An updated and quality controlled surface mass balance dataset for Antarctica
- 1UJF-CNRS, LGGE, 54 rue Molière, BP 96, 38402 Saint-Martin-d' Hères, France
- 2IRSTEA, UR ETGR Erosion Torrentielle Neige Avalanches, Domaine universitaire, 2, rue de la Papeterie, 38402 Saint-Martin-d'Hères, France
- 3Department of Geography, University of Liege, Belgium
Abstract. We present an updated and quality controlled surface mass balance (SMB) database for the Antarctic ice sheet. Importantly, the database includes formatted metadata, such as measurement technique, elevation, time covered, etc, which allows any user to filter out the data. Here, we discard data with limited spatial and temporal representativeness, too small measurement accuracy, or lack of quality control. Applied to the database, this filtering process gives four times more reliable data than when applied to previously available databases. New data with high spatial resolution are now available over long traverses, and at low elevation in some areas. However, the quality control led to a considerable reduction in the spatial density of data in several regions, particularly over West Antarctica. Over interior plateaus, where the SMB is low, the spatial density of measurements remains high. This quality controlled dataset was compared to results from ERA-Interim reanalysis to assess whether field data allow us to reconstruct an accurate description of the main SMB distribution features in Antarctica. We identified large areas where data gaps impede model validation: except for very few areas (e.g., Adelie Land), measurements in the elevation range between 200 m and 1000 m above sea level are not regularly distributed and do not allow a thorough validation of models in such regions with complex topography, where the highest scattering of SMB values is reported. Clearly, increasing the spatial density of field measurements at low elevations, in the Antarctic Peninsula and in West Antarctica is a scientific priority.