← Back to projects

Carnivore Habitat Monitoring

Some of eastern India’s most remarkable large carnivores are surviving alongside farms and villages. WINGS’ field research has identified at least 30 Indian grey wolves across four packs and 30 striped-hyaenas across six populations in industrial and mining landscapes. These animals are adapting to intensely human-shaped ecosystems, where livestock dependence, roads, and habitat fragmentation make understanding exactly how they use the landscape critical to their survival.

Habitat Robotics is leveraging WINGS’ camera trap data to better understand movement corridors and habitat usage patterns of these carnivores. We are integrating camera observations with geospatial data metrics, human activity and settlements to model temporal activity, detection intensity, and habitat use. For striped hyaenas, we are also exploring semi-automated identification from individual stripe patterns.

We combine WINGS’ field knowledge, camera trap data, and ecological context to reveal how wolves and hyenas use the landscape and guide better corridor protection over time.

Field Note

From camera traps to landscape intelligence

The Indian grey wolf (Canis lupus pallipes) is one of India’s most threatened large carnivores. A national assessment estimated approximately 3,170 adult wolves distributed across 500 potential packs and 364,425 km² of potential habitat. (Jhala et al., 2022)

In eastern India, the Chhota Nagpur Plateau and Lower Gangetic Plains form an important but comparatively under-studied part of the species’ range (Sharma et al., 2019). In Purba Bardhaman district, Durgapur WINGS deployed 49 camera traps between September and December 2025. The camera locations span a roughly 201 km² footprint, through a mosaic of forests and villages.

Habitat Robotics is partnering with Durgapur WINGS to turn this field material into a multimodal landscape-analysis workflow.

Using this analysis workflow, we are characterizing the monitored landscape across several ecological dimensions. In addition to aggregating the timing and distribution of wolf records & camera-level activity patterns, we are characterizing their co-occurrence patterns with people, vehicles and livestock. The team has been drawing correlations between wolf abundance and distance to villages and roads, and the wider wildlife community represented across the network, such as the Golden Jackal, Bengal Fox, Jungle Cat, and Grey Mongoose.

Our landscape analysis is derived from Sentinel-2 Level-2A surface-reflectance data. We mask clouds and no-data pixels using data-mask layers, then build temporal composites around the camera network. From these composites, we derive vegetation greenness, water-related indices, and bare-soil measures. These variables are summarized around each camera across different spatial footprints, allowing the analysis to examine how landscape signals change across spatial scales.

Through this work, Habitat Robotics aims to connect field ecology, remote sensing, spatial analysis, and conservation planning in one transparent system. Across eastern India, we are identifying the sensing and analysis workflows that repeat across landscapes and conservation teams, and finding ways to streamline repeatable steps so that frontline organizations can make better conservation decisions faster and at lower cost.