Transcript Sea Ice Model Update David Bailey and Marika Holland, NCAR • Community Ice CodE (CICE) 4.0 Base Code • Delta-Eddington Radiative Transfer (Briegleb.

Sea Ice Model Update
David Bailey and Marika Holland, NCAR
• Community Ice CodE (CICE) 4.0 Base Code
• Delta-Eddington Radiative Transfer (Briegleb and Light)
• Melt Pond Parameterization (Bailey and Holland)
• Multi-layer Snow Formulation
• Arbitrary Number of Tracers (for example - age added,
being tested)
• Albedo synchronization, different timestep
functionality, flexible domain decompositions, other
numeric/SE changes
• On table for CCSM4 - Possible snow aging
parameterization
Long-standing Arctic biases in CCSM
Ice Thickness
Jan Total Cloud
Observed
Moritz and Bitz, presentation, 2000
CCSM3 T42 Control
Sfc SW down
Moritz and Bitz, 2000
Sfc LW down
Moritz and Bitz, 2000
An interim step forward
CCSM3.5
CCSM3.5
CCSM3 T42
Arctic Sea Ice Thickness
CCSM3.5 Arctic Clouds
CCSM3
CCSM3
CCSM3.5
CCSM3.5
CCSM3.5
Arctic Surface
Radiation
CCSM3
CCSM3.5
CCSM3.5
Values shown for a
“SHEBA location” 20-yr
climatology
*=SHEBA data
CCSM3
Diamonds=other
observations
CCSM3.5
Antarctic
Sea Ice
CCSM3.5
CCSM3
Generally improved
Thinner Weddell
Sea ice
Less Extensive in
Atlantic sector
CCSM3.5-CCSM3
CCSM3.5 Runs
• Numerous polar climate improvements
as reported in June
• These still generally hold
• Although other problems have
been diagnosed and remain
CAM3.5
SSMI
20th
century run
Downwelling SW
Radiation in vicinity of
the SHEBA point
AMSR-E
High-Latitude North
Pacific Liquid Water Path
Zuidema and Joyce, JGR,
in press
CCSM3.5 - 20th century simulation
Ann 1980-1999 Thickness
JAS Ice Concentration
JFM Ice Concentration
Winter (JFM) Mean Ice Thickness
CCSM 31level CAM
Experiment
CCSM31lev
CCSM26lev
CCSM31lev-CCSM26lev
Good news: Ice
doesn’t really feel
the vertical
resolution / PBL
changes here.
Bad news: Big
differences with
UW PBL. Still in
development
stage.
CICE Physics Developments
• New radiative transfer in snow / ice.
• Sea ice melt ponds.
Delta-Eddington/Pond Runs
Pond parameterization: 15% of melt water accumulates on ice,
partitioned into area/depth based on SHEBA obs, no ponds if
snow present on ice category.
NH Ice Area
For these simulations:
The pond fraction remains
very small.
DE/Ponds Run
Control Run
NH Ice Volume
Increase in ice volume is
simulated.
Albedo generally increased.
Alternative Pond Formulation
Meltwater accumulation linear function of ice fraction (i.e. more water
remains on less concentrated ice); Ponds allowed to form on snow-free
ice fraction
June Pond Fraction
May Ice Thickness
• Larger pond
fractions present
• Leads to reduced
albedo
• Thinner ice cover
• Pond formulation is
highly “tunable”
• Looking into data to
help constrain
parameters
Summary
• Improvements in CCSM 3.5 mostly
due to changes in CAM and POP.
• Ongoing diagnostics of high-latitudes
in PBL and high-resolution runs.
• Continued CICE developments leading
to CCSM4.
• General PCWG interests include other
high-latitude aspects.
CICE Plans (i.e. Beyond CCSM4)
• Biogeochemistry
• Improved snow physics aging,
densification
• Thermo and dynamic processes
biology, soot, etc.
ridge disintegration
• Alternative dynamics schemes
(e.g. elastic-decohesive model)
• Geodesic infrastructure
• Hydrology
prognostic salinity
percolation
flushing, flooding, etc.
Requests from Users
• Fast ice
• Univ. College London rheologies
(directional leads)
• Air and ocean drag as F(ridges)
• Updated frazil/new ice
parameterization
• Regional configuration