Soil Moisture Retrieval Using Ground‐Based L‐Band Passive Microwave Observations in Northeastern USA

A field experiment was performed in a grassland at the NOAA‐CREST–Soil Moisture Advanced Radiometric Testbed (CREST‐SMART) facility, which includes a mobile L‐band dual‐polarized radiometer with an in situ soil temperature and soil moisture observation network, located near Millbrook, NY. During the day‐long field campaign, intensive spatiotemporal measurements of L‐band brightness temperatures, surface temperature, soil moisture, and soil temperature at 3‐, 7‐, and 12‐cm depths were collected during three passes at 0830, 1130, and 1430 h. During the second and third passes, half of the field was irrigated. Soil roughness and water content of the short grass that remained after mowing the study area were measured. The Tau‐Omega radiative transfer model was used to assess the performance of the soil moisture retrieval using measured soil temperatures at different depths. In addition, the collected microwave observations at the three different times of the day were used to assess the impact of the diurnal variation of soil temperature on the performance of soil moisture retrieval. Obtained results showed that the root mean square error (RMSE) decreased throughout the day to reach 0.03 m 3 /m 3 for the afternoon pass when 12‐cm soil temperature values were used in the radiative transfer model. In addition, during the three different passes, the lowest RMSE was consistently obtained when the 12‐cm soil temperature was used, which suggests that, for this investigated site, soil temperature at the 12‐cm depth can be a surrogate for soil effective temperature when L‐band microwave temperatures are used. In terms of diurnal variability, observations from the afternoon pass led to the highest agreement between observed and retrieved soil moisture values.