Zhang, X., Zheng, H., Lu, J., Cheng, D., Lou, Y., Delang, C.O., Gomez, C., Zhang, Z., He, H., Fan, Y., 2027. Environment-conditioned vegetation carbon-production performance informs restoration assessment across China. Environmental Impact Assessment Review 123, 108812 10.1016/j.eiar.2026.108812
Vegetation greenness and gross primary production (GPP) are widely used to assess large-scale ecological restoration. These measures describe how vegetation has changed, but their strong dependence on climate and terrain complicates comparisons among regions. We examined whether the widespread greening observed in China was accompanied by higher carbon-production performance relative to local environmental conditions. We developed a gross primary production performance index (GPI) that compares observed GPP with an output-oriented production frontier derived from environmentally similar observations within each climatic zone. Target-specific local frontiers were constructed using a fixed pooled 1982–2018 reference dataset. We applied GPI to 25-km grid cells across China from 1982 to 2018. National mean GPI rose from 0.408 to 0.453, but significant increases were confined to 28.9% of grid cells and 5.9% declined. The high-GPI state expanded from 29.7% to 37.8%, while 70.0% of cells remained in the same state. Annual correlations were stronger between GPI and GPP (ρ =0.69–0.88) than between GPI and the normalized difference vegetation index (NDVI; ρ =0.52–0.71), and GPI also produced different spatial rankings from GPP and NDVI in parts of China. Lagged Bayesian structural equation modelling associated precipitation with GPI mainly through soil moisture and vegetation state, but the pathways differed among climatic zones. China’s carbon-production performance improved during the study period, although the change was gradual and concentrated in particular regions. GPI identifies regional differences in carbon-production performance that are not fully captured by greenness or absolute productivity, providing additional information for restoration assessment.
