Submitted to: Journal of Environmental Quality
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/16/2012
Publication Date: 2/1/2013
Publication URL: http://handle.nal.usda.gov/10113/60618
Citation: Beeson, P.C., Sadeghi, A.M., Lang, M.W., Tomer, M.D. 2013. Evaluating the effect of digital elevation model resolution on sediment prediction in water quality models. Journal of Environmental Quality. 43(1):26-36. Interpretive Summary: Hydrological models are tools used to assess many water resource engineering problems from contaminant transport and soil erosion rates to flood or drought potential, to name a few. But these models need appropriate input data to perform well. One improvement to data collection is a highly accurate airplane mounted laser altimeter, which can be used for mapping the land surface rapidly at 10 times the resolution of prior methods. However, this finer resolution overestimates the amount of soil loss predicted. Users of this new data should be cautious of this consequence and calibrate the models accordingly.
Technical Abstract: Moderate resolution (30 m) digital elevation models (DEMs) are normally used to estimate slope for the parameterization of non-point source process-based water quality models. These models, such as the Soil and Water Assessment Tool, utilize the Universal Soil Loss Equation (USLE) and Modified USLE to estimate sediment loss. Among the critical parameters in USLE, is the slope length and steepness factor which has shown to have significant effect on sediment loss estimations. Depending on slope range, a twofold difference in slope estimation potentially results in either as little as 50% change or as much as 250% change in the LS factor and subsequent sediment estimation. Recently, the availability of much finer resolution (~2-3 m) DEMs derived from Light Detection and Ranging (LiDAR) data have increased. However, the use of these finer resolution data may not always be an appropriate choice, because slope values derived from fine spatial resolution DEMs are usually significantly higher, often more than twofold than those estimated from coarser DEMs, which results in considerable variability in model sediment output estimations. This paper addresses the implications of parameterizing models using slope values calculated from DEMs with different spatial resolutions (90, 30, 10, and 3 m). Overall, we observed a two and a half fold increase in slope from the 90m to 3m DEMs, which has a 130% increase in soil loss estimate from the USLE calculation. LiDAR application in water quality models should be used with caution, because it can result in unrealistically higher slope calculations and shorter slope lengths, which changes the amount of sediment loss.