Data Set ID:
NSIDC-0714

MEaSUREs Greenland Ice Mapping Project (GIMP) Land Ice and Ocean Classification Mask, Version 1

This data set, part of the NASA Making Earth System Data Records for Use in Research Environments (MEaSUREs) program, provides a complete land ice and ocean classification mask for the Greenland ice sheet, that was mapped using a combination of USGS Landsat 7 ETM+ panchromatic band imagery, and the Canadian Space Agency's (CSA) RADARSAT-1 Synthetic Aperture Radar (SAR) amplitude images.

See Greenland Ice Mapping Project (GIMP) for related data.

This is the most recent version of these data.

Version Summary:

Initial release

STANDARD Level of Service

Data: Data integrity and usability verified

Documentation: Key metadata and user guide available

User Support: Assistance with data access and usage; guidance on use of data in tools

See All Level of Service Details

Parameter(s):
  • LAND ICE/OCEAN CLASSIFICATION
Data Format(s):
  • GeoTIFF
Spatial Coverage:
N: 82, 
S: 60, 
E: -20, 
W: -70
Platform(s):LANDSAT-7, RADARSAT-1
Spatial Resolution:
  • 15 m x 15 m
  • 30 m x 30 m
  • 90 m x 90 m
Sensor(s):ETM+, SAR
Temporal Coverage:
  • 1 July 1999 to 30 September 2001
Version(s):V1
Temporal ResolutionNot applicableMetadata XML:View Metadata Record
Data Contributor(s):Ian Howat

Geographic Coverage

Other Access Options

Other Access Options

Close

As a condition of using these data, you must cite the use of this data set using the following citation. For more information, see our Use and Copyright Web page.

Howat, I. 2017. MEaSUREs Greenland Ice Mapping Project (GIMP) Land Ice and Ocean Classification Mask, Version 1. [Indicate subset used]. Boulder, Colorado USA. NASA National Snow and Ice Data Center Distributed Active Archive Center. doi: https://doi.org/10.5067/B8X58MQBFUPA. [Date Accessed].

Literature Citation

As a condition of using these data, we request that you acknowledge the author(s) of this data set by referencing the following peer-reviewed publication.

  • Howat, I., A. Negrete, and B. Smith. 2014. The Greenland Ice Mapping Project (GIMP) land classification and surface elevation data sets , The Cryosphere. 8. 1509-1518. https://doi.org/10.5194/tc-8-1509-2014

Created: 
24 April 2020
Last modified: 
30 April 2020

Data Description

This data set provides complete land ice and ocean classification masks posted at 15-m, 30-m and 90-m for the Greenland ice sheet. Ice cover was mapped using a combination of orthorectified panchromatic (band-8) imagery from the Landsat 7 Enhanced Thematic Mapper Plus (ETM+), distributed by the US Geological Society (USGS), and the Canadian Space Agency's (CSA) RADARSAT-1 Synthetic Aperture Radar (SAR) amplitude images. The Landsat imagery was acquired for the months of July through September in 1999, 2000 and 2001 (mostly 2000) and the RADARSAT imagery was acquired in fall of 2000.

Parameters

The parameter provided is Land Ice/Ocean Classification. Depending upon the specific mask, the data values (0 or 1) indicate glacier ice or ocean and all other terrain.

File Information

Format

Data are available in GeoTIFF (.tif) format.
The 15 and 30 m resolution masks are distributed as sets of 36 tiles, as shown in Figure 1. Additionally, a complete 90 m resolution ice-sheet-wide mask is also provided, and is comprised of the following: a grounded ice mask, an ice mask, and an ocean mask.

Naming Convention

The following tables explain the file naming convention for the mosaics included in this data set. See Table 1 for an explanation of the variables for file names for the 15 m and 30 m files and Table 2 for the 90 m files.

