Distribution Pattern, Ecological Determinants and Conservation Gaps of Model-Predicted Relative Probability of Occurrence Zones for the Tufted Deer (Elaphodus cephalophus) in China.

The tufted deer (Elaphodus cephalophus), a rare ungulate species endemic to China, faces mounting conservation concerns due to habitat fragmentation, climate change, and historical overhunting. However, its current patterns of model-predicted relative probability of occurrence and the environmental associations of its distribution remain poorly understood. In this study, we used 429 occurrence points and 28 environmental variables, refined to 11 key predictors, to predict the species' relative probability of occurrence across China using an optimized MaxEnt model. The model performed robustly, identifying six dominant environmental factors-temperature annual range, annual precipitation, mean temperature of the coldest quarter, slope, vegetation fractional cover, and human footprint index-that collectively contributed 91.6% to the model-predicted relative probability of occurrence. Model outputs indicated that the relative probability of occurrence was associated with moderate temperature variation (25.3°C-30.4°C), bimodal precipitation patterns (725-1324 mm and 1651-1898 mm), and cooler winter temperatures (-2.0°C-9.9°C), typically found in mountainous regions. Model-based analyses revealed that zones with moderate-to-high model-predicted relative probability of occurrence are concentrated in eight provinces, with Sichuan, Guizhou, and Yunnan contributing the largest zones. Despite these extensive zones with model-predicted relative probability of occurrence, our GAP analysis showed that 93.98% of them lie outside current protected zones overlapping with model-predicted relative probability of occurrence areas, indicating substantial conservation gaps. Even in core provinces such as Sichuan and Guizhou, only a small fraction (≤ 10.84%) of the zones with high model-predicted relative probability of occurrence are protected. These findings highlight the urgent need to re-evaluate and expand protected zones networks to include zones with high model-predicted relative probability of occurrence for tufted deer. Our study provided essential ecological insights and spatial data to guide habitat conservation, functional zoning, and long-term management strategies for tufted deer populations in China.