All sea-ice products on display here have been generated
from digital data provided by the National Snow and Ice Data Center NSIDC),
University of Colorado, Boulder. As to monthly sea ice concentrations
the Bootstrap algorithm has been used, while for daily near-real-time ice concentrations
the NASA-Team algorithm has been employed. The Arctic circular polar data hole
has been closed by interpolation.
The graphical presentation of these data sets aims to serve
the needs and interests of both navigation and climate research. Ice concentrations
are classified in accord with international standards applied to operational
ice charts by most national ice services: open water < 10 < very open
< 35 < open < 65 < close < 85 < very close ice. The refined
classification beyond 85 % appears sensible and necessary to visualize internal
ice distribution pattern outside the summer season.
In addition to geographical distributions of ice concentration
all charts contain information on absolute and relative frequency distributions
of ice concentration classes. In this context it is important to realize the
different concepts of ice extent (E) and actual ice area (A). The difference
E-A specifies the area of open water inside the ice boundaries, while the fraction
A/E gives the mean overall ice concentration within the ice limits.
The video machine affords animation of nearly arbitrary sequences
of monthly ice charts in both directions of time at various time steps. This
functionality makes it easy to look at very different aspects of the spatiotemporal
evolution of the sea ice distribution since 1978 in both hemispheres. Monthly
ice charts are continuously updated with a delay of 3 to 6 month.
Near real time ice charts are generated at 23:45 local time
for the previous day. The most recent 30 days are accessible through the video
machine. The charts for the Northern Hemisphere in addition show daily varying
distributions of continental snow cover.
Daily sea ice data have been filtered using monthly masks
compiled from the Bootstrap data set (1979-1999). If the local probability of
ice occurrence in the relevant month falls short of 35 % the ice concentration
of the associated grid cell is set to 0. This method eliminates spurious ice
concentrations, not only common in the vicinity of land but also in ice-free
ocean areas in stormy weather. In the presence of extreme ice extent some overfiltering
can occur, which then becomes visible in an abrupt transition from ice-free
conditions to very high ice concentrations.
Digital source data for our graphical sea-ice products have
been kindly provided by EOSDIS NSIDC Distributed Active Archive Center (NSIDC
DAAC), University of Colorado, Boulder. Extensive documentation on all data
sets can be found at NSIDC's website .
The monthly products are based on the Bootstrap data set
1 for the period October 1978 ? December 1999) and data set 2 thereafter. For
the daily near-real-time products use has been made of the NASA-Team data set
3 as to sea ice information and data set 4 as to continental snow cover.
1) National Snow and Ice Data Center. 1999. Bootstrap Sea
Ice Concentrations from Nimbus-7 SMMR and DMSP SSM/I. Digital data available
from firstname.lastname@example.org. Boulder, Colorado. NSIDC Distributed Active Archive Center,
University of Colorado at Boulder.
2) Comiso, J. 1990 - March 2000. DMSP SSM/I daily polar gridded
sea ice concentrations. Edited by J. Maslanik and J. Stroeve. Boulder, CO: National
Snow and Ice Data Center. Digital media.
3) Cavalieri, D., P. Gloerson, and J. Zwally. 1990 - March
2000. DMSP SSM/I daily polar gridded sea ice concentrations. Edited by J. Maslanik
and J. Stroeve. Boulder, CO: National Snow and Ice Data Center. Digital media.
4) National Snow and Ice Data Center. 1998. Near Real-Time
SSM/I EASE-Grid Daily Global Ice Concentration and Snow Extent. Digital data
available from email@example.com. Boulder, Colorado. EOSDIS NSIDC Distributed
Active Archive Center, University of Colorado at Boulder.