Spatio-temporal characteristics and driving mechanisms of cultivated land fragmentation, and implications for future land use: Empirical analysis from counties in China from 1990 to 2020

Cultivated land fragmentation (CLF) poses a great challenge to the sustainable utilization of farmland and food security in China. It is of great significance to systematically reveal the spatio-temporal characteristics and driving mechanisms of China's CLF for optimizing the utilization and management of cultivated land resources. However, there remains a gap in the literature on how CLF and its driving mechanisms change over continuous long-time series. This study identified these changes at the county scale in China and different agricultural regions from 1990 to 2020 by developing a new analytical framework for CLF integrating scale fragmentation, spatial fragmentation, and shape fragmentation. The results show that: (1) China's CLF exhibited a continuous and multi-dimensional strengthening trend during 1990–2020, with an average fragmentation index increasing by 3.17 %. The fractal attributes of CLF presented a pattern of "scale fragmentation > shape fragmentation > spatial fragmentation", with increases of 1.33 %, 3.85 %, and 5.04 % respectively. (2) The development of CLF and its fractal attributes in China in 1990–2020 both showed an increase in the east and a decrease in the west. Overall, economically developed areas with superior agricultural resource endowments such as the Huang-Huai-Hai Plain, and the central and eastern of the Middle-lower Yangtze Plain, etc. are higher-value clusters for CLF enhancement, with growth rates mostly exceeding 13.48 %. (3) CLF is the combined result of multiple factors, and the drivers of CLF vary significantly across different agricultural regions. The average explanatory power of various factor categories on CLF in 1990–2020 was ranked as "physical geographic environment > land use patterns > socio-economic development > spatial location conditions > land distribution process > distribution of linear features", with different impact intensities and trends. (4) The interaction between any pair-wise factors has a stronger impact on CLF than single factors. Approximately 62.54 % of nonlinear enhancement and 46.92 % of two-factor enhancement are both related to land use, highlighting the critical and complex impact of land use on CLF. Consequently, we appeal that the alleviation of CLF cannot be achieved through a single strategy, but especially the joint efforts on the policy system construction for optimizing utilization and management of key land resources in China, as well as the appropriate strategies for reducing CLF in typical agricultural regions based on regional differences. This study can provide essential insights for deepening the understanding of the spatio-temporal evolution mechanism of CLF at the national scale, and for policymakers to formulate differentiated CLF governance strategies.