In this Issue



Issue 18, Summer 1996


Antarctic Data Coordination Center Funded at NSIDC

NSIDC has received unofficial notification from the National Science Foundation's Office of Polar Programs that it intends to fund NSIDC's proposal to function as the U.S. Antarctic Data Coordination Center. The activities of the Antarctic Data Coordination Center will comprise identifying existing Antarctic data sets in all disciplines, locating points of contact and preparing data descriptions (DIFs) for these data, and coordinating submittal of the descriptions to the Antarctic Master Directory (AMD) at the International Centre for Antarctic Information and Research (ICAIR) in New Zealand, and the Global Change Master Directory (GCMD) in the U.S. Data descriptions would also be obtained from researchers as they prepare for or return from field work in Antarctica.

NSIDC developed the proposal for this effort based on the involvement of R.G. Barry and C.S. Hanson in the Scientific Committee for Antarctic Research (SCAR) - Council of Managers of National Antarctic Programs (COMNAP) and its ad hoc Planning Group on Antarctic Data Management, and after similar efforts for the Arctic (e.g., the Arctic Environmental Data Directory). NSIDC has requested three years of funding at approximately 1.0 FTE for the effort.

For further information contact: Greg Scharfen at (e-mail), or (303) 492-6197 (phone), or (303) 492-2468 (fax).

New Landsat Project to Map Ice Velocities of West Antarctica

The potential collapse of the West Antarctic ice sheet and the resulting sea level change that would ensue represents a major environmental threat. The result of this collapse would be an increase in current sea level by approximately six meters; some models predict that the rate of change could be rapid, up to ten times faster than the current estimated rate of change. Over the past 100 years the rate of rise in sea level has been relatively stable with an average growth rate of approximately 2 mm/yr. Even this relatively slow rate has caused hardships on coastal economies as well as wetlands areas.

Recently a grant proposal submitted by scientists at Goddard Space Flight Center, the University of Maryland, USGS, and the Cryospheric and Polar Processes Division /Cooperative Institute for Research in Environmental Sciences, University of Colorado (CPP/CIRES) was funded to map changes in the Antarctic ice sheet using Landsat imagery. This study is part of the continuing efforts of NASA's Mission to Planet Earth to understand and predict global climate change. The project will accomplish three objectives over the course of three years. The first objective will be to assess existing Antarctic ground control point data and produce sub-scenes of suitable control points for image geolocation. Included in this objective are plans to enhance and extend the ground control point library by "bridging" path-sequential images together over areas with no ground control. The second objective will be to generate an interpretation of the dynamic history of the potentially unstable West Antarctic ice sheet by mapping features on the Ross Ice Shelf. Flow- generated features survive for centuries within the ice shelf, therefore a 1000-year record of ice flow variations may be found in the Ross Ice Shelf which can then be used to determine other details of ice-sheet history. The final objective of this project will be to map ice outflow velocities of glaciers along the West Antarctic coast as far inland as possible, and look for changes in outflow velocity over the Landsat era. This last objective will be under the direction of Dr. Ted Scambos at CPP/CIRES and NSIDC.

The steps involved in the work to be accomplished at CIRES will begin with gathering available Landsat MSS and TM images and setting up a catalog of the best images for this research. The next task will be to use the images collected to map the outflow velocities using a cross-correlation feature-matching algorithm. This technique uses small surface features within an image as markers, then searches for similar features on a subsequent image and calculates the correlation index between the two. The displacement is then measured to subpixel accuracy by interpolating on the peak correlation values. Dividing the displacement by the time interval between the images gives the velocity. These measurements can then be used to produce an accurate mapping of the surface velocity field for that specific area. Accurate strain rate fields can also be mapped in areas with good velocity data. Included in the funded work are plans to enhance the capability and portability of the software. Subject to approval by NSIDC's User Working Group, this project plans to make available the processed images and ice velocity data for the West Antarctic coast. In addition, NSIDC plans to archive and distribute the expanded ground-control-point library (in conjunction with USGS).

