Soil organic carbon as a factor in passive microwave retrievals of soil water content over agricultural croplands

- Passive microwave SWC from PALS instrument compared to ground SWC during SMAPVEX12.
- Soil organic carbon explained significant difference between PALS and ground SWC.
- SOC explained greater relative anomalies in SWC in extremes of wetness and loam soils.

SummaryRemote sensing has the potential to deliver global soil water content (SWC) on vast scales with frequent revisit times for progress in the fields of climate, weather forecasting, agriculture and hydrology. Although surface roughness, vegetation and soil texture have been established as sources of variability in passive microwave interpretation, soil organic carbon (SOC) has not typically been considered as a factor that affects SWC estimation during field sampling campaigns. SOC was observed along with soil texture and bulk density during the Soil Moisture Active Passive Validation Experiment in 2012 (SMAPVEX12), the Soil Moisture Active Passive (SMAP) satellite algorithm development field sampling campaign held June 6 to July 19 in Southern Manitoba, Canada. Aerial measurements from the PALS (Passive Active L-band System) instrument were recorded over agricultural fields and forest areas from aircraft while SWC was measured simultaneously on the ground with resistance probes on 17 sampling dates. Additionally, fields were sampled for surface roughness, vegetation growth and water content, soil and vegetation temperature and soil physical characteristics. A soil core was collected on each field each sampling time to assess bulk density, soil particle size and SOC. SOC accounted for more variability in the anomalies between PALS and ground sampled SWC than sand, clay or bulk density, although all soil variables explained significant variability. With analysis by partial least squares multiple regression over 11 sampling dates and 39 fields where both ground and PALS data were well represented, only SOC contributed significantly to the regression of SWC beyond the variance all soil variables had in common. The significance of SOC in the relative SWC anomalies was highest in very wet and very dry conditions and in loam soil over all sampling dates, while bulk density was more significant in sand soils. This analysis suggests SOC is a simple variable that incorporates multiple aspects of soil hydrology and may improve the algorithms for remotely sensing estimation of SWC in various landscapes.