Modeling of Dissolved Oxygen Stratification Dynamics in Minnesota Lakes under Different Climate Scenarios

Abstract

A process-oriented, deterministic, one-dimensional, numerical dissolved oxygen simulation model is developed to make projections for lakes with a wide range of morphometries, trophic levels and climate scenarios. The model is needed to estimate the potential impact of projected global climate change on lake water quality and fish populations. The model, combined with a water temperature stratification model, simulates vertical dissolved oxygen profiles in stratified lakes in daily time steps throughout the open water season including met alimneti c oxygen maxima in oligotrophic lakes. The dissolved oxygen transport equation includes photosynthesis as a source term; biochemical oxygen demand and plant respiration are sink terms. Oxygen exchange through the air-water interface and sedimentary oxygen demand through the water-sediment interface are physical boundary conditions for the dissolved oxygen transport equation, but treated as source/sink terms in the numerical model. The model relates biological variables to specified trophic levels. Best values of biological parameters and rate coefficients are determined by sequential literature search, model calibration, validation and sensitivity analysis for an array of lakes. Average standard errors are 1.4 and 1.9 mg l for calibration and validation of model predictions compared to measurements, respectively. Model predictions of epilimnetic dissolved oxygen concentrations are found strongly sensitive to net photosynthetic production rate and surface gas transfer coefficient. Interaction between surface gas transfer, epilimnetic diffusion and photosynthetic productivity in the model is examined by an unsteady-state analysis. The model is applied to 27 lake classes based on lake surface area, maximum depth, and trophic levels under historical climate conditions in Minnesota. Daily dissolved oxygen profiles are simulated for each open water season from 1955 to 1979. A future climate scenario, as predicted by the GISS model for a doubling of atmospheric C02, is also applied. Simulated dissolved oxygen characteristics of lakes under the historical and the projected future climate scenario are interpreted and compared. Decreases of dissolved oxygen concentrations are predicted for most lakes under the projected future climate scenario.