Häftad, Engelska, 2027
1 622 kr
Kommande
AI and Remote Sensing for Monitoring, Prediction, and Mitigation of Urban Climate Risks provides a comprehensive exploration of the urban heat island (UHI) phenomenon, and the innovative technologies employed to address its challenges.The book begins by outlining the UHI crisis, highlighting the detrimental effects of urbanization on local climates, particularly the increased temperatures in urban environments. It emphasizes the role of remote sensing technologies and diverse data sources in analyzing UHIs, laying the groundwork for advanced research. The text then looks into AI-driven methodologies, showcasing how machine learning and deep learning techniques can enhance urban land classification, estimate urban 3D morphology, and retrieve land surface temperatures. Additionally, it discusses the application of deep learning for predicting UHIs and identifying the driving factors behind their formation, providing valuable insights into urban climate dynamics. In its final sections, the book addresses practical applications of these technologies, including the detection and assessment of heatwave events and urban heat health risk evaluations. It also emphasizes the importance of urban green infrastructure as a mitigation strategy and discusses the transition from data analysis to policy formulation. This essential resource illustrates how AI and remote sensing can be harnessed to monitor, predict, and mitigate urban climate risks effectively, fostering more resilient and sustainable cities.Examines AI-driven urban climate analysis methods with step-by-step workflows for applying machine learning (e.g., CNNs, transformers) to LST retrieval, fusion, and UHI prediction-with reproducible Python code and model weightsDetails policy-ready case studies with risk frameworks linking UHI patterns to actionable mitigation strategies (e.g., green infrastructure ROI, heat-health early warnings)Explains AI for UHI interpretability, including SHAP values, attention maps, and other AI tools to decode why models predict heat risks (e.g., "Park coverage reduces LST by 2�C vs. asphalt")