DYNAMIC SOIL STATE MODELS FOR MOISTURE AND TEMPERATURE PROFILE DETERMINATION FROM SEQUENTIAL REMOTE SENSING OBSERVATIONS

Authors: GALANTOWICZ, JOHN - MASS INSTITUTE OF TECH; ENTEKHABI, DARA - MASS INSTITUTE OF TECH; Jackson, Thomas

Submitted to: IEEE Transactions on Geoscience and Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/29/2001
Publication Date: N/A
Citation: N/A

Interpretive Summary: The development of robust techniques foe measuring and monitoring soil moisture using satellite instruments is needed before wide scale implementation is possible. This requires a thorough understanding of the physical principles and models that are capable of simulating the complex processes involved. A time-explicit coupled model of soil moisture and temperature for constraining retrieval from radiobrightness and thermal infrared measurements has been developed. The model was compared to in situ soil moisture and temperature measurements over a seven day period. Soil temperature dynamics and the diurnal mean soil moisture were accurately modeled whereas the diurnal soil moisture range in the top 5 cm was generally underestimated. Good soil temperature simulation implies that the effective emitting temperature in soil emission models can be accurately predicted. Good prediction of the mean surface soil moisture suggests that the deeper soil profile could be accurately modeled but more extensive experimental data are required to test this hypothesis. The results of this study contribute to the development of a comprehensive soil moisture retrieval using remote sensing. Products from satellite based soil moisture monitoring and mapping have a great potential benefits for hydrologic and agricultural water managers by providing improved assessments and forecasts.

Technical Abstract: A new time-explicit vadose zone coupled soil moisture and energy flow model for use in assimilation of sequential microwave and infrared remote sensing observations is introduced and evaluated. Bare soil microwave emission models was also evaluated for use in a data assimilation scheme where radiobrightnesses calculated from estimated soil temperature and moisture states are merged with observations. The models were applied to data from field site where simultaneous radiobrightness and in situ soil measurements were made. The soil state model surface boundary conditions are driven by micrometeorology and incident radiation observed over a seven day period. Model propagation of the soil moisture state and radiobrightness was compared in two operating modes: (i) an open-loop mode with no run-time hydrologic (precipitation or soil state) inputs and (ii) an updated mode with updates made directly from observed soil moisture. Simulated soil state and L and S band brightness temperatures were compared to measured values. We also compared the simulated L and S band radiobrightness dynamics of five emission parameterizations during diurnal cycles and gradual dry-down of the simulated soil.