Growing Resilience : A Land Use Decision-Making Framework Integrating Flood Risk Prediction and Land Use Prediction

Flooding has become more common due to global climate change, challenging urban resilience. However, there needs to be more research on hedge flood risk while conducting long-term series and refined land use development. To bridge this gap, we built a resilient land-use decision-making framework that integrates flood risk and land use prediction. For the Min Delta Urban Agglomerations (MDUA) in China, we used a multi-source data set, a random forest (RF) algorithm, and a patch-generating land use simulation model (PLUS) with coupled cellular automata (CA) to identify conflicting patches critical to urban expansion resilience. Three risk warning levels of low-risk, medium-risk, and high-risk locations, along with seven categories, were formed following the flood risk attributes. The best land use approach was determined by assessing quantity, spatial distribution, and risk-driven priority ranking of conflicting patches. The findings demonstrate that conflicting patches are mostly found in bare land within built-up areas, urban-rural intersection zones, and remote suburban areas with high traffic accessibility. These apply to three different land use patterns, including intra-city renewal, urban edge expansion, and the creation of derived cities. Throughout various stages of growth, each type of conflicting patch retains a fairly constant quantitative structure and spatial distribution characteristics. By integrating flood risk and land use prediction, this land-use decision-making framework can assist cities or potential urban regions that have yet to experience significant flooding in avoiding water problems and derivative issues. Different geographical areas and disasters can also benefit from the decision-making process for resilient land use