Combining Observations & Numerical Model Results to Improve Estimates of Hypoxic Volume within the Chesapeake Bay Aaron Bever, Marjy Friedrichs, Carl Friedrichs, Malcolm Scully, Lyon Lanerolle OUTLINE / SUMMARY 1.Relation to US-IOOS Modeling Testbed program and general methods. 2.Use 3D models to examine uncertainties in interpolating hypoxic volume. Observed DO has coarse spatial resolution = spatial error Observed DO is not a “snapshot” = temporal error 3.Use 3D models to improve EPA-CBP interpolations of hypoxic volume. p.1 of 22 Corresponding paper: JGR-Oceans, October 2013 issue

Relationship to US-IOOS Modeling Testbed: Part of Coastal & Ocean Modeling Testbed (COMT) Project headed by Rick Luettich (UNC), funded by NOAA US-IOOS Office COMT Mission: Accelerate the transition of scientific and technical advances from the modeling research community to improve federal agencies’ operational ocean products and services Phase I ( ): Estuarine Hypoxia, Shelf Hypoxia, and Coastal Inundation Modeling Testbeds; Cyber-infrastructure to advance interoperability and archiving Phase II ( ): Added West Coast Model Integration. (See Poster Session for initial results of Estuarine Hypoxia Phase II) Combining Observations & Numerical Model Results to Improve Estimates of Hypoxic Volume within the Chesapeake Bay p.2 of 22

Present Estuarine Hypoxia COMT Team: Virginia Institute of Marine Science: Marjy Friedrichs (lead PI), Carl Friedrichs (co-PI), Ike Irby (PhD student), Aaron Bever (past Post-Doc, now consultant), Jian Shen (collaborator), Cathy Feng (collaborator) Woods Hole Oceanographic Institution: Malcolm Scully (co-PI) Univ. Maryland Center for Environmental Studies: Raleigh Hood (co-PI), Hao Wang (PhD student), Jeremy Testa (collaborator), Wen Long (collaborator) NOAA Coastal Survey Development Lab: Lyon Lanerolle (co-PI), Frank Aikman (collaborator), Combining Observations & Numerical Model Results to Improve Estimates of Hypoxic Volume within the Chesapeake Bay p.3 of 22

Present Estuarine Hypoxia COMT Team: Virginia Institute of Marine Science: Marjy Friedrichs (lead PI), Carl Friedrichs (co-PI), Ike Irby (PhD student), Aaron Bever (past Post-Doc, now consultant), Cathy Feng (collaborator), Jian Shen (collaborator) Woods Hole Oceanographic Institution: Malcolm Scully (co-PI) Univ. Maryland Center for Environmental Studies: Raleigh Hood (co-PI), Hao Wang (PhD student), Jeremy Testa (collaborator), Wen Long (collaborator) NOAA Coastal Survey Development Lab: Lyon Lanerolle (co-PI), Frank Aikman (collaborator), Combining Observations & Numerical Model Results to Improve Estimates of Hypoxic Volume within the Chesapeake Bay p.3 of 22

Compare relative skill and strengths/weaknesses of various Chesapeake Bay models Assess how model differences affect water quality simulations Recommend improvements to agency operational products associated with managing hypoxia (DO < 2 mg/L) General COMT Estuarine Hypoxia modeling methods: p.4 of 22 Combining Observations & Numerical Model Results to Improve Estimates of Hypoxic Volume within the Chesapeake Bay 16 million people and > $1 Trillion in industries in CB watershed EPA Clean Water Act and a recent Presidential Executive Order both require a reduction of hypoxia in Chesapeake Bay Reducing hypoxia in Chesapeake through required nutrient reductions over next 15 years is expected to cost > $20 Billion Motivation for Better Hypoxia Modeling for Chesapeake Bay:

p.5 of 22 EFDC Shen VIMS CH3D Cerco & Wang USACE UMCES-ROMS Li & Li UMCES Five hydrodynamic models configured for Chesapeake Bay DC

