Grantee Research Project Results
Final Report: Scaling Up Spatially Distributed Hydrologic Models of Semi-Arid Watersheds
EPA Grant Number: R824784Title: Scaling Up Spatially Distributed Hydrologic Models of Semi-Arid Watersheds
Investigators: Tarboton, David G. , Cooley, K. , Seyfried, M. , Hanson, C. L , Flerchinger, Gerald N , Neale, C.M. U. , Slaughter, C. W.
Institution: Utah State University
EPA Project Officer: Packard, Benjamin H
Project Period: October 1, 1995 through September 1, 1998 (Extended to April 30, 2000)
Project Amount: $330,000
RFA: Water and Watersheds (1995) RFA Text | Recipients Lists
Research Category: Watersheds , Water
Objective:
Much of the Western U.S. rangeland is semi-arid. Water resources and water quality concerns are important environmental issues in this region. The goal of this project was to use data from the Reynolds Creek Experimental Watershed (RCEW) in southwest Idaho to understand interacting watershed processes and the water balance over a range of scales. The project was a collaborative effort involving faculty and students at Utah State University and researchers at the U.S. Department of Agriculture (USDA) Agricultural Research Service Northwest Watershed Research Center. The approach consisted of the development of a spatially distributed modeling framework that accounts for spatial variability in topography, vegetation, and soils, and facilitates physically realistic spatial integration of the complete water balance at a range of scales. The approach combined modeling, field measurements, and remote sensing.Summary/Accomplishments (Outputs/Outcomes):
The hydrology of Reynolds Creek Experimental Watershed, representative of much of the western rangeland in the United States is snowmelt driven; therefore, modeling and parameterizing the spatial variability of snow became one focus of this work. Detailed measurements were made of snow on a fine grid over a small subwatershed. These were compared to model simulations. The processes leading to snowpack variability include wind drifting and variable melt energy inputs due to the effect of topography on radiation. We showed (Luce et al., 1997; Luce et al., 1998) that representing the effects of subgrid variability on snow drifting is equally or more important than representing subgrid variability in solar radiation.This spatial variability of snow distribution is environmentally important because snow accumulated in drifts sustains streamflow later into the spring and summer, than would be sustained by a more uniform snowpack. This sustains plant communities and ecosystems that depend upon snowmelt, as well as being important for streamflow and water resources. Through this work, our better understanding of snowpack variability allows us to better model these environmental systems.
In the quest to scale up our models and apply them over larger watersheds, we explicitly incorporated a parameterization of subgrid variability in snow, through the use of a depletion curve into our snowmelt model (Luce et al., 1999). A depletion curve quantifies the relationship between snow covered area and snow water equivalent. Snow covered area is an important control on melt and surface water input rates because energy transfers to the snow pack are across the snow covered area. We also developed theoretical relationships between the depletion curve and distribution of snow at peak accumulation (Luce et al., 1999). To obtain the spatial distribution of snow accumulation over large areas, a wind blowing snow model was tested against Reynolds Creek data (Prasad et al., 2000b). This spatial distribution facilitates the derivation of depletion curves for the modeling of surface water inputs over large areas.
In addition to snow, this project also examined runoff and evapotranspiration processes. To correctly model runoff, it is necessary to account for the spatial distribution of snowmelt inputs due to snow drifting. In the past, we have accounted for this using a drift factor calibrated based on measurements at each grid element. A drift factor is an index of the snow accumulation accounting for drifting or scour at a particular location relative to an average or gage measurement. For scaling up to large areas, it was impractical to apply the grid model at each grid cell, so we developed an approach that divides a watershed into three zones based upon drift patterns, soil types, and vegetation (Flerchinger et al., 2000; Prasad et al., 2000a). We showed that these zones can be obtained from the distribution of calibrated drift factors at a small watershed. The timing of surface water input on the zone corresponding to deep drifts on the north-facing, leeward slope corresponds closely with the timing of streamflow at the outlet. A lumped hydrologic model was developed that consists of: (a) simple parameterization of evapotranspiration, (b) infiltration into the soil zone and recharge to the saturated zone, and (c) subsurface storage-discharge function. This model, applied to each of the three surface water input zones individually, was shown to be sufficient to parameterize the volume and timing of runoff from this watershed. To extend this approach to large areas requires a way to estimate drift factors and zones over large areas. Here, we again used the wind blowing snow model (SnowTran-3D), in collaboration with Glen Liston the model developer at Colorado State University. Results are reported in Prasad et al. (2000b). These results show that although there are discrepancies in pointwise comparisons that require further investigation, the wind blowing model provides a reasonable estimate of drift factors.
