Carl W. Chen, Engineer

Joel Herr, Engineer

Laura Ziemelis, Engineer

Systech Engineering, Inc., 3180 Crow Canyon Pl., Suite 260, San Ramon, CA 94583

Robert A. Goldstein, Project Manager

Electric Power Research Institute, P.O. Box 10412, Palo Alto, CA 94303

Larry Olmsted, Director of Scientific Services

Duke Energy Company, 13339 Hagers Ferry Rd., Huntersville, NC 28078

The Watershed Analysis Risk Management Framework (WARMF) has been developed to provide a road map that can help stakeholders organize themselves, develop a work plan, identify water quality issues, propose management alternatives, and evaluate their effects on water quality. Stakeholders can designate various parts of a river basin for beneficial uses such as drinking water supply, swimming, cold or warm water fisheries, and aesthetic value. Parameters such as water depth, flow velocity, fecal coliform, temperature, dissolved oxygen, suspended sediment, and chlorophyll level are used to set criteria for each beneficial use. WARMF contains dynamic catchment, river, and reservoir models. These models accept meteorological data, digital elevation maps, and land use information as inputs and simulate hydrology, nonpoint source loads, and water quality throughout the basin. Monitoring data is used to calibrate the models. The simulated results are processed to determine whether the water quality meets the specified criteria. The locations that meet the criteria are painted green and the locations that do not are painted red, in the GIS maps. WARMF, which makes intensive use of monitoring data, translates scientific measures of water quality into usabilities. The methodology has been applied to the Catawba River of North Carolina and South Carolina.

EPA’s concept of a watershed approach requires the involvement of local stakeholders in the development of a management plan or a TMDL. The stakeholders must identify water quality issues in the river basin, develop alternatives to solve the water quality problems, and reach consensus on a management plan. They need data to make informed decisions. The data must be brought to stakeholders in a logical and sequential manner and presented using management variables.

There are two types of data that can be used by stakeholders. One is monitoring data measured by scientists, and the other is the simulation results from a model. Both data are in scientific terms, such as the temperature of water in degree Celsius, the concentration of coliform bacteria in MPN per 100 ml, and the concentration of dissolved oxygen in mg/l. These data are not very meaningful to stakeholders who want to know whether the water is safe for swimming, acceptable for drinking water supply, not harmful to fish, and/or aesthetically pleasing.

WARMF is a multi-tasking decision support system that can calculate TMDLs, guide stakeholders through the consensus process, predict water quality improvement of a management scenario, and manage data for a watershed. The focus of this paper will be to show how WARMF can translate monitoring data and simulation data into usability terms.

The stakeholders can use WARMF to display where the pollutants come from, how these pollutants affect the water quality, and whether a management plan can render all sections of the river basin suitable for intended uses. All examples will be drawn from the application to the Catawba River Basin of North and South Carolina. Figure 1 shows the GIS map of the 5,000 square mile river basin. Both spatial and temporal data will be displayed with the basin map on the computer monitor in colors. Unfortunately, the colors cannot be seen in the black and white copies.

Field measurement is expensive. It is impractical to have sufficient monitoring stations and samples to characterize the entire watershed. Model simulation, on the other hand, can be performed for whatever spatial resolution or time scale desired. In WARMF, available monitoring data is used to calibrate the model. After a reasonable agreement is reached, the model is used to generate detailed simulation data for further analysis.

As a step of the consensus process, stakeholders need to identify the intended uses of their watershed. Stakeholders enter intended uses in a dialog box (see Figure 2). The intended uses in the Catawba basin include swimmable waters, warm water fish habitat, cold water fish habitat, water supply, and aesthetic enjoyment. Other uses such as ecosystem protection can be added to the list.

The stakeholders can select an intended use in the dialog box and then point and click at applicable locations on the GIS map. The applicable locations do not have to be contiguous. For example, various stream sections in the headwaters of the Catawba River can be chosen for coldwater fish habitat (see Figure 2).

In the next step, the stakeholders can specify the criteria for each intended use. To do this, the stakeholders will first select an intended use (e.g. swimmable) in the dialog box. They will then select a parameter (e.g. coliform), a concentration (e.g. 1 per 100 ml), a calculation method (e.g. 7 day minimum average), an exceedance rule (higher or lower), and a percent compliance (100, 90, or 80%).

For some uses, multiple criteria may be specified. For example, coldwater fish may require a 7 day minimum average dissolved oxygen of more than 6 mg/l and a 7 day average temperature of less than 22 degree Celsius. Figure 3 presents an example of the second criteria for coldwater fish habitat.

WARMF will process the simulated data according to the specified criteria. For an intended use, the sections that meet the criteria are painted green and the sections that do not meet the criteria are painted red. For an intended use with multiple criteria, a section has to meet all the criteria for it to be painted green. In a GIS map, the stakeholders can easily spot the sections with water quality problems (i.e. the red sections in Figure 4).

Figures 5 to 8 compare some of the observed and simulated parameters at various points in the Catawba River Basin. The figure heading shows the goodness of fit between observed data and simulation results using percent error and a correlation coefficient.

Some of the fits are exceptionally good, while the others are not. It is very difficult to match the timing of peaks and valleys to the observed data. A time delay of a day or two can make a big difference on the statistics, even though the model follows the pattern of observed data. We conclude that the simulated data is close to observed data and that the simulated data can be used to evaluate whether the water is suitable for intended uses.

Figure 9 shows the spatial distribution of phosphate loading for two management scenarios. The magenta is for point source and green is for nonpoint source loading. The key in the upper left indicates which simulations are represented by the loading bar Alternative 1 is the base case simulation which represents existing conditions. Alternative 2 represents a management alternative with buffer strip installation. With these displays, stakeholders can see where the pollution loads are and what impact the management scenario has on the pollution load.

Figure 10 shows the usability (aesthetic value) of water sections under base case conditions. Figure 11 shows the usability of water sections under the management scenario of buffer strip implementation. By comparing the two figures, stakeholders can see whether the management scenario has improved the water quality by changing the color of water sections from red to green on the GIS maps.

It is concluded that WARMF is an effective communication tool to present data to the stakeholders. The features of WARMF include: