Skip to content

Settlement Proximity Was the Strongest Model Predictor of Tiger and Leopard Livestock Depredation in Panna

Wild female leopard in Panna Tiger Reserve, India
Conservation · Human–wildlife conflict

A spatial analysis of livestock losses around Panna Tiger Reserve, India found that distance to human settlements was the strongest model predictor of depredation risk for both tiger (Panthera tigris) and leopard (Panthera pardus). The study used documented livestock attacks from 2017–2022 and MaxEnt modelling to identify environmental conditions associated with recorded attack locations. In practical terms, MaxEnt compares conditions around known events with the surrounding landscape and maps where similar combinations occur. The resulting patterns show statistical associations with recorded depredation, not proof that settlement proximity itself caused the attacks.

High-risk areas differed between the two species

The paper reports 866 livestock depredation incidents in the wider dataset, including 627 attributed to tigers and 189 to leopards. Records lacking usable GPS coordinates or falling outside the modelling area were removed before the spatial models were fitted. MaxEnt classified 12.3% of the study area as high relative depredation risk for tiger and 4.7% as high risk for leopard. These percentages describe the share of mapped land assigned to the high-risk class, not the proportion of livestock expected to be attacked.

Distance to settlements had the largest contribution to both models: 39.2% for tiger and 33.1% for leopard. These figures are the variables’ contributions to the fitted MaxEnt models, not the percentage of attacks caused by settlement proximity. Tiger risk was highest near open forest at elevations of roughly 350–500 m, while leopard risk was concentrated closer to bush vegetation and within about 400 m of settlements. Other important predictors included distance to open forest, altitude, density of small ruminants such as sheep and goats, and distance to cropland.

The maps discriminate known attack sites well

The models had AUC values of 0.98 for tiger and 0.96 for leopard. AUC measures how well a model separates known event locations from background locations across possible thresholds; values this close to 1 indicate very strong discrimination within the study data. They do not mean that the models will correctly predict 98% or 96% of future attacks. The authors therefore present the maps as tools for identifying parts of the human–forest interface where prevention and livestock protection can be prioritised, rather than as counts of future depredation events.

Sources and image credit

  1. Tripathi, S., Zarco-González, M. M., Singh, T., & Balbuena-Serrano, Á. (2026). Livestock Depredation by Large Carnivores in Panna Tiger Reserve, India: Spatial Analysis through Maxent. Journal for Nature Conservation, 93, 127351. https://doi.org/10.1016/j.jnc.2026.127351Primary research article published in 2026. The study analysed 866 livestock depredation incidents from 2017–2022 around Panna Tiger Reserve; 627 were attributed to tigers and 189 to leopards. Distance to settlements was the strongest model predictor for both species, and the models identified different high-risk habitat patterns for tiger and leopard.

Image: Ab0408 / Wikimedia Commons — wild female leopard in Panna Tiger Reserve, India, 26 March 2016; cropped to 16:9. Representative image for this story; not presented as an animal or scene from the reported study or event. · CC BY-SA 4.0 · Source

About this content: This story was produced with AI assistance within an editorial workflow developed by Wildlife Vagabond. Editorial responsibility remains with Wildlife Vagabond.How AI is used

Human–wildlife conflict

SPECIES IN THIS STORY

Species in this story

Tiger Panthera tigris Explore species Leopard Panthera pardus Explore species

Independent research and conservation news archive

Did you find this information useful?

Wildlife Vagabond is independently built and maintained. Voluntary support helps cover source verification, hosting and continued work on research and conservation news.

Support Wildlife Vagabond

The news archive will remain freely available.