Project Number: 6209-13000-014-21
Project Type: Reimbursable

Start Date: Sep 01, 2009
End Date: Aug 31, 2012

Objective:
The objectives of this research are to evaluate large aperture scintillometers to reliably monitor evapotranspiration (ET) at appropriate space and time scales to ground reference MODIS satellite data for water resources management. This will be accomplished by: 1) Monitoring ET using point ground-based methods (lysimetry) in different land use settings (rainfed and irrigated agriculture) and large scale ground-based methods (Large Aperture Scintillometers, LAS) in similar land use settings. 2) Applying a set of well calibrated EB-based ET models for semi-arid conditions (product of BEAREX07) to (a) derive ET from aircraft imagery and MODIS datasets; (b) evaluate various foot print models to improve ET mapping accuracy; and (c) assess various map disaggregation strategies to derive high resolution ET maps from low-resolution MODIS-ET maps.

Approach:
This study concentrates on the Texas High Plains Region. The LAS data collected during the BEAREX07 and BEAREX08 will be evaluated using lysimetric data for accuracy. The ET data derived from LAS measurements will be used for testing accuracy of regional ET maps derived from remote sensing data. In our study with MODIS, we will attempt to estimate surface temperature at 250 m pixels by using statistical variation in vegetation indices calculated for those pixels as a basis. Areas ranging from low to high surface temperature by vegetation type will be selected for this purpose. Similar analysis will be done with aircraft data. The airborne thermal scanner and ground based IRT surface temperature readings will be employed in the verification of downscaling accuracy. In addition to within sensor scaling, between-sensor scaling will also be examined. Surface flux maps from MODIS will be compared to path weighted flux measurements from scintillometers to evaluate effects of downscaling on estimating variations in the regional surface fluxes.