A satellite-based assessment of ERA5 soil moisture and land surface temperature interannual co-variability

Abstract Land surface temperature provides a fingerprint of the surface energy balance. Under similar radiative forcing, particularly during daytime and clear-sky conditions, its temporal variability is expected to be coupled to surface conditions such as availability of soil moisture and plant functional conditions. In this study, we use Earth observations to identify regions with a strong interannual co-variability between the diurnal amplitude of land surface temperature and surface soil moisture. Land surface temperature is taken from the European Organization for the Exploitation of Meteorological Satellites Applications Facility on Land Surface Analysis and soil moisture from the Copernicus Climate Change Service for the period 2004–2023 over the 0° Meteosat Second Generation region, encompassing Africa and Europe. We introduce a new metric to quantify the soil moisture-temperature interannual co-variability in the satellite data, which is then applied to evaluate the fifth reanalysis from the European Centre for Medium-Range Weather Forecasting (ERA5). Earth observations and ERA5 show similar spatial co-variability patterns across much of the domain, highlighting regions of transition between climate zones. However, ERA5 shows weaker co-variability in regions with large interannual vegetation variability. These findings suggest that accurately capturing the interannual variability of land surface temperature and its co-variability with soil moisture, in regions with large vegetation variability, may require a dynamic representation of vegetation status, beyond the mean annual cycles currently implemented in ERA5.