Frequently Asked Questions
1. How do the calibration coefficients differ between the Level-1A and Level-2A products?
2. How can I view AMSR-E data?
3. Why do I see coastline shifts in Level-2A brightness temperature data?
4. Why does the number of pixels per scan vary among different Level-2 products?
5. How can I receive a direct broadcast of Aqua/AMSR-E data?
6. How can I read AMSR-E Level-2 Soil Moisture Data?
7. What is NOSE and how does it work with AMSR-E data?
8. How and when will I be able to access the AMSR-E data from NSIDC?
9. How do AMSR and AMSR-E improve upon previous passive microwave radiometers?
10. What format are the data in?
11. What is the spatial resolution of AMSR-E products?
12. What do the various levels of data represent?
13. Is there a Level-1B AMSR-E product available?
14. What is EASE-Grid?
15. Why are there holes at the North and South Poles, and why are they different sizes in the ascending and descending grids?
The calibration coefficients and offsets for this data set are determined by JAXA. They are different from those of the AMSR-E/Aqua L2A Global Swath Spatially-Resampled Brightness Temperatures, which are determined by RSS. The Level-1A data have not been calibrated; specifically, they were not converted from instrument counts to antenna temperatures. Coefficients are written to the Level-1A data but not applied.
Note: These instructions do not apply to the Level-2B Soil Moisture data. See Step 6 for instructions on viewing Level-2B Soil Moisture data.
First, you can open AMSR-E data automatically with ENVI, which handles the HDF-EOS data format by selecting the variable you want to view, for example, 6.9V_Res.1_TB. (NSIDC performed these steps with ENVI 3.6 using example parameters from Level-2A data).
Otherwise, we suggest a free utility called hdp created by the National Center for Supercomputing Applications (NCSA) for dumping HDF data to binary format. This utility can also extract the header information to a text file so that you can easily view the variables in the data and their associated attributes.
You can then search the header information for the particular variable you are interested in viewing, for example, 6.9V_Res.1_TB, to view attributes associated with that variable. The dimensions of this attribute are listed as Dim0, which is the number of lines (y dimension), and Dim1, which is the number of samples or scans (x dimension). Example values are 2001 and 243 for the 6.9V_Res.1_TB variable. All of the data except for the 89 GHz have 243 scan pixels, but the 89 GHz data have 486 scan pixels. The data type of the variable is also listed as an attribute, for example, 16-bit signed integer. You can then use this information to view the binary output of this variable, extracted using the hdp utility mentioned above, using ENVI, ERDAS IMAGINE, or any other image viewing application. Note the scaling factor of 0.01 and offset of 327.679993 to convert the binary image values to brightness temperature units of kelvin. This scaling factor and offset are also listed in the variable's attributes, as Tb_scale and Tb_offset.
Complete guide documentation is available for all data sets. In particular, the Format section of each guide document describes the contents of the data. Also visit HDF-EOS Tools and Information for additional tools to read HDF-EOS data.
On 30 April 2004, the AMSR-E science team identified a navigation error in Level-2A brightness temperature data that reveals coastline shifts as large as 15 to 20 km, or about two 89 GHz pixels, when viewing animations over Florida, USA. Initial observations indicate that the errors:
- Occur early in AMSR-E processing.
- Vary from orbit to orbit in all channels.
- Are not related to the Aqua spacecraft. Errors do not appear with coincident MODIS data, further supporting the claim that the problem originated in early AMSR-E processing.
These errors primarily affect regional land surface and sea ice research, not global-scale studies. Immediate causes for the geolocation errors are not known, but the AMSR-E science team is investigating possible channel offsets -- the location of the channel footprints with respect to each other. Future releases of AMSR-E Level-2A data will include a more accurate geolocation.
A typical AMSR-E swath width consists of approximately 2000 scans, with 243 pixels per scan for the 6.9 GHz to 36.5 GHz channels, and 486 pixels per scan for the 89.0 GHz channel.
In 2002, the AMSR-E science team discovered distortion along the outer 23 to 24 pixels of each swath. Some algorithm developers, including those for the Level-2A brightness temperature product (AE_L2A) and the Level-2B rain product (AE_Rain) decided it was important to still keep those outer pixels in the final products. So AE_L2A has 243 pixels per scan (6.9 GHz to 36.5 GHz) or 486 pixels per scan (89 GHz). The Level-2B rain product (AE_Rain) is derived from 89 GHz data, so it also has 486 pixels per scan.
The Level-2B ocean product (AE_Ocean) algorithm developers, however, decided to trim the distorted pixels from their product, keeping only the center 196 pixels.
All algorithm developers for Level-2B and Level-3 products were recently asked to reevaluate whether they still want to keep the outer pixels or not. So the swath width may change among products in the near future.
The NASA Aqua spacecraft has a direct broadcast X-band downlink that allows sites with the proper reception hardware to receive real-time AMSR-E data. The International MODIS/AIRS Processing Package (IMAPP) software from the Space Science and Engineering Center at University of Wisconsin will soon be able to process AMSR-E direct broadcast data. Also see the Direct Readout Project for information about software to process direct broadcast data.
Regarding generation of science products from raw data, NASA requires algorithm developers to provide their software code to the Distributed Active Archive Centers (DAACs). This code is provided in files called Delivered Algorithm Packages (DAPs), which are subsequently archived and available upon special request from NSIDC User Services. The DAP to generate Level-2A brightness temperatures runs only on Windows platforms; other DAPs run only on Unix platforms. NSIDC is currently helping the National Oceanic and Atmospheric Administration (NOAA) generate Level-2A brightness temperatures from software within the Level-2A DAP.
