This data set, part of the NASA Making Earth System Data Records for Use in Research Environments (MEaSUREs) program, is an improved, enhanced-resolution, gridded passive microwave Earth System Data Record (ESDR) for monitoring cryospheric and hydrologic time series from SMMR, SSM/I-SSMIS and AMSR-E. The Calibrated Passive Microwave Daily EASE-Grid 2.0 (CETB) gridded data use the most mature available Level 2 satellite passive microwave records from 1978 to the present.
Get Data
DOWNLOADING DATA VIA HTTPS
To learn more about Earthdata Login and register for an account, please see How to Register with Earthdata Login. Once you have logged in, data can be downloaded via a Web browser, command line, or client. For help with downloading data, please see Options Available for Bulk Downloading Data from HTTPS with Earthdata Login.
NOTE: Reverb will be decommissioned in the coming months and replaced with Earthdata Search. All links to Reverb will be removed at that time.
MEaSUREs Calibrated Enhanced-Resolution Passive Microwave Daily EASE-Grid 2.0 Brightness Temperature ESDR, Version 1
Geographic Coverage
Spatial Coverage: |
|
---|---|
Spatial Resolution: |
|
Temporal Coverage: |
|
Temporal Resolution: | 12 hour |
Parameter(s): |
|
Platform(s) | AQUA, DMSP 5D-2/F10, DMSP 5D-2/F11, DMSP 5D-2/F13, DMSP 5D-2/F14, DMSP 5D-2/F15, DMSP 5D-2/F16, DMSP 5D-2/F8, DMSP 5D-3/F17, DMSP 5D-3/F18, NIMBUS-7 |
Sensor(s): | AMSR-E, SMMR, SSM/I, SSMIS |
Data Format(s): |
|
Version: | V1 |
Data Contributor(s): | Mary Brodzik, David Long, Molly Hardman, Aaron Paget, Richard Armstrong |
Metadata XML: | View Metadata Record |
Data Citation
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.
Brodzik, M. J., D. G. Long, M. A. Hardman, A. Paget, and R. Armstrong. 2016. MEaSUREs Calibrated Enhanced-Resolution Passive Microwave Daily EASE-Grid 2.0 Brightness Temperature ESDR, Version 1. [Indicate subset used]. Boulder, Colorado USA. NASA National Snow and Ice Data Center Distributed Active Archive Center. doi: http://dx.doi.org/10.5067/MEASURES/CRYOSPHERE/NSIDC-0630.001. [Date Accessed].Detailed Data Description
The Calibrated Enhanced Resolution Brightness Temperature (CETB) data set consists of gridded passive microwave brightness temperature data from the following instruments:
- Scanning Multichannel Microwave Radiometer (SMMR) on NIMBUS-7
- Special Sensor Microwave/Imager (SSM/I) on DMSP 5D-2/F8, F10, F11, F13, F14, F15
- Special Sensor Microwave Imager/Sounder (SSMIS) on DMSP 5D-2/F16, F17, F18
- Advanced Microwave Scanning Radiometer - Earth Observing System (AMSR-E) on AQUA
The data are a new, multi-sensor Level 3 Earth Science Data Record (ESDR) with recently released improvements in cross-sensor calibration and quality checking, modern file formats, better quality control, improved projection grids, and local time-of-day (ltod) processing. These data are gridded to the EASE-Grid 2.0 definition and include enhanced-resolution imagery, as well as coarse-resolution, averaged imagery.
Note: The data are being released in stages starting with AMSR-E in Fall 2016, followed by SSM/I and SSMIS data, with SMMR data released in 2017.
The data are in NetCDF (.nc) format, using CF 1.6 (Climate and Forecast) and ACDD 1.3 (Attribute Conventions for Dataset Discovery) metadata conventions.
Following is the file naming convention for this data set (Table 1) with an example file name.
