Evaluating reservoir model responsiveness to increased nutrient inputs from urbanization

Background/Question/Methods

This study determines the responsiveness of a reservoir model, Environmental Fluid Dynamics Code (EFDC), to changes in tributary loads to Converse Reservoir, a drinking water supply reservoir. The study determines the ability of the EFDC model to simulate how changes in upstream nutrient loads influence water quality at a drinking water in-take, which is located on a tributary roughly three miles from the mainstem of the reservoir. The reservoir model is then used to evaluate the influence of watershed urbanization on reservoir water quality.

Two model simulations are performed. The first simulation utilizes actual tributary nutrient levels to determine total nitrogen, total phosphorus and total organic carbon values at the in-take which is located along a branch of the reservoir and may not be influenced by changes to the mainstem. To test this, a second simulation utilizes increased nutrient loads to the uppermost mainstem tributary. Changes in phosphorus, nitrogen and organic carbon at the in-take are compared with baseline conditions to evaluate EFDC responsiveness. Water surface elevation and nutrient results are evaluated using Nash-Sutcliffe efficiency (NSE), percent bias (PBIAS), and ratio of the root mean square error to the standard deviation of measured data (RSR).

Results/Conclusions

The model simulated reservoir water surface elevation very well. Predicted water surface elevation resulted in an NSE (0.98), PBIAS (0.02%) and RSR (0.12) near optimal values.

To determine model performance in simulating water quality, results from the baseline scenario at the in-take are compared with measured data at the same location.  Monthly phosphorus (TP) and nitrogen (TN) samples were collected in 1991. Seventy-five percent of the TP measurements were reported as the detection limit (0.01) while model results were all less than 0.01, thereby limiting comparisons between measured and simulated results. PBIAS results suggest good model performance in simulating TN. NSE and RSR suggest unsatisfactory model performance for TN.  

TP, TN and total organic carbon concentrations at the in-take were greater during the urbanization scenario than the baseline scenario. Urban nutrient loads in the upstream most tributary increased nutrient concentrations at the drinking water in-take. Total phosphorus and total organic carbon concentrations during the urban scenario were approximately twice baseline concentrations indicating that future watershed urbanization could adversely impact source water.