ACCESS Ocean and Sea Ice Model CORE II

ACCESS Ocean and Sea Ice Model CORE II Simulations Simon Marsland CSIRO Marine and Atmospheric Research WGOMD, Exeter, May 1, 2009

ACCESS Australian Community Climate and Earth System Simulator • The ACCESS Model is being developed by the Centre for Australian Weather and Climate Research, a partnership between CSIRO and the Bureau of Meteorology Land Surface CABLE (+ CASA-CNP) Dynamic Vegetation (LPJ) Atmosphere (UM) Atmospheric Chemistry (UKCA) Coupler OASIS Sea Ice (Auscom – CICE) Ocean Biogeochemistry (Matear) CSIRO. WGOMD, Exeter, May 1, 2009 Ocean (Auscom – MOM 4)

ACCESS Ocean Model: MOM 4 p 1 + OASIS 3. 2. 5 + CICE 4 • 46 vertical levels • Nominal 1 degree global horizontal grid • Tripolar grid in Arctic region • Equatorial meridional refinement: 1/3 degree from 10 S to 10 N • Mercator grid in Southern Ocean: 1 degree at 30 S to 0. 25 degree at 78 S CSIRO. WGOMD, Exeter, May 1, 2009

ACCESS Implementation Plan • • 2009: buy a computer and couple the models • CSIRO. WGOMD, Exeter, May 1, 2009 2007: develop some expertise • The Atlantic, May 2009 2006: choose some models 2010: commence IPCC/CMIP 5 runs

How can a CORE II dataset be used by the “outsiders”? • Nobody (beyond the ocean modellers) expects much from CORE I • The opposite is true for CORE II. Everybody expects everything. • Interannual forcing means people want to see the real world • • • Modes of variability: ENSO, NAO, IOD, SAM, … sea ice, watermasses, convection, overturning, … Transports: Drake, ITF, … Seasonal variability, Interannual variability, Decadal variability etc Even some climate change, e. g. sea level rise • Highly likely people will use CORE II outputs as a guide when shopping for a model • Shoppers will look for what they want to see in a model • And then use the CORE II model run of choice as a control for whatever experiment they have been losing sleep over • And they will want extension of dataset through to 2008, and beyond in due course? • e. g. Arctic sea ice, Antarctic sea ice, … CSIRO. WGOMD, Exeter, May 1, 2009

10: Initialisation - Repeat cycles or spinup? Need to account for both initial shock and model drift • Do we want to spin up with climatology (CORE I) and then start CORE II? • Do we want to run repeat cycles of CORE II, and how many? COREv 2 interannual COREv 1 normal year CSIRO. WGOMD, Exeter, May 1, 2009

Model validation (I): Global Ocean Water mass transports CSIRO. WGOMD, Exeter, May 1, 2009

12: How can we make output more accessible to community • Central server versus individual repositories? • If we intend to individually serve our IPCC outputs to community, then this might be a good practice exercise • What do the AMIP community do? CSIRO. WGOMD, Exeter, May 1, 2009

Current Australian experiences with CORE II • We have chosen CORE for our tuning runs • Currently doing both CORE I and CORE II runs • • Laplacian versus Biharmonic KPP vs Chen vs GOTM Convective overturn vs enhanced diffusion Etc … CSIRO. WGOMD, Exeter, May 1, 2009

ACCESS ocean/sea ice model simulation 27 years 1979 - 2005 Sea surface temperature equatorial Pacific (5°N– 5°S; 170°W– 120°W) Upper ocean heat content equatorial Pacific (5°N– 5°S; 180°– 110°W; top 300 m) Aus. COM model observed CSIRO. WGOMD, Exeter, May 1, 2009

Equatorial thermocline - mean 1979 -2006 Aus. COM 1. 0: Long-term mean Levitus climatology: temperature along the equator Temperature along the equator Problems: • Thermocline too diffuse a common model problem • Western Pacific Warm Pool too hot CSIRO. WGOMD, Exeter, May 1, 2009

Pacific equatorial undercurrent (m/s) A variety of problems: • undercurrent too weak • undercurrent wrong structure Possible solutions: • improved Indonesian Throughflow? • need for higher resolution? • biharmonic versus Laplacian friction? Pers. Comm. Jaci Brown (CSIRO) CSIRO. WGOMD, Exeter, May 1, 2009

5: INSTANT Mooring Program • Susan Wijffels and Rebecca Cowley (CAWCR) • INSTANT. International Nusantara Stratification and Transport Program • 2004 -2006 continuous moorings in key straits • Allows direct comparison with ACCESS model transports Lombok Ombai Timor Total ITF CSIRO. WGOMD, Exeter, May 1, 2009

INSTANT/ACCESS comparison - Lombok St We are making incremental improvements to model ITF by tuning individual Strait transports against the observed intra-seasonal variability • momentum sink parameterisation is used to reduce flow where necessary • changes to bathymetry to increase flow where necessary CSIRO. WGOMD, Exeter, May 1, 2009