Example: NSIDC-0630-EASE2_N3.125km-F08_SSMI-1987304-37H-M-SIR-RSS-v1.0.nc
NSIDC-0630-EASE2_XXXXXkm-platform_sensor-yyyyddd-channel-pass-algorithm-input-version.nc
Variable | Description | Values |
---|---|---|
NSIDC-0630 | NSIDC unique data set identifier | NSIDC-0630 |
EASE2_XXXXXkm | EASE2-Grid 2.0 projection and grid resolution of data in the file |
Name = EASE2 |
platform_sensor | Satellite platform id and sensor | NIMBUS7_SMMR F08_SSMI F10_SSMI F11_SSMI F13_SSMI F14_SSMI F15_SSMI AQUA_AMSRE F16_SSMIS F17_SSMIS F18_SSMIS F19_SSMIS |
yyyyddd | Reference day | 4-digit year, 3-digit day of year |
channel | Channel ID (frequency and polarization - horizontal (H) or vertical (V)) | 2-digit frequency and 1-letter polarization, differs by sensor, e.g. 37H |
pass | The direction or local time of day of the satellite passes used | 1-letter code: A = Ascending (T grids only) D = Descending (T grids only) M = Morning ltod (N or S grids only) E = Evening ltod (N or S grids only) |
algorithm | Specifies the algorithm used for the image reconstruction | GRD = drop-in-the-bucket (25 km grids) SIR = radiometer version of Scatterometer Image Reconstruction (enhanced-resolution grids) |
input | Input data producer | CSU = Colorado State University RSS = Remote Sensing Systems |
version | Data set version number | vX.X for major/minor versions |
nc | NetCDF data formatting suffix | .nc |
The daily AMSRE file sizes range from 1 MB to 90 MB, with a full day of AMSR-E data ~4 GB. The variations in file sizes are a result of HDF5 internal compression and the different resolutions of the files.
The total volume for AMSR-E data is 6.8 TB. Volumes for remaining sensors will be included as the data become available.
Each data file contains data on one of three EASE-Grid 2.0 spatial coverages: Northern Hemisphere Lambert azimuthal equal-area, Southern Hemisphere Lambert azimuthal, and equal-area or cylindrical equal-area projection (bounded by +/- 67° latitude).
Spatial Resolution
Each channel is processed at conventional (25 km) and enhanced resolutions depending on frequency. The coarsest grid resolution is 25 km, with enhanced-resolution grids defined in a nested fashion in powers of 2, at 12.5, 6.25 and 3.125 km (see Figure 1.) (Brodzik and Long, 2016). The expected level of resolution enhancement for the CETB products is channel-dependent, at best 3.125 km (Long and Brodzik, 2016).

Projection and Grid Description
The data are gridded to EASE-Grid 2.0 projections, at various coverages and spatial resolutions as defined in Table 2. All channels are gridded to 25 km and the higher resolutions depend on channel frequency as follows: frequencies below 12 GHz are at 12.5 km, between 12 and 30 GHz, the resolution is 6.25 km, and above 30 GHz, the enhanced-resolution is at 3.125 km.
Name | Projection | Resolution (km) | Columns | Rows |
Latitude Extent (degrees) |
Longitude Extent (degrees) |
---|---|---|---|---|---|---|
EASE2_N25km | Northern Lambert Azimuthal | 25 | 720 | 720 | 0 - 90 | -180 - 180 |
EASE2_N12.5km | Northern Lambert Azimuthal | 12.5 | 1440 | 1440 | 0 - 90 | -180 - 180 |
EASE2_N6.25km | Northern Lambert Azimuthal | 6.25 | 2880 | 2880 | 0 - 90 | -180 - 180 |
EASE2_N3.125km | Northern Lambert Azimuthal | 3.125 | 5760 | 5760 | 0 - 90 | -180 - 180 |
EASE2_S25km | Southern Lambert Azimuthal | 25 | 720 | 720 | -90 - 0 | -180 - 180 |
EASE2_S12.5km | Southern Lambert Azimuthal | 12.5 | 1440 | 1440 | -90 - 0 | -180 - 180 |
EASE2_S6.25km | Southern Lambert Azimuthal | 6.25 | 2880 | 2880 | -90 - 0 | -180 - 180 |
EASE2_S3.125km | Southern Lambert Azimuthal | 3.125 | 5760 | 5760 | -90 - 0 | -180 - 180 |
EASE2_T25km | Cylindrical Equal-Area | 25.02526 | 1388 | 540 | +/-67.0575406 | -180 - 180 |
EASE2_T12.5km | Cylindrical Equal-Area | 12.51263 | 2776 | 1080 | +/-67.0575406 | -180 - 180 |
EASE2_T6.25km | Cylindrical Equal-Area | 6.256315 | 5552 | 2160 | +/-67.0575406 | -180 - 180 |
EASE2_T3.125km | Cylindrical Equal-Area | 3.128.15750 | 11104 | 4320 | +/-67.0575406 | -180 - 180 |
Temporal coverage varies by sensor. See Table 3 for the actual coverages.
