A CYGNSS-Based Algorithm for the Detection of Inland Waterbodies

The Cyclone Global Navigation Satellite System (CYGNSS) is a new constellation of eight low Earth orbiting spacecrafts that receive both direct and reflected signals from GPS satellites. Coherent reflection of the GPS signal from standing water over land results in a high surface reflectivity signal in the CYGNSS data. An image processing algorithm is presented, which leverages the surface reflectivity signal to produce a watermask of inland waterbodies at 0.01 degrees x 0.01 degrees spatial resolution. The watermask is compared to hand-drawn maps of inland waterbodies, as well as to the MODIS watermask product. We find that the CYGNSS watermask provides accurate, time-varying maps that are able to resolve changes in lake and river position and extent. With CYGNSS' short return time, watermasks can be generated using as little as half a month of data to produce near-real-time maps of flooding events.