Abstract
As a climate-aware and standardized classification scheme, the Local Climate Zone (LCZ) can be a research framework for analysis of effects of urban development on urban forms and Urban Heat Island (UHI) phenomenon in relation to urban planning. In this study, two new towns built with different urban planning in South Korea are selected as the study areas. In respect of the growing popularity of deep learning and advancements deep learning-based LCZ classification, we build a deep convolutional neural network architecture using a modified SE-ResNext50 backbone. As input data, we designed six schemes using different combinations of Sentinel-1 SAR and Sentinel-2 MSI imagery and thermal band from Landsat 9 is used for SUHI magnitude estimation and heat vulnerability. In addition, robust and quantitative data sampling using building surface fraction data and building height data was performed. The results indicate that combining SAR and multispectral data could increase LCZ classification accuracy and outline the capacity of polarimetric decomposition components. In addition, it was suggested that urban planning causes differences in the LCZ distribution of each new town, resulting in differences in SUHI magnitude and heat vulnerability. The research findings can help guide future urban development by considering the urban form and thermal environment according to the LCZ composition within the new plannedcity.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.