The experimental reach used in this study is located on the Virginia Tech Foundation’s Heth Farm within Blacksburg town limits, approximately 2.5 km downstream from the Virginia Tech Duck Pond, located on the main campus. The drainage area used in this case study includes 1,440 ha, almost all of which lie in the Town of Blacksburg. The reach was selected because it is experiencing severe bank retreat. The upstream terminus of the study reach is located just downstream of a small tributary and the modeled section extends 500 m downstream. To simplify modeling procedures, the study reach was chosen to avoid hydraulic structures and tributary inflows.

A natural levee separates the incised channel from floodplain areas. The immediate area surrounding the experimental reach is pasture grazed by non-lactating dairy cattle, but the overall contributing area is dominated by residential and urban development. Cattle have unlimited access to the stream corridor and contribute to streambank instability. The floodplain vegetation surrounding Stroubles Creek on Heth Farm is composed of cool season grasses. Pocket wetlands formed in abandoned oxbows harbor zones of wetland vegetation, such as rushes (Scirpus sp.). Abandoned meander bends are visible on high resolution aerial photography suggesting past channel straightening.

The Stroubles Creek watershed has been influenced by human activity since the 18th century. Upland sediments, exposed by logging and agricultural activities, were eroded and transported downslope to floodplain areas. Significant silt deposits are now visible in Stroubles Creek bank profiles, with deposit depths ranging from several centimeters to over 50 cm. These fine textured legacy sediments are particularly susceptible to weakening and erosion due to subaerial processes. A clearly defined layer of McGary silt loam is visible on exposed bank faces throughout the reach. Ongoing development and past disturbance have resulted in an experimental reach characterized by unstable actively eroding stream banks. Downcutting has disconnected Stroubles Creek from the floodplain creating banks as high as 2.0 m and resulting in increased flow velocities and boundary shear stresses. Portions of the channel have downcut to an underlying erosion resistant saprolite layer. Underlying bedrock now prevents further downcutting and limits bed sediment supply. This natural bed control forces the stream to adjust laterally as it attempts to convey larger peak discharges associated with urban development.

Urban development has significantly altered the natural hydrology of Stroubles Creek. Changes in discharge occur rapidly following periods of rainfall. Observations indicate a two-hour time of concentration. Urbanization has also intensified differences between baseflow discharge and peak discharges. Baseflow discharge ranges from 0.03-0.17 cms, while bankfull discharges are approximately 8.5-9.9 cms.

Objectives & Methodology

The objective of this study was to compare predicted sediment loadings from in-stream sources as predicted by GWLF, SWAT, and CONCEPTS. Model comparisons were also contrasted with channel degradation estimates derived from a system of erosion pins and scour chains.

GWLF Model:
This case study utilized ArcView GWLF version 2005b software for channel degradation simulation. GWLF simulation was broken into two phases: a channel erosion sensitivity analysis phase and a one-year channel erosion prediction phase, concurrent with monthly erosion pin measurements. The sensitivity analysis of GWLF to flow and channel parameters was performed before the GWLF case study simulation.

SWAT Model:
The July 2005 update of SWAT 2000 (B. Narasimhan, personal communication, August 3, 2006), was used in the model comparison case study. The July update of SWAT 2000 includes corrections to the sediment routing algorithms. The ArcView SWAT 2000 (AVSWAT 2000) GUI was used to parameterize the model and create the input files. Watershed data input requirements for SWAT are broken down into three groups based on watershed subdivision structure: watershed, subbasin, or Hydrologic Response Unit (HRU) level parameters (Neitsch et al., 2002). The SWAT parameter set used in this research was adapted from the Wagner (2004) SWAT simulation. Necessary modifications to the Wagner (2004) parameter set, such as subbasin redistribution, were made using data available in the Stroubles Creek TMDL (Mostaghimi et al., 2003). Before running the 2005-2006 SWAT simulation a sensitivity analysis was performed.

CONCEPTS Model:
The CONCEPTS reach-scale model was used to simulate channel degradation for the Stroubles Creek Heth Farm study reach and compare the output with GWLF and SWAT. The objective of this effort was to evaluate if a more process-based, reach-scale model, coupled with a watershed-scale model, more accurately predicts channel degradation. A sensitivity analysis of CONCEPTS channel parameters was performed before simulating the case study scenario.

