgado M, Pinchera F, Naves J, Fern\u0026aacute;ndez-Gil A, Kojola I, H\u0026auml;rk\u0026ouml;nen S, Norberg H, Frank J, Fedriani JM, Sahl\u0026eacute;n V, St\u0026oslash;en O, Swenson JE, Wabakken P, Pellegrini M, Herrero S, Sahl\u0026eacute;n V. (2016) Human behaviour can trigger large carnivore attacks in developed countries. Sci Rep-UK 6: 20552.\u003c/li\u003e\n \u003cli\u003ePettigrew M, Xie Y, Kang A, Rao M, Goodrich J, Liu T, Berger J (2012) Human\u0026ndash;carnivore conflict in China: a review of current approaches with recommendations for improved management. Integr Zool 7(2):210-26. https://doi.org/10.1111/j.1749-4877.2012.00303.x\u003c/li\u003e\n \u003cli\u003eProvost C, Lucas E, Coderre D (2006) Prey preference of \u003cem\u003eHyaliodes vitripennis\u003c/em\u003e as an intraguild predator: Active predator choice or passive selection?. Biol Control 37(2): 148-154. https://doi.org/10.1016/j.biocontrol.2006.01.005\u003c/li\u003e\n \u003cli\u003eR Core Team (2020) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/.\u003c/li\u003e\n \u003cli\u003eRamesh T, Kalle R, Milda D, Gayathri V, Thanikodi M, Ashish K, Giordano AJ (2020) Patterns of livestock predation risk by large carnivores in India\u0026rsquo;s Eastern and Western Ghats. Glob. Ecol. Conserv. 24: e01366. https://doi.org/10.1016/j.gecco.2020.e01366\u003c/li\u003e\n \u003cli\u003eRather TA, Kumar S, Khan JA (2020) Multi-scale habitat modelling and predicting change in the distribution of tiger and leopard using random forest algorithm. Sci Rep-UK 10(1): 1-19. https://doi.org/10.1038/s41598-020-68167-z\u003c/li\u003e\n \u003cli\u003eRiley SJ, DeGloria SD, Elliot R (1999) A terrain ruggedness index that quantifies topographic heterogeneity. Intermountain. J. Sci. 5(1-4): 23-27.\u003c/li\u003e\n \u003cli\u003eRodgers WA, HS Panwar, VB Mathur (2002) Wildlife Protected Area Network in India: A Review. Wildlife Institute of India publication, Dehradun, India, 44pp.\u003c/li\u003e\n \u003cli\u003eRostro-Garc\u0026iacute;a S, Tharchen L, Abade L, Astaras C, Cushman SA, Macdonald DW (2016) Scale dependence of felid predation risk: identifying predictors of livestock kills by tiger and leopard in Bhutan. Landsc. Ecol. 31(6): 1277-1298. https://doi.org/10.1007/s10980-015-0335-9\u003c/li\u003e\n \u003cli\u003eRouse J Jr, Haas RH, Schell JA, Deering DW (1974) Monitoring vegetation systems in the Great Plains with ERTS. NASA Spec. Publ. 351: 309.\u003c/li\u003e\n \u003cli\u003eSanderson EW, Jaiteh M, Levy MA, Redford KH, Wannebo AV, Woolmer G (2002) The Human Footprint and the Last of the Wild: The human footprint is a global map of human influence on the land surface, which suggests that human beings are stewards of nature, whether we like it or not. BioScience 52(10): 891\u0026ndash;904. https://doi.org/10.1641/0006-3568(2002)052[0891:THFATL]2.0.CO;2\u003c/li\u003e\n \u003cli\u003eSapkota S, Aryal A, Baral SR, Hayward MW, Raubenheimer D (2014) Economic analysis of electric fencing for mitigating human-wildlife conflict in Nepal. J. Resour. Ecol. 5(3): 237-243. https://doi.org/10.5814/j.issn.1674-764x.2014.03.006\u003c/li\u003e\n \u003cli\u003eSarkar MS, Ramesh K, Johnson JA, Sen S, Nigam P, Gupta SK, Murthy RS, Saha GK (2016) Movement and home range characteristics of reintroduced tiger (\u003cem\u003ePanthera tigris\u003c/em\u003e) population in Panna Tiger Reserve, Central India. Eur J Wildlife Res 62(5): 537-547. https://doi.org/10.1007/s10344-016-1026-9\u003c/li\u003e\n \u003cli\u003eSharma UK, Jarande IR (2020) A brief note on Tiger population dynamics and its future projection in Panna Tiger Reserve. https://www.pannatigerreserve.in/research.htm. Accessed 21 April 2022.\u003c/li\u003e\n \u003cli\u003eSeidensticker J (1987) Bearing witness: Observations on the extinction of \u003cem\u003ePanthera tigris balica\u003c/em\u003e and \u003cem\u003ePanthera tigris sondaica\u003c/em\u003e. In: Tilson R, Nyhus PJ (eds) Tigers of the world: The Biology, Biopolitics, Management, and Conservation of an Endangered Species. Noyes Publications, New York, pp 416-426.\u003c/li\u003e\n \u003cli\u003eShin RT, Ying JS (1994) Imposing linear homogeneity on Box\u0026ndash;Tidwell flexible functional forms. J Bus Econ Stat 12(2): 261-265.\u003c/li\u003e\n \u003cli\u003eSilveira L, Boulhosa R, Astete S, J\u0026aacute;como ATA (2008) Management of domestic livestock predation by jaguars in Brazil. Cat News 4: 26-30.\u003c/li\u003e\n \u003cli\u003eSoh YH, Carrasco LR, Miquelle DG, Jiang J, Yang J, Stokes EJ, Tang J, Kang A, Liu P, Rao M (2014) Spatial correlates of livestock depredation by Amur tigers in Hunchun, China: relevance of prey density and implications for protected area management. Biol Conserv 169: 117-127. https://doi.org/10.1016/j.biocon.2013.10.011\u003c/li\u003e\n \u003cli\u003eStewart M (2019) The Actual Difference Between Statistics and Machine Learning. Towards data science. https://towardsdatascience.com/the-actual-difference-between-statistics-and-machine-learning-64b49f07ea3\u003c/li\u003e\n \u003cli\u003eStore R, Jokim\u0026auml;ki J (2003) A GIS-based multi-scale approach to habitat suitability modeling. Ecol Model 169(1): 1-15. https://doi.org/10.1016/S0304-3800(03)00203-5\u003c/li\u003e\n \u003cli\u003eStruebig MJ, Linkie M, Deere NJ, Martyr DJ, Millyanawati B, Faulkner SC, Le Comber SC, Mangunjaya FM, Leader-Williams N, McKay JE, John FAS (2018) Addressing human-tiger conflict using socio-ecological information on tolerance and risk. Nat Commun 9(1): 3455. https://doi.org/10.1038/s41467-018-05983-y\u003c/li\u003e\n \u003cli\u003eSunarto S, Kelly MJ, Parakkasi K, Klenzendorf S, Septayuda E, Kurniawan H (2012) Tigers need cover: Multi-scale occupancy study of the big cat in Sumatran forest and plantation landscapes. Plos One \u003cstrong\u003e7\u003c/strong\u003e(1): e30859. https://doi.org/10.1371/journal.pone.0030859\u003c/li\u003e\n \u003cli\u003eSunquist M (2010) What is a tiger? Ecology and behavior. In: Tilson R, Nyhus PJ (eds) Tigers of the world: The Science, Politics, and Conservation of \u003cem\u003ePanthera tigris\u003c/em\u003e. Academic Press, London, pp 19-33.\u003c/li\u003e\n \u003cli\u003eThorn M., Green, M., Dalerum, F., Bateman, P. W., \u0026amp; Scott, D. M. (2012). What drives human\u0026ndash;carnivore conflict in the North West Province of South Africa? Biological Conservation, 150(1), 23-32.\u003c/li\u003e\n \u003cli\u003eTimm BC, McGarigal K, Cushman SA, Ganey JL (2016) Multi-scale Mexican spotted owl (Strix occidentalis lucida) nest/roost habitat selection in Arizona and a comparison with single-scale modeling results. Landsc Ecol 31(6): 1209-1225. https://doi.org/10.1007/s10980-016-0371-0\u003c/li\u003e\n \u003cli\u003eTreves A, Karanth KU (2003) Human‐carnivore conflict and perspectives on carnivore management worldwide. Conserv Biol 17(6):1491-9. https://doi.org/10.1111/j.1523-1739.2003.00059.x\u003c/li\u003e\n \u003cli\u003eTreves A, Martin KA, Wydeven AP, Wiedenhoeft JE (2011) Forecasting environmental hazards and the application of risk maps to predator attacks on livestock. BioScience 61(6): 451-458. https://doi.org/10.1525/bio.2011.61.6.7\u003c/li\u003e\n \u003cli\u003eTreves A, Naughton‐treves L., Harper, E. K., Mladenoff, D. J., Rose, R. A., Sickley, T. A., and Wydeven, A. P. (2004) Predicting human‐carnivore conflict: a spatial model derived from 25 years of data on wolf predation on livestock. Conserv Biol 18(1), 114-125. https://doi.org/10.1111/j.1523-1739.2004.00189.x\u003c/li\u003e\n \u003cli\u003eVenter O, Sanderson EW, Magrach A, Allan JR, Beher J, Jones KR, Watson JE (2018) Last of the wild project, version 3 (LWP-3): 2009 human footprint, 2018 release. NASA Socioeconomic Data and Applications Center (SEDAC), 10, H46T40JQ44, Palisades, NY.\u003c/li\u003e\n \u003cli\u003eWiens JA (1989) Spatial scaling in ecology. Functional ecology 3(4):385-397.\u003c/li\u003e\n \u003cli\u003eWeise FJ, Tomeletso M, Stein AB, Somers MJ, Hayward MW (2020) Lions \u003cem\u003ePanthera leo\u003c/em\u003e prefer killing certain cattle Bos taurus types. Animals 10(4): 692. https://doi.org/10.3390/ani10040692\u003c/li\u003e\n \u003cli\u003eWerner EE, Hall DJ (1974) Optimal foraging and the size selection of prey by the bluegill sunfish (\u003cem\u003eLepomis macrochirus\u003c/em\u003e). Ecology 55(5): 1042-1052.\u003c/li\u003e\n \u003cli\u003eWilkinson CE, McInturff A, Miller JR, Yovovich V, Gaynor KM, Calhoun K, Karandikar H, Martin JV, Parker‐Shames P, Shawler A, Van Scoyoc A (2020) An ecological framework for contextualizing carnivore\u0026ndash;livestock conflict. Conserv Biol 34(4):854-67.