Understanding Urban Redevelopment Patterns in city: A Case Study of Toronto, 2016-2023.

Keywords: Machine learning, Change detection, Urban redevelopment, Spatial logistic regression
Abstract Type: Paper Abstract

Abstract

Urban environments are in a constant state of transformation. This transformation inevitably brings along challenges due to growing population and economic development. Urban redevelopment as a key mechanism of this transformation involves the renovation, replacement or reuse of existing structures and is key to the changes in the urban environment. Toronto, Ontario has undergone rapid urbanization and development over the years, resulting in notable changes in land use and land cover (LULC). This study utilizes advanced remote sensing technology with Sentinel-2 multispectral satellite images and Light Detection and Ranging (LiDAR) information to investigate LULC changes from 2016 to 2023 using supervised machine learning classification. Remote sensing techniques including high-resolution enhancement and texture analysis are adopted to enhance classification accuracy. A spatial logistic regression is used to analyze the factors that drive the LULC change on a pixel level, based on the Alonso-Muth-Mills model, rent gap theory and real option theory. This study contributes to our understanding of the dynamics of urban redevelopment, offering empirical evidence that supports the theories and pertinent information for urban planners and policymakers for the sustainable development of the city.