Monitoring melt pond using Tianmu-1 GNSS-R Data: A Wind-concerned Model study

Melt ponds play a critical role in regulating the surface albedo of Arctic sea ice and accelerating its melt through the ice–albedo feedback mechanism. However, their high spatial heterogeneity and rapid temporal evolution make large-scale, continuous monitoring extremely challenging. Spaceborne optical remote sensing remains the primary technique for retrieving melt pond fraction (MPF), but its effectiveness is severely limited under persistent cloud cover and polar night conditions. Although GNSS-R provides all-weather observations with high temporal resolution, its potential for melt pond monitoring has not yet been systematically evaluated, nor have practical monitoring strategies been established. This study evaluates the potential of spaceborne Global Navigation Satellite System Reflectometry (GNSS-R) for melt-pond monitoring and characterizes the mechanisms through which melt-pond surface properties influence the reflected GNSS-R signals. An electromagnetic forward scattering model was developed to simulate GNSS-R reflectivity as a function of MPF and open water fraction (OWF) in representative summer sea ice scenes. The model was validated using observations from the Tianmu-1 GNSS-R satellite and the optical melt pond data. We evaluated the model performance using pan-Arctic data on three distinct dates representing different stages of melt pond development: June 15, July 1, and August 15, 2023. The modeled reflectivity shows strong agreement with GNSS-R observations, yielding Pearson correlation coefficients of interval means values of 0.99, 0.97, and 0.93, and corresponding unbiased RMSE (ubRMSE) values of 0.76 dB, 1.91 dB, and 1.18 dB, respectively. The results demonstrate the potential of using GNSS-R for melt pond monitoring, supporting the development of GNSS-R–based MPF retrieval algorithms and fusion approaches that integrate traditional remote sensing data.