Garissa County​ Ecological Conservation​: Mapping and Characterizing Invasive Shrub Encroachment across Kenyan Savannas to Protect Endangered Hirola Antelope Habitat​

The hirola (Beatragus hunteri) is a critically endangered antelope in Kenya with fewer than 250remaining. The rinderpest disease that originally decimated their population was eradicated, but research has shown that woody encroachment has hampered species recovery by reducing the rangeland habitat this species requires. However, research quantifying rates of encroachment are limited. An understanding of these woody encroachment trends is important for the Hirola Conservation Program (HCP), a nonprofit working to conserve the hirola. We used Earth observation data (Landsat 5 Thematic Mapper, Landsat 8 Operational Land Imager, Landsat 9 Operational Land Imager-2, NASA Shuttle Radar Topography Mission; 30 m pixels and Esri Wayback Imagery; 15 m pixels) to quantify encroachment over time. We used Landsat datasets to calculate Normalized Difference Vegetation Index and Tasseled Cap Transformation, which served as predictor variables in our random forest model, along with woody cover data from the HCP and woody cover training data. Our model estimated a 37% increase in woody vegetation within the hirola’s range from 1987–2025 and a 62% increase from 2017–2025. We calculated a Root Mean Square Error of 6.5% and an adjusted R² value of 0.913, indicating high prediction accuracy of the model. Due to a lack of training data for 1987, the prediction for 2017–2025 is more reliable. Our work showed that it is feasible to use remote sensing to map woody encroachment, but it has some limitations, mainly in mapping the spread of specific species. The results from this study will allow the HCP to assess encroachment trends over time within the hirola’s range to inform targeted restoration efforts.