Skip to main content
ARS Home » Southeast Area » Oxford, Mississippi » National Sedimentation Laboratory » Watershed Physical Processes Research » Research » Publications at this Location » Publication #254514

Title: Evaluation of the use of remotely sensed evapotranspiration estimates into AnnAGNPS pollution model

Author
item Momm, Henrique
item EASSON, GREG - University Of Mississippi
item Bingner, Ronald - Ron
item Wilcox, Darlene

Submitted to: Ecohydrology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/16/2010
Publication Date: 9/19/2011
Citation: Momm, H.G., Easson, G., Bingner, R.L., Wilcox, D.L. 2011. Evaluation of the use of remotely sensed evapotranspiration estimates into AnnAGNPS pollution model. Ecohydrology. 4(5):650-660. doi: 10.1002/eco.155.

Interpretive Summary: Effective management of runoff and erosion requires estimating soil moisture over many years for large watershed systems with computer models. Often there is very little data available to provide these estimates. The utilization of remote sensing data can be integrated with watershed models to provide better estimates of evapotranspiration (ET) throughout a watershed to account for this important component that affects soil moisture and runoff. The utilization of ET estimates derived from satellite remote sensing into watershed pollution model was investigated in this study. Modifications within watershed models were performed to allow the internal calculations of ET based on climate parameters, collected from ground-based measurements, to be replaced by ET estimates produced with the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor. Direct comparison of these sources of ET indicated that calculated ET estimates were 35% higher than the MODIS calculated ET estimates resulting in a 10% increase in runoff at the outlet of the Yalobusha River Watershed in Mississippi for 2004. Simulated monthly runoff values based on MODIS and simulated ET estimates for the year 2004 were compared to observed values from stream gages indicating slightly better agreement from the simulation using MODIS-derived ET. Analysis of the spatial distribution of the percent difference in runoff between watershed model-based ET and MODIS-based ET simulations showed higher differences at the most downstream and upstream of the watershed. These differences could be attributed to different soil properties or to the effect of distant field/weather station assigned to these locations. Although these findings should not be generalized to watersheds located in different ecosystems, the results indicate that the utilization of remotely sensed ET in hydrological modeling is feasible and that use of MODIS to estimate ET provided a more comprehensive spatial coverage than the often available through ground-based measurements. Utilization of remotely sensed ET information will reduce the uncertainty of this components effect on soil moisture and subsequently on watershed conservation planning involving runoff and erosion control practices.

Technical Abstract: The utilization of evapotranspiration (ET) estimates, derived from satellite remote sensing, into the Annualized Agricultural Non-Point Source (AnnAGNPS) pollution model was investigated. Modifications within AnnAGNPS were performed to allow the internal calculations of ET based on climate parameters, collected from ground-based measurements, to be replaced by ET estimates produced with the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor. Direct comparison of these sources of ET indicated that AnnAGNPS calculated ET estimates were 35% higher than the MODIS calculated ET estimates resulting in a 10% increase in runoff at the outlet of the Yalobusha River Watershed for 2004. Simulated monthly runoff values based on MODIS and AnnAGNPS ET estimates for the year 2004 were compared to observed values from stream gages indicating slight better agreement from the simulation using MODIS-derived ET. Analysis of the spatial distribution of the percent difference in runoff between AnnAGNPS-based ET and MODIS-based ET simulations showed higher differences at the most downstream and upstream of the watershed. These differences could be attributed to different soil properties or to the effect of distant field/weather station assigned to these locations. Although these findings should not be generalized to watersheds located in different ecosystems, the results indicate that the utilization of remotely sensed ET in hydrological modeling is feasible and that use of MODIS to estimate ET provided a more comprehensive spatial coverage than the often available through ground-based measurements.