ASAE Conference Proceeding
This is not a peer-reviewed article.
Linville Creek TMDL for a Benthic Impairment
G. Yagow, S. Mostaghimi, T. Dillaha, K. Brannan, J. Wynn, R. Zeckoski, and B. Benham
Pp. 395-400 in Total Maximum Daily Load (TMDL) Environmental Regulations II, Conference Proceedings, 8-12 November 2003 (Albuquerque, New Mexico, USA), ed. Ali Saleh. ,8 November 2003 . ASAE Pub #701P1503
A TMDL was developed using the reference watershed approach to address the benthic impairment identified in Linville Creek, Rockingham County, Virginia. A stressor analysis performed on existing monitoring data identified sediment as the major stressor for Linville Creek. The Upper Opequon Creek was selected as the TMDL reference watershed from a list of watersheds around the state with benthic monitoring data and a non-impaired status. The selection was based on the comparability of land use distribution, ecological, and sediment-generating characteristics with those of the Linville Creek watershed. Modeling was performed with a modified version of the GWLF model. Model inputs were developed for both watersheds considering surface runoff, stream bank and channel erosion, and point sources of sediment. Model simulations were performed over a 10-year period to generate average annual sediment loads from all sources within each watershed. The TMDL was defined as the average annual unit-area sediment load in the TMDL reference watershed times the area of the Linville Creek watershed. Allocation scenarios were developed using reductions from agriculture and from channel erosion sources related to livestock stream access.KEYWORDS. Benthic impairment, GWLF, Reference watershed approach, Sediment, Stressor analysis, TMDL.
A total maximum daily load (TMDL) is a pollution budget for an individual pollutant that accounts for all sources of an identified pollutant received by a given impaired stream segment. The TMDL furthermore defines the allowable load and load reductions needed for that segment to be in compliance with its applicable water quality standard. The TMDL program was created by the 1972 Clean Water Act to identify and remediate all stream segments in violation of states water quality standards. While a robust effort made by the states addressed impairments related to point sources of pollution, impairments due to nonpoint source (NPS) pollution were initially avoided. In the late 1990s, a number of different conservation and environmental groups and organizations sued the Environmental Protection Agency (EPA) for neglecting its responsibility to address NPS pollution under this Act. This litigation took place in 40 states and resulted in court orders or consent decrees in 22 of those states that specified hard and fast-paced schedules for TMDL plan development to remedy this past neglect. Virginia was one of these states.
Many states monitor streams for some type of biological impairment. In Virginia, biological monitoring is conducted by the Virginia Department of Environmental Quality (DEQ) using EPA’s Rapid Bioassessment Protocol II (RBP II) to assess the health of the benthic macroinvertebrate community (Barbour et al., 1999). When the assessment shows a reduction in the health of this community, the stream is declared to have a “benthic impairment”, and is deemed to be in violation of Virginia’s narrative General Standard for Aquatic Life (9 VAC 25-260-20). This standard states that “all state waters…shall be free from substances…which contravene established standards…or which are…harmful to human, animal, plant, or aquatic life”. In 1998, Linville Creek was placed on EPA’s 303(d) list as having a “moderate” benthic impairment based on the RBP II assessments. Agricultural NPS pollution was identified as the most probable cause of the impairment.
The most common causes of stress on the benthic community include excessive sediment, nutrients, toxics, organic matter, suspended solids, elevated temperatures, and extreme pHs, as well as channel or runoff modifications within the watershed. Virginia, like many other states, does not have numeric standards for many of these stressors. Therefore, an alternative approach must be used to establish a numeric TMDL load. The TMDL load for Linville Creek was defined using the reference watershed approach. This approach identifies a comparable watershed within the same physiographic region that has a healthy benthic community. The TMDL load (t/yr) is calculated as the modeled unit-area load of the identified pollutant from the benthically unimpaired TMDL reference watershed (t/ha-yr) times the area of the impaired watershed (ha), and becomes the target for pollutant reductions from the impaired watershed.
The impaired segment of Linville Creek is the entire main branch from its headwaters for 21.8 km (13.55 mi.) in a northeasterly direction to its confluence with the North Fork of the Shenandoah River (USGS Hydrologic Unit Code 02070006). The North Fork of the Shenandoah River is a tributary to the Potomac River, which flows into the Chesapeake Bay.