HDF-EOS defines three different data models – point, swath, and grid – each of which is constructed using standard HDF data types. While all other AMSR-E Level-2B products use the swath data model, the soil moisture product is unique in using the point data model. By using the point data model, each observation in the swath may be mapped to a location in the projected grid that is used for the Level-3 data.
HDF-EOS data files are supported by the HDF-EOS library, which allows data products to be created and manipulated in ways appropriate to each HDF-EOS data type without regard to the actual HDF objects underlying them. Thus, it is recommended that the HDF-EOS library be used for reading all HDF-EOS files. However, users who do not wish to download and install the HDF-EOS library may use the HDF library to read any HDF-EOS file. In addition, many commercial software packages such as IDL, ERDAS IMAGINE, and MATLAB have the HDF library built in, as well as the HDF-EOS library in the case of IDL.
To read AMSR-E Level-2B Soil Moisture data, the user must be aware of how the point data model is implemented. Specifically, the point data model uses the Vdata HDF object instead of the SDS data object to hold data. If using the HDF library as opposed to the HDF-EOS library, the routine needed is VSread. If using IDL, the EOS_PT_OPEN and related functions allow you access to the HDF-EOS specific interface for the point data type. The command-line utility HDP can also be used to dump the Vdata with the function dumpvd.
The HDF-EOS library is available for most common operating systems at: http://newsroom.gsfc.nasa.gov/sdptoolkit/toolkit.html
The HDF Reference Manaul can be found at: ftp://ftp.hdfgroup.org/HDF/Documentation/HDF4.2r1/HDF42r1_RefMan.pdf
The HDP utility is available from NCSA at: http://hdf.ncsa.uiuc.edu/hdp.html
AMSR-E Level-2A granules use a Nominal Orbital Spatial Extent (NOSE) to specify spatial coverage. These are predefined polygons based on the AMSR-E's viewing geometry and Aqua's orbital characteristics. The metadata shows the spatial coverage of a data granule by the Product Specific Attributes (PSAs) StartingPolygonNumber, EndingPolygonNumber, and NominalPassIndex. These values are not meaningful for users, but rather they are used by the system in spatial queries. So if you want to know which granules cover your area of interest, you must include this as a criterion in your search. Note that the Gringpointlatitude and Gringpointlongitude metadata attributes do not provide correct spatial coverage information. Also note that each data granule contains complete geolocation information in the form of pixel location arrays, which conform to HDF-EOS swath standards.
Users can order Level-1A AMSR-E data beginning 19 June 2003 and Level-2A data beginning 02 September 2003. Other products became available in March 2004. Users can order data through Reverb, or by contacting NSIDC User Services.
First, the spatial resolution of AMSR/ADEOS-II and AMSR-E/Aqua data doubles the resolution of Scanning Multichannel Microwave Radiometer (SMMR) and Special Sensor Microwave/Imager (SSM/I) data.
Second, AMSR and AMSR-E each combine into one sensor all the channels that SMMR and SSM/I had individually. Both have the lower channels which were only on SMMR, as well as a higher frequency channel which was only on SSM/I. SMMR has the following channel frequencies: 6.6 GHz, 10.7 GHz, 18 GHz, 21 GHz, and 37 GHz. SSM/I has the following channel frequencies: 19.3 GHz, 22.3 GHz, 37 GHz, 85.5 GHz. In comparison, each of the AMSR instruments has the following frequencies: 6.9 GHz, 10.7 GHz, 18.7 GHz, 23.8 GHz, 36.5 GHz, and 89 GHz. See the Comparative Operating Characteristics of SMMR, SSM/I, and AMSR table on the AMSR-E Instrument Description Web page.
The Level-1A products are in Hierarchical Data Format (HDF). The AMSR-E Levels-2 and Level-3 data products are in HDF-EOS.
HDF is the standard data format for all EOS data products. HDF is a multi-object file format developed at the National Center for Supercomputing Applications (NCSA) at the University of Illinois. For more information about the HDF-EOS format, tools for extracting binary and ASCII objects from HDF, and a list of other HDF-EOS resources, please visit NSIDC's HDF-EOS Web pages.
The spatial resolution of AMSR-E products varies from 6.25 to 57 km.
|Level-0||Reconstructed, unprocessed instrument/payload data at full resolution; any and all communications artifacts, for example, synchronization frames, communications headers, and d . . data removed.|
|Level-1A||Reconstructed, unprocessed instrument data at full resolution, time-referenced, and annotated with ancillary information, including radiometric and geometric calibration coefficients and georeferencing parameters. For example, platform ephemeris computed and appended but not applied to the Level 0 data.|
|Level-1B||Level-1A data that have been processed to sensor units. Not all instruments will have a
|Level-2||Derived geophysical variables at the same resolution and location as the Level-1 source data.|
|Level-3||Derived geophysical variables mapped on uniform space-time grid scales, usually with some completeness and consistency.|
|Level-4||Model output or results from analyses of lower level data. For example, variables derived from multiple measurements.|
An AMSR-E Level-1B product exists, but it is not available from NSIDC. For information about this product, please contact the Japan Aerospace Exploration Agency (JAXA).
The Equal-Area Scalable Earth Grid (EASE-Grid) consists of a set of three equal-area projections, combined with an infinite number of possible grid definitions. It is based on a philosophy of digital mapping and gridding definitions developed at NSIDC. For more information, please visit the EASE-Grid-Data Web site.
See the accompanying figure. The Aqua satellite travels along the dashed line marking the center of a scan. When that center point reaches its maximum latitude, the ascending half orbit (green dots) ends, and a new descending half orbit (orange) starts. All orbits will be split at this same latitude, all the ascending orbits will reach the same maximum latitude, and all the descending orbits will reach a different (and higher) maximum latitude. This creates pole holes with different radii. The South Pole is similar, except that ascending and descending are swapped.