Evaluation of a water quality model (CE-QUAL-R1) using data from a small Wisconsin reservoir

Abstract

Abstract The U.S. Army Corps of Engineers' 1-dimensional model (CE-QUAL-R1) was evaluated using data collected at a small (46 ha, mean depth of 3.7 m) eutrophic reservoir near Spring Valley, WI, U.S.A. The model predicts the gradients, in the vertical direction, of water quality variables including temperature, chemical species, and biological assemblages. Physical processes incorporated in the model include inflow, outflow, diffusion, convection, settling, adsorption and light extinction. Biological processes modeled are ingestion, egestion, decomposition, photosynthesis, respiration and mortality. Weekly or biweekly data, in the form of vertical profiles, were collected in 1981 and 1982 and were used as a basis for calibration and confirmation simulations. Over 3200 comparisons of measured versus predicted values, for 20 different variables, were made for each year. Comparisons of measured versus predicted flux values were also made. In order to better represent the reservoir and its dynamics, changes to the model were made representing inflow, outflow, and the compartments of algae, silica, macrophytes, zooplankton, and nitrite plus nitrate nitrogen. The average Reliability Index for each variable was between 1.06 and 4.86, indicating model precision was always better than a half order of magnitude. The average Reliability Index for all values was 2.57 for the final calibration simulation and 2.62 for the confirmation simulation. In addition, comparisons of measured and predicted flux values were satisfactory, helping to ensure that reasonable predictions were made for the correct reasons.