For the 15 and 30 m resolution masks, there are 72 files for each:  one GeoTIFF for the land ice mask and one GeoTIFF for the ocean mask for each of the 36 tiles. They are named according to the convention below:

Gimp[Ice/Ocean]Mask_[RR]m_tile[C_R]_v1.1.ext

Example File Names for 15 m and 30 m for tile 0_5:

  • GimpIceMask_15m_tile0_5_v1.1.tif
  • GimpOceanMask_15m_tile0_5_v1.1.tif
  • GimpIceMask_30m_tile0_5_v1.1.tif
  • GimpOceanMask_30m_tile0_5_v1.1.tif
String Description
Table 1. Naming Convention Description for 15 m and 30 m files
Gimp Greenland Ice Mapping Project
Ice/Ocean IceMask indicates glacier ice, OceanMask indicates ocean and all other terrain
RR Resolution (15m or 30m)
C_R Column and row of mosaic tile. Tile 0_0 is in the lower left corner, Tile 5_5 is in the upper right (see Figure 1)
v1.1 version 1.1
.tif File name extension for GeoTIFF-formatted file

For the 90 m files, the naming convention is as follows:

Gimp[GroundedIce/Ice/Ocean]Mask_[RR]m_tile[C_R].ext

File Names for 90 m:

  • GimpGroundedIceMask_90m.tif
  • GimpIceMask_90m.tif
  • GimpOceanMask_90m.tif
Table 2. Naming Convention Description for 90 m files
String Description
Gimp Greenland Ice Mapping Project
GroundedIce/Ice/Ocean GroundedIceMask indicates grounded glacier ice; IceMask indicates glacier ice, and OceanMask indicates ocean and all other terrain
RR Resolution (90m)
v1.1 version 1.1
.tif File name extension for GeoTIFF-formatted file

Spatial Information

Coverage

Southernmost Latitude: 60° N
Northernmost Latitude: 82° N
Easternmost Longitude: 20° W
Westernmost Longitude: 70° W

Resolution

The data are available in 15 m, 30 m, or 90 m resolutions.

Projection and Grid Description

Data are provided in standard NSIDC polar stereographic north grid (EPSG 3413) centered on Greenland.
The 15 m and 30 m masks are provided as sets of 36 tiles of 6 rows by 6 columns, numbered from 0-0 in the lower left corner to 5-5 in the upper right (see Figure 1). Each tile has dimensions of 249.3 by 450 km. These dimensions were selected because they are divisible by 15 m, which is the resolution of USGS's Landsat-7 Enhanced Thematic Mapper Plus (ETM+) panchromatic (band-8) imagery.

Figure 1. GIMP Landsat-7 ETM+ band-8 and RADARSAT-1 ice mask (90 m resolution) of Greenland with tile boundaries and tile numbers overlain. The map projection is polar stereographic (EPSG 3413).

Temporal Information

Coverage

These data cover the period from 1999 through 2001. See Data Acquisition Methods below for details.

Data Acquisition and Processing

Background

Land classification masks provide a method of co-registration of repeat imagery and elevation data to track changes of the ice surface, which can change with time, while areas of exposed bedrock provide a control for measuring the changes. In addition, masks provide an accurate delineation of ice boundaries, which become a benchmark for measuring future ice margin changes. Landsat-7 ETM+ data are commonly used for mapping snow and ice.

Instrumentation

This data set was produced from images acquired by the USGS's Landsat 7 Enhanced Thematic Mapper Plus (ETM+) and CSA's Synthetic Aperture Radar (SAR) from the RADARSAT-1 satellite.

Data Acquisition Methods

The mask files included in this data set were derived from the GIMP 2000 Image Mosaic, created using Landsat 7 ETM+ and RADARSAT-1 SAR imagery. The Landsat imagery was acquired for the months of July through September in 1999, 2000 and 2001 (mostly 2000) and the RADARSAT imagery was acquired in fall of 2000.

Processing Steps

The mask files included in this data set were created from manual digitization of the panchromatic and pan-sharpened multispectral GIMP 2000 Image Mosaic. Using manual methods, the ice margin can be difficult to locate visually in areas of abundant debris and snow cover. Margins of debris-covered ice were identified by breaks in surface slope, emerging melt water streams, color differences and the presence of small melt water ponds typical of debris-covered glaciers. Similarly, glaciers were differentiated from perennial snowfields by visible crevassing, surface moraines and the existence of a visible toe. Snowfields without these features were not classified as glaciers. Using the same method, the coastline is digitized to produce an ocean mask, with the null of the ice and ocean masks being ice-free terrain (Howat, et al., 2014). Please see Howat et al., 2014 for more information regarding processing methods.

