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 #304550
Title: A remote sensing-based dry and wet limit-reference evapotranspiration model for water use monitoring

Author
item XIA, T. - Tsinghua University
item Kustas, William - Bill
item WANG, Z. - Tsinghua University

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 4/9/2014
Publication Date: 4/23/2014
Citation: Xia, T., Kustas, W.P., Wang, Z. 2014. A remote sensing-based dry and wet limit-reference evapotranspiration model for water use monitoring [abstract]. Abs. 51. BARC Poster Day.

Interpretive Summary:

Technical Abstract: With increasing growth in human population, the demand for greater food production has exceeded the capability to provide a sustainable water supply for agriculture. This is exacerbated in areas suffering from prolonged drought conditions, particularly in water limited regions. Improving the management of water resources for sustainable agricultural production necessitates the capability to operationally monitor crop water use or evapotranspiration (ET) at watershed and regional scales. This requires the use of satellite remote sensing since it is the only technology that can provide the spatial and temporal information about vegetation and soil conditions necessary to quantify ET. Numerous remote sensing-based ET models incorporate two key surface states strongly linked to ET, land surface temperature (Ts) and fractional vegetation cover (fc) which is often derived from a remotely sensed vegetation index (VI). This information is used in various ways by ET models, however a popular approach is to derive from the satellite image dry and wet extremes where ET is minimum and maximum, respectively. While these approaches have significant advantages in reducing or eliminating errors caused by inaccuracies in the remote sensing and atmospheric forcing data, there is no guaranteed that a satellite scene will contain pixels representing extreme dry and wet conditions over the area of interest, and moreover it is well known that scene size and pixel resolution affect the magnitude of the extremes causing greater uncertainty in ET estimation. The satellite-based energy balance algorithm with reference dry and wet limits (REDRAW) model was developed to address this problem. The REDRAW method assumes in Ts~VI space that with sufficient pixels covering a full range in wet and dry conditions, a trapezoid would be formed with the four vertices representing a dry bare soil and dry densely vegetated surface, a wet bare soil and wet densely vegetated surface. These limits are derived theoretically with assumed surface resistances driven by observed meteorological forcing data and therefore are independent of the extreme temperatures and fractional vegetation cover conditions in the satellite image. The REDRAW model is applied to two agricultural regions, one in northern China and the other in central California with different land cover and climate conditions. Comparison with ground-based ET measurements show that REDRAW is capable of deriving reasonable ET estimates over these two regions, however, there are limitations on extending these estimates over a large area without adequate meteorological observations and environmental conditions where the application of the REDRAW model yields significant uncertainty in ET estimates. Current research is being focused on addressing these limitations.