A case study of the development, application, and acceptance of a water quality model for the lower Colorado River of Texas was performed. The new, comprehensive model was developed to assess the role that the nutrients nitrogen and phosphorus play in the growth of submerged rooted aquatic vegetation (SRAV).The model development and application was carried out as part of a larger study to determine if stream standards for those nutrients were required to protect the ecological health of the river. In addition to model development the case study also examined! the acceptance that the results received from various water resource agencies and the press. The water quality of the lower Colorado River below Austin became the focus of much public concern during a drought period in 1984 and 1985. The major concern was the prolific growths of submerged rooted aquatic vegetation, a condition commonly associated with eutrophic waters of high nutrient content. The belief was widely held that nitrogen and/or phosphorus, discharged with the municipal wastewater of Austin, caused the SRAV growths. A special committee appointed by the former Governor concluded that there was not enough information regarding the linkages between those nutrients and the SRAV to recommend nutrient stream standards. However because of the importance that the issue was accorded, the committee recommended stream standards be adopted, unless the study showed they were not needed. Although Texas has been at the forefront of water quality model development, the new model for the lower Colorado River was the first developed and tested for a Texas stream to include the ecological interactions of nutrients and SRAV. Included were subsystems for carbonaceous BOD, dissolved oxygen, four nitrogen species, two phosphorus species, total dissolved solids, attached spiphytic algae, and the SRAV. The mathematical relations between those subsystems, called kinetic equations, were based on previous studies performed in laboratory experiments and for other water bodies. The newly developed model was then calibrated by comparing its results to a set of field data collected during a low flow period. The constants and coefficients of the kinetic equations were adjusted to obtain agreement between the field data and model predictions. All of the constants and coefficients were adjusted to within ranges reported in scientific literature. After calibration, the model was then tested against another set of field data, and again was found to be in good agreement. The resulting set of kinetic relations showed that light penetration of the water, not the concentrations of nutrients, was the limiting factor for SRAV growth in the lower Colorado River. The primary variable to calibrate model predictions for SRAV areal density was that which represents the plants own attenuation of light due to the spreading of their leaf canopy. The density of plants was found to be very sensitive to this self shading coefficient. Although there was a dearth of literature regarding this self-shading coefficient, the final value was in the range reported. With the calibrated and tested model it was possible to perform a series of analyses to assess the utility of removing nutrients from The City of Austin's wastewater discharges as a means of controlling SRAV growths. Three scenarios were examined, the first with discharge flows set for the projected 1990 populations, the ammonia discharge was set at 2 mg/L, and no additional nutrient removal. ! This was considered a base case because The City of Austin is currently implementing treatment to essentially met the discharge conditions of this scenario. Scenario 2 was like Scenario 1 plus total phosphorus removal to 1 mg/L. Scenario 3 had phosphorus removal plus the further hypothesized addition of denitrification to remove nitrate nitrogen to 0 mg/L. Scenario 1 resulted in increases in river concentrations of phosphorus and nitrogen of +77% and +71%, respectively, as compared to the July 1986 SRAV calibration. these were accompanied by an estimated increase in SRAV density of+7.0%. The second and third scenarios resulted in reductions of phosphorus and nitrogen concentrations in the river of -73% and -49%, respectively. However, due to the importance of light in limiting the plant densities and the continued presence of upstream nonpoint sources, the nutrient reductions were not accompanied by similar SRAV reductions. The cited nutrient reductions were predicted t! o result in -3.6% and -13.9% reductions of SRAV densities, respectively. To accompany this assessment of the predicted benefits of nutrient removal, an analysis of the associated cost was also done. For each of the point source nutrient removal scenarios, the cost was also done. For each of the point source nutrient removal scenarios, the cost of implementing the necessary treatment was estimated. Total net present worth calculations were performed to represent both the initial capital expenditure and the operation and maintenance costs for a 20year period. The nutrient removal operations of Scenarios 2 and 3 were estimated to cost $63 million and $126 million more than the base case of Scenario 1. Although advanced treatment to remove nitrogen and/or phosphorus from Austin's wastewater result in some small reductions of SRAV, those reductions come at a high cost. The results of this research have proven to be controversial as they are counter to the widely held cause-effect relation between nutrients and SRAV growths. An assessment of the acceptance and criticisms of the modeling work by various water resource institutions and the press was performed. The main criticisms of the model were that the scope of the kinetic equations was too limited and the assumption of steady-state conditions was not valid for an inherently dynamic river. Water quality model development is an incremental process which relies on extensions of earlier documented, tested, and tractable kinetic relations. Added kinetic relations must be weighed against there likely refinement to the model. Because adding an abundance of new relations requires much more field data than was gathered, the additions may not increase the predictive ability of the model, and the new relations would be harder to check against established research, the level of complexity employed was felt to be adequate. The field data necessary to calibrate a fully dynamic model was shown to be at least an order of magnitude more than done in this research. Although a dynamic model may be theoretically desirable, the steady-state assumption is almost universally employed for wastewater simulations by both Federal and State agencies. Institutional and political issues were also shown to be possible contributors to the criticisms. There are different jurisdictional interests of the various institutions involved. Criticism to this study were made by an agency which would be a potential beneficiary of the nutrient removal operations. The agency position was also strongly stated prior to the study, and it may be difficult to publicly retreat from that statement regardless of the research results. An analysis of the portrayal of the model results in the press, showed that the foundation for important results of the research were not conveyed through the articles. The result that light limitation was controlling SRAV density rather than nutrient discharges was not adequately explained, with the result that the prior nutrient-SRAV, cause-effect linkage was again offered. This press coverage showed that the technical details which form the basis for the conclusions of this, and probably most other models, cannot be well understood outside the realm of the water quality scientists. Overall, the criticisms of the model and the acceptance that the results found in the press shed light on some important issues regarding water quality modeling. In order to properly develop and fully communicate the results of a model it is important to bear in mind the environment in which they will be viewed. Three things were found to be of overriding importance in this study: 1.) the previously held public opinion as to the root of the water quality problem, 2.) the potential roles of individuals or institutions as beneficiaries or payers, and 3.) previously stated positions before the results of the research. Due to the importance of water quality models in influencing the fiscal expenditures of municipalities, and this apparent dichotomy of understanding, an exploration of public official's views of water quality assessment was performed. This was accomplished through a mail survey to 98 mayors and city council members throughout the Colorado River basin of Texas. The survey results showed that there was only a moderate understanding of, and similar credibility placed in the methods of water quality assessment, among those city officials. The analysis of the survey results also found that neither population of the municipality nor the expenditure of money for wastewater treatment improvements increased the level of understanding. Another result was that increased information about the technical methods of water quality assessment should be made available to public officials, which would likely increase both their level of understanding, and the credibility they have of the methods. These efforts would lead to a more informed and possibly more cooperative atmosphere for water resources planning at the state level.