Some frequently asked questions:

What products are available?
How do scatterometers work?
What is the difference between a scatterometer and a radiometer?
Why scatterometers for climate studies?
Data from what instruments are available from this site?
What is the BYU .SIR file format?
How do I compute the latitude and longitude of a pixel of BYU .SIR file?
How do I read and display a BYU .SIR file?
What is the "Scatterometer Image Reconstruction (SIR)" algorithm?
What is the difference between SIR, AVE, and GRD?
How can I get information on the specific projection used for a BYU SIR file?
How to I convert BYU SIR images to GeoTIFF format?
How can I display a BYU SIR image in ENVI?
How can I semi automate my file downloads?
What tools are available to read a BYU SIR file into ArcGIS Raster format?
What tools are available for windows to read a BYU SIR file?
Where is the Lambert Azimuthal Equal Area projection used in BYU SIR files defined?
Can I use a script to download multiple SCP product files?
How can I use BYU .SIR files with the Matlab M_map package?
How is the SCP ftp (ftp.scp.byu.edu) site organized?

What products are available?

The Scatterometer Climate Record Pathfinder project develops and distributes innovative scatterometer and radiometer products designed to support climate studies. The core products are enhanced resolution radar backscatter (sigma-0) and brightness temperature (Tb) products. A brief list of available products is available.

Processed data from the Seasat Scatterometer (SASS), the NASA Scatterometer (NSCAT), the ERS-1 and ERS-2 AMI (scatterometer mode), and the SeaWinds on QuikSCAT and SeaWinds on ADEOS-II scatterometer data is available on this site. While the site emphasizes enhanced resolution image data, other data products are available.

The BYU .SIR image file format was developed by the Brigham Young University (BYU) Microwave Earth Remote Sensing (MERS) laboratory to store images of the earth along with the information required to earth-locate the image pixels. A "SIR" file consists of one or more 512-byte headers containing all the information required to read the remainder of the file, including projection information to map pixels to lat/lon on the Earth's surface. Image pixel values are generally stored as 2 byte (high order byte first) integers. Scale factors to convert the integer or byte pixel values to native floating point units are stored in the file header. The origin of the images is in the lower left corner of the displayed image. The earth location of a pixel is identified with its lower-left corner. The standard sir format supports a variety of image projections including: Rectangular array (no projection); A rectangular lat/lon array; Lambert equal-area; Polar stereographic; and EASE with various resolutions.

BYU .sir files include a 512 byte header that includes all of the map projection information to compute transformations between pixel grid location and latitude/longitude and between latitude/longitude and pixel grid location. There are two approaches: (1) download and run code (available in multiple languages) or (2) use location info "support images" provided on the SCP website whose pixel values are the latitude and longitude of each image. For the former start at the top of software downloads webpage. For the latter, goto the very bottom of the web page that lists the available data for the sensor of interest to find the table of "Support Images". Download the Latitude and Longitude images. These can also be obtained directly from the SCP ftp site in the "info" directory for each sensor's product.

Browse (small, reduced resolution) image files are available from the SCP web and ftp sites in GIF format for selected files; however, the best way to view a SIR file is download it and use a local program to display and analyze it. The image products are stored in the BYU MERS SIR file format in which the image is stored as a scientific (real valued) image file that includes both location and transformation information in a header. (Separate files containing the lat/lon locations of each pixel are also available.) Viewer and reader programs for the BYU MERS SIR file format are available on line from the BYU MERS web site the NASA Scatterometer Climate Record Pathfinder web site.

Some options include:

• Using matlab, IDL, PV-WAVE, ENVI, or other high level language. Reader and viewer utility routines are available for many of these.

• Using existing ustom viewer utility programs. These include modified xv program known as xv_BYU and the sirtool program. These must downloaded and compiled on the local machine, though executables for some platforms are available.

• Generate your own custom code. Read/write routines in c, c++, fortran, matlab, IDL, PV-WAVE are available.

