ICEYE owns and operates the world’s largest constellation of synthetic aperture radar satellites. This allows us to reliably capture imagery from unexpected natural catastrophe events such as floods, enabling rapid response. However, a single satellite acquisition might take tens of seconds while flood events might last up to days, leaving significant gaps in the timelines of said events. In this talk, we’ll talk about techniques to propagate spatial and temporal information from satellite acquisitions via fitting models to earth observation and auxiliary data streams such as social media and river gauge data. With modern software and hardware, we can scale our techniques to create the thousands of flood simulations necessary to find promising candidate simulations to fit our observations, while the combination of compute and independent direct observations can be used to fill in the gaps.