Greenland Ice Sheet surface roughness from Ku- and Ka-band radar altimetry surface echo strengths

Abstract. Surface roughness is an important factor to consider when modelling mass changes at the Greenland Ice Sheet (GrIS) surface (i.e., surface mass balance, SMB). This is because it can have important implications for both sensible and latent heat fluxes between the atmosphere and the ice sheet and near-surface ventilation. While surface roughness can be quantified from ground-based, airborne, and spaceborne observations, satellite radar datasets provide the unique combination of long-term, repeat observations across the entire GrIS and insensitivity to illumination conditions and cloud cover. In this study, we investigate the reliability and interpretation of a new type of surface roughness estimate derived from the analysis of Ku- and Ka-band airborne and spaceborne radar altimetry surface echo powers by comparing them to contemporaneous laser altimetry measurements. Airborne data are those acquired during the 2017 and 2019 CryoVEx (CryoSat Validation Experiment) campaigns, while the satellite data (ESA CryoSat-2, CNES–ISRO SARAL, and NASA ICESat-2) are those acquired in November 2018. Our results show GrIS surface roughness is typically scale-dependent. A revised empirical mapping between quantified radar backscattering and surface roughness gives a better match to the coincident laser altimetry observations than an analytical model that assumes scale-independent roughness. Surface roughness derived from the radar surface echo powers is best interpreted not as the wavelength-scale RMS deviation representative of individual features but as the continued projection of scale-dependent roughness behaviour observed at baselines hundreds of metres long down to the radar wavelength. This implies that the relevance of these roughness estimates to current SMB modelling efforts is limited, as surface roughness is treated as a homogenous and scale-independent parameter.