NOS Coastal Ocean Modeling Framework COMF and ROMS

  • Slides: 27
Download presentation
NOS Coastal Ocean Modeling Framework (COMF) and ROMS Applications in NOS Operational Forecast Systems

NOS Coastal Ocean Modeling Framework (COMF) and ROMS Applications in NOS Operational Forecast Systems Aijun Zhang 1, Mark Vincent 2, Frank Aikman 3 Eugene Wei 3 , Richard A. Schmalz, Jr. 3 , Lyon Lanerolle 3 Center for Operational Oceanographic Products and Services (CO-OPS), National Ocean Service/NOAA 2 Oceanic and Atmospheric Research/NOAA 3 Coast Survey Development Laboratory, National Ocean Service/NOAA 1 April 6, 2010

Outline • What and how NOAA does coastal operational forecast systems • Some examples

Outline • What and how NOAA does coastal operational forecast systems • Some examples of ROMS applications • Challenges and wish list

NOS Oceanographic Forecast Systems Requirements q Primary Mission: Support of safe & efficient navigation

NOS Oceanographic Forecast Systems Requirements q Primary Mission: Support of safe & efficient navigation • Water levels for under-keel clearance • Currents for right-of-way, maneuverability q Emergency response (provide circulation, T, S etc. fundamental information) • HAZMAT • Search & Rescue • Homeland Security q For environmentally sound management of the coastal zone • Ecosystem applications • Marine geospatial applications SST Salinity

Cook Inlet Great Lakes Columbia River (Mar. 2006) New York Harbor (Feb. 2003) San

Cook Inlet Great Lakes Columbia River (Mar. 2006) New York Harbor (Feb. 2003) San Francisco St. Johns River (Oct. 2005) Galveston Bay (Jun. 2004) Tampa Bay Gulf of Mexico Delaware Bay Chesapeake Bay (Aug. 2001)

New NOS OFS Strategy • Global/Basin Model from Navy and NWS/NOAA (e. g. NCOM,

New NOS OFS Strategy • Global/Basin Model from Navy and NWS/NOAA (e. g. NCOM, HYCOM, RTOFS, etc. ) • Shelf models to fill gaps between global/basin model and coastal/estuarine model (Gulf of Mexico Shelf, West Coast Shelf, East Coast Shelf, etc. ) , provide better boundary conditions for coastal/Estuarine models • Coastal/Estuarine model nested with Shelf models

FY 2010 – 2015 Development and Implementation Schedule for Operational Forecast Systems Fiscal Year

FY 2010 – 2015 Development and Implementation Schedule for Operational Forecast Systems Fiscal Year of Operational Delivery Water Body of new Operational Model FY 12 ROMS Delaware Bay ROMS Chesapeake Bay ROMS 5 Great Lakes POM (GLERL) Columbia River SELFE Northern Gulf of Mexico (New Orleans to Mobile) FVCOM Gulf of Mexico Region (Sabine Pass to New Orleans) ROMS or FVCOM San Francisco ROMS or FVCOM Narragansett Bay/Long Island Sound FY 13 Galveston Bay ROMS or FVCOM New York/New Jersey FVCOM ROMS or FVCOM Huron Erie Corridor (GLERL) or Prince William Sound FVCOM Eastern GOM-WFS West Coast Shelf # of Operational Models on NOAA High Performance Computers (HPC) Total # of Operational Models 3 8 11 3 10 13 1 14 15 18 18 20 20 22 22 0 ROMS or FVCOM Cook Inlet 0 5 Great Lakes FY 15 # of Operational Models on NOS Computers FVCOM Puget Sound St. Johns River FY 14 Hydrodynamic Model Selected Tampa Bay FY 10 FY 11 Water Body of Retrofitted Operational Model FVCOM (GLERL) ROMS or FVCOM

Coastal Ocean Modeling Framework (COMF) COMF is a set of standards and common tools

Coastal Ocean Modeling Framework (COMF) COMF is a set of standards and common tools to develop and operate NOS OFS. It is standardized for any model and any location • • PURPOSE: Efficient R&D, O&M for NOS OFS Easy Data Handling & Maintenance Single System for all Locations Single standard tool for Graphics and Web Pages Standard skill assessment and evaluation Tools Standard NOS Model products to Users Various Models Allowed ADCIRC, ECOM, EFDC, ELCIRC, FVCOM, MECCA, POM, ROMS, SELFE NOS selected ROMS and FVCOM as the core ocean models

