Obtaining and Importing Landsat Data

This document describes how to locate free, high quality Landsat imagery and how to import these data into the ENVI format. It assumes the user has a basic understanding of the Landsat program; including features such as the sensors used in the program, their spatial, spectral and temporal resolutions, when each of the sensors has operated, etc. While a brief description of the Landsat sensors is provided below, users are encouraged to review sensor information before proceeding with this guide. A few suggested sites are the USGS Landsat Program site, the NASA Landsat Tutorial, and the CEO User Guide. This document will also describe the process of obtaining free high-quality images from the GLOVIS and GLCF sites.

Worldwide Reference System

The Worldwide Reference System (WRS) is used to identify the path and row of each Landsat image. The path is the descending orbit of the satellite. Each path is segmented into 119 rows, from north to south. The Landsat MSS sensor had a swath width of 180 km and global coverage required 251 paths. The Landsat TM and ETM sensors have a swath width of 185 km and require only 233 paths for complete coverage. MSS and TM scenes share common rows, but in most cases the paths will be different. Because of this difference, MSS scenes are identified using WRS I while TM and ETM scenes use WRS II path/row designations. The data archive section of the CEO web site uses the WRS II designation for all path/row images. It is generally much easier to find images on the Internet when you know the path/row designation of the region you wish to work in. First, you should check out the CEO data archive. We may already have data for your area of interest. Even if we do not have data for your specific site, you may find WRS II paths and rows displayed on the archive maps of neighboring areas. We have an ArcGIS map on the CEO server that displays all of the WRS II paths and rows. You can view this by double clicking on the file N:\Landsat\PathRow.mxd. Navigate to your area of interest and zoom in until you can see the black grid of paths and rows. Use the ArcGIS Inquire tool to display the path and row of the cell you click on. Finally, the lab has a collection of 26 WRS index maps covering the earth. WRS I designations are on one side of these maps and WRS II designations are on the reverse side.

Basic Image Characteristics

There are three sensors used in the Landsat program. The Multi Spectral Scanner (MSS) mission ran from 1972 to 1993 and had four spectral channel covering the green, red, and (2) near infrared channels. The spatial resolution was 57 or 60 meters. The Landsat Thematic Mapper (TM) mission began in the mid-80s and Landsat 5 finally ended November 2012. This sensor features seven spectral channels at 30 meters spatial resolution. The Landsat 7 Enhanced Thematic Mapper (ETM) mission began in 1999 and is still operational. It features the same spectral channels as the TM sensor, with the addition of a second thermal channel
Copyright 2013 The Center for Earth Observation, Yale University http://www.yale.edu/ceo Page 1

and a 15 meter panchromatic channel. On May 31, 2003 the ETM scan line corrector failed and ETM images since that time are missing large portions each scene. On USGS sites these images are designated as SLC-Off and use of these images is generally not recommended.

Image Sources
There are many sites that you can use to locate and obtain Landsat satellite imagery. Three recommended sites are GLOVIS and EarthExplorer by the USGS and the Global Land Cover Facility (GLCF) at the University of Maryland. The most complete collection of Landsat data can be found at two USGS sites; GLOVIS and EarthExplorer. You will find a broad collection of Landsat data spanning the entire time of the program, beginning in the early 1970s. The user interface and download processes are a bit different for each site. More information about each is listed below. There are several international sources of Landsat images which typically charge $1,000 or more per scene. You may also find various government or non-profit organizations that maintain an archive of images for their region which can be shared with the public, or at least with research collaborators. Locating and accessing these sites is beyond the scope of this document.

GLOVIS (This should be your first choice)

The USGS Global Visualization Viewer GLOVIS site at: http://glovis.usgs.gov/ has Landsat data, as well as ASTER and some MODIS satellite images. All Landsat images at GLOVIS have recently had standard processing algorithms and terrain correction applied, making them very easy to use. When you enter the site your browser must allow popup windows so that the Visualization Viewer window can open. Select the appropriate image collection e.g. Landsat Archive Landsat 4 5 TM and then navigate to the region you are interested in. You can use the Prev Scene and Next Scene buttons to scroll through the available images. When you have located an image you wish to work with, click the Add button in the lower left to make it available for ordering. Users can select several images and place them in a cart for ordering. Some data are available for immediate download once they are Added to your cart. For these images, make sure you select the Level 1 Product. In other cases, the image request is submitted to the USGS and when the data are available the user will get an email with the FTP link to retrieve the data. This may take a few hours to a few days. Each user must register on this site before downloading or ordering images.