p.5 of 22 EFDC Shen VIMS CH3D Cerco & Wang USACE UMCES-ROMS Li & Li UMCES Five hydrodynamic models configured for Chesapeake Bay TODAY’S TALK  x ~ 2 km  z ~ 1-2 m  x ~ 1 km  z ~ 1-2 m  x ~ 0.5 km  z ~ 1.5 m

o ICM: EPA-CBP model; 27-component ecosystem model (multi P, multi Z, C/N/P/Si/DO, etc.) o BGCs: 3 NPZD-type (~10 component) models o 1eqn: Simple one equation respiration (includes SOD) o 1term-DD: depth-dependent respiration (not a function of x, y, temperature, nutrients…), surface DO = saturation o 1term: Same, but constant net respiration (constant with depth) Eight dissolved oxygen (DO) models configured for the Bay p.6 of 22

o ICM: EPA-CBP model; 27-component ecosystem model (multi P, multi Z, C/N/P/Si/DO, etc.) o BGCs: 3 NPZD-type (~10-component) models o 1eqn: Simple one equation respiration (includes SOD) o 1term-DD: depth-dependent respiration (not a function of x, y, temperature, nutrients…), surface DO = saturation o 1term: Same, but constant net respiration (constant with depth) TODAY’S TALK Eight dissolved oxygen (DO) models configured for the Bay p.6 of 22

Today’s talk = Four combinations: o CH3D + ICM  EPA model o CBOFS + 1term o ChesROMS + 1term o ChesROMS + 1term+DD Coupled hydrodynamic-DO models -- Physical models are similar, but grid resolution differs -- Biological/DO models differ dramatically -- All models run for 2004 and 2005 and compared to EPA Chesapeake Bay Monitoring Program’s DO observations p.7 of 22

-- The models all have significant skill (normalized RSMD < 1) in reproducing observed bottom dissolved oxygen (DO). -- EPA regulations require hypoxic volume be interpolated from observations. -- Unlike observations, model output is continuous in space and time. -- So use the continuous model output to estimate uncertainties caused by CBP interpolations of discontinous observed data and improve interpolation. Model skill: Bottom DO Total RMSD 2 = Bias 2 + unbiased RMSD 2 p.8 of 22

OUTLINE 1.Relation to US-IOOS Modeling Testbed program and general methods. 2.Use 3D models to examine uncertainties in interpolating hypoxic volume. Observed DO has coarse spatial resolution = spatial error Observed DO is not a “snapshot” = temporal error 3.Use 3D models to improve EPA-CBP interpolations of hypoxic volume. p.9 of 22 Combining Observations & Numerical Model Results to Improve Estimates of Hypoxic Volume within the Chesapeake Bay Aaron Bever, Marjy Friedrichs, Carl Friedrichs, Malcolm Scully, Lyon Lanerolle Corresponding paper: JGR-Oceans, October 2013 issue

Four Types of Hypoxic Volume Estimates Interpolation Method used for #1 - #3:  CBP Interpolator (inverse dist. weighting)  Hypoxic Volume (HV) = DO < 2 mg/L #1) Observations  Of 99 CBP stations (red dots), are sampled each “cruise”, each cruise takes 1 to 2 weeks #2) Modeled Absolute Match:  Same stations are “sampled” at same time/place as observations are available #3) Modeled Spatial Match:  Same stations are “sampled” in space, but samples are taken synoptically (i.e., all at once in time) #4) Integrated 3D Model:  Hypoxic Volume is computed from integrating over all model grid cells (“CBP” = EPA Chesapeake Bay Program) p.10 of 22

CH3D-ICM ChesROMS+1term Observations-derived = Absolute Match Hypoxic Volume Estimates When observations and model are interpolated in same way, the match is reasonably good p.11 of 22

CH3D-ICM ChesROMS+1term Data-derived = Absolute Match CH3D-ICM ChesROMS+1term Observations-derived Hypoxic Volume Estimates When observations and model are interpolated in same way, the match is reasonably good But interpolated HV underestimates actual HV for every cruise p.11 of 22

