In this Issue
Products and Services
NSIDC on the Web
National Antarctic Data Coordination Center at NSIDC
Arctic System Science (ARCSS)
NSIDC currently archives and distributes two sea ice data sets produced using the NASA Team Algorithm applied to satellite passive microwave data: DMSP SSM/I Daily and Monthly Polar Gridded Sea Ice Concentrations (here referred to as NSIDC) and Sea Ice Concentrations from Nimbus-7 SMMR and DMSP SSM/I Passive Microwave Data (here called GSFC).
Both data sets are derived from 25 km x 25 km gridded brightness temperatures in polar stereographic projection and provide users with daily- and monthly-averaged sea ice concentrations for both hemispheres. There are several differences between the two data sets and they are discussed below.
One significant difference between the two data sets is the period of time covered. The NSIDC sea ice concentration data set begins with the launch of the Defense Meteorological Satellite Program (DMSP) SSM/I F8 in July 1987. Data are currently available through June 1997. The GSFC data set extends from the launch of Nimbus-7 SMMR in October 1978 through December 1996. The remaining differences between the data sets result from differences in data processing.
The GSFC sea ice product aims to provide a consistent time series of sea ice extent and ice-covered area spanning the coverage of several passive microwave instruments. To aid in this goal, sea ice algorithm co-efficients are changed for each sensor during overlap periods to reduce overall differences in the derived sea ice extent. The algorithm tie-points were changed as follows. First the SSM/I F8 algorithm tie-points were changed based on comparisons of the daily brightness temperatures with those from SMMR during the overlap period. These linear relationships were then used to generate a set of SSM/I F8 tie-points that are consistent with the original SMMR sea ice algorithm tie-points. In addition to these transformations, the SSM/I F8 open water tie-points were tuned to help minimize the differences between the SMMR and SSM/I F8 sea ice extent and area during the overlap period. Data from the subsequent SSM/I F11 sensor were similarly matched with the adjusted SSM/I F8 data. The same procedure was applied to match the ice concentrations estimated from SSM/I F13 data to the concentrations generated using F11 data with the adjusted F11 tie-points.
In summary, most users may find the GSFC data set to be the best choice for time series studies due to the inclusion of SMMR and SSM/I data, minimization of mean differences in ice extent between the different sensors, and reduction in ice concentration errors. Because the NSIDC data set does not include additional filtering to remove land-contaminated pixels along coasts, the NSIDC ice concentrations may be better suited for detailed investigations of ice conditions along coastal margins. Since the tie-point adjustments used by GSFC are meant to reduce the mean effects of sensor-to-sensor differences, some regions may be affected to a greater or lesser degree by the adjustments. Users interested in regional ice conditions may wish to compare the unadjusted NSIDC product with the GSFC ice concentrations to assess the effects of these tie-point adjustments at specific locations.
NSIDC has recently completed processing daily and monthly total sea ice extent and total ice-covered area. The summaries of total area covered by sea ice and total area with some sea ice coverage are derived from sea ice concentrations from both the NSIDC and Goddard Space Flight Center (GSFC) data sets discussed above. These summaries are provided to aid investigations of interannual variability and trends in sea ice cover. The NSIDC data set is derived from the Special Sensor Microwave/Imager (SSM/I) data and spans the period July 1987 to June 1997. Total ice extent and ice-covered area are also derived from the GSFC sea ice concentration data set, which includes data from the Scanning Multichannel Microwave Radiometer (SMMR). The GSFC data set spans the period November 1978 through December 1996.
In computing the total ice-covered area and ice extent, pixels must have an ice concentration of 15 percent or greater to be included. Thus, total ice extent is computed by summing the total number of pixels with at least 15 percent ice concentration multiplied by the area per pixel (here, we assume a constant area per pixel of 25 km x 25 km). Total ice-covered area is defined as the area of each pixel with at least 15 percent ice concentration multiplied by the ice fraction in the pixel (0.15-1.00).
Total sea ice area and extent in the Northern and Southern hemispheres are computed for each day that contains data. Mean monthly values are computed by averaging the daily values in each month. Sea ice data are missing for some days in October 1978, December 1987 and January 1988. Therefore, in these months mean monthly values are given, but the values do not represent "true" monthly means since the sea ice data are incomplete.
Monthly anomalies of ice extent and ice-covered area have also been calculated throughout the period of study. These anomalies show the deviations from the mean monthly values averaged over the entire time period. Anomalies are computed from the monthly mean values for November 1978 through November 1987, and February 1988 through December 1996.
Data are available as tables and as plots in Post script, GIF, and Adobe Illustrator Encapsulated Post script.
NSIDC has recently created monthly ocean masks for use with the Sea Ice Concentrations from Nimbus-7 SMMR and DMSP SSM/I Passive Microwave Data and DMSP SSM/I Daily and Monthly Sea Ice Concentration Grids for the Polar Regions data sets. These masks are in the polar stereographic projection and were created for users who wish to have an idea of the maximum ice extent during a particular month, or for those who would like to mask out false ice concentrations resulting from weather effects and/or land contamination.
The ocean masks were derived for each month from the GSFC monthly averaged sea ice concentrations. In the first processing step, monthly ice concentrations from November 1978 through December 1996 were averaged over the entire time series to derive yearly- averaged monthly ice concentrations. To derive individual monthly ocean masks, each pixel containing a fraction of ice was flagged as ice and all other non-land pixels are flagged as open water. However, since the monthly-averaged ice fractions still contained false ice concentrations due to weather and land contamination, additional processing of the data was required. Using an ice threshold of one percent ice and a 3 x 3 moving window, most false ice concentrations due to weather were removed. In addition, any pixel adjacent to land containing less than two percent sea ice was considered open water.
