Abstract With the increasing intensity and frequency of flood events worldwide, the need for accurate and timely inundation mapping has never been more critical. Large‐scale flood extent estimations are vital for coordinating effective disaster response, facilitating recovery, and building future resilience. Traditional ground‐based and aerial monitoring methods are often impractical during major floods, limited by cost, safety, and their inability to capture the full scope of an event. Satellite‐based remote sensing provides the necessary large‐scale perspective with a unique vantage point to monitor extreme inundation events. This review assesses the potential of public satellite sensors to capture flood events using a novel analysis of the Dartmouth Flood Observatory (DFO) global flood database. Our analysis quantifies the major performance gaps between these sensors, demonstrating that no single instrument is sufficient for complete and continuous flood monitoring. Passive microwave radiometers are capable of capturing >95% of flood events, albeit at a coarse spatial resolution that may be unsuitable for detailed mapping or local risk assessment. In contrast, popular multispectral sensors such as Landsat and Sentinel‐2 capture no more than 30% of flood events. The number of sensors capable of capturing flood events doubled between 2015 and 2020, signaling immense potential for multi‐sensor integration. We examine how combining observations from multiple sensors can improve temporal coverage of flood events, however noting that temporal sampling along does not guarantee successful flood detection and how the rapid, dynamic nature of floods compounds the challenges inherent to satellite‐based monitoring.
Campo et al. (Sun,) studied this question.