• Convert the SIR file to a plain ASCII format and reading into another program (code and sample programs are available in c, c++, and fortran. These must be downloaded and compiled on the local machine)

• Some image processing programs (e.g., Adobe Photoshop) can be used to display file contents, though the scaling and header information are lost.

See the detailed software documentation available on the web and ftp sites under "software".

The Scatterometer Image Reconstruction (SIR) algorithm is a method for reconstructing images from raw scatterometer or radiometer data. The SIR algorithm is based on a multivariate form of block multiplicative algebraic recontruction. It generates enhanced resolution gridded images from irregularlly spaced measurements which may have a variable aperture function. It provides high resolution images from lower resolution images, assuming sufficiently dense sampling. Frequently mulitple satellite passes are combined to provided to ensure sufficient sampling. The theory behind the iterative SIR algorithm is described in D.S. Early and D.G. Long, "Image Reconstruction and Enhanced Resolution Imaging from Irregular Samples," IEEE Transactions on Geoscience and Remote Sensing, Vol. 39, No. 2, pp. 291-302, 2001. See also enhanced resolution imaging.

What is the difference between the SIR, AVE, and GRD algorithms?

The Scatterometer Image Reconstruction (SIR) algorithm is an iterative recontruction algorithm at creates high resolution images from irregullarly spaced measurements. The AVE algorithm is the first iteration of SIR and is computed by defining a high resolution pixel grid. A value for each pixel is created by weighted averaging all of the measurements that have a non-zero spatial response function for the pixel. The weighting function is the normalized value of each measurement's spatial response function for the particular pixel, see D.G. Long, P. Hardin, and P. Whiting, "Resolution Enhancement of Spaceborne Scatterometer Data," IEEE Transactions Geoscience Remote Sensing, vol. 31, pp. 700-715, 1993 and D.S. Early and D.G. Long, "Image Reconstruction and Enhanced Resolution Imaging from Irregular Samples," IEEE Transactions on Geoscience and Remote Sensing, Vol. 39, No. 2, pp. 291-302, 2001.

GRD stands for "gridded" image. This is a non-spatially enhanced image created by gridding the measurements into a fixed, low-resolution grid using "drop-in-the-bucket" (DIB) techniques. Each pixel value consists of the average of only the measurements whose center falls within the spatial extent of the particular pixel. GRD images have the coarsest resolution, but the lowest noise, while SIR images have the highest spatial resolution. AVE images fall in between. Typically the dimensions of AVE and SIR images are 1/5 the size of GRD images. The same projection is used for all three image types so each GRD pixel corresponds to a 5x5 area if AVE or SIR pixels.

The BYU .SIR file format includes all of the information needed to geolocate individual pixels. Software in various computer languages is available on the ftp/web site to read the file header and compute the forward and reverse map projections. In addition, the interactive C program sir_util can display individual pixel values and locations, as well as dump each pixel location to a text file. sir_util can print detailed map projection information and provide strings to use with standard proj4 and gdal_translate utilities. (Note: sir_util.exe is available as a precompiled windows executable. Source code [in tools] can be compiled for other platforms.).

The BYU .SIR file format includes all of the information needed to geolocate individual pixels. To convert these images to GeoTIFF format can be bit tricky due to the complexities of the GeoTIFF format. Some simple utilities have been written to simplify the conversion process. These are described in the page Conversion of BYU .SIR files to GeoTIFF format. The latest version of the sir_util2 utility includes geotiff conversion capability.

There are several ways to display BYU. SIR files in ENVI. One is to convert the image to a GEOTIFF first. Second, the IDL file readers can be used. Alternately, for just displaying the image, follow theses steps after starting ENVI: First, you have to determine the file size (X by Y) in pixels. This can be determined using the IDL reader code. Then,

• From the File menu select, "Open image file", and select the file to read after gunzipping it.

• A pop-up widow appears. Enter the X file size in "samples".

• Enter Y file size in "lines".

• Enter 1 in the "bands" and 512 in the "offset"

• Select "Unknown" for "File type" and "Network (IEEE)" for the "Byte Order".

