1. Habitat Suitability Modelling of White-Bellied Pangolin (Phataginus tricuspis) in Oluwa Forest Reserve, Ondo State, Nigeria
- Author
-
Adeniji Adebola Esther, Ejidike Bernadette Nwandu, and Olaniyi Oluwatobi Emmanuel
- Subjects
white bellied pangolin ,habitat suitability ,maxent modelling ,habitat suitability model ,machine learning ,Ecology ,QH540-549.5 - Abstract
Most endangered species face a significant threat from habitat loss. The destruction and degradation of natural tropical forest across West Africa has likely been the biggest threat to White-bellied Pangolin and has contributed to their decline as they depend on the habitat for different resources like food, water, and shelter. The current study investigated the habitat suitability of white-bellied pangolins in Oluwa Forest Reserve. The presence data of White-bellied pangolin was collected by taking the Global Positioning System (GPS) coordinates of the indirect signs observed. These data, along with the 19 bioclimatic variables, slopes, soil PH, soil texture, distance to rivers, distance to roads, and Normalized Difference Vegetation Index (NDVI), were used to generate habitat suitability maps using MaxEnt software. The MaxEnt analysis showed that out of 781 km2 available for White bellied Pangolin during dry season, 338 km2 was highly suitable, 209 km2 was suitable, 126 km2 was moderately suitable, 65 km2 was less suitable and 44 km2 was not suitable. During the wet season 235 km2 was highly suitable, 225 km2 was suitable, 164 km2 was moderately suitable, 100 km2 was less suitable and 57 km2 was not suitable habitat. The jackknife test of variable contribution revealed that during the dry season, NDVI was the most important predictor variable as measured by the gain produced by a one-variable model, followed by aspects such as distance to the river, slope, distance to the road, and temperature seasonality. During the wet season, the jackknife-cross-validation test showed the highest gain when NDVI was used in isolation. Aspects were found to be the second most important predictor variable as measured by the gain produced by a one-variable model, followed by distance to the road, slope, elevation, and the mean temperature of the wettest quarter.
- Published
- 2024
- Full Text
- View/download PDF