Sensor | Begin Coverage | End Coverage |
---|---|---|
AMSR-E | 1 June 2002 | 1 November 2011 |
SSM/I | 9 July 1987 | 19 November 2009 |
SSMIS | 1 November 2005 | to present |
SMMR | 25 October 1978 | 20 August 1987 |
Temporal Resolution
The grids are produced twice daily. T grids are separated by ascending/descending passes, and N and S grids are separated by local time of day.
The parameters for this data set are listed in Table 4.
Parameter | Description |
---|---|
TB | Brightness temperature |
TB_time | Average time of the measurements used to derive TB |
TB_std_dev | Standard deviation of the measurements used to derive TB |
TB_num_samples | Number of measurements used to derive TB |
Incidence_angle | Average incidence angle of the measurements used to derive TB |
Parameter Description
Brightness temperature depends on the emissivity and physical temperature of the observed target and varies with the frequency and polarization of the passive microwave sensors. The relationship between the measured brightness temperature and the effective physical temperature of the observed target is described by the Rayleigh-Jeans approximation of Planck's equation.
Sample Data Record
Figures 2-4 are examples of AMSR-E data from 27 September 2011. There is one image per grid (Southern, Temperate & Tropical, and Northern): two images (Southern and Northern) are at a spatial resolution of 3.125 km, and the Temperate & Tropical image is at 6.25 km.



Software and Tools
For a list of resources for accessing NetCDF files, see NetCDF Software Tools. Geolocation files and other tools for working with the data will be available shortly.
For a detailed description of the instruments used to acquire the data, see the following NSIDC web sites:
- Scanning Multi-channel Microwave Radiometer
- Special Sensor Microwave Imager (SSM/I)
- Special Sensor Microwave Imager/Sounder (SSMIS)
- AMSR-E Instrument Description
Table 6 provides a list of the channels for each instrument. All channels are gridded to 25 km and the higher resolutions depend on channel frequency. Frequencies below 12 GHz are at 12.5 km, between 12 and 30 GHz, resolution is 6.25 km, and above 30 GHz, resolution is 3.125 km.
Sensor | Channel Frequency (GHz) and Polarization |
---|---|
SMMR | 6H, 6V, 10H, 10V, 18H, 18V, 21H, 21V, 37H, 37V |
SSM/I | 19H, 19V, 22V, 37H, 37V, 85H, 85V |
SSMIS | 19H, 19V, 22V, 37H, 37V, 91H, 91V |
AMSR-E | 6H, 6V, 10.7H, 10.7V, 18H, 18V, 23H, 23V, 36H, 36V, 89H, 89V |
Data Acquisition and Processing
Table 7 lists the CETB input data sets.
Sensor | Temporal Coverage | Input Swath Data |
---|---|---|
SMMR | 1978-1987 | Nimbus-7 SMMR Pathfinder Brightness Temperatures, Version 1 (NSIDC-0036) |
SSM/I-SSMIS | 1987-present | CSU FCDR (http://rain.atmos.colostate.edu/FCDR/) |
AMSR-E | 2002-2011 | AMSR-E/Aqua L2A Global Swath Spatially-Resampled Brightness Temperatures, Version 3 (AE_L2A) |
There are two general processing steps in generating the CETB product. These include data set pre-processing for spatial and temporal selection, and gridding/reconstruction of the data. At this time, if there are no data available for a specific date, there is no file for that date.
Data set preprocessing
The first stage of processing ingests the raw swath TB and performs initial data and temporal selections. Only the highest quality TB measurements are used to ensure the most reliable data set. Swath data are mapped to output grids by measurement geolocation and local time-of-day (ltod).
Local Time-of-Day: All of the CETB passive microwave sensors fly on near-polar, sunsynchronous satellites, which maintain an orbital plane with an orientation that is (approximately) fixed with respect to the sun. Thus, the satellite crosses the equator on its ascending (northbound) path at approximately the same ltod. The resulting coverage pattern yields passes about 12 hours apart in ltod at the equator. Most areas near the pole are covered multiple times per day. Analysis shows that the data from a single sensor fall into two ltod ranges for polar measurements. The two periods are typically less than 4 hours long, spaced 8 or 12 hours apart. Significantly, due to the orbit repeat cycle, two succeeding days at any particular location may make measurements at different ltod, and therefore, at different times during the diurnal cycle (Gunn, 2007), introducing undesired variability (noise) into a time series analysis.
