Lab 8 Spectral signature analysis & resource monitoring

GEOG 338
Charlie Krueger
Lab 8: Remote Sensing of the Environment

Goal Background:

The goal of this lab for the class was to give us experience doing the tasks such as measuring spectral reflectance and interpret them. This would be done on many different types of materials that the Earth has to offer. The samples would be taken from satellite images of the counties of Eau Claire and Chippewa. The lab will instruct the students on how to collect spectral signatures from the samples, graph the samples in a program, and preform interpretations on them to see if they pass the spectral separability test. Another lesson that will be taught through this lab is the monitoring of the health of vegetation and soils using a band ratio technique. At the conclusion of this lab each student will be ready to analyze spectral signatures from samples that they collected from the Earth’s surface and monitor the health of soils and vegetation

Methods:

In the first part of this lab the class used a landsat ETM+ image to collect and then analyze the spectral signatures of the various Earth surfaces where in Eau Claire and Chippewa counties. First the program Erdas Imagine was opened then the surface image was opened on the screen. The tools that would be used in this section were in the spectral area of Imagine. The drawing tool section was selected then polygon tool was chosen so that an area inside one of the surfaces could be digitized. This mean a section was selected then a small shape was drawn on the surfaces that was to be sampled in the first section this was open water so Lake Wissota was selected for the class. Next in the raster tool area the supervised button was hit and that opened a drop down that showed all the options. The option that was selected for the lab was signature editor which created a new window were all the samples that would be collected would be shown. The class name was then changed to reflect the type of surface that was collected and in this case it was open water. The display mean plot button was selected next and this opened a window that showed a line of where the area that was collected had the highest spectral signatures and the lowest. This process was then continued to include 11 more different types of Earth surfaces. These were moving water, forest, riparian vegetation, crops, urban grass, dry soil, moist soil, rock, asphalt highway, airport runway, concrete surface. All of these were then compared on one of the graphs and shows the difference in all of the surfaces. See results for the graph.

The second part of the lab was performing simple band ration to see the health of vegetation and soils. The vegetation image was loaded into viewer and then the button under raster was selected and this was unsupervised and from that drop down NDVI was chosen. This opened the indices interface and this was where in the input image and output area were selected. The sensor was changed to Landsat 7 Multispectral. After all these changes were made the ok was given and a new image was created from the program that was run. The image was then taken and placed into ArcMap a map making program and the layers were then labeled so that a person on the street could understand the map. The same process was then run on the soil health in the same area and the vegetation in the first process. The only difference in the process was that under function ferrous minerals were selected because that was what was being looked at. Again the image was moved over to ArcMap was analyzing.

Results:

This was the first sample that was taken of the standing water and this is the graph that shows the signature mean of the spectral reflectance.

This is the image of all of the samples of the Earth surfaces on the signature mean plot graph. As someone could see there are many differences in the spectral reflectance of these samples. It was very interesting to see all the different samples next to each other.

This is the map that was created in ArcMap showing the difference in the vegetation in counties of Eau Claire and Chippewa. The map shows that the vegetation is higher in the eastern section were the development of bigger cities is very limited meaning that more vegetation is let grow.

This final image shows how common ferrous minerals are in the sample areas. The image shows that ferrous minerals are most common found in areas that are more developed than others. The area around the city of Eau Claire has more ferrous minerals because they were probably exposed during the development of the city.

Sources:

Satellite image is from Earth Resources Observation and Science Center, United States Geological Survey.

.