Assimilation of SMAP Brightness Temperatures: An Investigation into the Source of Near-Surface Temperature Forecast Errors in ECCC's Short-Range NWP Forecasts

Assimilation experiments performed over North America assimilating Soil Moisture Active Passive (SMAP) L-band brightness temperatures (TB) into Environment and Climate Change Canada’s (ECCC) short-range regional numerical weather prediction (NWP) model have identified a deterioration in near-surface dew-point temperature forecasts during the daytime. The area of degradation is centered over the Northern Great Plains of the US, a predominantly agriculture and cropland region noted for strong land-atmosphere coupling. The soils become progressively drier leading to low dew-point temperatures.

To further investigate the source of this excessive drying this study will focus upon a comparison of the simulated latent and sensible heat fluxes with those calculated from thermal infrared remote sensing. The Atmosphere-Land Exchange Inverse (ALEXI) model has been used extensively in previous studies and is driven principally by clear-sky derived land-surface temperature retrievals. When compared to the limited availability of in-situ flux measurements, ALEXI can provide continental-scale flux estimates. Each assimilation experiment consists of two components, an off-line assimilation cycle followed by a series of short-range NWP forecasts. The fluxes from two experiments, one assimilating SMAP TBs and the other screen-level temperatures to update soil moisture, will be compared against the ALEXI derived fluxes