Understanding the dynamics that drive human-wildlife conflict and identifying potential mitigation solutions requires understanding the spatial patterns of conflict. The juxtaposition of ecological preservation and economic growth has led to increased conflicts between humans and Asian elephants in the Rajaji-Corbett landscape of Uttarakhand, India, where the conversion of elephant habitat to agricultural land have increased over the last several decades. We investigated the predictors influencing household-level human-elephant conflicts (HECs) using binomial Generalized Linear Models (GLMs) collected from semi-structured questionnaire-based surveys of 266 households in the human-wildlife interface next to protected areas. Further, we modelled the landscape predictors that influence the spatial distribution of HECs by collecting occurrence data of HECs in 25 km grid units (N = 1473 grids) using Maxent software. We discovered that HECs are directly influenced by the diversity of major and minor crops planted and the proximity to agricultural land (conflicts decreased with increasing distance from the agricultural land). We also observed that the probability of HECs decreased with increasing elevation, increase in road networks, and with increasing slope in the study area; while HECs increased with increase in human population. We discovered that nearly one-fifth of areas sampled (3606.87 km) in the Rajaji-Corbett landscape were at high risk of HEC, especially flat, agrarian areas where most people reside. Farmers in the susceptible risk areas identified by our study could lessen the likelihood of crop damage and HEC incidents by cultivating highly profitable alternative crops that are less attractive to elephants. Additionally, implementing mobile-based Early Warning System in high HEC hotspot areas could mitigate crop raiding and potentially reduce the occurrence of HECs. The findings of our study can assist policymakers and park management in designing landscape-scale human-wildlife conflict mitigation strategies tailored to identified conflict hotspots.

中文翻译：

---

印度北阿坎德邦 Rajaji-Corbett 景观人类主导地区人象冲突热点的评估和预测

了解驱动人类与野生动物冲突的动态并确定潜在的缓解解决方案需要了解冲突的空间模式。在印度北阿坎德邦的 Rajaji-Corbett 地区，生态保护和经济增长的并存导致人类和亚洲象之间的冲突加剧，过去几十年里，大象栖息地转变为农业用地的情况有所增加。我们使用二项式广义线性模型 (GLM) 调查了影响家庭层面人象冲突 (HEC) 的预测因素，这些模型是从对保护区附近人类与野生动物接触面的 266 个家庭进行的半结构化问卷调查中收集的。此外，我们通过使用 Maxent 软件收集 25 km 网格单元（N = 1473 个网格）中 HEC 的出现数据，对影响 HEC 空间分布的景观预测因子​​进行了建模。我们发现，HEC 直接受到种植的主要和次要作物的多样性以及距农业用地的远近的影响（冲突随着距农业用地距离的增加而减少）。我们还观察到，HEC 的概率随着研究区域海拔的增加、道路网络的增加以及坡度的增加而降低；而HEC随着人口的增加而增加。我们发现，Rajaji-Corbett 景观中近五分之一的采样区域（3606.87 平方公里）存在 HEC 高风险，尤其是大多数人居住的平坦农业地区。我们的研究确定的易受影响地区的农民可以通过种植对大象吸引力较小的高利润替代作物来减少农作物受损和 HEC 事件的可能性。此外，在 HEC 热点地区实施基于移动设备的预警系统可以减轻作物劫掠，并有可能减少 HEC 的发生。我们的研究结果可以帮助政策制定者和公园管理人员针对已确定的冲突热点设计景观规模的人类与野生动物冲突缓解策略。