1. Assemble a data set of sediment chemistry and sediment pore water chemistry suitable for use in model calibration.
2. Develop a relationship between available spatial chemistry data and sediment oxygen demand that will permit use of spatially-variant SOD in more complex models.
3. Apply and calibrate existing biogeochemical models of sediment oxygen demand in Lake Erie sediments. Link these models to the ECOFORE water quality framework.
4. Apply and calibrate the more complex Biogeochemical Reaction Network Simulator (BRNS) model to provide predictions of the sediment oxygen demand and internal chemical loading to the hypolimnion of Lake Erie. Link this model to the ECOFORE water quality framework.

Clearly, the Lake Erie ecosystem has undergone significant and complex changes since P load control efforts achieved TP and chlorophyll a targets in the mid-1980s, including the introduction of dreissenid mussels, reduced lake water levels, enhanced water temperatures, and potentially enhanced non-point source inputs of P. Each of these changes, individually and in combination, may affect the extent and magnitude of hypoxia. These findings indicate that there is a need to better understand carbon and energy flow in Lake Erie and how the current trophic system affects oxygen concentrations in lake water. This proposed, relatively small project employs a modeling approach to provide improved estimates of sediment-oxygen demand and links the model to physical and ecosystem models being developed in the current NOAA ECOFORE (Ecological Forecasting) project.

The proposed work plan consists of 4 tasks:

1. All models require a calibration data set and this component of the work plan is crucial. In this task, existing data on pore water chemistry, sediment solids chemistry, and sedimentation rates will be compiled from each of the basins in Lake Erie to generate a calibration data set for each of the lake basins, and hopefully, for a higher spatial resolution.
2. This task will be to develop a relationship between available chemistry data and sediment oxygen demand that will enable higher spatial resolution SOD modeling and hence better hypoxia predictions. This will be accomplished by conducting a suite of simulations and a sensitivity analysis using a range of values in the key variables in the diagenetic and SOD models to develop approximations between sedimentation rate, carbon flux and SOD.
3. Apply simple SOD models to the Level 1 ECOFORE water quality model. The proposed work will include determining the spatial and temporal output and variables from the SOD model that can be accommodated by the water quality model and modifying the SOD model codes accordingly.
4. Apply the more complex BRNS model to calculate SOD using a technique that sums all oxygen consumption reactions in the sediment, including secondary reactions such as reduced iron phase oxidation. Once the code is adjusted to output SOD, it will be converted to FORTRAN and linked to the ECOFORE water quality model.