Modeling hypolimnetic dissolved oxygen depletion using monitoring data

Area of Science:

• Environmental Science
• Limnology
• Ecology

Background:

• Eutrophication leads to hypoxia in lakes and reservoirs, negatively impacting aquatic life.
• Existing quantitative models for hypoxia are often limited by data requirements, particularly temporally resolved dissolved oxygen data.
• Effective management strategies for hypoxia require accurate predictive models applicable to diverse lake systems.

Purpose of the Study:

• To develop and validate a hierarchical Bayesian model for predicting dissolved oxygen (DO) in lakes.
• To identify key environmental factors influencing DO depletion during summer stratification.
• To create a model applicable to lakes with both temporally resolved and single dissolved oxygen profiles.

Main Methods:

• Developed a hierarchical Bayesian model incorporating mechanistic understanding of hypoxia drivers.
• Included factors such as days since stratification, dissolved organic carbon, lake depth, and chlorophyll concentration.
• Validated the model using two distinct datasets: one with high temporal resolution and another with single profiles from numerous lakes.

Main Results:

• The model successfully predicted dissolved oxygen in lakes using limited and extensive datasets.
• Similar relationships between oxygen demand and chlorophyll were found across different datasets, validating the model's structure.
• Combining both datasets significantly enhanced the precision of the model's predictions.

Conclusions:

• The developed Bayesian model provides a robust tool for predicting lake hypoxia.
• The model's flexibility allows application to lakes with varying data availability, supporting broader management efforts.
• This approach improves our understanding of eutrophication's impact on dissolved oxygen dynamics in freshwater ecosystems.