文献类型：期刊文献

英文题名：Driving factors of fragmentation in urban landscapes: Local contributions, spatial relationships, and causal effects

作者：Jiang, Simin[1] Feng, Fei[1] Zhang, Xinna[1] Xu, Chengyang[1] Jia, Baoquan[2] Lafortezza, Raffaele[1,3]

第一作者：Jiang, Simin

通信作者：Lafortezza, R[1]

机构：[1]Beijing Forestry Univ, Res Ctr Urban Forestry, Key Lab Silviculture & Forest Ecosyst State Forest, Beijing 100083, Peoples R China;[2]Chinese Acad Forestry, Res Inst Forestry, Beijing 100091, Peoples R China;[3]Univ Bari Aldo Moro, Dept Soil Plant & Food Sci Di SSPA, Via Amendola 165-A, I-70126 Bari, Italy

年份：2025

卷号：174

外文期刊名：ECOLOGICAL INDICATORS

收录：;EI(收录号：20251518226006);Scopus(收录号：2-s2.0-105002308300);WOS:【SCI-EXPANDED(收录号:WOS:001468951100001)】；

基金：This study is funded by the 5 & sdot;5 Engineering Research & Innovation Team Project of Beijing Forestry University (BLRC2023B06) and by the National Natural Science Foundation of China (No. 32401370) , the Beijing Natural Science Foundation (8212031). Furthermore, we would like to thank Lanqing Zhang for her enthusiastic help in the research.

语种：英文

外文关键词：Landscape fragmentation; Urbanization; Driving factors; Causal effects

摘要：Rapid urbanization significantly alters landscape patterns, leading to fragmentation with implications for biodiversity, ecosystem function, and human well-being. Understanding the drivers of fragmentation is crucial for developing sustainable urban planning strategies. This study investigates the spatial patterns and driving mechanisms of landscape fragmentation in Nanchang, China, a rapidly urbanizing city. We analyze landscape fragmentation patterns for 2000, 2010, and 2022 and assess the influence of nine natural and anthropogenic factors using a geographically weighted random forest (GWRF) model, spatial autocorrelation analysis, and structural equation modeling (SEM). Our results reveal significant spatial heterogeneity in landscape fragmentation, with high-fragmentation hotspots concentrated in the southeastern and central regions. The intensity of land cover change and human activity emerge as the dominant human factors influencing fragmentation, exhibiting strong spatial correlations and causal effects. land cover change intensity and human activity intensity account for 44.58 % and 34.78 % of the highest ranking local feature importance, respectively. The human footprint also demonstrates a statistically significant positive spatial correlation with fragmentation (mean Moran's I = 0.5304). Furthermore, land cover change intensity exerts a direct positive influence on landscape fragmentation, as indicated by an average standardized path coefficient of 0.41. Slope and human footprint also play important roles, primarily through indirect effects (mean value above 0.3 and 0.35, respectively). The influence of impervious surface expansion intensity showed an inverted "U" shape over time in spatial correlation and causal effect, suggesting that while initial urban expansion increases fragmentation, advanced urbanization and green space restoration can mitigate these effects.