Many state and local governments in the U.S. are required to develop water quality management plans in partial fulfillment of the federally mandated National Pollutant Discharge Elimination System (NPDES) permit.  These plans also provide a framework for watershed restoration that enables the agency to achieve the greatest environmental benefit within the given watershed for the money available.

Two neighboring jurisdictions in the state of Maryland, Baltimore City and Baltimore County, are working together to meet the NPDES requirements and to improve water quality in the Chesapeake Bay Region.  In October 2002, the Baltimore City Department of Public Works and Baltimore County Department of Environmental Protection & Resource Management signed a historic watershed agreement that unified the city's and county's efforts to protect watersheds that are located in both jurisdictions.  This agreement calls for:

The Gwynns Falls Watershed Assessment

One of the first projects that resulted from this agreement was the Gwynns Falls Water Quality Management Plan. The Gwynns Falls watershed begins in western Baltimore County and drains into Baltimore City.  The streams drain to the tidal area of the Middle Branch of the Patapsco River in the western part of the Baltimore Harbor in downtown Baltimore, and ultimately into the Chesapeake Bay. 

PB is helping Baltimore County and Baltimore City in their effort.  The goals were to:      

• Identify and evaluate non-point source storm water pollution of the 170-km² (66-square-mile) watershed      
• Assess more than 214 km (133 miles) of streams
• Determine management and restoration measures to reduce non-point source pollution and reestablish stream stability
• Provide a watershed restoration framework and planning tool for capital projects for both jurisdictions.

County and City Watershed Characterizations.  Gwynns Falls is a highly urbanized watershed.  The county subsheds, particularly in the northern portion of the watershed, are a mix of residential, commercial and industrial land uses, although there still is a significant amount of forested area and park lands.  Much of this development was constructed when early storm water management guidelines were in place, so it has water quantity controls, but no water quality measures were installed. 

The city subsheds are characterized by the high imperviousness associated with dense residential (row houses), commercial and industrial development, the majority of which was also constructed before storm water management regulations were established.  Many of the city subsheds have been subject to significant storm sewer overflow events. 

Data Collection.  Two of our tasks, the stream stability assessment and the stormwater management (SWM) facility and outfall assessment, required extensive field data collection.  We developed standardized entry forms that we used on pocket PCs-an approach that enabled the field crews to cover approximately 1.6 km (1 mile) each day, much more than they could have without the PCs, and to spend less time in the office on data entry.

Stream Stability Assessment.  Cruising is defined as a team of two stream surveyors walking the entire length of a reach and performing rapid field assessments.  Our field teams cruised more than 214 km (133 miles) of stream channel and almost 900 stream reaches.  That was a lot of hiking! 

We took measurements using a stretched tape and surveyor's rod to assess bankfull width and depth (Figure 1) and many other parameters.  We collected information in four major categories:           

Results of our assessment showed the general stream channel characteristics to be as follows:                              

• Most are perennial, meaning they contain water at all times except during extremely dry weather      
     
• Bankfull widths are between 1.5 and 4.5 m (5 and 15 feet)
     
• 25 percent of the streams are Rosgen G stream types (gullies) and an additional 10 percent are Rosgen F stream types (over widened and entrenched) 
     
• More than 30 percent of the stream channel banks are unstable and subject to erosion      
     
• More than 50 percent of the streams are moderately or severely entrenched
     
• Almost half of the stream reaches have been altered due to urbanization.
     

In addition, continuous sewage leaks were evident and several of the streams had been posted for water quality concerns.

The typical stream buffer characteristics were as follows:

• More than half of the stream reaches have greater than 50 percent canopy cover.      
• Seventy-seven percent of the riparian buffers consist of deciduous over story with brush under story.
• More than half of the streams have more than 15 m (50 feet) of riparian buffer on each side.

