Satellite monitoring of the Arctic ocean and adjacent seas during IPY

1
Satellite monitoring of the Arctic ocean and
adjacent seas during IPY

• Stein Sandven, Lasse H. Pettersson et al.
• Nansen Environmental and Remote Sensing Centers
• Bergen, Norwayand St. Petersburg, Russia

2
The Nansen Group
Nansen Environmental and Remote Sensing Center,
Bergen, Norway Nansen International
Environmental and Remote Sensing Centre, St.
Petersburg, Russia Nansen Environmental
Research Centre - India, Cochin,
India Nansen-Zhu International Research Centre,
Beijing, China Nansen Scientific Society,
Bergen, Norway Terra Orbit AS Bergen,
Norway Arctica, Bergen, Norway
Bergen
St. Petersburg
Beijing
Cochin
3
Meteorological Oceanographic observations
4
Outline

• Background Arctic navigation, operations
  research
• Automated RS data products _at_ web-server
• Phytoplankton and Water quality
• Sea ice drift and concentration
• Surface Wind
• Detection of oil pollution at sea and in sea ice
• Sea ice area flux estimation
• Ice navigation support
• Assimilation of RS ice data in ocean models
• Iceberg detection and modelling
• Ice thickness - IceSAT and Cryosat
• Greenland Ice sheet elevation
• Dissemination - ArcticROOS

5
The LAstrolabe Expedition - 1991
6
Radar Satellites supports ARCDEV 1998
7
Marine transportation The Northern Sea Route
ICEWATCH
8
September minimum sea ice, 2002-2007
100 90 80 70 60 50 40 30 15
water land
9
Automated RS products _at_ web
Ocean and coastal winds Ice drift and coverage Water Quality Algae bloom monitoring system

10
Google 6. april 2008 - 7 days Phytoplankton
11
Early signs of the 2006 spring bloom - 7 DAYS
http//HAB.nersc.no
12
Satellite EO Monitoring and Modelling Algae
bloom - 27. March 2007
http//HAB.nersc.no http//wms.met.no/moncoze

13
ASAR wide swath coverage for sea ice monitoring
Period 18 - 22 February 2007
14
14000 scenes in browsable archive
15
Automated ice drift from Envisat ASAR Wide Swath
Mode
Franz Josef Land
Framstrait
Svalbard
Novaya Zemlya
16
Ice drift algorithm

• For two partly overlapping ASAR scenes
• projected to same coordinate system
• overlapping areas are extracted
• images are divided in sub-blocks
• a block from one image is correlated with
  blocks on the second image, shifted varying
  distance in varying directions
• best correlation gives ice drift vector
• several millions ice drift vectors are
  stored in a database
• validated versus manual ice drift vectors

17
Ice drift Area flux in the Fram Strait
Ice area flux across 79o N (in km2) derived from
SAR ice drift data 2004 - 2008
Multisensor analysis of ice drift from satellite
observations
18
Sea ice coverage SAR, IR and optical
MODIS image with SST over open water and albedo
over sea ice (25 April 2008)
ENVISAT ASAR image (25 April 2008)
19
Envisat ASAR WSM
AMSR-E 12.5 km Ice mask
NCEP GFS 0.5º wind direction
CMOD4
GTOPO30 land elevation
Ocean and coastal winds from Envisat ASAR
20
NCEP direction, ASAR speed,
NCEP (model) direction and speed
21
Fjord-jets when wind from the east
22
Comparison with QuikScat
23
Radar Imaging Model (RIM) RIM is aimed to
simulate on a quantitative level both the
background NRCS and radar signature of various
ocean phenomena - Surface current features -
Sea fronts - Long surface waves, - Internal
waves, - Surface contaminations (biogenic and
manmade) - Wind field
24
TOPAZ Arctic Ocean Modelling of Sea Ice
Properties

• Nested physical models
• 30km, 7km, and 4km resolution
• Operational products include
• Currents, SST, salinity, sea iceextent, typt and
  drift

25
Assimilation on 4th and 11th April Up to 10
days forecast
26
Forecast skills Barents Sea - ice concentrations
Average Winter 2007
Average Summer 2007
27
Icebergs in the Eurasian Arctic
North-eastern Barents Sea, April 16, 2006
North-eastern Barents Sea, April 17, 2006
North-western coast of Novaya Zemlya, April 17,
2006
FJL, Salm island, April 25, 2006
28
Iceberg detection in satellite SAR images
Sub-images of ENVISAT ASAR WS for September 15.
Icebergs are evident as bright spots among open
water background.
ENVISAT ASAR WS image of the area between FJL and
Novaya Zemlya for September 15, 2006. Open water
prevails
Map of iceberg distribution, composed from this
image
29
Iceberg detection using satellite ASTER images
2005
30
Modelling of iceberg drift in the Barents Sea
31
Elevation change rate from 1992 to 2007 from
merged ERS-1, ERS-2 and Envisat measurements
6.0 0.2 cm/year Area included 1,3106 km2
(77 of the ice sheet)
cm/year
32
Seasonally averaged time series of elevation
change
33
Correlation between winter elevation change of
Greenland ice sheet and NAO indices
r - 0.80
34
Arctic Regional Ocean Observing System (Arctic
ROOS) Chair of Arctic ROOS Stein Sandven Nansen
Environmental and Remote Sensing
Center www.arctic-roos.org
35
Summary

• Ice concentration and ice drift are produced
  operationally, and several products are available
  from OSI-SAF, Ifremer and NERSC
• Ice concentration retrieval from SSMI data has
  errors of less than 10 , except for the summer
  when the error is higher
• Ice concentration data are assimilated into the
  TOPAZ system, resulting in improved ice
  forecasting
• Multiyear ice retrieval from passive microwave
  data is improved by including backscatter data
  from Quikscat. Validation is done by comparison
  with SAR data.
• Ice drift is retrieved from scatterometer,
  passive microwave data, AVHRR and SAR. There is
  general good agreement between the different data
  sets in the high Arctic.
• In the Fram Strait, SAR data shows higher ice
  drift than the other data sets. SAR can provide
  year round ice drift data, which has been used to
  estimate ice area flux across 79 N
• Satellite-drived ice concentration and ice drift,
  and assimilation of these data in TOPAZ, will be
  further developed and validated in EC GMES MyOcean