Socioeconomic development, livestock industrial structure, and environmental pressure in China: spatiotemporal evolution and spatial associations of ammonia emissions

Background Provincial trends, regional disparities, and spatial associations in livestock-related ammonia (NH 3 ) emissions in China remain incompletely characterized. Methods A total ammoniacal nitrogen mass-flow inventory was constructed for 31 Chinese provinces during 2005–2024. Trends, disparities, spatial clustering, and conditional associations were assessed using the Mann–Kendall test and Sen’s slope, GE (0), Global Moran’s I , and a two-way fixed-effects spatial Durbin model (SDM). Results National emissions were 19.53% lower in 2024 than in 2005, although the national monotonic trend was not significant. Emissions decreased significantly in the high livestock-economic region and increased significantly in the medium region. GE (0) increased from 0.37 to 0.45, primarily because of widening within-region disparities, while geographical Moran’s I increased from 0.161 to 0.268. Baseline SDM associations of the manure nitrogen supply–demand ratio (MNSR) and livestock-oriented industrial structure with aggregate emissions weakened after controlling for standardized livestock units (SLU); their total-effect p-values were 0.078 and 0.969, respectively. With NH 3 emissions per SLU as the outcome, MNSR was insignificant and industrial structure was negatively associated with emission intensity. Although the spatially lagged covariates were jointly informative, the baseline spatial autoregressive coefficient was insignificant ( p = 0.577 ), and indirect associations were sensitive to the spatial-weight specification. Conclusion The results support regional divergence, widening within-region inequality, and spatial clustering, but not a stable interprovincial emission-spillover mechanism or scale-independent associations of MNSR and industrial structure. Region-specific assessment of manure nutrient balance and emission intensity may help inform, but does not itself establish, mitigation priorities.