The Linville Creek watershed is 11,998 ha in size and is located in Rockingham County, Virginia, in the Valley and Ridge physiographic region. Linville Creek flows through a mainly agricultural watershed, located in a rolling valley with the Blue Ridge Mountains to the east and the Appalachian Mountains to the west. Pasture is the main land use in the Linville Creek watershed, comprising 49% of the total area, with cropland and forest accounting for 21% and 16%, respectively. Residential and urban developments, the remaining 9%, are spread throughout the watershed with a slight concentration around the town of Broadway near the outlet. The predominant soil groups found in the Linville Creek watershed are the Frederick-Lodi-Rock outcrop, Endcav-Carbo-Rock outcrop, and Chilhowie-Edom soils. A USGS flow gaging station (01632082) is located near the mouth of Linville Creek at an elevation of 313.7 m (1,029.9 ft). Mean daily streamflow at this gage ranged from 11.2 to 294 cfs, with an overall daily mean of 36.8 cfs during the 17 years of record. Average annual precipitation in the watershed is 89.7 cm (35.3 in).
Because a benthic impairment is based on an assessment of benthic macroinvertebrates, rather than on specific pollutants, the cause of the impairment is not explicitly identified. The procedure used to identify the critical stressor for Linville Creek is outlined in the United States Environmental Protection Agency’s Stressor Identification Guidance Document (EPA, 2000), and is referred to as a stressor analysis. Stressor analysis involves examining water quality and bioassessment data to look for the most probable stressor – the pollutant or physical condition – causing the benthic degradation. The analysis is performed by looking at monitored and qualitative data that relate to each stressor, and assessing whether the data supports, refutes, or is inconclusive in its support of that stressor. Specific data are evaluated for each potential stressor. For example, the degree of embeddedness and total habitat scores from the RBP II habitat assessment, and observations of stream bank erosion and degradation in the watershed, are useful sources of information to gauge the impact of sediment. For some stressors, it is important to look at multiple pathways to make sure that all possible routes are explored in linking cause and effect. For instance, nutrients may be elevated above eutrophic sufficiency levels, and nitrates may even exceed their water quality standard, but unless these are accompanied by depressed dissolved oxygen readings, the nutrients are probably not the main cause of stress on the benthic community. Once the main stressor(s) of the benthic impairment are identified, a TMDL is developed for each specific stressor or pollutant.
In Linville Creek, data used for the stressor analysis were obtained primarily from DEQ’s ambient water quality monitoring program and were supplemented by project team observations during several watershed visits (Mostaghimi et al., 2003). Chemical and bacterial water quality in the watershed was monitored on a monthly basis from September 1993 through June 2001, and on a bimonthly basis from July 2001 through April 2002. RBP II biological sampling and habitat evaluations were performed semi-annually from October 1994 through May 2002.
Sediment was identified as the critical stressor, and therefore became the target pollutant in the benthic TMDL for Linville Creek. The evidence supporting sediment as the primary stressor came from several sources. Many of the scores for one of the benthic metrics (%haptobenthos) indicated poor habitat for functional groups requiring a coarse, clean sediment substrate. Linville Creek also received repeated low habitat scores for bank stability, substrate availability, bank vegetation, riparian vegetation, and embeddedness. Additionally, there was observed damage to stream banks from livestock trampling. Although not an overwhelming case, sediment was selected as the most likely stressor on the benthic community because it does play such a major role in the benthic community, because other potential stressors such as nutrients and organics are often associated with sediment and will be reduced when sediment is reduced, and because the above observations were consistent with an impairment due to sediment.
The reference Watershed Approach and the TMDL Load
With sediment identified as the stressor, the reference watershed approach was then used to define the target sediment load for the benthic TMDL (Mostaghimi et al., 2003). In this approach, watersheds were identified that were in the same Valley and Ridge physiographic region of Virginia as Linville Creek, and were classified as non-impaired based on the DEQ biological monitoring. The watershed characteristics in Table 1 were then compared in an effort to select the most comparable watershed. From the 7 potential watersheds identified, the Upper Opequon Creek watershed was selected as the TMDL reference watershed for Linville Creek. Land use distribution was the most important characteristic considered in this comparison. The Upper Opequon was the only watershed in the group that was predominantly agricultural with a small urban component. The Upper Opequon watershed is also located in the same Level III ecoregion – Central Appalachian Ridges and Valleys – as Linville Creek, and the majority of both watersheds lie in the same major Level IV ecoregion – Northern Limestone/Dolomite Valleys.
Table 1. Potential TMDL Reference Watersheds and Watershed Characteristics.
