Assessing urban environmental quality and optimizing governance path based on fixed-reference-frame
摘要
Scientific assessment of urban environmental quality underpins differentiated governance decisions.This study takes the environmental data of 31 major Chinese cities from 2011 to 2023 as its sample and,within the DPSIR framework,improves the method in three respects.First,a longitudinal entropy-weight TOPSIS with a fixed reference frame is constructed,in which the normalization bounds,indicator weights,and positive and negative ideal solutions are held fixed over the entire study period to ensure the cross-year comparability of scores;drawing on stochastic multicriteria acceptability analysis(SMAA),bootstrap resampling is applied to quantify the uncertainty of composite scores and ranks,thereby upgrading point rankings into robust rankings with 95%confidence intervals and an acceptability index for the excellent grade.Second,a dual-scale nested evaluation framework is proposed,in which city-level local indicators and provincial-level contextual indicators are scored separately,so that a city's local performance is distinguished from its provincial development background and the bias caused by scale mixing is corrected.Third,under a governance cost budget constraint,sequential linear programming is used to solve for the minimum cost and the corresponding indicator-improvement portfolio required for each city to reach the excellent threshold.The results show that the composite scores of the 31 cities range from 0.234 to 0.537 over the study period,and only 7 cities,including Guangzhou,Lhasa,and Hangzhou,robustly attain the excellent grade at the 95%confidence level.Environmental quality improves statistically significant improvement in 30 cities,with no significant deterioration observed in any city,yet inter-city relative gaps show no sign of convergence.The dual-scale decomposition identifies background-dividend cities such as Guangzhou,Hangzhou,and Beijing,whose high composite rankings are supported mainly by favorable provincial conditions while local governance still lags behind.Governance path optimization further indicates that the remaining 24 cities can all reach the excellent threshold within their feasible improvement space,and their minimum compliance costs,measured in normalized improvement distances,form a gradient from 0.05 to 0.30,which can be used to prioritize the allocation of governance resources.By extending multi-criteria evaluation from descriptive ranking to a governance decision-support framework that integrates uncertainty characterization,scale consistency,and prescriptive optimization,this study provides methodological and empirical references for differentiated decision-making in urban environmental governance.