Identification of urban waterlogging indicators and risk assessment based on MaxEnt Model: A case study of Tianjin Downtown

Abstract

Waterlogging is one of the world's most dangerous climatic hazards, seriously limiting the safety and sustainable development of cities. Analysis of the factors influencing urban waterlogging and disaster risk assessment are of great importance for the prevention and control of waterlogging. The study constructs a framework for assessing Urban Waterlogging Risk (UWR) from four dimensions: natural condition, social capital, infrastructure and built environment, emphasizing the need to understand and address the vulnerabilities and risks faced by cities in terms of waterlogging. On this basis, the MaxEnt model is used to rank the contribution rate of the indicators and identify positive and negative correlations, as well as to assess the risk of waterlogging and identify waterlogging-prone areas, taking Tianjin Downtown as the research object. The results show that: (1) MaxEnt model has strong applicability for waterlogging risk assessment. (2) The elements with the highest impact on urban waterlogging are population density, impervious surface, precipitation, etc., and prove that the regional waterlogging risk increases with the increase of population, building density, and impervious surface. (3) The high-risk area of waterlogging in Tianjin Downtown is concentrated in the central part of the city, which is characterized by a high degree of development and construction and a dense population; the low-risk area is mainly the area with a low degree of construction at the edge of the city, as well as the area with a large green area or water bodies. In this study, a comprehensive framework for assessing UWR was developed, key factors affecting the spatial distribution of UWR were identified, and the MaxEnt model was used to perform the risk assessment with a high degree of accuracy. The study results can provide a theoretical reference for the urban disaster warning and land use optimization.