Contiguous Cropland in China Promotes Novelty in Crop Planting Structures

Abstract

Understanding how cropland restructuring contributes to agricultural system transformation requires a clear representation of changes in crop planting structures. This study introduces “crop planting structure novelty” (hereafter, crop novelty), defined as the degree to which the current planting‐area structure of major crops deviates from a historical reference set. Unlike crop diversity, which reflects crop richness and area‐allocation evenness within the same period, crop novelty identifies whether current planting structures lack close historical analogs. The spatial patterns of contiguous cropland in China from 2010 to 2020 were analyzed, and crop novelty was quantified for rice, wheat, maize, and soybean. Extreme gradient boosting models with Shapley additive explanations identified factors associated with crop novelty, while partial least squares structural equation modeling quantified region‐specific direct and indirect pathways. Crop novelty was higher in eastern and northeastern China (0.13 and 0.08, respectively) than in northwestern and southwestern China (0.04 and 0.03), indicating greater departures from historical planting structures. Grid‐level contiguous cropland share had the largest predictive contribution and showed a threshold‐like pattern around 50%. In eastern and northeastern China, a positive direct effect was partially offset by a negative total indirect effect; both effects were positive elsewhere. Higher crop novelty indicates greater historical departure but does not necessarily imply greater crop diversity or improved sustainability. By complementing diversity metrics with a measure of historical non‐analogy, this framework distinguishes routine adjustments from more substantial structural transformations and supports region‐specific policies that account for restructuring capacity and environmental and management trade‐offs.