Error Estimates

Uncertainty in these classification masks arise from three sources of error: (1) image pixel resolution, (2) image geo-registration and (3) erroneous selection or non-selection of pixels (i.e., mapping error). All error sources are expected to vary randomly in space. However, there is likely a systematic component of error source (2) over distances equivalent to the size of a single image (e.g., 185 km for Landsat 7). This is due to errors in the registration model used to orthorectify the image, which typically is ±5 m, or one third of a pixel for L1T-processed imagery.

Error source (1) contributes a random error of one pixel for each ice boundary pixel. The position of any point of the ice margin has an uncertainty of 21 m while the total error for a given area of ice is then (8N)1/2x2, where N is the number of boundary pixels and x is the pixel posting in meters. Erroneous selection or non-selection of pixels (error source 3) can be due to debris cover, shadows and misidentification by the operator, as well as the ambiguity of delineating an ice boundary at glacier fronts ending in packs of icebergs. Without ground control, delineation of the ice edge in areas of debris cover, terminal moraines and persistent snow cover is subjective. These errors are difficult to quantify. Due to ambiguity in the ice edge and operator error, estimations were done by comparing mappings performed by three different operators over the same area. On average, each operator identified 24.21 km (1614 pixels) of ice margin over the common area, with a 660 m (44 pixels) difference between the maximum and minimum mappings, giving an estimated error of ±3 percent. This error, however, is expected to vary widely by particular location and size of area considered (Howat et al., 2014).

Quality Assessment

Please see Howat et al., 2014 for more information regarding the quality and processing methods used to produce this data set.

Software and Tools

A variety of Geographical Information System (GIS) software packages will work with GeoTIFF files, including ArcGISENVIGDAL, and QGIS.

Version History

Version 1.1 was released April 2017. Refer to Table 3 for this data set's version history:

Table 3. Version History
Version Description
V1.1 This version corrects the location of the upper left pixel in each GeoTIFF file. The previous version incorrectly specified the location of the upper left pixel as the distance in meters from the north pole to the center of the upper left pixel. The corrected location of the upper left pixel is the distance in meters from the north pole to the upper left corner of the upper left pixel. This correction effectively shifted the location by one-half pixel to the upper left relative to their positions in the previous version. Only the geolocation metadata in each GeoTIFF has changed; the data array is the same as in the previous version.
V1 Initial version (February 2017)

Related Data Sets

Related Websites

Contacts and Acknowledgments

Ian Howat
Ohio State University
Byrd Polar Research Center

These data were generated through a grant from the NASA Making Earth System Data Records for Use in Research Environments MEaSUREs Program.

References

Howat, I. M., A. Negrete, and B. E. Smith. 2014. The Greenland Ice Mapping Project (GIMP) land classification and surface elevation datasets. The Cryosphere8(4): 1509-1518. doi: 10.5194/tc-8-1509-2014.
Joughin, I., B. E. Smith, I. M. Howat, T. Scambos, and T. Moon. 2010. Greenland flow variability from ice-sheet-wide velocity map- ping. Journal of Glaciology56: 415–430.

How To

Programmatic Data Access Guide
Data from the NASA National Snow and Ice Data Center Distributed Active Archive Center (NSIDC DAAC) can be accessed directly from our HTTPS file system or through our Application Programming Interface (API). Our API offers you the ability to order data using specific temporal and spatial filters... read more
Filter and order from a data set web page
Many NSIDC data set web pages provide the ability to search and filter data with spatial and temporal contstraints using a map-based interface. This article outlines how to order NSIDC DAAC data using advanced searching and filtering.  Step 1: Go to a data set web page This article will use the... read more
How do I reproject a geoTIFF from polar stereographic to geographic lat/lon?
We recommend using the Geospatial Data Abstraction Library (GDAL) or GIS to reproject geoTIFF files. Here we outline command line or python options for using GDAL, and instructions for QGIS and ArcMap to... read more
How do I convert a GeoTIFF into a NetCDF file?
We recommend using the Geospatial Data Abstraction Library (GDAL) to convert GeoTIFF files into a different format. If you want to reproject to lat/lon as well, then we recommend reprojecting before converting to NetCDF (see the FAQ "How do I reproject a GeoTIFF to from polar steroegraphic to... read more

FAQ

Do you have sample code to read GeoTiffs?
Sample image ouput from Python code. Below you will find  sample code in IDL, MATLAB, and Python to read in a GeoTIFF file, extract the metadata, and create an image. The code has been tested with... read more