GSA Annual Meeting in Denver, Colorado, USA - 2016

Paper No. 310-12
Presentation Time: 4:50 PM


GRIFFIN, Eleanor R. and FRIEDMAN, Jonathan M., U.S. Geological Survey, 3215 Marine St, Suite E-127, Boulder, CO 80303,

Ecologic or hydrologic disturbances in semiarid regions can increase the susceptibility of hillslope, channel, and floodplain surfaces to erosion caused by runoff or streamflow. Riparian areas treated for vegetation control can have an increased risk of erosion during subsequent high flow, similar to the increased potential for erosion and flooding following wildfire. In September 2003, a 12-km segment of the lower Rio Puerco arroyo beginning 0.5 km down-valley from the confluence with the Rio San Jose was sprayed with an herbicide by helicopter for the purpose of saltcedar (Tamarix spp.) control. Woody vegetation on channel banks and floodplain surfaces was weakened or killed by the spraying, reducing plant strength and resistance to flow. High-resolution satellite imagery acquired in November 2006, January and February 2014, and November 2014 combined with aerial LiDAR survey data collected in March 2010 show the effects of subsequent channel and overbank flow on stability of the Rio Puerco. Large floods in August 2006 (176 m3/s) and September 2013 (255 m3/s) caused considerable erosion in the sprayed reach, but not in a 13-km untreated segment downstream. Since the aerial spraying of herbicide, there have been a total of 5 meander cutoffs in the sprayed reach that reduced channel centerline length by >4 km (21%) and none in the untreated reach downstream. The August 2006 flood increased average channel width in the sprayed reach by 84%. Between November 2006 and February 2014, average channel width increased an additional 7.9 m (31%) in the reach from 1.5 to 8.0 km down-valley from the confluence with the Rio San Jose. Sediment eroded from channel banks and floodplain surfaces was transported downstream and contributed to channel narrowing by an average of 2.4 m (12%) in the lower 4 km of the sprayed reach. There was no detectible change in average channel width in the untreated reach downstream. Our results show that channel change following the initial ecologic disturbance was minor until there was a second substantial disturbance, i.e., the largest flood since 1972. Remote sensing data indicate that the channel in the sprayed reach remained unstable >10 yr after the initial disturbance.