探究中国林地细碎化的演变特征及影响因素对推动林业高质量发展、提升森林生态系统服务价值具有重要意义。论文基于1992—2022年土地覆被数据，利用多维景观格局指数构建了林地细碎化指标体系，并通过空间统计和机器学习模型分析了中国林地细碎化的时空演变趋势及驱动因素。研究发现：① 1992—2022年间中国林地细碎化指数以2004年为分界点，呈现“缓慢下降—波动中上升”的变化趋势。林地斑块的规模性、边缘密集性和形状规则性均呈现“下降—轻微上升”的趋势，空间离散性则表现出“低—高—低—高”的阶段性波动特征。② 在空间格局上，林地细碎化呈现“南高北低”的特征，东南低山丘陵林区的细碎化水平最高，热带和西南高山林区次之，西北高山和东北内蒙古林区的细碎化水平最低；在省级层面上，林地细碎化指数呈现以贵州省为中心向四周递减的分布格局，广西、广东、湖北、湖南、云南等省细碎化趋势也较为明显。③ 林地细碎化的驱动因素具有阶段性和多样性。在研究期内，森林抚育对林地细碎化具有减缓效应，而人口密度和气温上升对林地细碎化有加剧效应；1992—2001年气温和木材产量是影响林地细碎化的主要因素；2002—2011年耕地面积和集体林面积对林地细碎化影响的重要性上升；2012—2022年集体林面积是影响林地细碎化的首要因素。④ 林地细碎化驱动因素的交互效应分析表明，在降水量少、森林抚育强度低的地区过度造林会加剧林地细碎化，降水量较大、气温较高的地区林地细碎化趋势更强，而造林活动在一定程度上可以缓解耕地扩张对林地细碎化的负面影响。研究有助于深化对中国林地空间格局演变规律的理解，识别不同区域林地细碎化的驱动特征，为优化林地管理政策、推进集约化经营以及提高森林生态系统服务功能提供理论依据和决策支持。

Exploring the evolution characteristics and influencing factors of forest land fragmentation in China is of great significance for promoting high-quality development in the forestry sector and enhancing the ecosystem service value of forests. Based on land cover data from 1992 to 2022, this study constructed a comprehensive forestland fragmentation indicator system using multidimensional landscape pattern metrics and analyzed the spatial-temporal trends and drivers of forestland fragmentation across China with spatial statistics and machine learning models. The findings are as follows: 1) In the study period, China's forestland fragmentation indices showed a shift around 2004, with a trend of gradual decline-fluctuating increase. Forest patch size, edge density, and shape regularity exhibited a decline-slight increase trend, while spatial dispersal followed a cyclical low-high-low-high fluctuation. 2) Spatially, forestland fragmentation was high in the south and low in the north, with the highest levels in Southeast China's low mountains and hilly forests, followed by tropical and southwestern mountainous forests, and the lowest in the northwestern mountains and the Inner Mongolia and Northeast forest regions. At the provincial level, fragmentation values peaked in Guizhou, gradually declining outward, with significant fragmentation also observed in Guangxi, Guangdong, Hubei, Hunan, and Yunnan. 3) The drivers of forestland fragmentation varied across periods and regions. Throughout the study period, forest management practices generally mitigated fragmentation, while increasing population density and rising temperatures intensified it. From 1992 to 2001, temperature and timber production were the primary influences on fragmentation; from 2002 to 2011, cropland area and collective forest area became more impactful; and from 2012 to 2022, collective forest area became the leading factor. 4) An examination of the interaction effects among the drivers revealed that in areas with low precipitation and limited forest management intensity, excessive afforestation can exacerbate fragmentation, while fragmentation trends are stronger in regions with higher rainfall and temperatures. Afforestation helps mitigate the adverse impacts of cropland expansion on forestland fragmentation.This study deepens the understanding of forestland spatial pattern evolution in China, identifies regional drivers of fragmentation, and provides a theoretical basis and policy support for optimizing forestland management, promoting intensive use, and enhancing forest ecosystem services.