Spatial-temporal dynamics of grain yield and the potential driving factors at the county level in China

Understanding the spatial-temporal dynamics of grain production and the influencing factors at the county level in China may promote the knowledge of land-use management and local policymaking, which are conducive to food security and the sustainable development of society. This study aims to evaluate China's grain yield (GY) from 2000 to 2014 and investigate the potential driving factors (PDFs) that affect the spatial-temporal dynamics of GY, including land, labor force, capital, and macro-background. Specifically, the locational Gini coefficient and exploratory spatial data analysis (ESDA) were used to characterize the spatial patterns of GY and its correlations with PDFs. Spatial regression models (SRMs) were employed to investigate the spatial dependence of GY on each PDF in 2000, 2005, 2010 and 2014. Results reveal that China's grain production has been on the rise with high-yield regions distributed mainly within the northeastern agricultural regions. Moreover, the proportion of counties in the northeastern agricultural regions with high grain yield has increased, while the number of low-yielding counties has increased in other agricultural regions. This finding highlights the increasing trend of spatial polarization in grain production. The significant bivariate Moran's I (p < 0.05) further revealed a global spatial spillover effect in the spatial correlation of GY and four PDFs. The spatial correlations could be categorized into four types: high GY and high PDFs, high GY and low PDFs, low GY and high PDFs, and low GY and low PDFs. SRMs were capable of quantifying the spatial dependence of GY on various PDFs, thereby revealing that land factors had a substantial effect on the grain production dynamics nationwide. The exploration of the spatial relationships between GY and PDFs provide a reference for formulating scientific and reasonable agricultural policies. (C) 2020 Elsevier Ltd. All rights reserved.