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 #280027
Title: Using satellites to provide reliable daily water use estimates at field scales

Author
item Cammalleri, Carmelo
item Anderson, Martha
item Gao, Feng
item Kustas, William - Bill

Submitted to: BARC Poster Day
Publication Type: Abstract Only
Publication Acceptance Date: 3/26/2012
Publication Date: 4/23/2012
Citation: Cammalleri, C.N., Anderson, M.C., Gao, F.N., Kustas, W.P. 2012. Using satellites to provide reliable daily water use estimates at field scales [abstract}. BARC Poster Day. Poster No. 10.

Interpretive Summary:

Technical Abstract: The ability to accurately map daily crop water use over agricultural landscapes at scales resolving individual farm fields is a significant challenge, but it is increasingly relevant in a future scenario of reduced water availability and increased atmospheric demand. Remote sensing provides a robust means to determine spatially distributed evapotranspiration (ET) over large areas in a cost effective manner, capitalizing the strong interconnection between remotely-observed land-surface temperature (LST) and ET fluxes. Nevertheless, there is an inherent trade-off in satellites collecting thermal infrared (TIR) imagery used to map LST. In fact, due to technical and budget limitations the currently available data are generally characterized by low spatial resolution (10^3-10^4 m) and high repeatability (1 day to 15 min), such as MODIS (MODerate resolution Imaging Spectroradiometer) and GOES (Geostationary Operational Environmental Satellite), or by moderate/high spatial resolution (10^1-10^2 m) and low temporal frequency (16 days), as for instance with the Landsat series. In order to obtain reliable daily ET estimates at field scale, the best characteristics of each dataset should be combined. With this aim, we developed a multi-scale, multi-sensor modeling framework. The Atmosphere-Land EXchange Inverse (ALEXI) model uses GOES data to obtain coarse (10-km) maps at the U.S. continental scale, while the Dis-ALEXI procedure disaggregates ALEXI ET to finer spatial scales using MODIS 1-km (daily) and Landsat 30-m (16 day) LST imagery. These latter estimates are combined by means of a data fusion algorithm to retrieve daily and seasonal 30-m ET maps over areas on the order of 100 km in dimension. In this work, ALEXI/DisALEXI was applied to GOES, MODIS and Landsat images acquired in 2002 over central Iowa, the site of the Soil Moisture Experiment of 2002 (SMEX02). The time series of 30-m maps obtained in the period May-August were validated by means of 7 micrometeorological flux towers, which provide a unique and extensive observational dataset of ET time series in corn and soybean fields during a period of rapid crop development. The MODIS-Landsat fused results were also compared to a benchmark case, obtained using Landsat images only. The results indicate that significant improvement in agreement between modeled and observed ET can be obtained in some cases by fusing MODIS and Landsat data, particularly when a rainfall event occurs between Landsat image acquisitions. The fused ET maps had an accuracy of about 0.5 mm/d at daily timesteps (10% of the observed values) and 5% for seasonal total water use, which is suitable for detailed analysis and management of crop water consumption.