Spatial temporal trends and inequality in agricultural eco-efficiency under carbon constraints in China

Eco-efficiency in cultivated land use is crucial for ensuring food security and promoting sustainable agriculture amidst the challenges posed by urbanization and climate change. Distinct from much of the existing literature, this paper adopts a super-efficiency EBM model that integrates both radial and non-radial perspectives to estimate the eco-efficiency of cultivated land use and identify sources of inefficiency under carbon emission constraints, focusing on 180 prefecture-level cities in China's major grain-producing regions. Additionally, spatial kernel density estimation is applied to analyze the spatiotemporal dynamics and long-term trends in eco-efficiency, while the Dagum Gini coefficient is used to examine the causes of spatiotemporal disparities. The key findings include (1) The Songhua River Basin shows significantly higher eco-efficiency (mean: 0.718) than the Yellow River (mean: 0.559) and Yangtze River Basins (mean: 0.587), with distinct evolutionary patterns; (2) Long-term evolution reflects a bipolar pattern with spatial agglomeration disparities; (3) Positive spatial spillover effects are evident in regions with efficiency levels between 0.4 and 0.9, with inter-regional differences and super-variable density as significant spatial variation sources. This study reveals a "long-term increase - short-term decline" trend in ecological efficiency, highlighting carbon emissions as the primary limiting factor, although this constraint is gradually diminishing.