Wildfires can cause dramatic increases in runoff and erosion from forested watersheds. A paired watershed approach was used to monitor post-fire runoff and sediment yields from two small watersheds in the June 2000 Bobcat Fire west of Fort Collins in the Colorado Front Range. One of the watersheds was treated with rehabilitation treatments consisting of contour-felled logs, mulching, and seeding (Bobcat); the other remained untreated (Jug Gulch). Streamflow water levels are continuously monitored and direct and indirect discharge measurements used to develop stage-discharge rating curves. Suspended sediment samples were collected during runoff events with automatic pumping samplers and grab samples. Streamflow-sediment rating curves were used to estimate storm-specific suspended sediment yields.

Immediately after the fire in the summer of 2000, rainfall and streamflow measurements started. Summer storms are variable in space and time, and characterization of precipitation intensities and depths required additional instrumentation. A network of ten rain gages in or near the watersheds was in place by summer 2001. During that summer, there were 72 rainfall events in Bobcat and 85 in Jug Gulch. Storms with maximum 30-min intensities of 23 and 32 mm hr^-1 generated peak runoff rates of 4.7 and 6.3 mm hr^-1 in the treated and untreated watersheds, respectively. Maximum 30-min rainfall intensities explained > 87% of the variability in the peak discharges.

Suspended sediment concentrations approached 42,000 mg L^-1 and turbidity approached 28,000 NTU. Maximum storm-specific suspended sediment yields were 670 and 1,200 kg ha^-1 for the treated and untreated watersheds, respectively. Sediment yields were correlated with rainfall intensity and storm erosivity. Although sediment yields were generally greater in the untreated watershed, variability in rainfall and with no precipitation events from summer 2002, no conclusions about the effectiveness of rehabilitation treatments at the watershed-scale can be made yet.