INSTANT/ACCESS comparison - Lombok St We are making incremental improvements to model ITF by tuning individual Strait transports against the observed intra-seasonal variability • momentum sink parameterisation is used to reduce flow where necessary • changes to bathymetry to increase flow where necessary CSIRO. WGOMD, Exeter, May 1, 2009

INSTANT/ACCESS comparison - Timor St • Model/data collaboration We are making incremental improvements to model ITF by tuning individual Strait transports against the observed intra-seasonal variability • momentum sink parameterisation is used to reduce flow where necessary • changes to bathymetry to increase flow where necessary CSIRO. WGOMD, Exeter, May 1, 2009

Opening Timor St - massive earthworks required • Model/data collaboration CSIRO. WGOMD, Exeter, May 1, 2009

INSTANT/ACCESS comparison - ITF (old) INSTANT (obs) CSIRO. WGOMD, Exeter, May 1, 2009

INSTANT/ACCESS comparison - ITF (new) INSTANT (obs) CSIRO. WGOMD, Exeter, May 1, 2009

INSTANT/ACCESS comparison - watermasses Importance of resolving correct watermasses in transports Requires tuning of individual straits Recent work indicates enhanced mixing may help (e. g. tides) CSIRO. WGOMD, Exeter, May 1, 2009

INSTANT/ACCESS comparison - vertical structure Resolving wind driven Strait transport reversals Work in progress CSIRO. WGOMD, Exeter, May 1, 2009

The INSTANT project was 2004 -06: same years as the recent bug report LONGWAVE SHORTWAVE Delivery is delayed to these recipients or distribution lists: oar. gfdl. core-dataset-users@noaa. gov Subject: This message has not yet been delivered. Microsoft Exchange will continue to try delivering the message on your behalf. Reporting-MTA: dns; relay-central. nems. noaa. gov (tcp_intranet-daemon) Final-recipient: RFC 822; oar. gfdl. core-dataset-users@noaa. gov Action: delayed. Status: 5. 4. 0 X-Supplementary-Info: < #4. 4. 7> CSIRO. WGOMD, Exeter, May 1, 2009

4: Baseline metrics: Validating an Ocean Climate Model • Start with the Griffies et. al CORE I baseline metrics • Global Diagnostics • Meridional overturning circulation (Global and Atlantic) • • • Strength and depth of North Atlantic Deep Water cell Atlantic outflow at 30 S Antarctic Bottom Water Cell Southern Ocean Cell Poleward heat transport (Global and Atlantic) Gulf Stream and Kuroshio currents Drake Passage transport Sea ice concentration and thickness (winter and summer) North Atlantic and Southern Ocean convection Tropical Pacific velocity structure CSIRO. WGOMD, Exeter, May 1, 2009

ACCESS Atlantic Meridional Overturning Circulation (Sv) Equatorial cells North Atlantic Deep Water (NADW) Depth of interface Antarctic Bottom Water (AABW) CSIRO. WGOMD, Exeter, May 1, 2009

ACCESS Global Meridional Overturning Circulation (Sv) Antarctic Intermediate Water (AAIW) Southern Ocean cell CSIRO. WGOMD, Exeter, May 1, 2009

ACCESS barotropic streamfunction (Sv) Gulf Stream Kuroshio Current Aghulus Drake Passage transport Indonesian Throughflow CSIRO. WGOMD, Exeter, May 1, 2009

Drake Passage transport (Sv) COREv 2 interannual COREv 1 normal year • Common Ocean Ice Reference Experiment forcing (CORE) • corrected version of NCEP forcing (Large and Yeager, 2008) • COREv 1 climatology - too weak? • COREv 2 interannual - too strong? CSIRO. WGOMD, Exeter, May 1, 2009

Aus. COM high latitude deep convection sites Northern Hemisphere Southern Hemisphere • Mixed layer depth shows main sites of deep convection • Importance for water mass exchange • Importance for overturning circulation CSIRO. WGOMD, Exeter, May 1, 2009

Aus. COM Arctic sea ice thickness (m) February (winter) August (summer) • Arctic sea ice too thin in summer • Too strong ocean vertical mixing? • Sea ice initial condition? Collapse of Arctic halocline. CSIRO. WGOMD, Exeter, May 1, 2009

Aus. COM Antarctic sea ice thickness (m) February (summer) August (winter) • Too little summer sea ice • Poor representation of Western Weddell Sea • Nice representation of East Antarctic coastal polynyas CSIRO. WGOMD, Exeter, May 1, 2009

Southern Hemisphere Sea Ice Concentration Year 200; climatological forcing CSIRO. WGOMD, Exeter, May 1, 2009

Southern Hemisphere Sea Ice Thickness (m) Year 200; climatological forcing CSIRO. WGOMD, Exeter, May 1, 2009

Surface Temperature and Salinity Initial Condition Year 100 Year 200 S A L I N I T Y T E M P E R A T U R E CSIRO. WGOMD, Exeter, May 1, 2009

Zonal Average Temperature and Salinity: Loss of Antarctic Intermediate Water Initial Condition AAIW Year 100 Year 200 S A L I N I T Y T E M P E R A T U R E CSIRO. WGOMD, Exeter, May 1, 2009