摘要
高密度城市扩张过程中,极端气候事件频发,已严重威胁居民健康。厘清该过程中绿色基础设施(GI)降温效应的时空异质性及其关键影响因素是建设凉爽城市的重要依据。以上海市(2005—2020年)为例,基于GI降温效应与城市扩张特征量化结果,级联回归树与SHAP模型探究GI变化类型中,影响其降温效应的关键因素与阈值。结果显示,GI呈现植被净初级生产力(NPP)下降面积提升与NPP提升形态破碎2种主要类型。不透水面占比、建筑密度与人口密度是影响GI降温效应的关键要素,阈值区间分别为30%~50%、50%~65%和10%~20%。相关结果旨在为城市的可持续发展与推动高温韧性规划提供理论依据与实践指导。
Rapid urbanization over the past few decades has led to the extensive conversion of natural landscapes such as forests,wetlands,and lakes into impervious artificial surfaces.This transformation contributes to the irreversible loss of soil,biodiversity,and ecosystem functions,while exacerbating the Urban Heat Island(UHI)effect,which poses a significant threat to public health.Green Infrastructure(GI),encompassing the network of natural and semi-natural green and blue spaces,is widely recognized for its cooling benefits.However,highdensity urban expansion often leads to the fragmentation of GI,diminishing its cooling efficacy.This issue is particularly pronounced in rapidly developing regions like China's Yangtze River Delta.While existing research has explored the spatial patterns and mechanisms of GI's cooling effects,a comprehensive understanding of the dynamic,multi-dimensional impacts of urban expansion over time remains limited.This study addresses this gap by examining Shanghai,a megacity that has undergone dramatic urbanization,as a case study.It investigates the dynamic influence of three-dimensional urban expansion on the cooling effect of GI from 2005 to 2020.First,it identified five typical patterns of GI landscape change during this period using K-medoid clustering based on variations in GI area,spatial configuration(fractal dimension),and quality(Net Primary Productivity,NPP).These patterns were categorized as:1)low-quality expansion compensation,2)cooling-gain fragmentation,3)comprehensive degradation-induced heating,4)reduction-regularization-induced heating,and 5)edge-shrinkage-induced warming.Subsequently,the study developed a quantitative framework of urban expansion indicators,encompassing both construction-related metrics(impervious surface area,building density,floor area ratio)and socio-economic factors(population density,nighttime light index).By integrating a Gradient Boosting Regression Tree(GBRT)model with the Shapley Additive exPlanations(SHAP)algorithm,it quantified the relative importance of these urban expansion factors on GI's cooling intensity for each of the five change patterns and identified their critical impact thresholds.The results reveal significant spatial heterogeneity in the evolution of GI and its cooling effect.The cooling intensity was observed to decrease in 49.1%of the study area,particularly in newly developed towns and mixed-use commercial-residential zones.The key drivers weakening the cooling effect varied across different GI change patterns.Overall,the proportion of impervious surface area,building density,and population density were identified as the most significant influencing factors.The analysis revealed critical thresholds for these factors;for instance,the negative impact on cooling becomes substantially more pronounced when the increase in impervious surface area exceeds a range of 30%-50%,the increase in building density surpasses 50%-65%,and the increase in population density goes beyond 10%-20%.Furthermore,for areas where GI is becoming more regularized in shape but reduced in size,the study found that controlling building density increases to below 65%while simultaneously increasing the floor area ratio by over 25%can effectively enhance the cooling effect by promoting more compact,vertical development.The nighttime light index also emerged as a crucial secondary factor,particularly in areas with degrading GI,highlighting the role of anthropogenic heat emissions.This study provides a quantitative,evidence-based framework for developing targeted urban planning and management strategies to mitigate urban heat.By understanding the specific thresholds of key urban expansion indicators under different GI evolution scenarios,policymakers can implement more precise and effective interventions,such as controlling impervious surfaces,optimizing building layouts,and managing anthropogenic heat,to build cooler,more resilient cities.Future research should incorporate multi-temporal data and explore the synergistic effects of heatwaves and different GI types to further refine these strategies.
作者
黄俊达
王云才
HUANG Junda;WANG Yuncai(College of Architecture and Urban Planning,Tongji University,Shanghai 200092;Tongji University,Shanghai 200092)
出处
《中国园林》
北大核心
2025年第8期29-35,共7页
Chinese Landscape Architecture
基金
国家自然科学基金重点项目(52238003)。