Individual Model systems NOS Coastal Ocean Modeling Framework (COMF) REAL-TIME DATA INGEST QA/QC (COMF)

Individual Model systems NOS Coastal Ocean Modeling Framework (COMF) REAL-TIME DATA INGEST QA/QC (COMF) Data Tank on CCS: • Atmospheric Forcing • Coastal Boundary Conditions • Riverine Fresh Water Inputs OPERATIONAL MODELS (COMF) on CCS Products and archives Linux Server in CO-OPS FORECAST MODEL GUIDANCE (water level, water temp, currents, & salinity) PRODUCTS (web pages and digital pt. & gridded data) FOR USERS tidesandcurrents. noaa. gov QA/QC (CORMS) 24 x 7

NOS Operational Forecast System Evaluation (Model Skill Assessment) § Objective - Measure the performance

NOS Operational Forecast System Evaluation (Model Skill Assessment) § Objective - Measure the performance of model simulations (including tidal simulation, hindcast, nowcast, and forecast) by comparing with observations - All models should be assessed and pass the NOS skill assessment standards before transferring to operation § Functions – Data acquisition and process (observations and model outputs) • Time interval conversion and Gap-filling • Concatenation of model outputs • Filtering • Tidal harmonic analysis and prediction • Extracting extremes/events and slack waters – Compute statistical variables – Generate skill assessment score tables – Harmonic constants comparison

Continuous Operational Real-Time Monitoring System (CORMS) CORMS is the primary system to perform 24

Continuous Operational Real-Time Monitoring System (CORMS) CORMS is the primary system to perform 24 x 7 real-time QA/QC on all real-time data, systems, and products in the CO-OPS domain

World-Class Supercomputing AT NCEP • IBM Power 6 p 575 – 69. 7 Teraflops

World-Class Supercomputing AT NCEP • IBM Power 6 p 575 – 69. 7 Teraflops Linpack • #36 Top 500 Nov 2008 – 156 Power 6 32 -way Nodes – 4, 992 processors @ 4. 7 GHz – 19, 712 gigabytes memory – 170 terabytes of disk space – 100 terabyte tape archive – Two identical systems locate in different locations. One is production, the other is backup. – NOAA’s atmosphere operational forecast systems (e. g. NAM, GFS, etc) – NOAA’s ocean forecast systems (e. g. RTOFS, HYCOM, GLOFS, CBOFS, DBOFS, TBOFS, etc. )

NOS OFS PRODUCT FLOW CHART http: //tidesandcurrents. noaa. gov Forcing Data on CCS CORMS

NOS OFS PRODUCT FLOW CHART http: //tidesandcurrents. noaa. gov Forcing Data on CCS CORMS • Real-Time Observations • Model reanalysis and forecast products NCEP Central Computer System (CCS – IBM Super Computer) Sftp using secure key Server “ofsprod” at CO-OPS • sftp access model outputs on CCS • Produce graphic products. • Produce NOS OFS web pages on CO-OPS web. • Archive Operational Products (NAS) All NOS OFS NWS Web Operations Center (WOC) and Distributed Brokered Networking (DBNet) Public THREDDS Data server Opendap / webservices

NOS OFS Nowcast/Forecast Schedule ay st d 3 a 2 - rec Fo 6

NOS OFS Nowcast/Forecast Schedule ay st d 3 a 2 - rec Fo 6 hour nowcast 00 z 06 z (now) 12 z 18 z

NOS OFS Forcing OBC Surface Forcing River Forcing Tides from ADCIRC Subtidal WL from

NOS OFS Forcing OBC Surface Forcing River Forcing Tides from ADCIRC Subtidal WL from ETSS adjusted by real-time observed SWL RTMA USGS Real-time river observations (discharges, T, S) NAM and GFS as backup Nowcast Climatologic daily mean as backup T & S from NCOM adjusted by real-time observations Tides from ADCIRC +Subtidal WL from ETSS Forecast T & S from NCOM NAM GFS as backup NWS river forecast Products Or Persistence from previous nowcast cycle

