DESTABILIZATION OF A SAND/SILT-BED CHANNEL BY VEGETATION CONTROL, RIO PUERCO ARROYO, NEW MEXICO

An open question in the field of fluvial biogeomorphology concerns how much time is required for vegetation recovery sufficient to stabilize channel banks and inhibit erosion during future floods. This work examines the lasting (>10 yr) effects of vegetation control along a sand/silt-bed stream in the semi-arid southwestern United States. Saltcedar (Tamarix spp.) control efforts by aerial spraying of an herbicide in September 2003 along a 12-km reach of the lower Rio Puerco arroyo killed much of the woody vegetation on the channel banks and floodplain. A large flood in August 2006 caused extensive channel bank erosion in the sprayed reach but not in a 12.7-km untreated reach downstream. By April 2010, saltcedar and willow (Salix spp.) on channel banks and floodplains in the sprayed reach had begun to regenerate, but woody bank vegetation remained sparse.

A sequence of low-to-moderate flow years followed the 2006 flood until 15 September 2013, when >7.6 cm of precipitation over much of the basin during the preceding week caused flow in the Rio Puerco near Bernardo, New Mexico to increase from <28 m³/s to a peak discharge of 255.4 m³/s. This peak was 45% greater than the August 2006 flood peak (175.9 m³/s) and was the second highest peak since 1947. Observations from an aerial LiDAR survey in March 2010 and high-resolution satellite imagery acquired in January and February 2014 show that channel bank erosion in the sprayed reach continued at high rates compared to bank erosion in the untreated reach downstream. Channel width in the sprayed reach increased from an average of 26.1 m in March 2010 to an average of 28.2 m in January/February 2014, while average channel width in the downstream reach remained the same at 11.6 m. An unexpected finding after the 2013 flood was that the extent of arroyo wall erosion in the downstream reach was about twice that in the sprayed reach. In the sprayed reach, wall erosion was dominantly within sharp bends where channel flow eroded the base of the unprotected wall. In contrast, wall erosion in the downstream reach included channel bends as well as straight arroyo segments, where the down-valley flood flow was constricted within limited open paths, including those between dense shrub patches and the arroyo wall.