WATERSHED.ORG
Formerly the Watershed Management Council

Guidelines for Monitoring

Lee MacDonald
Watershed Science Program, Colorado State University

Figures for this article are only available on the paper copy.

The escalating debate and growing scientific literature on water quality monitoring lends credence to the generalization that monitoring is a complex undertaking. We all want an inexpensive, universally-applicable technique that yields an instantaneous evaluation of water quality. Furthermore, we want the resulting measurement to be highly sensitive to the management activities of concern, yet reliable and unvarying over time. Although such a technique is likely to remain the Holy Grail of land managers and regulators for some time to come, our likelihood of finding such a technique has a similar probability.

The absence of ideal monitoring procedures often leads us to fall back on those measurements with which we are most familiar. Often the relevance of these measurements to the issues at hand, and the likelihood of obtaining the desired answer, are not rigorously evaluated prior to initiating a water quality monitoring project. In the words of one land manager "90% of the water quality monitoring in my area is hobby-horse s---".

In the last decade EPA has recognized that nonpoint sources are the major cause of impaired water quality, and that better data are needed to justify continuing expenditures. In 1989 Region 10 of EPA commissioned the Center for Streamside Studies at the University of Washington to develop a set of guidelines for monitoring the effects of forestry activities on streams. The resulting 162-page document (Monitoring Guidelines to Evaluate Effects of Forestry Activities on Streams in the Pacific Northwest and Alaska) and the accompanying expert system (PASSSFA) are now available from EPA.

The purpose of the Guidelines is to assist people in the design of monitoring projects and the selection of monitoring parameters. As implied by the title, the Guidelines focus on the effects of forestry activities on streams in the Pacific Northwest and Alaska (i.e., Washington, Oregon, Idaho, and Alaska). However, the thought process and much of the information is more widely applicable. For example, the Guidelines consider other management activities that often occur in forested areas (e.g., grazing, mining, and recreation), as these directly affect water quality and their effects may be difficult to distinguish from forestry-related activities. Similarly, the Guidelines focus on streams, but recognize that the design of monitoring projects and selection of monitoring parameters must consider downstream designated uses. Finally, water quality is defined in a very broad sense, and the Guidelines consider not only the traditional physical and chemical parameters such as pH, dissolved oxygen, and suspended sediment, but also include discharge parameters (e.g., size of peak flows), aquatic organisms, riparian conditions, and channel characteristics (e.g., width/depth ratios, habitat types, bed material, bank stability, and large woody debris).

Part I provides guidance in the design of monitoring projects and the selection of the most appropriate monitoring parameters. The legal basis for water quality monitoring is reviewed, and seven different types of monitoring are defined depending upon the objectives. A structure for developing water quality monitoring projects is developed, and this is used to demonstrate the importance of pilot projects and feedback loops. One chapter is devoted to statistical design and evaluation because water quality monitoring is inherently a process of sampling selected variables in space and over time. Careful statistical design is essential if quantitative statements are to be made.

The selection of monitoring parameters is presented as a function of the designated uses of the water body being monitored, the type of management activities, and the cost of monitoring. Monitoring costs are broken into the frequency of sampling, the range of flow conditions needing sampling, data collection time, equipment costs, and analytic costs. All 30 parameters evaluated in the Guidelines are qualitatively ranked with regard to each controlling factor in a series of tables. Other factors-particularly access and the physical environment-also are discussed but could not be qualitatively ranked.

These qualitative rankings then are combined to provide an overall evaluation of the usefulness of each parameter for the 10 different management activities considered in the Guidelines. A final table evaluates the interrelationships between each possible pair of parameters, and this helps to identify interacting or overlapping parameters. Figure 1 illustrates this process of parameter selection.

Part II of the Guidelines is a technical review of each of the parameters evaluated in Part I. This is designed to provide an overview of each parameter and to serve as a reference section. For each parameter there are seven subsections: (1) definition; (2) relation to designated uses (i.e., how changes in the parameter affect the designated uses of water), (3) effect of management activities on the parameter; (4) measurement concepts, (5) standards, (6) current uses, and (7) assessment. The assessment section is an overall summary of the parameter as it relates to water quality monitoring-particularly for forestry activities-in streams in the Pacific Northwest and Alaska.

The expert system (PASSSFA, or PArameter Selection System for Streams in Forested Areas) is a computerized version of the parameter selection system developed in Part I (Fig. 1. This allows users to quickly identify appropriate monitoring parameters through an interactive series of questions and answers. A confidence level assigned to each recommended parameter provides a relative indication of the likely usefulnes given a particular set of management activities, designated uses, and monitoring constraints. A "what if" function allows the user to quickly alter the response to a particular question and then generate a revised list of recommended monitoring parameters. PASSSFA runs on IBM-compatible computers, and a VGA color monitor is preferred but not necessary.

Copies of the Guidelines and PASSSFA can be obtained by writing to;

U.S. EPA Region 10, NPS Section WD-139
1200 Sixth Ave., Seattle, WA 98101.

To obtain the expert system it is necessary to include a diskette formatted in MS-DOS with at least 225K of available space.

Lee can be reached at (303) 491-6307