Operational Environmental Prediction Nearshore Water Quality in the

Operational Environmental Prediction: Nearshore Water Quality in the Great Lakes David J. Schwab NOAA Great Lakes Environmental Research Laboratory Ann Arbor, MI

Factors Contributing to Nearshore Water Quality in the Great Lakes Climate – Meteorology – Hydrodynamics – Biology/Chemistry

Beach Closings or HABs Meteorology Change in Land-use Forecasting Beach Closings Hydrology/Water Flow Bacterial Fate Circulation and Bacterial Fate

Outline 1. Lake Michigan tributary modeling using nested -grid hydrodynamic models - application to beach water quality forecasting 2. Lake Erie coupled physical/biological model application to HAB and hypoxia forecasting

Beach Closures • Major health risk of microbial contamination by bacteria, viruses and protozoa in recreational waters • E. Coli requires a 24 hour incubation period – People may unintentionally swim in contaminated water

Lake Michigan Beach Quality Forecasting Lakewide grid (POM model) Coupled models nested grids + Burns Ditch nested model grid

Princeton Ocean Model (Blumberg and Mellor, 1987) - Fully three-dimensional nonlinear Navier-Stokes equations - Flux form of equations - Boussinesq and hydrostatic approximations - Free upper surface with barotropic (external) mode - Baroclinic (internal) mode - Turbulence model for vertical mixing - Terrain following vertical coordinate (-coordinate) - Generalized orthogonal horizontal coordinates - Smagorinsky horizontal diffusion - Leapfrog (centered in space and time) time step - Implicit scheme for vertical mixing Nested grid considerations: - Arakawa-C staggered grid - Fortran code optimized for vectorization - 3 d boundary condition for u, v, and T Application to the Great Lakes - No open boundary - No tides - Uniform salinity - Seasonal thermal structure - Uniform rectangular grid - XDR used for input and output interpolated from coarse grid at each boundary point - Vertically integrated velocity is specified for external mode - Internal mode velocity and temperature are specified from 3 -d boundary condition for inflow, use radiation condition for outflow - Water level is adjusted to maintain zero mean in nested grid subdomain

Nested grid hydrodynamic models in Lake Michigan

Burns Ditch 100 m computational grid 24 km 6 km

Web site: www. glerl. noaa. gov/res/glcfs/bd

Great Lakes Coastal Forecasting System - Operational Nowcast 20 day sample using vertically averaged currents

Lake Erie Coupled Physical/Biological model

The Problem: - Excessive nutrient loading in the 1960’s led to massive algal blooms, oxygen depletion, and diminished water quality in Lake Erie. - 1972 Water Quality Agreement between the US and Canada limited P loads from municipal, industrial, and agricultural sources. - With controls, P levels decreased to acceptable levels and water quality improved. - In recent years, P levels in Lake Erie appear to be increasing, despite controls.

The Problem: - Excessive nutrient loading in the 1960’s led to massive algal blooms, oxygen depletion, and diminished water quality in Lake Erie. - 1972 Water Quality Agreement between the US and Canada limited P loads from municipal, industrial, and agricultural sources. - With controls, P levels decreased to acceptable levels and water quality improved. - In recent years, P levels in Lake Erie appear to be increasing, despite controls. Our Approach: - Incorporate phosphorus transport and fate dynamics into high resolution (hourly time scale, 2 km horizontal resolution) hydrodynamic model of Lake Erie as a first step toward spatially explicit model of entire lower food web

Lake Erie Physical Characteristics: Surface Area: 25800 km 2 Volume: 480 km 3 Mean Depth: 18. 6 m Throughflow ~ 6000 m 3 s-1 Retention time: 2. 5 yrs

Ecosystem Forecasting of Lake Erie Hypoxia • What are the Causes, Consequences, and Potential Remedies of Lake Erie Hypoxia? • Linked set of models to forecast: – changes in nutrient loads to Lake Erie – responses of central basin hypoxia to multiple stressors • P loads, hydrometeorology, dreissenids – potential ecological responses to changes in hypoxia • Approach – Models with range of complexity – Consider both anthropogenic and natural stressors – Use available data – IFYLE, LETS, etc. – Will assess uncertainties in both drivers and models – Apply models within an Integrated Assessment framework to inform decision making for policy and management

Hypoxia Forecasting Modeling Approach • Model ranging in complexity – Correlation-based models – 1 D hydrodynamics with simple mechanistic WQ model • Vertical profiles extracted from full hydrodynamic model • TP, Carbon, Solids – 3 D hydrodynamics with simple mechanistic WQ model • Physics from full hydrodynamic model – 3 D hydrodynamics with complex mechanistic WQ model • WQ framework similar to Chesapeake Bay ICM model • Multi-class phyto- and zooplankton, organic and inorganic nutrients, sediment digenesis, etc • Addition of zebra mussels and other improvements

Chapra, S. C. 1980. J. Great Lakes Res. 6(2): 101 -112.

Effect of Phosphorus Controls on Lake Erie Central Basin Springtime P Concentration (Ryan et al. , 1999)

Lake Erie 1994 physical/biological model Hydrodynamics - Great Lakes version of POM - 20 vertical levels, 2 km horizontal grid (~6500 cells) - Hourly meteorology (1994, JD 1 -365) - Realistic tributary flows - Accounts for ice cover Mass balance for P - POM hydrodynamics (2 d for now) - Realistic P loading - Constant settling velocity (for now)

Computer animation of model results: -Starts in January, 1994 -Uses 2 d currents from hydrodynamic model -Time dependent P loads -Combination Lax-Wendroff and upwind advection scheme -No horizontal diffusion -Initial condition: C = 10 ug/L -Settling velocity = 6. 8 E-7 m/s (21 m/yr)

Questions?