The CETB azimuthal (Northern or Southern) grids are split into two images per day based on the ltod approach of Gunn and Long (2008). This ensures that all measurements in any one image have consistent spatial/temporal relationships. The CETB adopts the ltod division scheme for the Northern and Southern hemispheres. In the equatorial regions of the EASE2-T grids, ltod is equivalent to division by ascending vs. descending passes,
Each file includes gridded arrays of the following variables: brightness temperature, number of contributing measurements, as well as the average time, standard deviation, and average incidence angle of contributing measurements used to derive the TB at each pixel. This enables investigators to explicitly account for the ltod temporal variation of the measurements included in a particular pixel.
Known Data Problems
When no swath measurement center locations were mapped to the area of the a gridded pixel, GRD images will occasssionally have single pixels with no data. Normally, rSIR images do not suffer from this problem, because the rSIR gain threshold is set to a value that almost always ensures at least one component measurement that can be used to derive the pixel brightness temperature. However, beginning 4 Nov 2004, the AMSR-E 89 GHz A-horn developed a permanent problem that resulted in a loss of observations for the remining life of AMSR-E. After this date, the rSIR 3.125 km 89 GHz data does occasionally have missing pixels. (Personal corresondence, NSIDC USO, Dec. 5, 2016)
Gridding/Reconstruction
CETB products are generated on coarse resolution grids for all channels using a low-noise “drop-in-the-bucket” average, and enhanced-resolution grids using rSIR (radiometer version of Scatterometer Image Reconstruction) image reconstruction techniques (Long and Brodzik, 2016). For enhanced-resolution grids, the effective resolution depends on the number of measurements and the precise details of their overlap, orientation, and spatial locations. See the Derivation Techniques and Algorithm section for more information.
Antenna Pattern and Measurement Spatial Response
For image reconstruction processing, information about the antenna gain pattern, the scan geometry, and integration period are required to compute the effective measurement response function (MRF). The MRF describes how much the emissions from a particular receive direction affect the observed TB value. For each sensor and channel, the MRF is modeled as a two-dimensional Gaussian using the 3-dB footprint size (Long and Brodzik, 2016). See Table 8 for the field-of-view values.
Sensor | Frequency (GHz) | Semi-major (km) | Semi-minor (km) |
---|---|---|---|
SSMI | 19 H, V | 69 | 43 |
SSMI | 22 V | 60 | 40 |
SSMI | 37 V | 37 | 28 |
SSMI | 37 H | 37 | 29 |
SSMI | 85 H, V | 15 | 13 |
SSMIS | 19 H, V | 72 | 44 |
SSMIS | 22 V | 72 | 44 |
SSMIS | 37 H, V | 44 | 26 |
SSMIS | 91 H, V | 15 | 9 |
AMSR-E | 6 H, V | 75 | 43 |
AMSR-E | 10.7 H, V | 51 | 29 |
AMSR-E | 18 H, V | 27 | 16 |
AMSR-E | 23 H, V | 32 | 18 |
AMSR-E | 36 H, V | 14 | 8 |
AMSR-E | 89 (H or V) | 7 | 4 |
AMSR-E | 89 (H or V) | 6 | 4 |
SMMR | 6.6 H, V | 121 | 79 |
SMMR | 10.7 H, V | 74 | 49 |
SMMR | 18 H, V | 44 | 29 |
SMMR | 21 H, V | 38 | 24 |
SMMR | 37 H, V | 21 | 14 |
The following sections describe CETB gridding algorithms. Please refer to Long and Brodzik (2016) for the theory of reconstruction techniques and complete details of rSIR. Figure 5 provides a graphical representation of the enhanced resolution of rSIR measurements.

Coarse Resolution (GRD) Gridding Algorithms
The CETB coarse resolution gridding procedure is a simple, “drop-in-the-bucket” average. The resulting data grids are designated GRD data arrays. For the "drop-in-the-bucket" gridding algorithm, the key information required is the location of the measurement. The center of each measurement geolocation is mapped to an output-projected grid cell. All measurements within the specified time period that fall within the bounds of a particular grid cell are averaged. This is the reported TB value for this pixel. Ancillary variables contain the number and standard deviation of included samples. The effective spatial resolution of the GRD product is defined by a combination of the pixel size and spatial extent of the 3dB antenna footprint size (Long and Brodzik, 2016). Figure 6 provides a graphical representation of the coarse resolution of GRD measurements.

Reconstruction Algorithm
Reconstruction algorithms use the effective Measurement Response Function (MRF). The MRF is determined by the antenna gain pattern (which is unique for each sensor and sensor channel, and may vary with scan angle), the scan geometry (notably the antenna scan angle), and the integration period. The latter “smears” the antenna gain pattern due to antenna rotation over the measurement integration period. The MRF describes how much the emissions from a particular receive direction contribute to the observed TB value.