CH3D-ICM ChesROMS+1term Observations-derived Hypoxic Volume Estimates p.12 of 22 When observations and model are interpolated in same way, the match is reasonably good But interpolated HV underestimates actual HV for every cruise Much of this disparity could be due to temporal errors (red bars)

When observations and model are interpolated in same way, the match is reasonably good But interpolated HV underestimates actual HV for every cruise Much of this disparity could be due to temporal errors (red bars) Same pattern across all 4 models for both 2004 & 2005 p.13 of 22

Spatial errors show interpolated HV is almost always too low (up to 5 km 3 ) The temporal errors from non-synoptic sampling can be as large as spatial errors (~5 km 3 ) Similar patterns across all 4 models for both 2004 & 2005 p.14 of 22

OUTLINE 1.Relation to US-IOOS Modeling Testbed program and general methods. 2.Use 3D models to examine uncertainties in interpolating hypoxic volume. Observed DO has coarse spatial resolution = spatial error Observed DO is not a “snapshot” = temporal error 3.Use 3D models to improve EPA-CBP interpolations of hypoxic volume. p.15 of 22 Combining Observations & Numerical Model Results to Improve Estimates of Hypoxic Volume within the Chesapeake Bay Aaron Bever, Marjy Friedrichs, Carl Friedrichs, Malcolm Scully, Lyon Lanerolle Corresponding paper: JGR-Oceans, October 2013 issue

Blue triangles = 13 selected CBP stations Improving observation-derived hypoxic volumes  Reduce Temporal errors: 1.Choose subset of 13 CBP stations 2.Routinely sampled within 2.3 days of each other 3.Characterized by high DO variability p.16 of 22

Improving observation-derived hypoxic volumes p.17 of 22 Integrated 3D HV = (1 + CF) (Interpolated HV) Smoothed approximation of CF vs. HV  Reduce Spatial errors: 1. For each model and each cruise, derive a Correction Factor (CF) as a function of interpolated HV that “corrects” this 13-station Spatial Match HV to equal the Integrated 3D HV.

 Reduce Spatial errors: 1. For each model and each cruise, derive a Correction Factor (CF) as a function of interpolated HV that “corrects” this 13-station Spatial Match HV to equal the Integrated 3D HV. 2. Apply smoothed CF (as a function of HV) to HV time-series 3. Scaling-corrected “interpolated” HV more accurately represents true HV Before Scaling After Scaling Improving observation-derived hypoxic volumes p.18 of 22

Blue triangles = 13 selected CBP stations Improving observation-derived hypoxic volumes  Reduce Temporal errors: 1.Choose subset of 13 CBP stations 2.Routinely sampled within 2.3 days of each other 3.Characterized by high DO variability But why 13 stations? p.19 of 22

Improving observation-derived hypoxic volumes p.20 of 22 Modeled Integrated 3D vs. Spatial Match for Different Station Sets (a) 2004 (b) 2005 After being “scaling-corrected”, an interpolation based on these 13 stations did especially well in reproducing 3D HV.

Interannual ( ) corrected (i.e., scaled) time series of observed Hypoxic Volume p.21 of 22  Time-series of corrected hypoxic volume for are provided within JGR article (annual maximum HV, annual duration of HV, annual cumulative HV), and corrected HV for every CBP cruise is provided in JGR electronic supplement.

 Information from multiple models ( ) has been used to assess uncertainties in present CBP interpolated hypoxic volume estimates Temporal uncertainties: up to ~5 km 3 Spatial uncertainties: up to ~5 km 3  These are significant, given maximum HV is ~10-15 km 3  A method for correcting interpolated HV time series for temporal and spatial errors has been presented, based on the 3D structure of multiple model DO results 13 stations (sample in 2 days) do as well for HV as or more Corrected HV for are downloadable from JGR website Summary/Conclusions Combining Observations & Numerical Model Results to Improve Estimates of Hypoxic Volume within the Chesapeake Bay p.22 of 22