Impact of Coincidental Precipitation on the Accuracy of SMAP Soil Moisture Retrievals

The measurement of soil moisture during and right after rain has a special role in the determination of the temporal evolution of soil moisture. Remotely sensed soil moisture improves the performance of many hydrological applications from weather forecast and climate modeling to flood prediction and drought mitigation. Precipitation events represent critical time periods with enormous impact on such applications. However, such events create a vertically uneven distribution of soil moisture in the top layer of soil. After a precipitation event it will take time for the soil to reach a relative equilibrium again, which will depend upon the characteristics of both the precipitations and the local conditions. An assumption of most satellite retrieval algorithms is that soil moisture is homogeneous in the first few centimeters. Moreover, many in situ sensors used for soil moisture validation measure the average water content in the surface soil but often this measurement does not include the first couple of centimeters as the probes are installed horizontally at 5 cm depth. Hence, the usual assumption is that both the algorithm and the representativeness of the in situ measurement are compromised during and right after the rain event and omitted from traditional validation metrics. The actual performance of the retrieved soil moisture would provide a way to use these retrievals in an informed manner.

The objective of this investigation is to quantify the impact of precipitation on soil moisture retrieval error. The analysis uses the Soil Moisture Active Passive (SMAP) Level 2 enhanced soil moisture product based on the L-band passive microwave radiometer (L2SMPE). The reference data consists of the SMAP core validation sites where a number of permanently installed stations determine soil moisture and precipitation over a representative SMAP retrieval domain. Additionally, a hydrological land surface model (LSM) provides an estimate of the vertical distribution of water in the top layers of soil. The LSM accounts for variability in infiltration and redistribution using soil texture information to determine the location-specific retrieval uncertainty during rain events. The modeled retrieval uncertainty is time-dependent and driven by the local rain gage measurements; it is used in interpretation of the soil moisture retrieval uncertainty based on the in situ soil moisture measurements. Precipitation impacts only a fraction of the retrievals reducing the number of available data points but the over three-year data record collected by SMAP allows a robust investigation into this issue.

Acknowledgement: The research described in this abstract was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration.