Multi-decadal Coupled Sea- ice/Ocean Numerical Simulations of the Bering Sea Kate Hedstrom, ARSC/UAF Enrique Curchitser, LDEO Al Hermann, PMEL January, 2006
Dec 30, 2015
Multi-decadal Coupled Sea-ice/Ocean Numerical
Simulations of the Bering Sea
Kate Hedstrom, ARSC/UAFEnrique Curchitser, LDEO
Al Hermann, PMELJanuary, 2006
Bering Sea
• Motivation and background• Bering sea model implementation• Results:
– Circulation– Sea-ice cover– Interannual variability and trends
• Conclusions and future work
Motivation
• A yardstick for climate change (sea ice)
• High primary productivity• Significant commercial fisheries (Pollock)
• Comparison with other sub-Arctic seas (e.g., Barents)
The Model• ROMS ocean
– F90– Now with an adjoint and tangent linear for
data assimilation
• Ice model from Paul Budgell– EVP dynamics (Arakawa C grid)– Mellor-Kantha thermodynamics– Oceanic molecular sublayer under the ice
for improved behavior
NEP Implementation
• 10 km average horizontal resolution• Run1:
– 30 vertical layers– IC’s and BC’s from NPac– Daily fluxes from NCEP hindcast (modified)– 1996-2002, 1960-1970
• Run2:– 42 vertical layers– IC’s and BC’s from CCSM (POP)– Six-hourly fluxes from CCSM hindcast– 1958-2000+, up to 1974 so far
Surface velocities - NCEP
Sea ice concentration: January 1996
ROMS - NCEP SSM/I+
Sea ice concentration: January 1997
ROMS - Run 1
SSM/I+
Sea ice concentration: January 1998
ROMS - Run 1
SSM/I+
Lessons from a “bad” simulation: The global warming scenario
NCEP (tweaked)
NCEP
Sea Surface Temperature
ROMS - NCEP ROMS - CCSM
Sea-ice Concentration: April 1966
ROMS - NCEP ROMS - CCSM
Sea-ice Concentration: April 1969
ROMS - NCEP ROMS - CCSM
Sea-ice Concentration: April 1963
ROMS - NCEP ROMS - CCSM
Future Plans
• Finish the 10 km run driven by CCSM forcing
• Run on the 4 km grid– Four tidal constituents from Mike Forman– Closer to eddy resolving– Estimate 200 hours/year on 48 pwr4 processors
• Add an ecosystem model (and modeler)
Conclusions• The model reproduces the seasonal and interannual variability in the sea-ice conditions as well as the major circulation features
• The CCSM forcing fields are a significant improvement:– Higher time resolution– No hack to the heat fluxes
• We hope to use these simulations to improve our understanding of the circulation and ecosystem in the Bering Sea
Surface Velocityweekly means 1960-1963