Impact of the New ERA5 Reanalysis in the Computation of Radar Altimeter Wet Path Delays

Abstract

Satellite altimetry allows the estimation of accurate water surface heights only with accurate determination of all involved terms, namely, the wet tropospheric correction (WTC) or its symmetric value, the wet path delay (WPD). WPD is best determined from onboard microwave radiometer (MWR) measurements; however, the corresponding WPD retrievals become invalid close to land (e.g., coastal and inland waters). Alternative WPD sources are numerical weather models (NWMs), e.g., from the European Centre for Medium-Range Weather Forecasts (ECMWF). NWMs provide the parameters at 6-h intervals; however, ERA5 (the latest ECMWF reanalysis) provides hourly atmospheric parameters at $0.25^{\circ } \times 0.25^{\circ }$ . The best spatial resolution is provided by the ECMWF operational model at $0.125^{\circ } \times 0.125^{\circ }$ . Motivated by this new and improved temporal resolution, the focus of this paper is the global assessment of the impact of different temporal resolutions of ERA5 in the WPD computation. The same assessment is also performed concerning the spatial resolution of ERA5 and operational models. Aiming to identify the best compromise between spatial/temporal resolutions, accuracy, and computational time, WPDs computed using various combinations of these resolutions were inter-compared and compared with MWR-derived WPD. The results show that the global root mean square (RMS) of the differences between MWR and ERA5 is 1.2 cm and the effect of using 1-h intervals instead of 6-h intervals is small, significant only for latitudes 30°–60° S and 30°–60° N. Hourly intervals do not have a significant impact on the WPD from ERA5, being a temporal resolution of 3 h high enough to ensure the same accuracy of 1 h, showing that ERA5 cannot map the WPD short space and time scales.