The benthic TMDL for the Linville Creek watershed was developed using sediment loads generated by the Generalized Watershed Loading Functions model (GWLF; Haith et al., 1992). Even though GWLF was originally developed for use in ungaged watersheds, the BasinSim adaptation of the model (Dai et al., 2000) recommends hydrologic calibration of the model. Because observed daily flow data were available at both Linville Creek and its TMDL reference watershed, and because preliminary calibrated model results for Linville Creek watershed showed an 18% reduction in the percent error between simulated and observed monthly runoff, hydrologic calibration was performed on both watersheds. To ensure comparability between the impaired and TMDL reference watersheds, GWLF parameters for both watersheds were calibrated for hydrology in a consistent manner. Hydrologic calibration was accomplished through adjustments to the recession coefficient, the seepage coefficient, the area-weighted seasonal evapo-transpiration coefficients, and a curve number multiplier (Mostaghimi et al., 2003).
Sediment was modeled within GWLF in the following manner. In-stream sediment in the watershed was generated by surface runoff from both pervious and imperious areas, by channel erosion, and from permitted discharges. Pervious area sediment loads were modeled through sediment detachment and modified USLE erosion algorithms, and a sediment delivery ratio was applied to the pervious loads to calculate watershed outlet loads. Impervious area sediment loads were modeled using an exponential buildup-washoff algorithm. Channel erosion was modeled within GWLF using the algorithms included in the AVGWLF adaptation of the GWLF model (Evans et al., 2001). In these equations, channel erosion was calculated as a function of daily stream flow volume and a regression coefficient. This regression coefficient was calculated as a function of the percentage of developed land, animal density, watershed-averaged soil erodibility, the watershed-averaged runoff curve number, and total stream length. Daily loads were aggregated to monthly loads for each land use category.
Sediment loads from point sources were calculated using TSS concentrations and flow volumes. For existing loads from permitted Virginia Pollutant Discharge Elimination System (VPDES) facilities, available monthly discharge monitoring report (DMR) data for each facility (maximum concentration and maximum daily flow) were used to calculate average daily TSS loads. Sediment loads from 1000 gallon per day (gpd) general permit facilities were calculated as the number of facilities multiplied by the annual permitted TSS load for each facility.
The data used to evaluate model parameters were obtained from a variety of sources. Digital data were used wherever possible to enable parameter evaluation using ArcView 3.3 GIS software. Land use was obtained from the 1992 MRLC data layer with modifications for land use classifications as used by Virginia in its 2002 statewide NPS pollution assessment (Yagow et al., 2002). County level soil surveys, 30-meter DEMs, and USGS National Hydrography Dataset (http://nhd.usgs.gov) stream layers were all used as part of the parameter evaluation process along with the guidance provided in the GWLF user’s manual (Haith et al., 1992). Daily temperature and rainfall were obtained from a statewide database of NWS/NOAA gage sites assembled for use with Virginia’s 2002 NPS assessment. Daily flow data from USGS were obtained for use in calibrating hydrologic parameters in the GWLF model. Livestock populations and locations were obtained for the purpose of assessing livestock stream access and its impact on channel erosion.
TMDL model simulations were performed using the GWLF model of each watershed over a 10-yr period from January 1988 to December 1997. Sediment loads for existing conditions are listed in Table 2 for each watershed by land use category and percent of total load. The target TMDL load for the impaired Linville Creek watershed (34,549.3 t/yr) was defined as the 10-yr average annual unit-area sediment load (t/ha-yr) from the TMDL reference watershed multiplied by the area of the Linville Creek watershed (ha.).
Table 2. Existing Sediment Loads in Linville and Upper Opequon watersheds
The TMDL for Linville Creek, based on sediment loads, consists of the three required components – point source loads (WLA), nonpoint source loads (LA), and a margin of safety (MOS) – as quantified in Table 3. The margin of safety was explicitly calculated as 10% of the TMDL load. The WLA load was calculated from the maximum permitted flows and total suspended solids (TSS) concentrations for all point sources in the watershed. The allowable LA load was calculated as the remainder (LA = TMDL – WLA – MOS).
Table 3. Linville Creek TMDL Sediment Loads (t/yr)
TMDL Allocation Scenarios
Because little land use change was expected in the future for Linville Creek watershed, TMDL allocations were based on existing sediment loads. Allocation scenarios were created to suggest combinations of various source category reductions that could be used to reach the TMDL target load. The allowable sediment load for allocation among the modeled sediment source categories is the sum of the WLA and LA loads in Table 2 (31,094.3 t/yr). To develop the allocation scenarios, sediment sources were grouped into four categories: Agriculture, Urban, Channel Erosion, and Point Sources, as shown in Table 3. Because all Point Source sediment loads are permitted, and because Urban sources contributed an insignificant amount of sediment, no reductions were taken from these two categories. The three alternative allocation scenarios shown in Table 4, therefore, were developed with varying percentage reductions from the remaining Agriculture and Channel Erosion categories. The benthic TMDL for Linville Creek will require an overall reduction of 12.3% from existing sediment loads.
