Development and Validation of the Water Quality Model MINLAKE96 with Winter Data

Abstract

A deterministic, one-dimensional lake water temperature and dissolved oxygen model (MINLAKE96) has been developed. The model operates on a time step of one day, makes year-round simulations, includes winter ice covers, and uses weather data, lake morphometry and Secchi depth as input. MINLAKE96 is modified from the water quality model MINLAKE95, which has been applied to simulate water temperature and dissolved oxygen characteristics of lakes in the state of Minnesota under past and projected future climate scenarios. Modifications made include the decay of albedo of the winter snow cover after snowfall, and ice growth upwards from the ice surface. Modifications are based on recent field studies conducted in Ryan Lake, Minnesota, and observations in other lakes. A sensitivity analysis for model input parameters is performed. Simulated ice thicknesses are weakly sensitive to the attenuation coefficient of ice, not sensitive to the attenuation coefficient of snow, and weakly sensitive to the density of snow. Parameter and coefficient values used in MINLAKE96 are therefore the same as in MINLAKE95. MINLAKE96 has been validated against the same data sets used for validation of MINLAKE95 and additional data obtained in Ryan Lake during the winter of 1994/1995. The data sets include freeze-over dates, ice/snow thicknesses, water temperatures and dissolved oxygen concentrations. Average standard errors between simulated and measured water temperatw'e and dissolved oxygen concentrations are 1.4°C and 1.9 mg/I, respectively. The results are not significantly different from those obtained with the original version - MINLAKE95. MINLAKE96 will be used in the future for regiona11ake studies because it includes surface ice growth and better validated physical parameters for ice and snow covers.