Let's step back and consider the position of a water quality manager, who, confronted by various kinds of information, has to make a decision about whether modeling would be a useful tool for assessment. Water quality issues may have been brought to the water quality manager's attention in various ways, including:
Informal Assessment: For example, issues might be identified after receiving a public complaint or observing a nuisance condition. Issues might also be identified as a result of a decline in resource condition or quality perceived by general public or by experts.
Formal Assessment: Section 106(e)(1) of the Clean Water Act requires each state to establish appropriate monitoring methods and procedures necessary to compile and analyze data on the quality of waters of the United States. Under Section 305(b), each State, Territory, and Interstate Commission provides a biennial report of the current status of water quality to EPA. These 305(b) lists contain a formal assessment of water quality issues identified in monitoring. For instance, monitoring might reveal a chronic occurrence of a standard violation, a declining trend in water quality, or a documented impairment of aquatic or wildlife habitat.
Proactive Need: Issues also arise from a desire to protect existing resources and prevent future degradation. For example, a local community may want to ensure protection of its drinking water supply for the next 50-100 years, or a wildlife agency may wish to protect critical habitat.
Since issues are identified in such various ways, our background knowledge of the issues may vary considerably. In carrying out an assessment of a given problem, the manager needs to ask a number of questions to identify what is already known about magnitude, sources and causes of problems or threats to water quality:
Having asked and answered these questions to the extent possible with existing information, the manager must decide if further assessment is required, and if so, what type of assessment. Often the manager will be confronted by choice: monitor or model? Modeling is useful for many purposes, but it may not always be the best tool for a given situation. If one has the choice, real monitoring data are always preferable to model predictions (and, of course, the quality of any modeling effort depends a great deal on the quality and quantity of data available to it).
The first step in choosing a model for watershed assessment is to step back and decide if a model is needed at all, or whether another assessment tool might be better for the given situation. The next several screens present questions which will often clarify whether a model would be useful.