TZAI-HUNG WENLiao, Hao YuHao YuLiaoYang, Kai LingKai LingYangChen, Tzu Hsin KarenTzu Hsin KarenChen2023-05-182023-05-182021-01-019781119625865https://scholars.lib.ntu.edu.tw/handle/123456789/631173Under the impact of global warming, mosquito-borne diseases, such as dengue fever, are expanding geographically in subtropical metropolises. Rain-induced standing water bodies could become mosquito larvae-favorable environments and trigger disease transmission in highpopulated areas. However, it would be a labor-intensive campaign to survey possible mosquito breeding sites, such as water-filled containers and facilities. In this chapter, we propose a multilevel image analysis framework integrating multisource remote sensor data of weather and built environments from various spatial scales to identify urban standing water environments. It was used to locate potential standing water after heavy rain and also identify water-prone urban environmental characteristics along with different rainfall patterns. The multilevel modeling results indicate that urban areas with neighboring areas with low-densities, low-heights of buildings, or high-densities of greenness are more susceptible to having after-rain standing water. Moreover, a larger rainfall quantity under these building characteristics could promote the occurrence probability of standing water, whereas a higher rainfall intensity reduces the probability. In summary, our research framework has demonstrated the feasibility of integrating multiscale remote sensing imagery to detect after-rain standing water. The findings could also provide health authorities with reference guidelines to geographically prioritize the targets of mosquito larval control in urban areas.[SDGs]SDG3[SDGs]SDG11book part10.1002/9781119625865.ch192-s2.0-85152811182https://api.elsevier.com/content/abstract/scopus_id/85152811182