Detailed inventory of rainfall-induced landslides triggered by Typhoon Doksuri in July 2023 and county-scale spatial variability characteristics

Abstract In July 2023, persistent extreme rainfall induced by Typhoon Doksuri triggered widespread clustered landslides across North China. Based on a previously established high-resolution inventory containing more than 100,000 rainfall-induced landslides, this study investigates the spatial heterogeneity and statistical scaling characteristics of landslide inventories at the county scale. Sixteen counties with concentrated landslide occurrences were selected for systematic comparison of landslide abundance, spatial clustering, and landslide-size composition. The analysis combines kernel density estimation with cumulative frequency–area and probability density approaches to quantify both spatial clustering and statistical scaling behavior of landslide inventories. The results reveal pronounced spatial variations in landslide occurrence among the counties. The largest landslide numbers were observed in Mentougou District, Laishui County, Fangshan District, Huailai County, and Laiyuan County, whereas landslide densities showed substantial variations after normalization by county area. Landslide hotspots were predominantly distributed continuously along mountainous valleys and middle-to-high elevation hillslopes, exhibiting an overall spatial pattern characterized by dense landslide occurrence in mountainous areas and sparse occurrence across plains. The cumulative frequency–area distributions of landslides in all sixteen counties exhibited robust power-law scaling behavior (R 2 = 0.845–0.933), while the probability density distributions of landslide area were consistently well fitted by the Inverse Gamma distribution (R 2 > 0.96). These results indicate that county-scale landslide inventories share a common statistical scaling framework despite pronounced differences in power-law exponents, shape parameters, and scale parameters. Specifically, Huailai County, Quyang County, and Lingqiu County exhibited relatively smaller proportions of large landslides, whereas Mancheng District and Fanshi County were characterized by comparatively higher proportions of large landslides.