Overall the goal of this project is to produce data that can be used as a benchmark for future studies concerning the stability of the West Antarctic ice sheet.

For further information concerning this project, please contact: Ted Scambos (


New CD-ROM: Russian Arctic Ocean Drift Station Data

NSIDC, in collaboration with the Polar Science Center (PSC), University of Washington, has just released a new CD-ROM containing snow, meteorological, and solar radiation data from the Soviet North Pole drifting stations. Arctic Ocean Snow and Meteorological Observations from Drifting Stations, 1937, 1950-1991, Version 1.0 contains: snow depth, snow surface temperature, snow density, snow water equivalent; meteorological data (air temperature, relative humidity, surface pressure, wind speed and direction, U-V components of wind, total and low cloud amounts, precipitation type and amounts); and solar radiation data (diffuse, direct, global, reflected, and net radiation, and albedo.) North Pole (NP)-1 operated from 22 May 1937 through 19 Feburary 1938. NP-2 began on 22 April 1950, and at least one station, sometimes two, operated from that date until the end of NP-31 on 31 March 1991.

Figure 1 represents the daily positions of the Soviet North Pole stations 3 through 31, which were manned continuously from 1954 to 1991.

Data on the CD-ROM are daily averages, or 3-hourly measured values; spatial resolution is dependent on the movement of the ice floes on which the stations were located. Both original data, as digitized from station logbooks at the Arctic and Antarctic Research Institute, (AARI), St. Petersburg, Russia, and daily averaged data are included. Position data were linearly interpolated from the original irregular position observations.

Meteorological instrument descriptions, including technical specifications regarding accuracy at low temperatures, are included in the documentation, which is provided on the CD-ROM. All the data are in ASCII files and are readable on any type of computer platform.

Rescue and organization of the meteorological data at AARI and PSC/University of Washington, and the CD-ROM development portion of this project, were funded by the U.S. National Oceanic and Atmospheric Administration (NOAA) as a part of the NOAA National Environmental Satellite, Data and Information Service (NESDIS) Environmental Science Data and Information Management (ESDIM) Program. Additional funding for snow data rescue and preparation was provided by the U.S. National Aeronautics and Space Administration (NASA). Funding for the rescue, publication and access to the solar radiation data included on this CD-ROM was provided by the U.S. National Science Foundation, Division of Atmospheric Sciences. This CD-ROM is an Arctic Climate System (ACSYS) contribution.

The CD-ROM is available at no charge. For further information or to request the CD-ROM, contact us at:

NSIDC User Services
National Snow and Ice Data Center
University of Colorado
Boulder, CO 80309-0449  USA
phone: +1 303.492.6199
fax: +1 303.492.2468
form: Contact NSIDC User Services

SSM/I F11 Brightness Temperatures Volume 16

The final CD-ROM volume in the SSM/I F11 brightness temperatures series is now complete, and was shipped early in September to all subscribers. The time period covered is only one month. We elected to finish out the F11 SSM/I series with volume 16 even though there are not many data files, and to begin a new series for the F13 SSM/I. Let us know if you have any questions or if you did not receive volume 16 as expected.

Correction: Program LOCATE FOR for SSM/I Grids

There is a typographical error in a subroutine of the FORTRAN program 'locate.for' which is used with the DMSP SSM/I Brightness Temperatures and Sea Ice Concentration Grids. The error will not affect the output of values when the program is used with these data sets, but it may make a difference if the program is used with data in other projections.

The program 'locate.for' was written for use with the SSM/I brightness temperatures and sea ice concentration data sets, and is used to transform X,Y coordinates of the SSM/I polar stereographic grid to latitude/longitude coordinates, and to convert geodetic latitude/longitude to X,Y coordinates. Two subroutines 'mapll.for' and 'mapxy.for' originally written by a source outside NSIDC, are used in conjunction with 'locate.for'. An equation (see below) in the most recently revised version (1990) of 'mapll.for' has a typographical error.

