The Wissahickon Creek Watershed Stormwater Basin Inventory was implemented by the Philadelphia Water Department (PWD) as part of a larger watershed-wide planning initiative under way in the Wissahickon Creek Watershed (WCW). PWD began initiating intergovernmental, multijurisdictional partnerships in 1999 to create and implement watershed-focused strategies for ecological restoration and water quality improvement.
These initiatives recognize that watersheds do not stop at municipal borders. Development trends in upstream communities dramatically affect downstream waterways and natural resources. PWD initiated a watershed planning process within WCW in 2005. The Wissahickon Basin Inventory initiative presented an opportunity for implementing a mutually beneficial data gathering process between upstream and downstream municipalities.
WCW has a total drainage area of approximately 64 square miles. According to the Delaware Valley Regional Planning Commission's (DVRPC) year 2000 land-use data, more than 65 percent of this watershed area has been developed. This development has resulted in land cover that is mostly impervious to water. Furthermore, large portions of WCW experienced extensive land development prior to the implementation of stormwater management controls required by the Pennsylvania Stormwater Management Act of 1978. According to the Wissahickon Creek River Conservation Plan (2002), approximately 60 percent of the watershed area was developed prior to the adoption of runoff control ordinances that limit impervious areas or require detention of stormwater runoff.
The consequences of this unmanaged land development process have included severe stream bank erosion, flooding, and impairment of aquatic communities. More than 90 percent of stream segments in WCW are listed by the Pennsylvania Department of Environmental Protection (PA DEP) as impaired due to siltation. In 2003, a Total Maximum Daily Load (TMDL) for siltation was established by the U.S. Environmental Protection Agency (EPA), Region III. The TMDL program determines the safe level of loading for a pollutantin this case TSSand allocates that load equitably among dischargers through the stormwater permitting process.
Municipalities needed to identify sources of TSS loading and find ways to reduce the load. The Wissahickon Basin Inventory has given municipal partners updated information about their stormwater infrastructure as well as a tool that assists in planning overall TSS load reduction.
The Grant Program and Project Scope
The Wissahickon project was intended to develop an approach for generating a stormwater management facilities inventory that could be replicated. It would include development of a process for inventorying basins and prioritizing retrofit activities with structural and nonstructural stormwater best management practices. These activities would enhance water quality treatment of stormwater runoff and increase infiltration where conditions allow.
Prioritization of inventory of sites would identify where retrofits would be most beneficial in reducing stormwater runoff impacts to the creek and increasing stream baseflow. The project resulted in the production and dissemination of a prioritized inventory and a final report document, implementation cost estimates, and many geospatial data layers. This program was initiated in May 2006 and completed by September 2007.
Desktop Analysis and Fieldwork Preparation
Some existing GIS data sources were used for initial desktop stormwater basin screening. The City of Philadelphia contracted with Sanborn, Inc., in 2004 to provide high-resolution orthophotography and planimetrics including topographic contours interpolated at two-foot intervals. This data was visualized, along with other basemapping data such as hydrologic features, roads, and municipal and watershed boundaries, using ArcMap to create a point shapefile of stormwater basin locations. Many of the data layers used for the analysis, including high-resolution orthophotography, are available to the public at no charge through the Pennsylvania Geospatial Data Clearinghouse (www.pasda.psu.edu/).
Initial Basin Identification
GIS technicians manually scanned the watershed area at approximately a 1:20,000-foot scale searching for topographic contour patterns characteristic of stormwater basins (i.e., concentric rings). Many features were positively confirmed as stormwater basins upon closer inspection of the high-resolution aerial photography. In photographs, concrete headwalls and/or outfall control structures were clearly visible. Other possible basins that could not be positively identified were attributed as potential basins in the shapefile.
Other attributes added during the desktop data gathering phase were approximate basin area, municipal jurisdiction, receiving waters, stream order, soil type, and ownership category. The Montgomery County Planning Commission shared parcel data and a partial stormwater basin inventory conducted within a tributary subwatershed. The desktop analysis was completed in approximately two weeks and yielded 215 points for subsequent field investigation.
Customized Data Input Forms
The stormwater basin retrofit prioritization design demanded a long list of descriptive attributes that would be used to rank the basins according to their potential for retrofit. The stormwater basin point shapefile attribute table was modified to accommodate this data. Collecting this data on paper forms would be time consuming and error prone so a custom data entry form was created using ArcPad Application Builder. The form ensured that descriptive data was properly entered for each site. The customization options in ArcPad Application Builder allowed data entry to be restricted to predefined categories and performed basic data validation. The extensive use of listbox controls eliminated tedious data entry and potential spelling and transcription errors by supplying predefined categories with drop-down lists containing appropriate values.