• Select "Integer" for the "Data Type" (leave "BSQ" for "Interleave").

• Hit OK.

• Select the "Gray Scale" radio button and press "Load Band".

Note: the file will be displayed flipped vertically since ENVI defines the image origin (pixel 1,1) at the upper-left rather than lower-left..

file download automation can be easily semi automated depending upon how comfortable you are with scripting. Create a simple text file, ftp.script, which contains the ftp paths of the files you wish to download. The SCP ftp site follows a very standard organization, which can be determined by examining the html files on the web site. Here is an example of a simple ftp script:

anonymous

get data/qscat/1999/sir/qush/Arc/202/a/qush-a-Arc99-202-202.sir.gz
get data/qscat/1999/sir/qush/Arc/203/a/qush-a-Arc99-203-203.sir.gz
get data/qscat/1999/sir/qush/Arc/204/a/qush-a-Arc99-204-204.sir.gz

For *nix systems pipe this file to your ftp command
ftp.scp.byu.edu < ftp.script
and type in the password [your email address] at the prompt.

For Microsoft Windows XP and similiar systems bring up a command line.
1. Click the Start Menu.
2. Select Run.
3. Type "cmd", and click run.
4. In the black command line window that appears type:
ftp -i -s:C:\ftp.script ftp.scp.byu.edu
4a. "C:\ftp.script" needs to be replaced with the path of your script.
5. Type in the password at the prompt.

This simple semi automation can be further modified as the directory structure of scp.byu.edu is very predictable and consistent, see file site organization.

Tools such as wget can be used, though use these with caution since the entire site is over 40 TB in size!

What tools are available to read a BYU SIR file into ArcGIS Raster format?

The BYU .SIR file format includes all of the information needed to geolocate individual pixels. The Marine Geospatial Ecology Tools (MGET) project (also known as GeoECho Python) has developed an open source geoprocessing tool box that enables conversion of BYU SIR files into ArcGIS raster format files.

In reponse to requests, a simple program that reads BYU SIR files is available from the SCP web/ftp site. This pre-compiled program, sir_util.exe, can print individual pixel and location values or dump the contents of the file to a text file. Geometric transformations are supported as is conversion of a SIR file to a BMP or GIF image file. Source code for the program is included in the regular c code reader distribution in the tools subdirectory. Source code for the latest version (sir_util2.c) is now available as is a windows executable sir_util2.exe. The new version includes color tables and conversion options to TIFF and GeoTIFF files. Note: The binary executable on the web site is compiled for Windows XP. It should run in Windows 7. After downloading, right click on the code and click the "Properties" tab. On the "Properties" sheet, click on the "Compatibility" tab, then click on the box "Run this program in compatibility mode for:" and select "Windows XP SP3" from the dropdown list.

Note: when the program is run, the output string for gdal_translate does not include the double quotes the should surround the argument to the -a_srs command option of the gdal_translate program. The double quoted string begins just before the +a and ends after the +lon_0=(number) argument.

Pages 157-173 in the report J.P. Synder, "Map projections used by the U.S. Geological Survey," U.S. Geological survey, Bulletin #1632, 1982. Available from pubs.usgs.gov/bul/1532/report.pdf. Source code for the software used to compute the map projection for all projections used in BYU .SIR files are available from the SCP ftp site www.scp.byu.edu/software.

Yes. The ftp site uses a very regular form that lends to scripting. Using the website file browser to determine the name of the particular type file you want to download. Then download and modify the windows batch or linux bash script in the software library to download the desired files.

NOAA's M_map is a simple but powerful package for plotting maps. When using the Matlab map package M_map recommend using M_map version 1.4 with GSHHS installed. When combined with the SCP Matlab file reader and map projection package, the Matlab script m_loadsir.m can load SCP product files into Matlab for use with M_map. An example script is available at m_sirtest.m. Test files can be downloaded from the SCP site.

The SCP ftp site is organized in a simple and consistent way to ease download of its large datasets. See file site organization.