USE OF A RAINFALL SIMULATOR TO ASSESS CHANGES IN RUNOFF AND SEDIMENT PRODUCTION OVER TIME, COLORADO FRONT RANGE

Wildfires in the Colorado Front Range cause large increases in runoff and erosion, but the variability in rainfall inputs makes it difficult to determine the role of individual site factors, or rigorously assess the rate of recovery over time. Use of a rainfall simulator allows a more uniform means for comparing post-fire runoff and erosion rates. The objectives of this project are to: (1) determine whether runoff and sediment yields are significantly related to key site variables; and (2) assess when post-fire runoff and erosion rates return to background levels. The study sites are the June 2000 Bobcat fire and the November 1999 Lower Flowers prescribed fire. Twenty-six rainfall simulations were conducted on 1 m2 plots in 2000, 23 simulations in 2001, and we are now collecting a third year of data. For each simulation rainfall is applied at 70-80 mm hr-1 for 60 minutes.

Fire severity does not significantly affect the runoff rate, and this presumably is due to the relatively high rainfall rate. The runoff ratio for sites burned at high severity in the Bobcat fire declined from 66% in 2000 to 46% in 2001. Runoff ratios for sites burned at moderate severity decreased from 58% in 2000 to 40% in 2001. In contrast, there was no significant difference in runoff ratios with fire severity or between years for the simulations at Lower Flowers. Runoff ratios in both years were significantly related to soil water repellency and soil moisture.

Sediment yields varied significantly with fire severity but did not significantly decrease from 2000 to 2001. For the Bobcat fire, sediment yields from high severity sites averaged 1.2 kg m-2, or seven times the value for sites burned at moderate severity and 24 times the value for sites burned at low severity and unburned plots. At Lower Flowers the mean sediment yield for high severity sites was 0.5 kg m-2, or four times the value for sites burned at moderate severity and 17 times the mean value for low severity and unburned sites. The variability in sediment yields in 2001 increased relative to 2000, and this is probably due to the greater variability in percent cover. Percent bare soil is the primary control on sediment yields, although soil water repellency, slope, and soil moisture were also significant. Preliminary results from 2002 suggest similar runoff rates to 2001 and increasing variability in sediment yields.