Storm Water Management Facility and Outfall Assessments:  We assessed 39 storm water management facilities and 73 major storm drain outfalls in the county and city in order to identify:

• Those storm water management facilities that had the potential to be converted and enhance water quality      
• Structural deficiencies in the storm water management facilities and downstream channel instabilities
• Opportunities for best management practices creation at existing storm drain outfalls      
• Structural deficiencies at the storm drain outfalls (Figures 2 and 3).

The results, which were based also on water quality modeling discussed below showed that converting standard SWM detention facilities to extended detention facilities would provide water quality treatment and reduce the peak discharges that enter the stream channel.  Standard detention facilities typically have a 24-hour detention time.  Extended detention facilities typically hold the flood volumes for a 48-hour period and often have pretreatment facilities associated with them, such as forebays.  The lower discharges and velocities would reduce stream power and, consequently, channel erosion. 

Based on these findings, we recommended 81 sites in the county and city for the following water quality retrofits and improvements: 

• SWM facilities.  In addition to conversion to extended detention, typical retrofits included addition of forebays or other pretreatment options and the creation of shallow marshes.       
• Storm drain outfalls.  Typical retrofits included energy dissipation devices, such as plunge pools, creation of shallow floodplain marshes, creation of bioretention facilities, channel daylighting, and stabilizing the downstream channels.  (Daylighting refers to taking closed systems, such as storm drain pipes and converting them to natural, open channels.)

Water Quality Modeling:  Hydrologic and non-point source pollutant modeling was conducted using the EPA Storm Water Management Model (SWMM).  Existing and ultimate land use conditions were modeled in order to develop watershed management recommendations.  Twelve pollutants were analyzed, including nutrients, metals, BOD/COD and fecal coliform.  (BOD, or biological oxygen demand, is the amount of oxygen that would be consumed if all the organics in one liter of water were oxidized by bacteria and protozoa,  COD, or chemical oxygen demand tests indirectly measure the amount of organic pollutants in water.  Both BOD and COD are measures of water quality.)

The Gwynns Falls watershed was divided into 505 catchments with an average catchment size of 0.34 km² (83 acres).  The model simulation ran from 1980 through 2002, a 23-year period.  XPSWMM and ArcView were used to analyze watershed data.

Forest Assessment.  A rapid field assessment was conducted of 34 forest patches within the watershed.  The goal was to look at habitat and connectivity of forested areas throughout the watershed and stream buffers. 

Watershed Evaluation and Restoration Opportunities

Following the comprehensive assessments described above, we defined management and restoration goals for each subshed-the county's and the city's.  Potential projects were identified using the results of the field assessments and the management goals for each subshed.  The complexity of data collected during this study led us to develop the following two-tiered approach to aid in project identification:

We identified more than 120 water quality improvement projects throughout the watershed (Table 1) with a total cost of approximately $30 million.  These projects included more than:

The proposed stream restoration and riparian enhancement projects restore 25 percent of the watersheds' most unstable streams.  Stabilizing these streams reduces sediment loads within the watershed by approximately 26 percent. 

Prioritization.  The city and county face limited budgets for water quality projects, so we developed a comprehensive ranking system that incorporates environmental benefits and cost.  In developing the ranking system, our aim was to have one that was well defined and balanced and allowed a direct comparison of competing projects.  We used the following five criteria:      

Project Achievements

The key to developing a successful plan for the Gwynns Falls watershed was to examine the needs of the entire watershed first and then work to meet the needs of the individual constituents.  With the completion of the Gwynns Falls Water Quality Management Plan, the city of Baltimore and Baltimore County now have a comprehensive water quality plan that will enable them to make the most efficient use of the limited funds available.  The plan serves as a framework and provides multiple tools for the agencies to select the best project, based on cost, subwatershed area or overall pollutant reductions.  By combining capital projects with community education and outreach, the city and county can work towards achieving their water quality goals. 

We have subsequently been awarded another contract by the Baltimore County Department of Environmental Protection and Resource Management to perform a series of smaller, more detailed subwatershed assessments.  We have recently completed the Hunt Valley Stream Stability Assessment in 2005 and the Pretty Boy Stream Stability Assessment during the summer of 2006.