Land Surface Modeling at Hyper-Resolution in the Context of SMAP Cal/Val

In the context of SMAP Cal-Val, this study evaluates the performance of Environment Canada's Surface Prediction System (SPS) with respect to soil moisture with an emphasis on spatial variability. To do so, SPS is run at hyper-resolution (100m) over a small domain in southern Manitoba, Canada, where an intensive measuring campaign took place in the summer of 2012 to facilitate comparison to observations. It is shown that SPS is able to simulate the near-surface soil conditions with high accuracy, but that, at such high resolution, the quality of the geophysical fields has a large impact on modeled results, particularly on the spatial variability.