In the subroutine there are two equations to compute the variable 'rho' - one is applied when the parameter 'slat' (standard latitude -no distortion) is equal to 90 degrees, and the other is used when standard latitude is not 90°(as in the case of the SSM/I brightness temperatures and sea ice concentration grids where the standard latitude is 70°). The error in 'mapll.for' is in the first equation.





Because the second equation has been used for the calculation of latitude and longitude values in the brightness temperature and sea ice data sets, the incorrect equation does not apply. However, those users who may want to employ this program with other data should make note of the correct equation.

The original equation can be found in J.P. Snyder, 1982. Map Projections Used by the U.S. Geological Survey. U.S. Geological Survey Bulletin 1532. U.S. Government Printing Office, p. 162, equation 17-33.

For more information on the gridding methods used for the SSM/I data set at NSIDC, please see DMSP SSM/I Brightness Temperatures and Sea Ice Concentration Grids for the Polar Regions. Users Guide, 1990, rev. 1996, p. 31.

The latest versions of the subroutines are available via ftp.

Your Feedback is Solicited!

You may have received by mail a questionnaire regarding HDF (Hierarchical Data Format) products at NSIDC, or EASE-Grid SSM/I products at NSIDC. Or you may have seen these questionnaires on our Web site (/). We have received many responses, and we are tabulating them during September. If you have not responded and wish to do so, we will still gladly accept Web, e-mail, fax, or regular mail responses. Results of the two surveys will be reported in the Winter 1996 issue of NSIDC Notes.

Antarctic Buoy Data Planning

The International Programme for Antarctic Buoys (IPAB) held its first meeting at the Scott Polar Research Institute, Cambridge, UK, on 1-3 August 1996. This federation of scientists from about 15 countries is undertaking a focused program of sea ice motion studies using instrumented buoys deployed on ice floes (or in some cases in open water) in the Southern Ocean, as near to the coastline of Antarctica as is practicable. The study is ongoing, and is a part of the World Climate Research Programme of the World Meteorological Organization. The aim is to provide observations around the entire Antarctic continent. The U.S. contribution will be in the form of technical advice to participants from the National Ice Center, Suitland, MD, and data archiving and distribution services from WDC-A/NSIDC. We will collaborate with the Programme Coordinator, Dr. I. Allison of the Antarctic Climate Research Centre, Hobart, to help publicize the availability of data from the buoys, and to ensure data access for users outside the IPAB. C. Hanson of WDC-A/NSIDC attended the Cambridge meeting.

Global Geocryological Database

WDC-A/NSIDC has been funded by the National Science Foundation to implement a pilot "Global Geocryological Database" (GGD). This effort provides start-up funding for the assembly of priority permafrost and frozen ground datasets in Russian archives, and for WDC-A/NSIDC to inventory, retrieve, and organize priority datasets identified by other members of the International Permafrost Association (IPA).

Several members of the Data and Information Working Group (R.G. Barry, Chair) and Jerry Brown, IPA Secretary General, met at WDC-A/NSIDC in July to review the results of preliminary inventories, and to plan future data activities. In preparation for the 1998 Seventh International Permafrost Conference in Yellowknife, Northwest Territories, Canada, the group reconfirmed two planned activities: (1) a data workshop organized by Dr. M.J. Clark (Geodata Centre, Southampton, UK) to be held during the conference, and (2) preparation of a CD-ROM containing permafrost data and information.

The planned CD-ROM will be called CAPS - "Circumpolar Active-Layer Permafrost System: A Contribution to Global Change Research". IPA Adhering Members are being asked to contribute at least one long-term dataset for the CD-ROM, and individuals and IPA Working Groups are also invited to contribute important data to the project. Nominations for inclusion are needed by 1 December 1996.

In order to maintain a reasonable rate of progress, the IPA proposes to hold a series of small regional meetings over the next 6-12 months. The first, focused on North American data holdings, will take place in Denver/Boulder on 12-13 December 1996. Another is proposed for Pushchino, Russia, in late April 1997. There are numerous conferences in 1997 and we hope to use these as opportunities to review data submissions.

