The goal of this lab was to work with the measurement and analysis of spectral reflectance signatures of different features on the Earth's surface. The analysis is done through satellite images that captured different materials on the Earth's surface. This lab works with the collection of spectral signatures for remotely sensed images, the graphing of these signatures, and the analysis of the graphs. This tool is necessary for proper image classification. This skill can be used to monitor the health of vegetation and soils.
Methods:
Part 1: Spectral signature analysis
This part of the lab works with a year 2000 ETM+ image of the Eau Claire area and other nearby areas in Wisconsin. In the image spectral signatures are analyzed of various Earth surface and near surface features. I started by opening up the image in an Erdas Imagine viewer. The first spectral signature I collected was for Lake Wissota located in the north-northeast portion of Eau Claire. I zoomed the image into Lake Wissota. I selected the Drawing tab on the main Erdas Imagine interface and selected the Polygon tool form the displayed area of interest tools. I digitized a portion of Lake Wissota. I then clicked the Raster tab on the main Erdas Imagine interface and selected the Supervised tool dropdown menu and selected the Signature Editor option from the dropdown list. The Signature Editor window opened. I clicked the Create New Signature(s) from AOI icon from the toolbar heading. My spectral plot of my signature of Lake Wissota was added to the window. I changed the Signature Name from Class 1 to Standing Water and changed the color of the line to a more visible color, I used blue. To examine the spectral curve I clicked on the Display Mean Plot Window tool in the main toolbar. This opened up a window displaying the spectral plot of the signature I collected from Lake Wissota called the Signature Mean Plot. To change the background color to white to make the graph easier to read I clicked the Edit heading on the Signature Mean Plot window and selected Chart Options from the drop down list and changed the Chart Background color drop down to white. I selected close apply and close to make the changes. I completed the steps I took for the creation of the standing water spectral analysis for 11 other materials, moving water, vegetation, riparian vegetation, crops, urban grass, dry soil (uncultivated), moist soil (uncultivated), rock, asphalt highway, airport runway, and concrete surface (parking lot). I created signature mean plots for each of the surfaces and created one signature mean plot of all of the surfaces. All of the signatures were stored in one Signature Mean Plot Window.
Part 2: Resource Monitoring
This part of the lab worked with performing simple band ratios to monitor the health of vegetation and soils.
Section 1: Vegetation health monitoring
This section performed a simple band ratio by using the normalized difference vegetation index (NDVI). I started by opening an Erdas Imagine viewer and bringing in the image ec_pw_200.img. I clicked the Raster tab on the main interface and selected the Unsupervised toolset and selected NDVI from the dropdown menu. The Indices window opened. I made sure the input image was the one I had loaded in the viewer and I set a location and name for my file to save as. I set the sensor to Landsat 7 Multispectral from the dropdown menu, and under Select Function I made sure NDVI was highlighted. I clicked OK to create the NDVI image. After the completion of the model run, I clicked Dismiss and brought the newly created image into the Erdas Imagine viewer. I imported the image into ArcMap and created a map of the image. I changed the Classified symbology type and set it to an equal interval classification style with 5 classes. I added a legend, key, north arrow, and scale bar to my map. I exported my map as a JPEG.
Section 2: Soil health monitoring
This section performed a simple band ratio by using the ferrous mineral ratio. I started by opening an Erdas Imagine viewer and bringing in the image ec_pw_200.img. I clicked the Raster tab on the main interface and selected the Unsupervised toolset and selected Indices from the dropdown menu. The Indices window opened. I made sure the input image was the one I had loaded in the viewer and I set a location and name for my file to save as. I set the sensor to Landsat 7 Multispectral from the dropdown menu, and under Select Function I selected FERROUS MINERALS. I clicked OK to create the image. After the completion of the model run, I clicked Dismiss and brought the newly created image into the Erdas Imagine viewer. I imported the image into ArcMap and created a map of the image. I changed the Classified symbology type and set it to an equal interval classification style with 5 classes. I added a legend, key, north arrow, and scale bar to my map. I exported my map as a JPEG.
Results:
Part 1: Spectral signature analysis
Signature Editor Window:
All of the signature analyses were stored in one window.
Figure 1
Signature Mean Plots: In interpreting the graphs, the materials with the most water content display the least reflectance. Water absorbs energy, do the less water the more reflectance. Water and reflectance have an inverse relationship.
Standing Water: Lake Wissota The spectral channel that displays the highest reflectance is 0.45-0.52 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
Figure 2
Moving Water:
The spectral channel that displays the highest reflectance is 0.45-0.52 micrometers and the spectral channel that displays the lowest reflectance is 10.40-12.50 micrometers.
The spectral channel that displays the highest reflectance is 0.45-0.52 micrometers and the spectral channel that displays the lowest reflectance is 10.40-12.50 micrometers.
Figure 3
Vegetation:
The spectral channel that displays the highest reflectance is 0.77-0.90 micrometers and the spectral channel that displays the lowest reflectance is 10.40-12.50 micrometers. Vegetation gave the highest and lowest reflectance at the spectral channels specified in question 3 because, in the visible spectrum there is high absorption in the visible spectrum because chlorophyll in plant leaves use the viable light energy in photosynthesis, so there is low absorption in the visible spectrum in plants. The near infrared channel is the highest because of the structure of plant leaves, they reflect the NIR energy to not harm cell structure and protect the protein content.
