Skip to main content
ARS Home » Research » Publications at this Location » Publication #263649
Title: An intercomparison of available soil moisture estimates from thermal-infrared and passive microwave remote sensing and land-surface modeling

Author
item HAIN, CHRISTOPHER - National Oceanic & Atmospheric Administration (NOAA)
item Crow, Wade
item MECIKALSKI, JOHN - University Of Alabama
item Anderson, Martha
item Holmes, Thomas

Submitted to: Journal of Geophysical Research Atmospheres
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/16/2011
Publication Date: 8/9/2011
Citation: Hain, C.R., Crow, W.T., Mecikalski, J.R., Anderson, M.C., Holmes, T.R. 2011. An intercomparison of available soil moisture estimates from thermal-infrared and passive microwave remote sensing and land-surface modeling. Journal of Geophysical Research Atmospheres. 116:D15107.

Interpretive Summary: Currently three independent techniques exist for measuring soil moisture over large continental-scale regions: water balance modeling using rainfall observations, microwave remote sensing and thermal remote sensing. This paper presents the first attempt to inter-compare all three techniques over the continental United States between 2003 and 2008. The comparison allows us to learn important information about the relative merits of all three approaches and provides a valuable overview of the geographic areas where each performs best (i.e. where each technique provides the most accurate estimate of soil moisture variations). Better information about the accuracy of each approach will eventually lead to optimal systems for fusing soil moisture information obtained from multiple data sources and a better characterization of agricultural drought extent and magnitude.

Technical Abstract: Remotely-sensed soil moisture studies have mainly focused on retrievals using active and passive microwave (MW) sensors whose measurements provided a direct relationship to soil moisture (SM). MW sensors present obvious advantages such as the ability to retrieve through non-precipitating cloud cover which provide shorter repeat cycles. However, MW sensors offer coarse spatial resolution and suffer from reduced retrieval skill over moderate to dense vegetation. A unique avenue for filling these information gaps is to exploit the retrieval of SM from thermal infrared (TIR) observations, which can provide SM information under dense vegetation cover and at significantly higher resolutions than MW. Previously, an intercomparison of TIR-based and MW-based SM has not been investigated in the literature. Here a series of analyses are proposed to study relationships between SM products during a multi-year period (2003-2008) from a passive MW retrieval (AMSR-E), a TIR based model (ALEXI), and a land surface model (Noah) over the continental United States. The three analyses used in this study include: (a) a spatial anomaly correlation analysis, (b) a temporal correlation analysis, and (c) a triple collocation error estimation technique. In general, the intercomparison shows that the two datasets (TIR and MW) provide complementary information about the current SM state. TIR methods can provide SM information over moderate to dense vegetation, a large information gap in current MW methods, while serving as an additional independent source of SM information over low to moderate vegetation. The complementary SM information from the two retrieval methods provides a potential for integration within an advanced data assimilation system towards the improved prediction of SM evolution in land surface models.