2009 Portland GSA Annual Meeting (18-21 October 2009)

Paper No. 1
Presentation Time: 9:00 AM-6:00 PM

WATERSHED EROSION MODELLING: A NEW APPROACH TO INCLUDE GULLY EROSION


SANG, Joseph K.1, ALLEN, Peter M.2 and DUNBAR, John A.1, (1)Department of Geology, Baylor University, One Bear Place #97354, Waco, TX 76798-7354, (2)Department of Geology, Baylor University, One Bear Place #97354, Waco, TX 76798, Joseph_Sang@Baylor.edu

Currently the main challenge of hydrology and watershed management is prediction and control of soil erosion. Soil erosion has impacts both on agricultural lands and water reservoirs. The economic cost of erosion is in the range of millions of dollars. Despite the development and subsequent modification of Universal Soil Loss Equation (USLE) and other related models, simulation of gully erosion has not yet been adequately addressed. This new approach to erosion simulation attempts to account for gully erosion and their impacts.

The study sites are Post Oak and Deep creeks in McCulloch County (Texas) and Sandstone creek in Beckham and Roger Mills counties (Oklahoma). The methodology involves Strahler order type analysis of gullies distribution and characteristics based on field survey and interpretation of aerial photographs. Secondly, the reservoir bottoms were imaged using acoustic profiling system (APS) and dating of reservoir core samples to determine sedimentation rate. This is being used for calibration and validation of the simulation models. Finally, scenario based hydrological modeling using Soil and Water Assessment Tool (SWAT) and SWAT-Deg determines the contribution of gullies to the overall erosion under different climatic conditions.

Initial results have shown that drought induced erosion occurred in the study sites. Sediment fluxes were initial high immediately after drought but reduced considerably later. The effective sediment yields from the watershed were in the range of 23 to 4,785 ton/km2/yr. This provided crucial information for calibrating hydrological models. This is crucial where there is no monitoring of erosion rates. Modeling results indicates that inclusion of gully erosion increase the accuracy of prediction of watershed erosion. Further the number of number of gullies could be determined based on the Strahler order. GIS based mapping with high resolution DEM could provide a spatial distribution of gullies within the unit of analysis.

This study concludes that strahler type gully ordering, GIS and hydrological modeling could improve prediction and scenario analysis watershed erosion, which is crucial for management decision support. However, this new approach requires further tests to ascertain a possible worldwide applicability.