关键词: 庞泉沟国家级自然保护区, 活动节律, 时空分布, 物种共存, 集合模型

Abstract

Aims: This study aims to reveal the daily activity rhythms, spatial distribution, environmental responses, and coexistence patterns of roe deer (Capreolus pygargus) and wild boar (Sus scrofa), filling the research gap regarding the spatiotemporal distribution and coexistence patterns of ungulate species in Pangquangou National Nature Reserve. This is of great theoretical and practical significance for understanding interspecific interactions in North China and guiding the ecosystem management of protected areas. 

Methods: Based on the monitoring data collected by 40 infrared cameras deployed in Shanxi Pangquangou National Nature Reserve from August 2023 to November 2024, this study systematically analyzed the spatiotemporal distribution and coexistence patterns of sympatric roe deer and wild boars using multiple methods, including ensemble models, kernel density estimation, daily-discrepancy index (α) and diurnal-nocturnal index (β). 

Results: (1) A total of 40 infrared cameras operated continuously for a cumulative 15,880 camera-days, capturing 7,116 independently identifiable valid photos, including 2,344 of roe deers and 192 of wild boars. (2) Both species showed a diurnal bimodal activity pattern (roe deer: β = 0.63; wild boars: β = 0.71). The monthly differences in daily activity patterns were not significant (P > 0.05), but there was a significant difference in the overlap of daily activity rhythms (Δ₄ = 0.85; 95% CI: 0.81–0.87; P < 0.05). The morning activity peak of roe deer occurred 1–2 hours later than that of wild boars, while their afternoon activity peaks were nearly synchronous. This temporal shift indicates that the two species reduce interspecific competition through temporal niche differentiation. (3) The ensemble models showed good performance in predicting the distribution of both species (TSS > 0.7, AUC > 0.9). The suitable habitats of both roe deer and wild boars were mainly concentrated in the central part of the reserve. The suitable habitat area for roe deer was 71.64 km², accounting for 67.06% of the total reserve area; the suitable habitat area for wild boars was 59.08 km², accounting for 56.91%. The niche overlap between the two species was high (Schoener’s D = 0.70). (4) The human influence index, normalized difference vegetation index (NDVI), and isothermality were identified as the key factors influencing the distribution of both species. Specifically, the probability of presence of both roe deer and wild boars decreased significantly with increasing isothermality, was positively correlated with NDVI, and exhibited a non-linear relationship with the human influence index (initially decreasing before increasing). 

Conclusion: Regarding the temporal niche, although both species exhibited a diurnal bimodal activity patterns with high niche overlap, a significant fine-scale differentiation in peak activity—specifically, the 1–2 hour delay in the morning peak of roe deer—constitutes a key behavioral adaptation to reduce direct competition for critical resources. Spatially, suitable habitats for both species were concentrated in the central reserve, and they demonstrated similar responses to the environmental factors driving their distribution. In conclusion, this study highlights that conservation strategies for roe deer and wild boars must prioritize the central core habitat quality and address the nonlinear impacts of environmental factors and human activities on species distribution and interspecific relationships.

Key words: Pangquangou National Nature Reserve, activity rhythm, spatiotemporal distribution, species coexistence, ensemble model