Multidimensional evaluation and driving factors detection of cultivated land fragmentation based on plot-household scale

Cultivated land fragmentation is often formed to divide the limited cultivated land resources into the "plots". Their distribution to "households" can follow the "combining fertile and infertile plots and pairing distant and nearby ones". Thus, the "plots" and "households" can serve as the smallest independent farming and ownership units in the current agricultural production, respectively. They are also the most direct evaluation units to represent the cultivated land fragmentation. However, existing studies focus predominantly on the degree and causes of the cultivated land fragmentation from a single landscape perspective or that of farmers. Particularly, the refined remediation is required to formulate the spatial differentiation of the cultivated land fragmentation from multiple dimensions. In this study, an evaluation framework was constructed for the cultivated land fragmentation from the resource scale, spatial agglomeration, production and ownership division. The "plot-household" was taken as the smallest connection unit to represent the cultivated land fragmentation. An empirical study was also conducted using multi-source survey data on the land use in Minqing County, Fujian Province, China. The results were as follows: 1) There were the favorable spatial agglomeration degree and production, while the cultivated land resource scale was relatively poor, and the ownership division was rather severe. It infers that the solid foundation was provided for the modern agriculture. The ownership division was the key factor contributing to the cultivated land fragmentation. 2) The fragmentation index of the cultivated land was ranged from 0.320 to 0.643. There were the substantial differences in the quantitative structure and regional distribution of the cultivated land fragmentation over different dimensions. Consequently, it was essential to adopt the appropriate remediation models, such as the "paid withdrawal", "spatial replacement", "merging small plots into large ones", and "one household, one plot". Additionally, it was very necessary to implement some policies, like converting sloping farmland into terraced fields, returning farmland to forests, and forest-farmland replacement. Engineering measures should be utilized to optimize, such as the field roads, irrigation and drainage systems. Ownership adjustment and cultivated land transfer should be innovated to effectively solve the multi-dimensional cultivated land fragmentation. 3) Cultivated land fragmentation in the different dimensions exhibited the distinct spatial agglomeration. Natural background, socio-economy situation, policy and institutional factors shared the varying degrees of influence on the cultivated land fragmentation. Moreover, the interaction among these influencing factors was mainly characterized by nonlinear enhancement and bi-variable enhancement. Some recommendations were also proposed to effectively promote the remediation of the cultivated land fragmentation. A land transfer market can be cultivated to balance the fairness and efficiency. A two-way interactive fragmentation remediation model can be established for the shift of the mindset regarding cultivated land fragmentation remediation from a passive-response to an active-response mode. This research can offer the valuable references to explore the diverse models of the cultivated land fragmentation remediation in the hilly regions of the southeastern coast.