Skill Assessment of Multiple Hypoxia Models in the Chesapeake Bay and Implications for Management Decisions Isaac (Ike) Irby - Virginia Institute of Marine Science Marjorie Friedrichs – VIMS Carl Friedrichs - VIMS Cathy Feng – VIMS Raleigh Hood – UMCES Jeremy Testa – UMCES

Project Group U.S. IOOS Coastal Ocean Modeling Testbed (COMT)Support for Scenario-based Ecological Forecasts To improve understanding and operational forecasts of extreme events and chronic environmental conditions affecting the U.S. Five Science Teams Chesapeake Bay Estuarine Hypoxia Forecasting Integration of West Coast Operational Coastal & Ocean Models Puerto Rico/U.S. Virgin Islands Inundation & Wave Forecasting Northern Gulf of Mexico Ecological Forecasting Cyberinfrastructure

Project Group U.S. IOOS Coastal Ocean Modeling Testbed (COMT)Support for Scenario-based Ecological Forecasts To improve understanding and operational forecasts of extreme events and chronic environmental conditions affecting the U.S. Chesapeake Bay Estuarine Hypoxia Forecasting VIMS: Marjy Friedrichs (lead PI) Carl Friedrichs (VIMS-PI) Ike Irby (funded student) Aaron Bever (consultant) Jian Shen (collaborator) Cathy Feng (collaborator) WHOI: Malcolm Scully (WHOI-PI) UMCES: Raleigh Hood (UMCES-PI) Hao Wang (funded student) Jeremy Testa (collaborator) Wen Long (collaborator) NOAA-CSDL: Lyon Lanerolle (NOAA-PI) Frank Aikman (collaborator)

Chesapeake Hypoxia Objective Assess the readiness and maturity of a suite of existing estuarine ecological community models for determining past, present, and future hypoxia events within the Chesapeake Bay, in order to accelerate the transition of hypoxia model formulations and products from academic research to operational centers. Operational Centers: Chesapeake Bay Ecological Prediction System (CBEPS) Short-term forecasts (R. Hood) NOAA/NOS/CO-OPS Short-term forecasts (L. Lanerolle) EPA Chesapeake Bay Program (CBP) Scenario-based forecasts (M. Friedrichs, C. Friedrichs, I. Irby)

Scenario-based Forecasting Project Can community-based models achieve a similar skill to the Chesapeake Bay Program’s regulatory model in representing dissolved oxygen concentrations? Would a similarly skilled model predict similar water quality standard attainment as the Chesapeake Bay Program’s regulatory model from a given level of nutrient reduction? How can these models be used to define uncertainty in regulatory model predictions?

Hypoxia Model Comparisons Statistically compare output from four Bay models Three ROMS-based models with varying biological complexity ROMS – RCA ChesNENA ChesROMS – BGC CBP regulatory/operational biologically sophisticated model CH3D – ICM Examine how well each model reproduces the mean and spatial/seasonal variability of: Temperature Salinity Stratification Dissolved Oxygen Biological Complexity Chlorophyll-a Nitrate

Hypoxia Model Comparisons Compare simulations to observations at 10 main-stem stations for ~16 cruises in 2004 and 2005

Model Skill Assessment

Model Comparison Results Overall skill of all four models (temporal + spatial variability) are: highest in terms of temperature similar to each other in terms of T, S, stratification and DO Different in terms of chlorophyll and nitrate 2004

Model Comparison Results High Flow Normal Flow

Summary of Results Regardless of complexity, models do similarly well in terms of reproducing observations of T, S, and DO, and similarly poorly in terms of stratification All models reproduce DO better than variables that are typically thought to be primary influences on DO (stratification, chlorophyll, and nitrate) This is because seasonal DO variability is sensitive to T (solubility effect), and the models reproduce T very well Modeled DO simulations may be very sensitive to any future increases in Bay temperature  Hypoxia scenario forecasting is possible with simple biological formulations

Moving to Scenario-based Forecasts EPA/CBP model ChesNENA Ensemble of Implementations EPA/CBP Projected Water Quality Model Ensemble Projected Water Quality Nutrient Reduction Scenario Estimate Uncertainty in Projections Conducted with assistance from the CBP

Session goals… “… how we can make the various ecological models/forecasts that are being created through R&D more sustainable and accessible for application to important ecosystem-based management decisions and to be improved through time” Community-based models to support and define uncertainty in regulatory models Easy to run/can be quickly updated to tell us when the regulatory model needs be updated

Thank You

Skill Assessment Project Results: 2004 Seasonal Variability Surface Chlorophyll Surface Nitrate CBP A A B B C C

Skill Assessment Project Results: 2004 Seasonal Variability CBP A A B C B C