Table 4. Alternative Sediment Load Allocation Scenarios for Linville Creek Watershed.
The Linville Creek stream segment was assessed as having both benthic and bacteria impairments. Although a separate TMDL must be developed for each impairment, changes in land use management called for in one TMDL may have implications for the pollutant loads being addressed by other concurrent TMDLs in the same watershed. Such is the case in Linville Creek. The bacteria TMDL for Linville Creek called for a 100% reduction in livestock access to streams as part of its required reductions (Mostaghimi et al., 2003). Since restricting livestock access to streams has a major impact on stream bank stabilization and sediment generation in those areas, the reductions called for by the bacteria TMDL will have a synergistic effect on the benthic TMDL. The Channel Erosion load in Scenario 3 was calculated to reflect the reduction in livestock stream access areas called for in the bacteria TMDL. The reduction percentage from restricting livestock access to streams was calculated as the product of the percentage of total stream length with livestock access (46.2%), the percentage reduction of livestock access corresponding with the bacteria TMDL, and an estimated 50% effectiveness of the livestock access restriction practice. Scenario 3 is the recommended scenario for the benthic TMDL, because it minimizes the total reductions called for from Agriculture, by crediting mutually beneficial reductions in Channel Erosion from the concurrent bacteria TMDL. This effectively reduces the additional reductions called for in the benthic TMDL from 12.3% to 7.8% of the existing sediment loads.
The Linville Creek benthic TMDL (Mostaghimi et al., 2003) was developed based on the unit-area sediment load from a comparable TMDL reference watershed – the Upper Opequon Creek. The TMDL was developed to take into account all sediment sources in the watershed from both point and nonpoint sources. The sediment loads were averaged over a 10-year period to take into account both wet and dry periods in the hydrologic conditions, and the model inputs took into consideration seasonal variations and critical conditions related to sediment loading. An explicit 10% margin of safety was added into the final TMDL load calculation. Final TMDL allocations were developed for major land use categories with consideration of mutually beneficial reductions from a concurrent bacteria TMDL.
Barbour, M. T., J. Gerritsen, B. D. Snyder, and J. B. Stribling. 1999. Rapid bioassessment protocols for use in streams and wadeable rivers: Periphyton, benthic macroinvertebrates, and fish. Second Edition. EPA 841-B-99-002. U. S. Environmental Protection Agency. Washington, DC.
Dai, T., R. L. Wetzel, Tyler R. L. Christensen, and E. A. Lewis. 2000. BasinSim 1.0, A Windows-Based Watershed Modeling Package. User’s Guide. Special Report in Applied Marine Science and Ocean Engineering #362. Virginia Institute of Marine Science, College of William & Mary. Gloucester Point, Virginia.
EPA. 2000. Stressor identification guidance document. EPA 822/B-00/025. U. S. Environmental Protection Agency, Office of Water, Office of Research and Development. Washington, DC. (www.epa.gov/ost/biocriteria/stressors/stressorid.pdf ).
Evans, B. M., S. A. Sheeder, K. J. Corradini, and W. S. Brown. 2001. AVGWLF version 3.2. Users Guide. Environmental Resources Research Institute, Pennsylvania State University and Pennsylvania Department of Environmental Protection, Bureau of Watershed Conservation.
Haith, D. A., R. Mandel, and R. S. Wu. 1992. GWLF. Generalized Watershed Loading Functions, version 2.0. User’s Manual. Department of Agricultural and Biological Engineering, Cornell University. Ithaca, New York.
Mostaghimi, S., B. Benham, K. Brannan, T. Dillaha, J. Wynn, G. Yagow, and R. Zeckoski. 2003. Total Maximum Daily Load Development for Linville Creek: Bacteria and General Standard (Benthic) Impairments. Biological Systems Engineering Department, Virginia Tech, Blacksburg, Virginia. (www.deq.state.va.us/tmdl/tmdlrpts.html ).
Yagow, G., S. Mostaghimi, and T. Dillaha. 2002. GWLF model calibration for statewide NPS assessment. Virginia NPS pollutant load assessment methodology for 2002 and 2004 statewide NPS pollutant assessments. January 1 – March 31, 2002 Quarterly Report. Submitted to Virginia Department of Conservation and Recreation, Division of Soil and Water Conservation. Richmond, Virginia.