Abstract: The natural protected areas in the Taihang Mountains play a critical role in ecosystem restoration, biodiversity conservation, and natural heritage protection, with consequence on the construction of ecological civilization in the north China. In this study, we analyzed the spatial patterns of the number and size of 300 natural protected areas (including national, provincial, and municipal levels) in Taihang Mountains using spatial analysis methods, including Moran’s index, multi-distance spatial clustering, and kernel density analysis. The multi-scale geographically weighted regression (MGWR) model was used to analyze the spatial heterogeneity of influencing factors, aiming to provide a scientific basis for the spatial optimization of natural protected areas. The results showed that: (1) The number of natural protected areas was significantly clustered in spatial distribution (I_global=0.9707). Wetland parks (I_wetland=0.7130) and scenic areas (I_scenic=0.5031) had a stronger degree of clustering than other types. (2) The area of natural protected areas formed a “double-core and one-belt” clustered pattern in space. Specifically, it was manifested as the northeastern highdensity core areas centered on Baoding City (Lushui County, Yixian County), the northwestern high-density core areas centered on Shijiazhuang City (Lingshou County) and Xinzhou City (Wutai County), and the core-belt formed by the linking of Xingtai County, Huguan County, and Jiyuan City as the core. (3) The influencing factors of the spatial pattern of the area of natural protected areas had significant spatial heterogeneity. Vegetation coverage rate and road network density had the strongest positive impact, while elevation and construction land area had the strongest negative impact. Therefore, the attribute characteristics of different types of natural protected areas should be distinguished in the planning and construction of natural protected areas. While the improvement of vegetation coverage in natural protected areas should be emphasized, we should pay attention to the positive impact of road network density and the inhibitory effect of construction land on natural protected areas in urbanized regions. In high-altitude areas, the number and scale of natural protected areas should be designed moderately.

Key words: natural protected area, spatial differentiation, pattern optimization, multi-scale geographically weighted regression (MGWR), Taihang Mountains