Date of Award

December 2016

Degree Type


Degree Name

Doctor of Philosophy


Freshwater Sciences

First Advisor

J. Val Klump

Committee Members

James T. Waples, Sandra L. McLellan, Clare E. Reimers, Hector Bravo, Joseph DePinto


Benthic Respiration, Environmental Modeling, Green Bay, Hypoxia, Natural Tracers, Primary Production


Green Bay, Lake Michigan is a large freshwater estuary that has experienced seasonal hypoxia for decades. Hypoxia, or dissolved oxygen concentrations less than 2 mg L-1, is a problem in coastal ecosystems around the world because it has a negative impact on ecosystem health by decreasing biodiversity and fisheries. In order to create adequate management policies for hypoxia, it is important to understand the sources and sinks of oxygen within Green Bay. This study utilizes a number of traditional and novel field methods to measure the production and respiration of oxygen within lower Green Bay, defined as south of Chambers Island, which is the area that experiences hypoxia. Primary production was measured using light-dark bottles and via in-situ diel oxygen fluctuation calculations. The epilimnetic waters are slightly net autotrophic during the summer months, meaning that they accumulate organic matter that can drive oxygen respiration in the hypolimnion. Hypolimnetic oxygen consumption was calculated as the loss of hypolimnetic oxygen inventories between two time periods. It was also determined that the two major processes consuming oxygen within the hypolimnion are sediment oxygen demand (SOD) and water column respiration (RH). SOD was measured using core incubations and eddy covariance. RH¬ was estimated as the difference between hypolimnetic oxygen consumption and SOD. In shallow waters, close to the Fox River mouth, SOD dominates the oxygen consumption, while in mid-bay waters oxygen respiration is divided between SOD and RH. Based on natural tracer results, cool bottom waters flows southward from the Lake Michigan-Green Bay gap and begins to lose oxygen once it reaches the mid-bay. This somewhat oxygen depleted water is further pushed down the bay into shallow waters where benthic respiration consumes more oxygen and drives water hypoxic.

222Rn and CH4 were also used as natural tracers to estimate advective flow in mid-bay bottom water and apparent methane production, respectively. Advective flow was estimated at ~3 km d-1, which agrees with current profiler velocities of 1.8 km d-1. This velocity can vary though, depending on upstream 222Rn activity. Apparent methane production trends match apparent oxygen utilization trends, confirming that methane is produced when oxygen is depleted.

Finally, a biogeochemical model for Green Bay was created to predict what level of nutrient reductions would sufficiently reduce hypoxia, both now and under future scenarios. This biogeochemical model was part of an integrated modeling effort. The model was successfully formed and can be used to evaluate responses, although the baseline line model needs to be much better calibrated to better replicate observations.