Abstract:
The NASA Soil Moisture Active Passive (SMAP) mission has been collecting global L-band (1.4 GHz) brightness temperature observations every 1-3 days since 31 March 2015. These observations are sensitive to soil moisture and soil temperature in the near-surface layer. They are used in the Goddard Earth Observing System version 5 (GEOS-5) land data assimilation system to generate the Level 4 Soil Moisture (L4_SM) product, which provides 3-hourly, 9-km resolution, global estimates of surface (0-5 cm) and root zone (0-100 cm) soil moisture along with related land surface states and fluxes with a latency of ~3 days. This presentation investigates the soil moisture analysis underlying the L4_SM product. The soil moisture analysis converts the differences between model-predicted brightness temperatures and SMAP observations into adjustments (or increments) to the modeled soil moisture state based on error correlations between the modeled and observed variables. The analysis (i) uses an ensemble approach, (ii) is spatially distributed, (iii) performs downscaling from the resolution of the observations to the of the model, and (iv) respects the relative uncertainties of the modeled and observed brightness temperatures, thereby resulting in soil moisture estimates that meet the root-mean-square error target of 0.04 m3/m3 (after removal of the long-term mean differences). The early results presented here focus on an in-depth investigation of the observation-minus-forecast residuals, which reveals modest biases in the assimilation system. Furthermore, the investigation demonstrates where the assimilation system overestimates or underestimates the actual errors in the system. Finally, the increments that are applied to the model states are investigated.