|Zhang, Yu - USEPA|
|Shuster, William - USEPA|
Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: December 9, 2005
Publication Date: December 9, 2005
Citation: Bonta, J.V., Zhang, Y., Shuster, W., Hicks, N.S. 2005. Detectability and uncertaintites in the interpretation of the impacts of land disturbances on catchment runoff [abstract]. American Geophysical Union 2005 Fall Meeting, San Francisco, CA. Dec. 5-9, 2005. Technical Abstract: Ensembles of hypothetical spatial arrangements of impervious areas were coupled with a runoff model to address the potential uncertainties in the model-projected impacts of imperviousness on catchment runoff response and to examine the underlying causes of the uncertainties. The methodology was implemented for a pair of small (< 1 ha), headwater catchments in Coshocton, Ohio. For each catchment, eight spatial fractions of 2-m square impervious elements were imposed, and 50 hypothetical scenarios of spatial distribution were randomly generated for each fraction. These scenarios were incorporated into the Gridded Surface Subsurface Hydrologic Analysis (GSSHA), a spatially explicit, Hortonian runoff model that had been calibrated against 40 historical, warm season runoff events. These events were re-simulated under each hypothetical scenario to examine the uncertainties in association with spatial distribution of impervious elements, rainfall structure, soil, and basin morphology. It was found that, for both simulated runoff peaks and depths, the among-scenario variations are relatively minor in comparison with dependence on the fractional imperviousness on an event-mean basis. Yet, for an individual event the variations can be significant, particularly when the proportional areas of pervious and impervious elements are comparable (40%-60%). The projected increase in simulated runoff peaks following increased fraction of imperviousness vary considerably across events, mostly in response to the inter-event variations in rainfall structure and the resultant differential antecedent soil moisture. Two quantities, averaged flow distance of impervious areas from the outlet and the averaged number of downstream pervious elements, were used to characterize the spatial arrangement of impervious elements in each scenario. The dependence of simulated runoff on these quantities, however, was found to be highly variable across catchments, for their effects can be confounded by the effects of spatial variability of soil permeabilities. The practical implications of the work were discussed on the basis of the observations.