Chesapeake Bay Operational Forecast System (CBOFS) • 4 cycles per day • 48 hours

Chesapeake Bay Operational Forecast System (CBOFS) • 4 cycles per day • 48 hours forecasts of Water levels, current, T &S Dimensions: 291 x 332 x 20 Res. 50 m – 5 km Bathymetry from NOS sounding

Delaware Bay Operational Forecast System (DBOFS) • 4 cycles per day • 48 hours

Delaware Bay Operational Forecast System (DBOFS) • 4 cycles per day • 48 hours forecasts of Water levels, current, T & S Dimensions: 119 x 732 x 10 Res. 100 m – 3 km Bathymetry from NOS sounding

Tampa Bay Operational Forecast System (TBOFS) • 48 -hour forecasts of Water levels, current,

Tampa Bay Operational Forecast System (TBOFS) • 48 -hour forecasts of Water levels, current, T & S (4 cycles per day) Dimensions: 176 x 290 x 11 Res. 100 m – 1. 2 km Bathymetry from NOS sounding

Water Level forecasts at CBBT and Kiptopeke from CBOFS

Water Level forecasts at CBBT and Kiptopeke from CBOFS

Synoptic Hindcast : water level validation Observations from NOS/CO-OPS monitoring stations

Synoptic Hindcast : water level validation Observations from NOS/CO-OPS monitoring stations

Synoptic Hindcast : currents validation Observations from NOS/CO-OPS/CMIST monitoring stations

Synoptic Hindcast : currents validation Observations from NOS/CO-OPS/CMIST monitoring stations

Synoptic Hindcast : T error summary • Left panel show RMSE and Right panel

Synoptic Hindcast : T error summary • Left panel show RMSE and Right panel show mean error • Look at surface, 15 -feet (4. 6 m) and bottom • Surface, 15 -feet similar & largest errors at bottom • Horizontally, errors mainly along axis of the Bay • Most mean errors in [-1 o. C, +1 o. C] range and RMSE in [0 o. C, 2 o. C] • CBOFS 2 excessively cool at surface and excessively warm at bottom

Synoptic Hindcast : S error summary • Left panel show RMSE error and Right

Synoptic Hindcast : S error summary • Left panel show RMSE error and Right panel show mean error • Look at surface, 15 -feet (4. 6 m) and bottom • Surface, 15 -feet similar & largest errors at bottom • Horizontally, errors mainly along axis of the Bay • Most mean errors in [-2 , +3 ] range and RMSE in [0 , 4 ] in PSU • CBOFS 2 excessively salty at surface and excessively fresh at bottom

Water Level time series from TBOFS.

Water Level time series from TBOFS.

Current time series at Sunshine Skyway Bridge and Old Port Tampa from TBOFS.

Current time series at Sunshine Skyway Bridge and Old Port Tampa from TBOFS.

Challenges and Wish List • Reliable and stable operational version of ROMS (community model).

Challenges and Wish List • Reliable and stable operational version of ROMS (community model). Reliability > 99% • Perfect Restart • Wetting and Drying (test in Cook Inlet) • Data Assimilation (e. g. for shelf model) • Composite grids and grid refinement (shelf model + Estuary model) • Coupling with Wave, Ecological and biological modules (hydrodynamic + eco. ) • Better vertical mixing scheme in coastal shallow waters (vertical stratification) • Future Collaborations with ROMS Community in transitioning research to operation.

Thanks !! Questions ?

Thanks !! Questions ?

Data Resources used for NOS OFS Data tank on CCS – NCEP Central Computer

Data Resources used for NOS OFS Data tank on CCS – NCEP Central Computer System CORMS – Continuous Operational Real-Time Monitoring System COMF – Coastal Ocean Modeling Framework Real-Time Observations NAM – North American Mesoscale model Water Levels, River Flow GFS – Global Forecast System Temperature, Salinity Meteorological variables Model reanalysis and forecast products RTMA – Real-Time Mesoscale Analysis RUC – Rapid Update Cycle NDFD – National Digital Forecsat Database ETSS – Extra-Tropical Storm Surge Model NCOM – Navy Coastal Ocean Model NAM, GFS, RTMA, RUC, NDFD ADCIRC – ADvanced CIRculation model ETSS, NCOM, RTOFS – Real-Time Ocean Forecast System ADCIRC tide constituents WOA – World Ocean Atlas ROMS – Regional Ocean Modeling System