Modeling scour and deposition in ephemeral channels after wildfire

The area burned by wildfire in the states of Arizona and New Mexico in the southwestern US has been increasing in recent years. In many cases, high severity burns have caused dramatic increases in runoff and sediment yield from burned watersheds. This paper describes the potential and limitations of the HEC6T sediment transport model to describe changes in channel scour and deposition following the Cerro Grande fire near Los Alamos, New Mexico. Following the fire, Pueblo Canyon, near Los Alamos, was subject to a peak flow two orders of magnitudes higher than any discharge in the 7-year period of record, and twice the initial post-fire estimate of the 100-year event. HEC6T requires that the limits of scour and deposition on a cross-section be specified prior to application. This was achieved by using geomorphologic principles, predicted post-burn hydrology and long-term estimates of channel change derived from air photos, to estimate post-fire channel widths. Because significant quantities of silt and clay were present in the runoff, erosion shear stress and erosion rate parameters for cohesive sediments had to be obtained experimentally. After a sensitivity analysis, an optimization routine was used to estimate the optimal model parameter values for sensitive parameters. HEC6T was able to accurately model the change in cumulative sediment volume change derived from Airborne Laser Swath Mapping (ALSM, often called Lidar) taken before and after the large post-fire event. One discrepancy between the HEC6T model prediction and the ALSM-estimated change was that the ALSM-estimated change showed the greatest amount of deposition in a portion of the canyon with increasing slope, which the HEC6T model did not predict. Any sediment transport model will predict increased sediment transport capacity with increasing energy slope, so that it was considered to be beyond the capability of any sediment transport model to predict this deposition. Therefore, HEC6T simulated the overall changes in scour and deposition within reasonable expectation of the capabilities of physically-based sediment transport modeling indicating that it is capable of modeling sediment transport in ephemeral channels following wildfire.