气候变化领域集成服务门户(CCPortal): Fine-scale modeling of the urban heat island: A comparison of multiple linear regression and random forest approaches

CCPortal

DOI	10.1016/j.scitotenv.2021.152836
	Fine-scale modeling of the urban heat island: A comparison of multiple linear regression and random forest approaches
	Oukawa, Gabriel Yoshikazu; Krecl, Patricia; Targino, Admir Creso
发表日期	2022
ISSN	0048-9697
EISSN	1879-1026
卷号	815
英文摘要	Characterizing the spatiotemporal variability of the Urban I lent Island (UHI) and its drivers is a key step in leveraging thermal comfort to create not only healthier cities, but also to enhance urban resilience to climate change. In this study, we developed specific daytime and nighttime multiple linear regression (MLR) and random forest (RF) models to analyze and predict the spatiotemporal evolution of the Urban I lent Island intensity (UHII), using the air temperature (T-air) as the response variable. We profited from the wealth of in Situ T-air data and a comprehensive pool of predictors variables - including land cover, population, traffic, urban geometry, weather data and atmospheric vertical indices. Cluster analysis divided the study period into three main groups, each dominated by a combination of weather systems that, in turn, influenced the onset and strength of the UHII. Anticyclonic circulations favored the emergence of the largest UHII (hourly mean of 5.06 degrees C), while cyclonic circulations dampened its development. 'I'he MLR models were only able to explain a modest percentage of variance (64 and 34% for daytime and nighttime, respectively), which we interpret as part of their inability to capture key factors controlling T-air. The RE models, on the other hand, performed considerably better, with explanatory power over 96% of the variance for daytime and nighttime conditions, capturing and mapping the fine-scale T-air spatiotemporal variability in both periods and under each cluster condition. The feature importance analysis showed that the meteorological variables and the land cover were the main predictors of the Urban planners could benefit from these results, using the high-performing RF models as a robust framework for forecasting and mitigating the effects of the UHI.
英文关键词	Multiple linear regression; Random forest; Machine learning; SHAP values; Explainable artificial intelligence
语种	英语
WOS研究方向	Environmental Sciences
WOS类目	Science Citation Index Expanded (SCI-EXPANDED) ; Social Science Citation Index (SSCI)
WOS记录号	WOS:000743240100010
来源期刊	SCIENCE OF THE TOTAL ENVIRONMENT
文献类型	期刊论文
条目标识符	http://gcip.llas.ac.cn/handle/2XKMVOVA/280857
作者单位	Universidade Tecnologica Federal do Parana; Universidade Tecnologica Federal do Parana
推荐引用方式 GB/T 7714	Oukawa, Gabriel Yoshikazu,Krecl, Patricia,Targino, Admir Creso. Fine-scale modeling of the urban heat island: A comparison of multiple linear regression and random forest approaches[J],2022,815.
APA	Oukawa, Gabriel Yoshikazu,Krecl, Patricia,&Targino, Admir Creso.(2022).Fine-scale modeling of the urban heat island: A comparison of multiple linear regression and random forest approaches.SCIENCE OF THE TOTAL ENVIRONMENT,815.
MLA	Oukawa, Gabriel Yoshikazu,et al."Fine-scale modeling of the urban heat island: A comparison of multiple linear regression and random forest approaches".SCIENCE OF THE TOTAL ENVIRONMENT 815(2022).