Climate and Cryosphere CliC: Cryosphere Reanalyses

1
Climate and Cryosphere (CliC)Cryosphere
Reanalyses
(Agenda item 10.3)
2
Reanalyses of, or relevant to, the
cryosphere(almost certainly an incomplete list)

• Global models that include an ice model
  presentation by Detlef Stammer
• Regional models
• Arctic System Reanalysis (ASR) - coupled
  ice-ocean, land surface and other models. Under
  development.
• Pan-Arctic Ice-Ocean Modeling and Assimilation
  System (PIOMAS), 1978 2004.
• SNODAS at NWS/NOHRSC (not a reanalysis)
• Experimental models
• Product reprocessing
• Sea ice extent and concentration
• Snow cover
• Snow/ice surface temperature and albedo
• Winds (not cryosphere, but relevant)

3
Cryosphere Variables

• Snow- Snow water equivalent, depth, extent,
  state, density, snowfall, solid precipitation,
  albedo
• Lake and River Ice- Freeze-up/Break-up,
  thickness, snow on ice
• Sea Ice- extent, concentration, open water,
  type/age, thickness, motion, icebergs, snow on
  ice
• Glaciers, Ice Caps, Ice sheets- mass balance
  (accumulation/ablation), thickness, area, length
  (geometry), firn temperature, snowline/equilibrium
  line, snow on ice
• Frozen Ground/Permafrost- soil
  temperature/thermal state, active layer
  thickness, borehole temperature, extent, snow
  cover

4
Assimilation of Snow and Ice Data(Incomplete
list)

• Snow/ice
• Sea ice motion (sat) experimental, climate
  model, PIOMAS (U. Washington)
• Sea ice extent (sat) operational NWP and
  reanalyses, U.S. Navy PIPs model PIOMAS Canada
  others?
• Sea ice concentration (sat) experimental
  forecast model
• Snow cover (sat surf) operational NWP, NOHRSC
  Canada others?
• Snow water equivalent (sat surf) operational
  NWP, NOHRSC
• Ice surface temperature (sat, surf)
  experimental.
• Polar Atmosphere
• Polar winds (sat) operational NWP models (11
  worldwide). Reanalysis use soon (JRA, ERA,
  others)

5
PIOMAS

• (Jinlun Zhang, University of Washington)

The Pan-Arctic Ice-Ocean Modeling and
Assimilation System (PIOMAS) is a coupled
ocean-ice capable of assimilating ice
concentration and velocity data. It generates ice
thickness, concentration, motion, and snow depth.
Monthly mean model simulated sea ice thickness
(m) and satellite observed ice edge (white line)
for 9/1979 and 9/2003.
6
Experimental Assimilation of Sea Ice Motion

• (Todd E. Arbetter, National Ice Center)

Satellite-derived observations greatly increase
inventory of ice motions SMMR, SSM/I, AVHRR,
RGPS, QuikScat, AMSR Much improved spatial and
temporal coverage These observations can be
used with data assimilation to enhance model
simulations. Correct ice motion to observed
values Reduce propagation of errors in calculated
fields Identify errors in model parameterizations
7
Optimal Interpolation of Ice Motion Data

• Baseline (without Data Assimilation), model
  processes external forcing as usual, predicts ice
  motion due to external/internal momentum balance
• With Data Assimilation, satellite and
  buoy-derived (observed) ice motions are blended
  with modeled (predicted) ice motions before
  advection step

8
Fractional Ice-Covered AreaBasin Mean

• Baseline simulation
• Strong interannual variability
• Less MY Ice in 1990s

• Assimilation simulation
• Less MY Ice overall,
• Ridged ice has stronger seasonal amplititude
• More summertime melt each year

Open Water First-Year Ice Multi-Year Ice
Ridged Ice
9
Assimilation of Snow DataSNODAS at NOHRSC (U.S.
NWS)

• SNODAS combines all available data, including NWP
  model output coupled with meteorological and snow
  observations, to generate a best estimate of
  gridded snow water equivalent in near real-time.
  SNOWDAS includes
• data ingest and downscaling procedures,
• a spatially distributed energy-and-mass-balance
  snow model that is run once each day, for the
  previous 24-hour period and for a 12-hour
  forecast period, at high spatial (1 km) and
  temporal (1 hr) resolutions, and
• data assimilation and updating procedures.
• The snow model is driven by downscaled analysis
  and forecast fields from a mesoscale, NWP model,
  surface weather observations, satellite-derived
  solar radiation data, and radar-derived
  precipitation data. It is updated with satellite
  and surface observations of snow extent, snow
  depth, and snow water equivalent.

10
(No Transcript)
11
Additional Products for Assimilation or
Verification

• A few current reprocessing efforts
• Sea ice extent and concentration for SSMR and
  SSM/I period (NSIDC and Ocean and Sea Ice group
  of EUMETSAT's Satellite Application Facility)
• Sea Ice Charts of the Russian Arctic, 1933-2006
  (AARI and NSIDC)
• Snow reanalysis (NSIDC and Rutgers U.)
• Historical AVHRR polar winds (NOAA and U.
  Wisconsin)
• Snow/ice surface albedo

12
MODIS and AVHRR winds filling observing system
void
13
Historical Polar Winds from AVHRR
A 20 year dataset of wind vectors (speed,
direction, height) in both polar regions has been
generated from AVHRR data for the period
1982-2002. It is being reprocessed to add
additional satellites and to bring it up to date.
Daily composite of wind vectors derived from
NOAA-17 AVHRR GAC data on March 17, 2003 over
Antarctica. The South Pole is at the center of
the image. The background is the AVHRR 11 micron
brightness temperature image. Wind vectors are
grouped into three height categories (for
illustration only) below 700 hPa (yellow), from
400 to 700 hPa (cyan), and above 400 hPa.
14
Satellite-derived Surface Broadband Albedo
15
New Product Potential for Reanalyses Low-level
Atmospheric Temperature Inversion Strength from
HIRS
Inversion strength from raobs vs the HIRS
2-channel retrieval (asterisks) and the TOVS
Path-P temperature profile retrievals (diamonds).