The above scheme was found to remove most of the spurious ice concentrations on the grids. However, some false ice concentrations remained along the coast-lines. These were mainly along the coasts of Norway and Great Britain, Hudson Bay and part of the Siberian coast. To further remove these ice fractions, some editing by hand was required. Comparisons between the ocean masks and ice charts from the National Ice Center helped determine which pixels required further editing. In doing the editing, NSIDC chose to err on the side of overestimating the potential ice coverage rather than removing these sea ice areas. Thus, some coastal errors are likely to remain.
NSIDC has created QuickTime movies of Arctic and Antarctic monthly mean sea ice concentrations offering viewers a snapshot of sea ice concentration change on a monthly basis. Users may find the movies particularly useful as a browse tool, or as a quick means of examining seasonal patterns of sea ice change. NSIDC made them for each year from 1979 through 1996 from sea ice GIF images provided by Goddard Space Flight Center(GSFC). The QuickTime movies use GSFC's standard color bar that describes sea ice concentrations in four percent intervals. Ice concentrations less than 12 percent are considered open water.
The animations were created from the GIF files using the SGI utility, dmconvert. Each animation file is approximately 8.31 megabytes and can be displayed by most web browsers. QuickTime viewers which provide additional capabilities, such as the ability to step forward and backward, are available for most computers,and can be downloaded from the World Wide Web. New QuickTime movies will be added as data processing continues.
For more information please see the Sea Ice Concentrations from Nimbus-7 SMMR and DMSP SSM/I Passive Microwave Data Help Page.
The Moderate Resolution Imaging Spectroradiometer (MODIS) Instrument Science Team has officially endorsed the production and distribution of the MODIS snow and ice products in a polar grid in order to meet the needs of the cryospheric user community. The MODIS Team has specified the NSIDC EASE-Grid polar projections for this purpose. The polar grids would supplement the MODIS Integerized Sinusoidal Grid which is the standard format for MODIS products.
NASA's Earth Science Data and Information System (ESDIS) Project must approve the additional production hardware and network resources needed at NSIDC to implement the polar grids. The MODIS Team has provided an assessment of the needed resources, but a decision by ESDIS Project has not been made. Preliminary analysis indicates that planned storage capacity at NSIDC is probably adequate. Science software development to support the polar grids is already underway by the snow and ice product developers and the MODIS Science Data Support Team. For more information, contact NSIDC User Services.
This is the first in a series of articles discussing aspects of NSIDC's World Wide Web site and the resources available there.
NSIDC's Snow on the Web page provides links to a wide variety of snow and ice sites, covering a broad range of snow topics. General information about snow includes a glossary of snow terms, frequently asked questions, and links to snow crystal information. Snow and winter storm safety sites include information about surviving severe winter weather as well as links to emergency assistance agencies. An avalanche site provides links to a wide variety of professional and informational pages, providing information about avalanches, safety, equipment, and mountain weather updates. Ski and snowboard sites links to a single commercial site where you can search for general information, such as ski area snow reports, ski resorts, and more. El Niño and snow sites links to NOAA's Climate Diagnostics Center which has general information about this climate phenomenon.
One of the most recent additions to the Snow on the Web page is NSIDC's new Avalanche Awareness page that went online in January 1998, supplementing our other educational publications online. This new page provides general avalanche information and promotes understanding of the how, why, and where of avalanches (with the help of local sources of expertise, the Colorado Avalanche Information Center and NSIDC snow expert Richard Armstrong). Recognizing the conditions that cause avalanches can help people avoid them. Avalanche safety measures, explanations of rescue equipment, survival tips from experts, further avalanche resources in print, and links to sources on the web are included.
Artifically released powder snow avalanche, north of Silverton, Colorado. Photo: R.L. Armstrong, NSIDC
The international effort to create an electronic web-based directory of scientific data collected in the Antarctic is continuing.
NSF's Office of Polar Programs (OPP) is participating in the Antarctic Master Directory (AMD) and has funded the National Antarctic Data Coordination Center (NADCC) at NSIDC to coordinate U.S. entries. For more information please contact NSIDC User Services.
In 1993 the U.S. Navy awarded the scientific community the use of its Sturgeon class submarines as a research support platform to acquire information about the Arctic Ocean and its ice pack in order to increase our fundamental understanding of processes in the Arctic Ocean. This program, Scientific Ice Expeditions (SCICEX), focuses on six broad scientific categories: air-ice-ocean interaction, ocean circulation, ocean biology and chemistry, ice-acoustic interaction, ice electromagnetic interaction, and marine geology and geophysics. The ARCSS Data Coordination Center at NSIDC has been asked to be the long-term archive for data and information acquired from this project. As the submarines track through the Arctic, scientists gather water samples, temperature or sound velocity profiles taken by expendable probes, Conductivity, Temperature, Depth (CTD) profiles taken by expendable probes, CTD data mounted on the submarine sail, bathymetry recorded by installed fathometers, ice profile data from upward-looking narrow-beam sonar, ice image data from upward-looking video and sidescan sonar, water sampling from ships' seawater system, and navigation data at a non-classified level. The project runs through 1999.
Matthew Cross and Chris McNeave met with SCICEX investigators at the American Geophysical Union Oceans Meeting in San Diego during the second week of February to coordinate data management issues. For further information please contact NSIDC User Services.
The Greenland Ice Sheet Project 2 (GISP2) and the Greenland Ice Core Project (GRIP) data are now online and available to the scientific community. The collection represents all data archived from the two campaigns, as well as information on the drilling procedures, core analysis and scientific results. To get access to this landmark data collection, go to ARCSS.
The Greenland Summit Ice Cores CD-ROM is currently in production. More information about its release will be given in the next issue of NSIDC Notes. If you have questions please contact NSIDC User Services.