For implementation in the CETB, fine map grid resolutions were selected for each channel according to Table 9.
Sensor | Frequency (GHz) | Enhanced Resolution Grid (km) | ||
---|---|---|---|---|
12.5 | 6.25 | 3.125 | ||
SMMR | 6 | X | ||
SMMR | 10 | X | ||
SMMR | 18 | X | ||
SMMR | 21 | X | ||
SMMR | 37 | X | ||
SSM/I | 19 | X | ||
SSM/I | 22 | X | ||
SSM/I | 37 | X | ||
SSM/I | 85 | X | ||
SSMIS | 19 | X | ||
SSMIS | 22 | X | ||
SSMIS | 37 | X | ||
SSMIS | 91 | X | ||
AMSR-E | 6 | X | ||
AMSR-E | 10.7 | X | ||
AMSR-E | 18 | X | ||
AMSR-E | 23 | X | ||
AMSR-E | 36 | X | ||
AMSR-E | 89 | X |
Table 10 compares current data sets with the CETB product.
Variables | SMMR EASE-Grid | SSM/I-SSMIS EASE-Grid | Enhanced Resolution SSM/I & AMSR-E Polar TBs | Polar Stereographic Gridded TBs | AMSR-E Gridded TBs | CETB |
---|---|---|---|---|---|---|
Data Set ID | NSIDC-0071 | NSIDC-0032 | NSIDC-0464 | NSIDC-0001 | NSIDC-0301; NSIDC-0302 | NSIDC-0630 |
Spatial Coverage | Global | Global | Global | Limited Arctic/Antarctic regions | Global | Global |
Spatial Coverage (Gridded) | NSM EASE-Grids | NSM EASE-Grids | NSM EASE-Grids | NS Polar Stereographic, limited to high latitudes | NSM EASE-Grids and 1/4-degree lat/lon | NST EASE-Grid 2.0 |
Spatial Resolution(s) | 25 km | 25 km (all); 12.5 km (85GHz only) |
SSM/I: 12.5 km (19 and 22 GHz; 7.5 km (37.0 GHz); 2.5 km (85.5 GHz) |
25 km (all); 12.5 km (85GHz only) | 25 km and 1/4-degree | 25 km (all); enhanced resolution up to ~3 km depending on channel |
Temporal Coverage | 1978-1987 | 1987-present | SSM/I: 1995-2008 AMSR-E: 2002-2011 |
1987-present | 2002-2011 | 1978-present |
Interpolation (image reconstruction method) | ID2 | BG (tuned for low noise) or ID2 (F13 only) | SIR | DITB | ID2 | DITB at 25 km (all); SIR for enhanced resolution |
Image Reconstruction Tuning Parameters | N/A | BG tuned to match resolution of all channels to 19V; ID2: N/A | Independently optimized resolution for each channel | N/A | N/A | SIR tuned to optimize resolution for each output channel |
TB Files per Channel per Day | 2 (asc and des) | 2 (asc and des) | 2 local time of day (ltod) | 1 (daily averages) | 2 | 2 (ltod for NS and asc/des for T grids) |
Overlapping Orbits | Choose one orbit | Choose one orbit | Combine orbits (with ltod) | Averaged | Choose one orbit | Combine orbits (with ltod or asc/des) |
Variable Metadata | Time of sample used | Time of sample used | ltod? | N/A | Time of sample used | Ltod, measurement counts, TB standard deviation, TB time |
Algorithm Documentation | Limited to User's Guide | Limited to User's Guide | Limited to User's Guide | Limited to User's Guide | Limited to User's Guide | ATBD and Reconstruction White Paper |
Satellites Included | Full SMMR operations | F08, F11, F13, F17 with limited overlap periods | F13, AMSR-E | F08, F11, F13, F17 with limited overlap periods | Full AMSR-E operations | All available satellite data included (SMMR, F08, F10, F11, F13, F14, F15, SSM/I, F17, F18 SSMIS, AMSR-E) |
Input Data Calibration | Best available (Njoku, 2003) | Older RSS calibration | Older RSS calibration | Older RSS calibration | Best available | Best available |
References and Related Publications
Document Information
DOCUMENT CREATION DATE
November 2016
DOCUMENT REVISION DATE
Access complete Knowledge Base
Questions? Please contact:NSIDC User Services
Phone: 1 303 492-6199
Email: nsidc@nsidc.org