5A.2 Utilizing Multimedia Modeling and Machine Learning to Assess Dissolved Oxygen as a Proxy for Hypoxia in Lake Erie

Tuesday, 14 January 2020: 1:45 PM
156A (Boston Convention and Exhibition Center)
Christina Feng Chang, Univ. of Connecticut, Storrs, CT; and M. Astitha, V. Garcia, C. Tang, P. Vlahos, D. Wanik, and J. Yan

The prediction of hypoxia in lakes is important because healthy levels of dissolved oxygen (DO) is vital for many aquatic organisms. In fact, many aquatic organisms require DO concentrations of at least 5 mg/L for survival. Concentrations of DO less than 5 mg/L are known to lead to fish kills and changes in fish behavior. Urban areas, industries, and agricultural activities have undoubtedly contributed to an increased loading of nutrient pollution into Lake Erie, which has led to frequently occurring harmful algal blooms (HABs). The overgrowth of algae likely triggers hypoxia in the lake as a result of the biological oxygen demand required for breakdown process by microbes. In this study, we use a suite of physical modeling systems with in-situ measurements of DO in Lake Erie to serve as a proxy to identify hypoxia. Observations are provided by the Lake Erie Committee Forage Task Group (LEC FTG) and the Great Lakes National Program Office (GLNPO) for the period 2002-2012. Physical modeling systems involved are the: 1) Weather Research and Forecasting Model (WRF); 2) Variable Infiltration Capacity Model (VIC); 3) Community Multiscale Air Quality Model (CMAQ); and 4) Environmental Policy Integrated Climate Model (EPIC). Meteorological weather variables from the WRF model, hydrological variables from the VIC model, nitrogen deposition from the CMAQ model, and agricultural management practice variables from the EPIC model are used to fit random forest models that predict concentrations of DO at multiple depths of the lake. Multiple predictive models have been developed for DO that are capable of explaining more than 72% of the variance in lake depths that experience DO concentrations less than 5 mg/L. The importance of explanatory variables is evaluated, and the contribution of each covariate in the model is analyzed with Accumulated Local Effect (ALE) plots to better understand the occurrence of hypoxia.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner