Work

Quantifying the mitigating effect of urban eco-efficiency on canopy urban Heat Islands in China: Evidence from multi-source remote sensing data

Year: 2026

Type: article

Abstract: Urban heat exposure is intensifying in rapidly urbanizing regions, but whether urban eco-efficiency (UEE) can measurably reduce canopy urban heat island intensity remains insufficiently understood. Using a panel dataset of 279 Chinese prefecture-level cities from 2005 to 2020, this study constructs a dataset anchored in remote sensing observations by integrating gridded surface air temperature, MODIS land cover and NDVI products, nighttime light data, socioeconomic statistics, and pollutant emission indicators. Canopy urban heat island intensity, hereafter referred to as UHI intensity, is measured using a refined Dynamic Equal-Area method, while UEE is estimated with a super-efficiency SBM model that incorporates undesirable outputs. We combine two-way fixed effects and instrumental-variable models with moderation, component-level, spatial econometric, directed investment-network, MGTWR, and XGBoost-SHAP analyses to examine average effects, mechanisms, spatial spillovers, heterogeneity, and nonlinear drivers. Nationally, UEE increased by 15.6%, whereas overall UHI, daytime UHI, and nighttime UHI intensity rose by 42.6%, 46.91%, and 34.02%, respectively, indicating that improvements in UEE have not fully offset rising urban heat intensity. Baseline regressions show that higher UEE is significantly associated with lower UHI intensity, with the strongest cooling effect observed at night. Mechanism tests indicate that labor productivity strengthens the cooling effect of UEE, energy-related pollution intensity weakens it, and NDVI plays a context-dependent role in ecological regulation. Spatial models confirm a robust local cooling effect but reveal channel-dependent spillovers: peer effects may facilitate coordinated heat mitigation, whereas directed secondary-industry investment linkages suggest a possible redistribution of heat burdens from core cities to peripheral cities. XGBoost-SHAP results further show that daytime UHI is dominated by anthropogenic and morphological amplification, while nighttime UHI is more strongly constrained by hydroclimatic conditions. These findings suggest that UEE is a necessary but insufficient condition for urban thermal resilience. Effective mitigation requires cleaner energy use, productivity upgrading, targeted green infrastructure, and regionally coordinated heat governance.

Source: Ecological Indicators

DOI: 10.1016/j.ecolind.2026.115068

Authors: Jizhong Deng, Kai-sen Nong, Jiaju Guo, Youyu Luo, Jiaan Lin, Dongqi Sun