We present a new method to create spatial data using a generative adversarial network (GAN). Our contribution uses coarse and widely available geospatial data to create maps of less available features at the finer scale in the built environment, bypassing their traditional acquisition techniques (e.g. satellite imagery or land surveying). In the work, we employ land use data and road networks as input to generate building footprints and conduct experiments in 9 cities around the world. The method, which we implement in a tool we release openly, enables the translation of one geospatial dataset to another with high fidelity and morphological accuracy. It may be especially useful in locations missing detailed and high-resolution data and those that are mapped with uncertain or heterogeneous quality, such as much of OpenStreetMap. The quality of the results is influenced by the urban form and scale. In most cases, the experiments suggest promising performance as the method tends to truthfully indicate the locations, amount, and shape of buildings. The work has the potential to support several applications, such as energy, climate, and urban morphology studies in areas previously lacking required data or inpainting geospatial data in regions with incomplete data.
29 November 2022 in ISPRS Journal of Photogrammetry and Remote Sensing
Flood maps are essential for disaster preparedness, insurance and climate adaptation, yet US official maps cover only one-third of river channels. Here we present a deep-learning framework that learns from existing records to produce a spatially complete 30-m flood hazard map for the contiguous US, filling unmapped regions and updating legacy extents. We estimate that the 2023 national database omits 11 million people and 4.1 million buildings in flood zones, adding 69% and 81% beyond the official baseline respectively. Crucially, our socioeconomic analysis of 917 cities reveals that entirely unmapped areas disproportionately harbor highly vulnerable demographics, particularly the elderly and children, exposing them to unrecognized risks and compounding existing inequalities. By solving the trilemma between accuracy, scale, and cost in national scale flood mapping, our open-source method broadens public access to risk information, paving the way for targeted and equitable climate adaptation.
The mapping of flood-prone areas is crucial for disaster response and adaptation. With a flood map, communities, insurers, and planners can prepare and be insured against risk damages. However, hindered by high resource demand, open-access flood maps, such as the FEMA National Flood Hazard Layer (NFHL) in the US, suffer from incomplete information. Currently, only one-third of all river channels in the country is complete, leaving population and structures in the other two-thirds vulnerable. To address this data gap, we present a methodology for completing the missing flood zones in the FEMA NFHL based on generative AI. The model learns patterns from the mapped areas and generates flood maps at 30 meters resolution in the unmapped areas with up to 81% and 65% in national scale studies. With the new flood data covering all river channels, we found that there are 11 million exposed population, and 4.1 million buildings located in flood zones that are unmapped by the NFHL as of 2023. This quantifies the inequitable access to open flood information across the country. Additionally, we found that 40% of all urban areas are inadequately mapped by the NFHL, with 103 out of the total 917 Core-based statistical areas having little to no FEMA maps even though there are at least more than 1% of the population at risk. More alarmingly, demographic analyses have revealed that these cities also have the highest percentage of elderly and young populations, who are more vulnerable to crises. This study highlights the use of generative AIs in scaling up mapping processes. With the completed NFHL flood map, exposed communities that were previously left out of the NFHL can begin planning for flood resilience today.