Assessment of Environmental Variables in the Ensemble Model and Their Impact on Species Distribution
In the construct of this study, nine environmental variables were integrated into the predictive modeling framework. Of these, the precipitation during the coldest quarter (bio19), the Normalized Difference Vegetation Index of March (NDVI0321), and the topographic variable of slope emerged as pivotal, each contributing in excess of 10% to the model’s predictive capacity, as elucidated in Table 2. This denotes their substantial relevance and influence within the ecological modeling construct. In contrast, the other variables demonstrated a relative importance below the 10% benchmark, suggesting a more marginal role in the model’s overall predictive accuracy. Notably, the leading trio of environmental parameters exhibited a significant positive correlation with the probability of habitat suitability for Manis pentadactyla (Chinese pangolin), as illustrated in Figure 2. In comparison, the secondary environmental variables did not exhibit marked correlations with the species’ distribution probabilities, underscoring the differential impact of various ecological factors on pangolin habitat suitability.