Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Hydrology and Remote Sensing Laboratory » Research » Publications at this Location » Publication #309929
Title: Daily field-scale ET mapping using a data fusion approach

Author
item Yang, Yun
item Anderson, Martha
item Semmens, Kathryn
item Gao, Feng
item Kustas, William - Bill
item HAIN, C. - University Of Maryland
item Schull, Mitchell

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 9/22/2014
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Land-surface evapotranspiration (ET) transfers water from soil and vegetation into atmosphere and is affected by drought, disease, insects, vegetation amount and phenology, and soil texture, fertility and salinity. The ability to accurately map daily ET at field scale will provide useful information for water management, regarding spatial distribution of water use and vegetation moisture deficits at the scale water and landus is actively being managed. The data fusion approach presented here, using thermal infrared (TIR) and reflectance images at multiple scales from multiple sensors, can improve spatial and temporal sampling in satellite-based ET retrievals. The retrieval approach is built on the two-source energy balance (TSEB) model, which partitions the composite surface radiometric temperature into soil and canopy temperatures. Coupling TSEB with an atmospheric boundary layer model, the Atmosphere-Land Exchange Inverse (ALEXI) model maps daily fluxes at continental scales. The related DisALEXI algorithm can then be used to spatially disaggregate ALEXI results to finer spatial scales using TIR imagery from the Moderate Resolution Imaging Spectroradiometer (~daily revisit, 1-km resolution) and Landsat (~bi-weekly to monthly revisit, sharpened to 30-m resolution. The, Spatial Temporal Adaptive Reflectance Fusion Model (STARFM) is the applied to get daily field scale ET estimation by fusing ET timseries maps estimated using MODIS and Landsat. This data fusion approach has been applied to study areas with various climate conditions, different crop types and different water management strategies. Model simulated results have a good correlation with flux tower observations and can clearly show the difference between different water management strategies. The comparison between the data fusion results and interpolation using only high resolution data shows the data fusion approach can capture precipitation influences which might be missing by using only high resolution data.