An Estimate of Effective Resolution for CYGNSS Surface Water Detection Revealed by Comparison with Multi-sensor Optical Data

The location of water on the Earth’s surface iscritical for a wide range of science and application uses. GlobalNavigation Satellite System reflectometry (GNSS-R) is a remotesensing technique with demonstrated potential for identifyingareas of the Earth’s surface that are inundated, given that GNSSR signals are sensitive to smooth surfaces, allowing for the detection of surface water. However, the signal is known to saturateeven at relatively low quantities of surface water within anobservational footprint, complicating the interpretation of GNSSR observations. To date, the minimal detectable size of a waterbody within the GNSS-R signal footprint has remained elusive.This manuscript examines this sensitivity of one GNSS-R mission,the Cyclone Global Navigation Satellite System (CYGNSS), byusing a high resolution surface water extent product as truth(Dynamic Surface Water Extent). This analysis reveals that forthe Tonle Sap Lake region in Cambodia, the lower limit of waterarea detectable by CYGNSS was 0.62 km2of water surfacearea within a 0.05 degree grid cell, or 2% surface water cover.This finding indicates that at least for this region, a probabilisticinterpretation of CYGNSS observations may be more appropriatethan a binary classification. Such an interpretation establishes afoundation for synergistic use of CYGNSS GNSS-R with othersurface water remote sensing technologies.