Monitoring Data:
In conjunction with the stream degradation modeling effort, an erosion pin monitoring study was conducted to quantify streambank retreat within the experimental reach. Streambank retreat rates were measured using erosion pins and scour chains on the 500-m long reach. Pins made from #316, 6-mm diameter stainless steel rod stock measuring 50 cm in length were marked on one end with 2.5 cm of brightly colored heat-shrink tubing and placed on actively eroding stream banks in a systematic 10-m horizontal and 0.3-m vertical grid. In total, 268 erosion pins and 7 scour chains were initially installed along the 500-m reach.

The point data collected through erosion pin monitoring was transformed into monthly mass sediment loadings to the stream. Point data were extrapolated to create an average predicted retreat surface. In an effort to limit extrapolation error, pin data were only applied to the continuous bank surface surrounding the pin, i.e. the directly connected, actively eroding portion adjacent to the pins. A measuring tape was used to estimate bank segment height and length. The erosion pin measurements were compiled and averaged for each individual bank. The average segment-specific retreat rate was multiplied by an estimated segment surface area. The soil volume was then multiplied by an average bulk density of 1300 kg/m3 as reported by Wynn, 2004 and Henderson, 2006.

Results and Conclusions

Sediment loading to the stream from bank retreat was estimated as 41 tonnes/yr, based on erosion pin measurements. GWLF, SWAT, and CONCEPTS predicted stream channel sediment contributions of 8 tonnes/yr, 1500 tonnes/yr and 4 tonnes/yr, respectively. Theil-Sen non-parametric simple linear regression was used to test agreement between monthly model predictions and erosion pin estimates. No significant agreement was found between any model predictions and measured retreat, using a conservative α-value of 0.2. GWLF model predictions underpredicted measured channel degradation, but most closely approximated observed data. This result is likely due to similarities in climate and watershed characteristics for the Stroubles Creek watershed and the Pennsylvania watershed used in the empirical model development. SWAT predicted retreat rates exceeded measured values by two orders of magnitude. This result is explained by the inability of SWAT to predict daily flow and sediment discharge. Highly sensitive channel degradation parameters and the lack of calibration data also contributed to SWAT simulation error. CONCEPTS simulation predicted monthly retreat rates slightly less than GWLF. The lack of agreement between CONCEPTS simulation and observed data was mainly the result of limited input data availability. SWAT daily discharge predictions were used as CONCEPTS input data and likely contributed to poor model agreement. Poor estimation of sensitive sediment input parameters may have also contributed to underpredictions by CONCEPTS. Results showed the potential of screening-level watershed models in channel degradation prediction and the importance of flow and sediment time series discharge data in detailed process-based simulation. The limited flexibility of the GWLF channel degradation algorithm makes it unsuitable for evaluating the effects of stream restoration. SWAT and CONCEPTS should only be used for evaluation if appropriate input data are available.

References

Henderson, M. B. 2006. Changes in streambank erodibility and critical shear stress due to surface subaerial processes. MS thesis. Blacksburg, VA: Virginia Tech. Department of Biological Systems Engineering.

Mostaghimi, S., Brian Benham, Kevin Brannan, Theo A. Dillaha, III, Rachel Wagner, Jeff Wynn, Gene Yagow, Rebecca Zeckoski. 2003. Benthic TMDL for Stroubles Creek in Montgomery County, Virginia. Prepared for VADEQ, Virginia Department of Environmental Quality. Richmond, Virginia.

Neitsch, S. L., J. G. Arnold, J. R. Kiniry, R. Srinivasan, J. R Williams. 2002. Soil and Water Assessment Tool, User’s Manual. Blackland Research Center, USDA-ARS, Temple, Texas.

Wynn, T. M. 2004. The effects of vegetation on stream bank erosion. PhD dissertation. Blacksburg, VA: Virginia Tech. Department of Biological Systems Engineering.

Additional Project Information

More detailed information on this case study can be found in the project’s final report available here.