Further information on the GGD project is given in the IPA Newsletter Frozen Ground, no. 18, December, 1995, p. 12; also see the IPA Home Page:

Contact: R.G. Barry, WDC-A/NSIDC Director, or Claire Hanson, GGD/CAPS project at WDC-A/NSIDC, on contributing data to the GGD.

World Data Center-A for Glaciology, GGD/CAPS Project, CIRES, 449 UCB, University of Colorado, Boulder, CO 80309-0449 USA, R.G. Barry, +1 303.492.5488 (phone), or Claire Hanson, +1 303.492.1834 (phone), or +1 303.492.2468 (fax) or (e-mail).


GISP Data Acquisitions and CD-ROM

Greenland Ice Sheet Project 2 (GISP2) data continue to arrive at the ARCSS Data Coordination Center and the GISP2 Science Management Office at the University of New Hampshire. We expect all data from the GISP2 project be downloaded to NSIDC by the end of November. The data will available as soon as possible via ftp from the ARCSS site. A CD-ROM of the GISP2 and GRIP data will be released early in 1997 in conjunction with the publication of the special Journal of Geophysical Research issue devoted to the GISP2/GRIP project.

For further information, contact NSIDC User Services.

Arctic Radiation Data CD-ROM

Mark Serreze, an ARCSS-funded investigator at NSIDC, is nearing commpletion of a new Arctic downwelling short-wave radiation climatology. In conjunction with this climatology, we are currently compiling all of the NSIDC Arctic radiation data to produce a composite radiation data CD-ROM. This will include short-wave, long-wave and net radiation values from stations above 50°North as well as the new Serreze climatology.

For further information, contact NSIDC User Services.

Educational Program Development at the ARCSS Data Coordination Center

The ARCSS Data Coordination Center at NSIDC is helping bridge the gap between high-level science and secondary level (grades 9-12) studies by collaborating with Brenton Burnett, a high school teacher from Denver, Colorado, in developing high-school science curricula using ARCSS data and information. This activity provides research results to students and teachers through a WWW interface and CD-ROM publication. The release of Greenland Ice Sheet Project 2 (GISP2) data in the near future provides a timely database for the initial curriculum development, with expansion to other ARCSS data in the future.

This innovative way to bring high level research results to the 9-12 community is summed up by Mr. Burnett:

We want to get students as close to real science as possible. Educational resources on the Internet could provide an ideal source for broad-scale communication between scientists and high school students. However, today many sites still use a 1950s book model of teaching instead of an interactive activity and thinking-based model which is the pedagogy of the nineties. Other sites succumb to the temptation of technoflash and lack substance and quality pedagogical grounding. Our ARCSS education mission is to provide current scientific results to students so they can manipulate, analyze and relate the science to their own lives.

This project is an important element in our long-range outreach to all communities interested in ARCSS and NSIDC data, information and products. All inquiries can be directed to Dr. David McGinnis at (303) 492-1160, (phone) or (303) 492-2468 (fax) or (e-mail).


Investigating Mixed Errors in SSM/I Data Processed with the AES/York Sea Ice Concentration Algorithm

Eight years of Special Sensor Microwave Imager (SSM/I) data were analyzed to investigate the interference which is inherent in data processed with the Atmospheric Environment Service (AES)/York sea ice concentration algorithm. By viewing the 1987-1994 SSM/I Northern Hemisphere images, significant mixel errors (pixels with a mixture of land and water, resulting in a mixed signature value) are observed along the coasts. The SSM/I daily averaged CD-ROM product distributed by NSIDC was used in conjunction with the AES/York sea ice algorithm to produce the images analyzed. The AES/York algorithm uses a linear combination of brightness temperatures from the 37 and 19 GHz channels which produces an estimation of total ice concentration and edge location. As only the CD-ROM raw brightness temperature data were used in this investigation, an assumption is necessary to conduct this analysis: the region within the study areas represents completely ice free conditions in the summer images and therefore, all sea ice concentration values in the images are treated as mixel errors.

