Abstract

Data collected from 2002 through 2008 by the U.S. Geological Survey, in cooperation with the U.S. Environmental Protection Agency, were used to assess geomorphic characteristics and geomorphic changes over time in a reach of Minebank Run, a small urban watershed near Towson, Maryland, prior to and after its physical restoration in 2004 and 2005. Data collection included continuous-record streamflow and precipitation; surveyed elevations of the channel bed, water surface, and bank features; surveyed cross sections; pebble counts from the channel bed; measurements of bed elevation over time; and high-water marks from storm runoff events.

To provide hydrologic context to the geomorphic monitoring that was conducted during the study period, precipitation, discharge intensity, time of concentration, flood frequency, and rainfall-runoff relations were investigated. Hydrologic conditions were found to be comparable for the pre- and post-restoration monitoring periods. Flood frequency analysis indicated that the largest peak flows that occurred in the Minebank Run watershed during 2002–08 could represent events of lower probability when compared to a longer peak-flow record in the adjacent West Branch Herring Run watershed.

Longitudinal profiles of the channel bed and water surface indicated differences in the distribution and location of riffles, pools, and runs during 2002–08. During both the pre- and post-restoration monitoring periods, the analyses indicated that on average, the stream is maintaining the overall slope of the channel bed and water surface at about 1 percent, despite considerable changes in the percentages of riffles, pools, and runs, and changes in the distribution and location of these features.

Post-restoration, lateral erosion has been reduced with fewer indications of channel widening. Flood flows can now inundate sections of the overbank area, and also bypass the main channel in small sections of the study reach, which was not possible before restoration. Much of the post-restoration geomorphic variability is due to alternating patterns of sediment storage and removal, and shifting of the channel thalweg, in contrast to channel degradation and widening, and lateral erosion from receding cut banks observed during the pre-restoration monitoring. Reduced variability in cross-sectional area and mean depth between surveys conducted from 2006 through 2008 indicates that the stream channel could be establishing a dynamic equilibrium and a more stable geometry after the initial geomorphic response observed in 2005 and 2006, just after restoration in 2004 and 2005.

Most channel geometry variables had moderate to strong linear relations with discharge. Pre-restoration, the strongest relation was between cross-sectional area and discharge. Post-restoration, the strongest relation was between mean flow velocity and discharge.

Composite particle-size analyses of the channel bed from pebble counts over time indicated that sources of fine sediment, possibly from bank erosion, still exist in the watershed despite restoration of the stream channel. The abundance of relatively small bed material sizes in combination with flashy streamflow from urban and suburban runoff likely contributes to the considerable changes in grain-size distribution and alternating periods of storage and transport of sand and gravel.

Sequential measurements of bed elevation over time in three selected locations of the study reach indicated a period of considerable vertical adjustment between September 2005 and December 2006 based on post-restoration monitoring. One of the three selected locations was in one of the most geomorphically unstable sections of the study reach before restoration, and was found to be considerably less prone to rapid and extreme changes in bed elevation after restoration of the stream channel.

An analysis of boundary shear stress was conducted by use of measured channel-geometry variables and water-surface slopes in the vicinity of the continuous-record streamgage. The results indicated that, post-restoration, larger increases in mean velocity were required to initiate sediment transport in the stream channel.

Pre-restoration, sediment volume computations indicated that sediment was being removed from the study reach, with the largest sediment volume and rate of removal occurring during December 2002 through July 2003. Post-restoration, an alternating pattern of sediment removal and storage was observed, with the largest sediment volume and rate of removal occurring during April 2005 to December 2005, just after completion of the restoration. The largest volumes and rates of sediment storage during the post-restoration period occurred between December 2005 and September 2006. The storm and flood of June 25, 2006 was a likely factor in the volume of sediment stored in the study reach during this period.

Comparing an early 1960s channel bed profile to those developed during 2002–08 confirmed that the stream channel has maintained an overall slope of about 1 percent for greater than 40 years. Comparisons of bed elevations between the early 1960s and 2002 indicate that the channel bed degraded within an approximate range of 1.3 ft (feet) to 5.0 ft, and with possible local scour of up to 5.6 ft near a sanitary sewer pipe that was exposed within the channel bed. In the post-restoration condition, the channel bed is still about 0.8 ft to 4.8 ft lower than bed elevations from the early 1960s at approximately equivalent locations along the profile.

First posted February 14, 2012