Field documentation included
Field Data Collection
The project team used an Xplore Technologies i104c2 Ruggedized Tablet PC running ArcPad and equipped with a GPS module for field data collection. Using ArcPad, teams could directly edit the basins point shapefile while in the field and benefited from access to the basemap layers in a familiar GIS interface.
Tablet PCs have several advantages as a platform for mobile field computing and data collection. The large display provided ample room for data entry without requiring a multi-tabbed form and allowed for GPS navigation at high resolution without frequent screen redraws. The GPS module, which received correctional data from the Wide Area Augmentation System (WAAS), delivered two- to five-meter accuracy. The Tablet PCs had an added benefitthe team was able to use the GPS and tablet display for road navigation while traveling between sites.
Data Gathering Performance
The data entry form was tested for two days, then modified to update listboxes with additional categories. Field visits were then conducted over the two-week period between June 13 and 27. A two-person field crew inventoried roughly 25 basins per day.
During this field investigation, many potential basins originally identified through the desktop evaluation were eliminated because they were not actually stormwater management facilities, cutting the number of basins in the inventory from 215 to 178. The project team was well satisfied with the efficiency and quality of the field data collection phase. Because the GIS was edited directly in the field, no additional steps were needed before the team began processing and prioritizing the basins.
Basin Ranking Process and Results
Basins were ranked using EVAMIX, a sophisticated, spreadsheet-based multicriteria evaluation program designed to handle both quantitative and qualitative data in a mathematically rigorous fashion. Fifteen criterion were developed by PWD for use in the basin inventory prioritization scheme. Each criterion was assigned an individual weight because some criteria were more critical to basin retrofit potential than others. The criteria weights ranged from 2 to 17 and the sum of all weights was 100. The matrix tool was executed at the watershed-wide level to produce a watershed-wide prioritization and again at the individual municipal level to produce a ranking for municipalities so each municipality could use this information when creating an implementation approach.
The prioritized output was sorted and categorized as high (1–20), medium (20–60) and low (all other ranks) based on priority for retrofit potential. However, a lower ranking assigned in this prioritization process did not mean a basin should not be considered for retrofit. Rather, basins in the high and medium categories were deemed a higher priority in this first level of screening. The "biggest bang for the buck" comes by clustering retrofits within an individual tributary subshed, so it makes sense to implement retrofits on lower-priority basins as a complement to retrofits on medium- and higher-priority basins.
Data produced through this initiative was made available to the U.S. EPA, PA DEP, and municipalities in the Wissahickon watershed, along with extensive updated infrastructure and basemap data at both the municipality and watershed levels. To provide the most value and universal applicability to the municipal partners, all geospatial data produced was provided in shapefile and Keyhole Markup Language (KML) formats. KML is compatible with free Web-based viewing applications such as Google Earth, Microsoft Virtual Earth, and NASA WorldWind. Spatial data shared with Wissahickon municipalities included
Microsoft Excel spreadsheets of the Basin Inventory data were also shared along with the spatial data so new records can be added and data can be sorted and queried as necessary. Photos taken during PWD field visits were shared as high-resolution JPEGs and compiled into HTML files. Additionally, a spreadsheet ranking tool was packaged and shared with partners to give users of this data the opportunity to continue to add data to the evaluative matrix and prioritize basins for retrofit potential.
Retrofits of two municipally owned and maintained stormwater basins have been selected for funding under the Pennsylvania Growing Greener Watershed and Flood Protection Grant program. PWD has also begun development of a follow-up initiative that will take publicly owned basins identified through this inventory through a site-specific evaluation and retrofit design process. With engineering designs in place, these basins will be ready to acquire funding for construction of stormwater best management practices. PWD and the U.S. Geological Survey will continue to monitor streamflow and water quality in Wissahickon Creek and its tributaries in the hope of someday seeing results of stormwater best management practices.
For more information, contact
About the Authors
Marissa Barletta is an environmental planner with D.S. Winokur Associates. She has spent the last three years as a contract employee with the Philadelphia Water Department working on the city's Integrated Watershed Management Planning program. She has worked on developing watershed-wide implementation approaches that seek to bridge political divides to improve ecological health.
Jason Cruz has been an aquatic biologist with the Philadelphia Water Department for the last six years. His interests include algae and its effects on stream metabolism, urban stream restoration, GIS, life history strategies of aquatic invertebrates, and new technologies to share data and improve watershed-based education.
Voogd, H. (1982), Multi-Criteria Evaluation with Mixed Qualitative and Quantitative Data, Environment and Planning, vol. 9, 221–236.
Maimone, M. (1985), An Application of Multi-Criteria Evaluation in Assessing Municipal Solid Waste Treatment and Disposal Systems, Waste Management and Research, vol. 3, 217–231.