Monthly averaged sea ice concentration images from 1987 to 1994 were used to investigate mixel error variations found in the images. The month of September typically represents the minimum Arctic sea ice concentration time period, and therefore the September images were used exclusively in this analysis, with the exception of 1994. By means of visual inspection, systematic interference was identified in the September 1994 image and consequently, the October image was substituted.

Two regional study areas were selected for detailed analysis: Hudson Bay and Scandinavia (Figure 2). These two areas were chosen due to the high intervention of mixel errors and the varying shoreline types. The Hudson Bay study area presents a more distinct boundary between land and water, whereas the Scandinavian coastline is highly complex with many small inlets and waterways (fjords). Although different, a string of pixels outlines both coastlines with anomalous sea ice concentration values. This phenomenon is most evident in the late summer/early fall images, when the sea ice extent is at a minimum. Only summer images were investigated for mixel errors, however a certain degree of interference may still occur in the winter months, masked by the appearance of sea ice. Studies should be conducted to identify whether the same effects occur with the land/sea ice boundary as with the land/water boundary. Detailed research in Hudson Bay and Scandinavia reveals that after the shoreline has been dilated the mixel errors in the images are reduced. By identifying mixel errors in the sea ice concentration images, correction procedures can be developed to improve data accuracy.

The Hudson Bay study area is located within the seasonal sea ice zone. This area produces a high proportion of mixels, yet includes a large body of water after the shoreline has been dilated the mixel errors in which contains valid sea ice concentration values in the winter months. The northern region of the Scandinavian study area is within the seasonal sea ice zone, while the southern region is free of sea ice year around. The Scandinavian study area includes part of the Atlantic Ocean coast, various inlets and gulfs, resulting in an irregular shoreline. As a result of this complex coastline, a large number of mixels occur in this region. Basic statistical measures were calculated for both study regions, however the percentage of mixel error variable describes the overall trends of the analysis (Table 1).

Table 1. Percentage of mixel errors with original land mask applied.
Year Hudson Bay Scandinavia

This calculation was derived by determining the number of pixels containing false sea ice concentration values within the study areas, then dividing by the total number of pixels representing ocean.

The original land mask distributed by NSIDC, along with the SSM/I data, was dilated to investigate the effect of mixels concentrated along the land/water interface. National Institutes of Health Image 1.51 software was used to dilate the binary land mask image; by increasing the shoreline one pixel in width, the percentage of mixel errors is significantly reduced (Table 2). The average percentage of mixel errors reduced in the Scandinavian study area was 71.8%, and 77.8% in the Hudson Bay study area.

Table 2. Percentage of mixel errors reduced with dilated land mask.
YearHudson BayScandinavia

These results reflect the regional differences in the shoreline and spatial distribution of mixel errors. The majority of mixels in the Hudson Bay study area occur at the land/water boundary; whereas in the Scandinavian study area, the mixels extend into the small waterways along the coast. Therefore, it is expected that the percentage reduction in mixels would be less in the Scandinavian than the Hudson Bay region. Results from both areas show that significant reductions in mixel errors can be obtained when eliminating values within one pixel of the shoreline. Within this investigation, an overall average of 74.8% of the mixel interference is reduced from the SSM/I sea ice concentration images when the dilated land mask was applied. For presentation or statistical purposes, the sea ice concentration images must be manipulated to reduce mixel interference. The results of this study suggest that any investigations conducted with the SSM/I sea ice concentration images should discount values within one pixel of the coast for any comparative studies.

For further information contact: Jennifer Sokol, 219 Strathcona, Ottawa, Ontario, Canada K1S 1X7, (613) 563-0738, (phone) or Ellsworth LeDrew, University of Waterloo, (e-mail), (519) 888-4567 (phone).

NOTE: The recent article, Maslanik, J.A., M.C. Serreze and R.G. Barry, 1996. Recent decreases in Arctic summer ice cover and linkages to atmospheric circulation anomalies. Geophysical Research Letters, 23(13), p.1,677-1,680, used an expanded land mask to minimize the effect of different magnitudes of land contamination between SMMR and SSM/I.

NSIDC is currently investigating methods of identifying and masking land-contaminated pixels.