The spectral channel that displays the highest reflectance is 0.77-0.90 micrometers and the spectral channel that displays the lowest reflectance is 10.40-12.50 micrometers. Vegetation gave the highest and lowest reflectance at the spectral channels specified in question 3 because, in the visible spectrum there is high absorption in the visible spectrum because chlorophyll in plant leaves use the viable light energy in photosynthesis, so there is low absorption in the visible spectrum in plants. The near infrared channel is the highest because of the structure of plant leaves, they reflect the NIR energy to not harm cell structure and protect the protein content.
Figure 4
Riparian Vegetation:
The spectral channel that displays the highest reflectance is 1.55-1.75 micrometers and the spectral channel that displays the lowest reflectance is 10.40-12.50 micrometers.
The spectral channel that displays the highest reflectance is 1.55-1.75 micrometers and the spectral channel that displays the lowest reflectance is 10.40-12.50 micrometers.
Figure 5
Crops:
The spectral channel that displays the highest reflectance is 1.55-1.75 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
The spectral channel that displays the highest reflectance is 0.77-0.90 micrometers and the spectral channel that displays the lowest reflectance is 0.63-0.69 micrometers.
The spectral channel that displays the highest reflectance is 1.55-1.75 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
The spectral channel that displays the highest reflectance is 1.55-1.75 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
The spectral channel that displays the highest reflectance is 1.55-1.75 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
The spectral channel that displays the highest reflectance is 1.55-1.75 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
The spectral channel that displays the highest reflectance is 0.63-0.69 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
The spectral channel that displays the highest reflectance is 0.45-0.52 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
The two spectral signatures that are the most different are the moving water and the airport runway. The airport runway shows high reflectance throughout and the moving water shows low reflectance throughout. The airport runway is made of black tar, the color black reflect energy, while the moving water is a lighter color, blue/clear and absorbs energy. Also moving water would give off diffuse reflectance because of the rough surface, giving it a lower reflectance than the airport runway which would give off specular reflection, giving it a higher reflection. The two spectral channels that are the most similar are the moving water and the standing water. They are the most similar because their lines on the graph are almost overlaid on top of each other. I think they are the most similar because they are made of the same material H20. They both show high reflectance in the blue giving it its blue color and low in the near infrared channel, because water absorbs most energy and has minimal reflection.
The two soils compare very different in the visible range, layers one to three. This is because of the water content difference in the soils. Water absorbs energy, so the more water the less absorption. The near infrared layers are very similar in amount of reflectance.
The spectral channel that displays the highest reflectance is 1.55-1.75 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
Figure 6
Urban Grass:The spectral channel that displays the highest reflectance is 0.77-0.90 micrometers and the spectral channel that displays the lowest reflectance is 0.63-0.69 micrometers.
Figure 7
Dry soil (uncultivated):The spectral channel that displays the highest reflectance is 1.55-1.75 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
Figure 8
Moist soil (uncultivated):The spectral channel that displays the highest reflectance is 1.55-1.75 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
Figure 9
Rock:The spectral channel that displays the highest reflectance is 1.55-1.75 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
Figure 10
Asphalt highway:The spectral channel that displays the highest reflectance is 1.55-1.75 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
Figure 11
Airport runway:The spectral channel that displays the highest reflectance is 0.63-0.69 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
Figure 12
Concrete surface (parking lot):The spectral channel that displays the highest reflectance is 0.45-0.52 micrometers and the spectral channel that displays the lowest reflectance is 0.77-0.90 micrometers.
Figure 13
All Signature Mean Plots:The two spectral signatures that are the most different are the moving water and the airport runway. The airport runway shows high reflectance throughout and the moving water shows low reflectance throughout. The airport runway is made of black tar, the color black reflect energy, while the moving water is a lighter color, blue/clear and absorbs energy. Also moving water would give off diffuse reflectance because of the rough surface, giving it a lower reflectance than the airport runway which would give off specular reflection, giving it a higher reflection. The two spectral channels that are the most similar are the moving water and the standing water. They are the most similar because their lines on the graph are almost overlaid on top of each other. I think they are the most similar because they are made of the same material H20. They both show high reflectance in the blue giving it its blue color and low in the near infrared channel, because water absorbs most energy and has minimal reflection.
Figure 14
Soil Comparison:The two soils compare very different in the visible range, layers one to three. This is because of the water content difference in the soils. Water absorbs energy, so the more water the less absorption. The near infrared layers are very similar in amount of reflectance.
Figure 15
Part 2: Resource MonitoringSection 1: Vegetation health monitoring
Figure 16
Figure 16 shows a normalized vegetation index (NDVI) of the Chippewa and Eau Claire counties. The areas of bright white in the NDVI image represent areas
with high vegetation. I would expect to
find forests there. In
the black areas there is zero vegetation and in the grey areas there is moderate amounts of
vegetation. In the NDVI most of the
areas that are black are water bodies and have no vegetation making them black. I would expect the grey areas to be areas
with a little of vegetation, like grasslands.
Figure 17
Figure 17 is a map of the NDVI image pictured in Figure 16.
Section 2: Soil health monitoring
Figure 18
Figure 18 shows a ferrous mineral ratio image of the Chippewa and Eau Claire counties. The white areas represent areas with higher concentration of ferrous minerals. The black areas represent areas with no ferrous minerals and the grey areas represent areas with moderate concentrations of ferrous minerals. The distribution of ferrous minerals in Eau Claire and
Chippewa counties show that ferrous minerals are present highest along the
banks river waterways. The ferrous
mineral concentration is lowest in forested areas and areas with high amounts
of vegetation.
Figure 19
Figure 19 is a map of the ferrous minerals image pictured in Figure18.
Sources:
Satellite image is from Earth Resources Observation and Science Center, United States Geological
Survey.