Floods are some of the most devestating natural disasters worldwide. While the U.S. has a robust flood warning system built upon a network of stream gaging stations, many other countries are not so lucky. It may be possible to use the increasing number of satellites to monitor and predict storms and associated flood risk in these data-scarce areas. Many satellites, however, may only pass over a specific part of the globe once every several days or weeks — much too long to monitor rapidly changing storm systems. Coordinating multiple satellites to examine the same area would increase the frequency of incoming data, but it is still unclear how best to schedule a network of satellites for this task.
We are working with a group at NASA on developing such a scheduling algorithm. Well, they are working on the algorithm but we are contributing the flood modeling expertise. We are determining what sort of information a satellite might “see” that would indicate flooding could occur in the near future. In addition, we are modeling flood events to show how having more frequent satellite data could improve model and forecast accuracy.