Abstract
Reducing agricultural carbon emissions (ACE) requires understanding how urban–rural transformation and land reallocation jointly reshape agricultural production. Using county-level data from Hubei Province, China, for 2019–2021, this study examined the spatial and nonlinear associations among urbanization (URB), agricultural land transfer (ALT), and ACE through spatial autocorrelation analysis, spatial panel models, and geographical convergent cross-mapping. URB, ALT, and ACE exhibited significant spatial dependence, with high-value clusters concentrated mainly in central Hubei. ALT showed a significant level-dependent nonlinear association with URB. URB and ALT were also significantly associated with ACE in a level-dependent nonlinear manner, with negative linear and positive quadratic components, whereas their joint term showed the opposite pattern. The directional-linkage analysis further revealed both recurring and year-specific connections between multiple dimensions of URB and ALT. Taken together, these findings show that the relationships among URB, ALT, and ACE are spatially interdependent, multidimensional, and dependent on local development conditions rather than adequately represented by constant linear relationships. Agricultural carbon mitigation should therefore combine cross-county spatial coordination with differentiated policies that reflect local URB and ALT conditions, particularly in spatially clustered areas.