HR: 0800h
AN: ED51C-0035    [Abstracts]
TI: Salish Kootenai College Student Internship With the Landsat Data Continuity Mission: A Student's Perspective
AU: * Fisher, R
EM: trjj@ronan.net
AF: Salish Kootenai College, P.O. Box 70, Pablo, MT 59855 United States
AB: Hello my name is Richard D. Fisher. I was very fortunate to be picked to travel to Washington DC in July 2004, to complete a five week internship at NASA. My internship project is located on the Flathead Reservation and the area is called the Jocko-Spring Creek. My project was to complete land cover classification and land cover change detection. In order for me to accomplish my goals I had to learn how to use two new computer programs, MultiSpec and ENVI (Environment for Viewing Images) for remote sensing processing. Computer use does not come easy to me because, I lack the training most people take for granted. However, I did not let this lack of training get me down. The first step was to acquire two Landsat images. The first image was from the Landsat 7, landsat satellite in 1999 and the other was from the Landsat 5 satellite in 1987. The path row for my study area is 41-27. Once the images were acquired I had to combine the different color bands to make one image and perform a blue band correction. The blue band correction takes the blue haze out of the images making them clearer. The visible bands are blue, green, red and three bands of infrared. Once these color bands are together you can change the color of the image to help you look for different features, because each different color band will show you something different. After I put the images together I used ENVI to do the land cover classifications. The next step was to subset my project area to a smaller size. I cut both images in exactly the same coordinates. With help from my NASA mentor scientist, Rich Irish from the Landsat Data Continuity Mission, and I used photo shop Adobe PhotoShop to do the subsetting of both images. We were able to then link the two images together using ENVI software. After that I started to analyze the different pixels and their colors. I classified each image starting with the areas I knew from the fieldwork. After the classifications on both images were complete and I felt confident in the final classification, I needed to work with the shadows from the mountains in the image. We performed change detection from subtracting the two images. These computer programs are fun to use and were very useful especially when combining Landsat images. I would recommend these programs to anyone.
DE: 0800 EDUCATION
DE: 0810 Post-secondary education
SC: Education and Human Resourcese [ED]
MN: 2004 AGU Fall Meeting