Copyright 2013 The Center for Earth Observation, Yale University http://www.yale.edu/ceo Page 2

Earth Explorer
The Earth Explorer site at: http://earthexplorer.usgs.gov/ includes many types of data in addition to Landsat images. Begin by selecting the type of data you want, as you would for GLOVIS. Next select the location and time frame you are interested, as well as the number of results for the search. Then click on the Search button to begin the data search. You may need to modify the dates and/or the number of results to find appropriate data. Click on the Results button to view images that meet your criteria. From this page you can look at browse images, inspect the meta data, download an image, or place an order. As with the GLOVIS site; you must register on the site to access the images.

USGS Data Import

Landsat data from either site are distributed as a single file in an archived and zipped .TAR.GZ format. After downloading the file move it to a separate folder in your user section of the server. Double click on it to load the program 7-Zip, showing the .tar file. Right-click on the .tar file and select Open Inside to display the detail data files. Click on the blue Extract icon and select the destination folder to extract the individual files that comprise the entire image. Each data layer is a separate TIF image file. There are also two text files with the same base filename but ending with _GCP.TXT and _MTL.TXT. This file structure is referred to as GeoTIFF with Meta data. You can open each data layer individually, then create a layer stack, and finally save the result as a new file. ENVI, however, can easily open data in this USGS format. From the ENVI main menu bar, select File Open Image File and navigate to the _MTL.TXT file. ENVI will automatically open the Landsat image with all bands in the correct order. The six reflective bands are placed in one file, the thermal band(s) in another file, and for ETM data there will also be a 15m panchromatic file. While you can work with these data as they are, ENVI has only created a temporary layer stack that is constantly resampled as you move around the image. You need to save each file as a new dataset. From the ENVI main menu, select File Save File As ENVI Standard. Select the file you wish to save, then navigate to your work area and enter an appropriate file name. Once this is saved as a new file, the uncompressed .TIF files and the .tar file can be deleted.

GLCF Landsat Images

The Global Land Cover Facility (GLCF), at the University of Maryland, is a source of free high-quality satellite images and many satellite-derived products. You should take time to explore the documentation on this site, especially information regarding image citations in your work. This paper will focus on the Landsat imagery. The GeoCover images are the highest quality Landsat images that are generally available here. These have been selected for minimal cloud cover, during peak greenness, and have been precisely orthogeoreferenced. When searching for Landsat data on this site, favor data that have this attribute. There will be one TM, and ETM scene for each Landsat path/row.

Copyright 2013 The Center for Earth Observation, Yale University http://www.yale.edu/ceo Page 3

Once you have identified the path and row of your region of interest, go to the GLCF Data & Products page, select the Earth Science Data Interface link, and then click on the Path/Row Search icon. Now define your search criteria on this page. Enter the start path and row for a single scene. If you are searching for a range of data include an end path/row as well. Select WRS-1 for MSS scenes or WRS-2 (the default) for TM and ETM scenes. You can restrict your search to GeoCover scenes only by selecting this in the Require section. After entering this information click on the Submit Querry button. The window will display the number of scenes found that meet your criteria. Click on the Preview and Download button to explore the defined images. Within the preview window you can select each image by clicking the ID number in the first column. This will display a small browse image. Click in the browse image to see a larger view in a new window. If you wish to use this image, click on the Download button to open a download window. Before downloading any data create a unique folder on your U:\ drive for each image. Now right click on each file and save them in your new folder.

GLCF Data Import

Each data layer will be a TIF image in compressed .GZ format. You can double-click each layer to uncompress it. As with the USGS datasets, ENVI can open all bands for the image in the proper order by simply opening the metadata file. This would be the file with the common base name and the file extension .MET. Reflective bands, thermal band(s), and the panchromatic band if available, all have different spatial resolutions and are treated as separate files. Save these as new ENVI files as is described in the USGS section above. Once this is saved as a new file, the uncompressed .TIF files can be deleted.

Copyright 2013 The Center for Earth Observation, Yale University http://www.yale.edu/ceo Page 4