A 10-year water balance was computed for a 26 ha watershed by dividing the watershed into three zones based upon drift patterns, soil types, and vegetation (Flerchinger et al., 2000). It was shown that approximately 450 mm of precipitation is necessary to generate runoff from the watershed; above this threshold, runoff increases somewhat linearly with precipitation. An estimated 46 mm, or approximately 10 percent of the annual precipitation, was lost to deep percolation losses through fractures in the basalt underlying the watershed. Water percolating beyond the root zone, as simulated by the Simultaneous Heat and Water (SHAW) model, was directly related to measured runoff (R2=0.90). Above a threshold of about 50 mm, 67 percent of the water percolating beyond the root zone produced runoff. This can have important ramifications in addressing subsurface flow and losses when applying a snowmelt runoff model to simulate runoff and hydrologic processes in the watershed.
In terms of evapotranspiration, the spatial variability of surface vegetation properties is important. The Utah State University airborne videography system was used to acquire high resolution multi-spectral remote sensing imagery. Ground based measurements of leaf area index have been used to relate vegetation parameters such as leaf area index and plant type and height to the remote sensing data (Crosby et al., 2000a). These relationships facilitate the derivation of hydrologic model inputs (Crosby et al., 2000b). The remote sensing data, together with a spatially distributed energy balance model for the estimation of evaporation was used to quantify the scale of variability associated with the surface energy balance (Artan et al., 2000). An innovative aspect of this study was that model testing included comparisons of model surface temperature against spatially distributed thermal imagery from the airborne remote sensing system. This provides a rigorous spatially distributed check of energy balance model performance. The scaling analysis comprising statistical analysis of the spatial fields at different resolutions suggest a grid scale of 10 x 10 m2 for the modeling of surface energy balance processes.
Five graduate students (3 Ph.D. and 2 M.S.), namely, Charlie Luce, Rajiv Prasad, Guleid Artan, Greg Crosby, and Kevin Williams, worked on this project. Six refereed papers have been published. Three papers are under review. There have been three nonrefereed conference papers and multiple conference presentations.
Overall, this study has led to a better understanding of water balance processes in western semi-arid rangeland watersheds, and improved modeling methodology for simulation of hydrologic processes in this region. Many of the papers and models from this research are available at the web site below.
Journal Articles on this Report : 7 Displayed | Download in RIS Format
Other project views: | All 20 publications | 8 publications in selected types | All 7 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Artan GA, Neale CMU, Tarboton DG. Characteristic length scale of input data in distributed models: implications for modeling grid size. Journal of Hydrology 2000;227(1-4):128-139. |
R824784 (Final) |
Exit Exit |
|
Flerchinger GN, Cooley KR, Hanson CL, Seyfried MS. A uniform versus an aggregated water balance of a semi-arid watershed. Hydrological Processes 1998;12(2):331-342. |
R824784 (Final) |
Exit |
|
Flerchinger GN, Cooley KR. A ten-year water balance of a mountainous semi-arid watershed. Journal of Hydrology 2000;237(1-2):86-99. |
R824784 (Final) |
Exit Exit |
|
Luce CH, Tarboton DG, Cooley KR. The influence of the spatial distribution of snow on basin-averaged snowmelt. Hydrological Processes 1998;12(10-11):1671-1683. |
R824784 (1998) R824784 (Final) |
Exit Exit |
|
Luce CH, Tarboton DG, Cooley KR. Sub-grid parameterization of snow distribution for an energy and mass balance snow cover model. Hydrological Processes 1999;13(12-13):1921-1933. |
R824784 (1998) R824784 (Final) |
Exit Exit |
|
Prasad R, Tarboton DG, Liston GE, Luce CH, Seyfried MS. Testing a blowing snow model against distributed snow measurements at upper Sheep Creek. Water Resources Research 2001;37(5):1341-1356. |
R824784 (Final) |
Exit Exit |
|
Williams KS, Tarboton DG. The ABC's of snowmelt: a topographically factorized energy component snowmelt model. Hydrological Processes 1999;13(12-13):1905-1920. |
R824784 (Final) |
Exit |
Supplemental Keywords:
watersheds, scaling, hydrology, modeling, remote sensing, western, Utah, UT, Idaho, ID, EPA Region 8., RFA, Scientific Discipline, Water, Geographic Area, Water & Watershed, Hydrology, State, Environmental Monitoring, Wet Weather Flows, Watersheds, EPA Region, scaling, collaborative hydrologic modeling, spatially distributed hydaulic models, water balance, Idaho (ID), streams, arid watersheds, climate change, vegetation, Region 8, infiltration, aquatic ecosystems, remotely sensed data, Reynolds Creek Experimental Watershed, semi-arid watersheds, snowmelt, climate variabilityRelevant Websites:
David Tarboton ExitProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.