Habitability functions are developed to estimate the probability of a building becoming uninhabitable due to coastal flooding. These functions are created by combining a Hurricane Irma flood model with cell phone data showing which buildings people returned to following Irma. We find that modeled flood depth is the best predictor of building habitability. By quantifying the dependence of building habitability on flood hazards, this work improves how coastal communities prepare for flood events.

We present a standardized, bottom-up framework to estimate slum populations at neighborhood level across 129 Global South countries. Our approach addresses underestimations in prior studies that rely heavily on slum geometry. Our dataset offers the first comprehensive inventory in data-sparse settings. This research offers valuable insights to support sustainable urban development goals, inform humanitarian aid distribution, and enhance the well-being in underserved communities.

Deep learning methods are increasingly used as surrogates for spatio-temporal flood models but struggle with generalization and speed. Here, we propose a multi-resolution approach using graph neural networks that predicts dike breach floods across different meshes, topographies, and boundary conditions with high accuracy and up to 1000× speed-ups. The model also generalizes to larger more complex case studies with just one additional simulation for fine-tuning.

There are indications that typhoon intensities may increase under global warming. However, further research on these projections and their uncertainties is necessary. We study changes in typhoon intensity under SSP5-8.5 for seven events affecting the Pearl River Delta using Pseudo-Global Warming and a storyline approach based on 16 CMIP6 models. Results show intensified wind speed, sea level pressure drop and precipitation levels for six events with amplified increases for individual storylines.

Tropical cyclones (TCs) often cause high economic loss due to heavy winds and rainfall, particularly in densely populated regions such as the Pearl River Delta (China). This study provides a reference to set up regional climate models for TC simulations. They contribute to a better TC process understanding and assess the potential changes and risks of TCs in the future. This lays the foundation for hydrodynamical modelling, from which the cities' disaster management and defence could benefit.