1
Development of U.S. Navy Coupled Atmosphere-Ocean-Sea Ice Earth System Model for Polar Predictions N.P. Barton 1 ([email protected]) E.J. Metzger 2 , P.G.Posey 2 , B.C. Ruston 1 , A.J. Wallcraft 2 , C. Reynolds 1 , J. Richman 3 , and M.W. Phelps 4 NAVY’S EARTH SYSTEM MODEL (NESM): 1 Naval Research Laboratory - Marine Meteorology Division; 2 Naval Research Laboratory – Oceanography Division; 3 Florida State University; 4 Jacobs Technology SPONSORS: LOOSELY COUPLED DATA ASSIMILATION: GOFS 3.1/NCODA COMPARISON NESM NAVGEM Mediator/Coupler (ESMF, NUOPC) Atmosphere (NAVGEM) Sea Ice (CICE) Ocean (HYCOM) The Marine Meteorology and the Oceanography divisions of the Naval Research Laboratory are developing its new global coupled model that builds on the existing operational models: NAVy Global Environmental Model (NAVGEM) for the Atmosphere, and Global Ocean Forecasting System (GOFS) 3.1, which consist of the HYbrid Coordinate Ocean Model (HYCOM) for the ocean and the Community Ice CodE (CICE) for the sea ice. MORE REALISTIC SEA ICE FOR THE ATMOSPHERE MODEL: MODEL DESIGNED FOR SUB-SEASONAL TO SEASONAL PREDICTIONS: Long-range sea ice predictions have been produced using the Navy’s ESPC coupled system in 2015 and 2016 through an international effort called the Sea Ice Prediction Network (https://www.arcus.org/sipn) to forecast the September minimum Arctic sea ice extent months in advance. Shown are predictions from July initial conditions. The Navy ESPC model performed well and was comparable with the more mature Navy GOFS 3.1 and other international models (green bar lines below). NESM NAVGEM NESM minus NAVGEM Data Assimilative Runs July to October 2015 NAVGEM v1.4 NESM NESM is configured to run within a loosely coupled data assimilation cycle, which uses the mature atmosphere (NAVDAS-AR) and ocean/ice (NCODA) Navy data assimilation schemes, but information between the DA schemes is not exchanged. NESM produces very similar results to the operational GOFS 3.1. The currently uncoupled NAVGEM uses simple static sea ice properties with the surface albedo being constant (right figure). In NESM, sea ice properties are from the sea ice model, CICE, which are much more sophisticated and result in more realistic sea ice properties (left figure). NAVGEM/NAVDAS-AR COMPARISON Atmospheric results between NESM and stand-alone NAVGEM differ the most over Antarctica for these runs due to using the CICE sea ice model . Region NESM GOFS 3.1 Arctic 24.9 24.9 Greenland/ Norwegian 21.4 21.2 Barents/ Kara 21.2 21.2 Laptev/ E. Siberian 28.1 27.9 Bering/ Chukchi/ Beaufort 25.8 25.7 Canadian Archipelago 26.7 27.2 NESM GOFS 3.1 Sea Ice Thickness (m) Sea Ice Concentration (%) NESM GOFS 3.1 Ice Edge Error (km) relative to NIC ice edge Observed Value 4.72m km 2 NESM Prediction: 4.8m km 2 Predictions from July NH September 2016 Sea Ice Prediction Network July (left) and August (right) Reports Predictions from August NESM GOFS 3.1 NESM GOFS 3.1

Development of U.S. Navy Coupled Atmosphere-Ocean-Sea Ice Earth System Model … · 2017. 8. 21. · Model (NAVGEM) for the Atmosphere, and Global Ocean Forecasting System (GOFS)

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

  • Development of U.S. Navy Coupled Atmosphere-Ocean-Sea Ice Earth System Model for Polar Predictions

    N.P. Barton1 ([email protected]) E.J. Metzger2, P.G.Posey2, B.C. Ruston1, A.J. Wallcraft2, C. Reynolds1, J. Richman3, and M.W. Phelps4

    NAVY’S EARTH SYSTEM MODEL (NESM):

    1Naval Research Laboratory - Marine Meteorology Division; 2Naval Research Laboratory – Oceanography Division; 3Florida State University; 4Jacobs Technology

    SPONSORS:

    LOOSELY COUPLED DATA ASSIMILATION: GOFS 3.1/NCODA COMPARISON

    NESM NAVGEM

    Mediator/Coupler (ESMF, NUOPC)

    Atmosphere (NAVGEM)

    Sea Ice (CICE)

    Ocean (HYCOM)

    The Marine Meteorology and the Oceanography divisions of the Naval Research Laboratory are developing its new global coupled model that builds on the existing operational models: NAVy Global Environmental Model (NAVGEM) for the Atmosphere, and Global Ocean Forecasting System (GOFS) 3.1, which consist of the HYbrid Coordinate Ocean Model (HYCOM) for the ocean and the Community Ice CodE (CICE) for the sea ice.

    MORE REALISTIC SEA ICE FOR THE ATMOSPHERE MODEL:

    MODEL DESIGNED FOR SUB-SEASONAL TO SEASONAL PREDICTIONS:

    Long-range sea ice predictions have been produced using the Navy’s ESPC coupled system in 2015 and 2016 through an international effort called the Sea Ice Prediction Network (https://www.arcus.org/sipn) to forecast the September minimum Arctic sea ice extent months in advance. Shown are predictions from July initial conditions. The Navy ESPC model performed well and was comparable with the more mature Navy GOFS 3.1 and other international models (green bar lines below).

    NESM NAVGEM NESM minus NAVGEM

    Data Assimilative Runs • July to October 2015 • NAVGEM v1.4 • NESM

    NESM is configured to run within a loosely coupled data assimilation cycle, which uses the mature atmosphere (NAVDAS-AR) and ocean/ice (NCODA) Navy data assimilation schemes, but information between the DA schemes is not exchanged. NESM produces very similar results to the operational GOFS 3.1.

    The currently uncoupled NAVGEM uses simple static sea ice properties with the surface albedo being constant (right figure). In NESM, sea ice properties are from the sea ice model, CICE, which are much more sophisticated and result in more realistic sea ice properties (left figure).

    NAVGEM/NAVDAS-AR COMPARISON

    Atmospheric results between NESM and stand-alone NAVGEM differ the most over Antarctica for these runs due to using the CICE sea ice model .

    Region NESM GOFS 3.1

    Arctic 24.9 24.9

    Greenland/ Norwegian 21.4 21.2

    Barents/ Kara 21.2 21.2

    Laptev/ E. Siberian 28.1 27.9

    Bering/ Chukchi/ Beaufort 25.8 25.7

    Canadian Archipelago 26.7 27.2

    NESM GOFS 3.1

    Sea

    Ice

    Th

    ickn

    ess

    (m)

    Sea

    Ice

    C

    once

    ntra

    tion

    (%)

    NESM GOFS 3.1

    Ice Edge Error (km) relative to NIC ice edge

    Observed Value

    4.72m km2

    NESM Prediction: 4.8m km2

    Predictions from July

    NH September 2016 Sea Ice Prediction Network July (left) and August (right) Reports

    Predictions from August

    NESM

    GOFS 3.1

    NESM

    GOFS 3.1

    Slide Number 1