\u003c/li\u003e\n \u003cli\u003eWilson SM, Madel MJ, Mattson DJ, Graham JM, Burchfield JA, Belsky JM (2005) Natural landscape features, human-related attractants, and conflict hotspots: a spatial analysis of human-grizzly bear conflicts. Ursus 16(1): 117-129. https://doi.org/10.2192/1537-6176(2005)016[0117:NLFHAA]2.0.CO;2\u003c/li\u003e\n \u003cli\u003eWood SN (2017) Generalized additive models: an introduction with R. CRC press, Florida.\u003c/li\u003e\n \u003cli\u003eYirga G, De longh HH, Leirs H, Gebrehiwot K, Berhe G, Asmelash T, Gabrehiwot B, Bauer H (2012) The ecology of large carnivores in the highlands of northern Ethiopia. Afr J Ecol 51: 78\u0026ndash;86. https://doi.org/10.1111/aje.12008\u003c/li\u003e\n \u003cli\u003eYoung SP, Goldman EA (1944) The wolves of North America. Dover, New York.\u003c/li\u003e\n \u003cli\u003eZarco-Gonz\u0026aacute;lez MM, Monroy-Vilchis O, Alan\u0026iacute;z J (2013) Spatial model of livestock predation by jaguar and puma in Mexico: conservation planning. Biol Conserv 159: 80-87. https://doi.org/10.1016/j.biocon.2012.11.007\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"human-carnivore conflict, ecological predictors, prey, cover, domestic livestock, Panna","lastPublishedDoi":"10.21203/rs.3.rs-1754621/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-1754621/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eContext: \u003c/strong\u003eSpatial modelling of human-carnivore conflict has recently gained traction, and predictive maps have become a great tool to understand the distribution of present and future conflict risk. However, very few such studies consider scale and use appropriate spatial modelling tools. \u003c/p\u003e\u003cp\u003e\u003cstrong\u003eObjectives: \u003c/strong\u003eWe aimed to understand the ecological predictors of human-tiger (\u003cem\u003ePanthera tigris\u003c/em\u003e) conflict and predict livestock predation risk by reintroduced tigers in Panna Tiger Reserve, Central India. By modelling livestock kill as a function of various tiger relevant ecological variables at multiple scales employing spatially explicit statistical tools. \u003c/p\u003e\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eWe used\u003cstrong\u003e \u003c/strong\u003egeostatistical modelling to create raster layers of covariates (prey, cover, human activities), following which we did univariate scaling. We then modelled livestock loss by tiger using spatial Generalized Additive Model (geoGAM), predicted and mapped conflict risk probability. \u003c/p\u003e\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eWe found that prey and shrub cover, both selected at a fine scale, were key ecological determinants of human-tiger conflict. Prey showed an inverse relationship with livestock predation and shrub nonlinear; livestock predation increasing with an increase in shrub cover but decreasing beyond a certain point. Thus, in habitats where optimum ambush cover is available but prey presence is low at fine-scale, carnivores are more likely to depredate domestic livestock since livestock have lost most of their anti-predator behaviours. \u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConclusions: \u003c/strong\u003eLivestock kill by tiger is a culmination of predator choice and foraging tactics, and prey vulnerability and defence mechanism. The spatially explicit predation risk map produced in this study can guide adequate human-tiger conflict prevention measures.\u003c/p\u003e","manuscriptTitle":"Multiscale spatially explicit modelling of livestock depredation by reintroduced tiger (Panthera tigris) to predict conflict risk probability","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2022-06-17 20:03:55","doi":"10.21203/rs.3.rs-1754621/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"105882ff-aeda-4938-af0c-d860165f1d2c","owner":[],"postedDate":"June 17th, 2022","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2022-07-30T00:14:07+00:00","versionOfRecord":[],"versionCreatedAt":"2022-06-17 20:03:55","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-1754621","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-1754621","identity":"rs-1754621","version":["v1"]},"buildId":"J0_U0BvcaRcwD8yVFaRlm","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}