Quantifying Land Cover Changes in Oklahoma

STUDENT HANDOUT

IntroductionChange detection tools enable us to compare satellite data from different times to assess damage from natural disasters, characterize climatic and seasonal changes to the landscape, and understand the ways in which humans alter the land. In this exercise, you will study a real-world situation in which change detection techniques are applied to track regional growth in part of the Tulsa Metropolitan Statistical Area (MSA). The Tulsa MSA is comprised of seven counties: Creek, Okmulgee, Osage, Pawnee, Rogers, Tulsa and Wagoner. According to Tulsa Where Business Grows (http://www.growmetrotulsa.com/general/496/demographics) the MSA has a population of approximately 946,000 representing 25% of the population in the state of Oklahoma. In the past few years Tulsa’s economy has grown steadily resulting from a strong energy sector (http://ww3.tulsachamber.com/upload/file/Economic%20Development/2013%20Economic%20Profile.pdf).

Municipal and regional planning officials often need to understand the urban growth pattern for a regional area in order to manage urban policies regarding utilities and services. Additionally, planning officials want to predict urban growth trends in order to prepare for future growth and development. Satellite data, combined with GIS data, provide a means for quantifying past urban growth and can be used to develop models to predict future growth.

In this exercise, you will use Landsat data for the Tulsa MSA to look for changes in urbanization growth over a ten year period. The scenes for this exercise have already been downloaded for you in the data package that comes with this module, along with a shapefile defining the study area. You will be using Landsat 5 scenes from 2004, and 2010 so that you can quantify changes over that time period.

Part 1: Prepare your data set.

Step 1. Download the Data Create a “Landcover” folder in your workspace.

Create two new folders within the Landcover folder called LT50270352004251PAC01 and LT50270352010123PAC01.

Download the LT50270352004251PAC01.tar.gz and LT50270352010123PAC01.tar.gz files to their associated folders that you just created.

Developed by the Integrated Geospatial Education and Technology Training (iGETT) project, with funding from the National Science Foundation (DUE-1205069) to the National Council for Geographic Education. Opinions expressed are those of the author and are not endorsed by NSF. Available for educational use only. See http://iget-t.delmar.edu for additional remote sensing exercises and other teaching materials. Created 2014.

Hint: Having trouble understanding the band file names? This guide will help.

Notice the satellite images come with very long file names. The file names are long for a reason. They contain information about the image. Figure 1 illustrates what each of the letters and numbers mean. The first two letters indicate the satellite and sensor. LC – Landsat Combined – image combines both the MSS and Thermal bands.LT – Landsat Thematic Mapper – image taken from Thematic Mapper sensorLO – Landsat OLI –image taken from the Operational Land Imager sensorLE – Landsat ETM+ - image taken from the Enhanced Thematic Mapper + sensor

The number following the first two letters indicates the satellite number. For example, LT5 refers to a Landsat 5 Thematic Mapper image.

The next six digits refer to the row and path number. Tulsa’s row and path number are 027 and 035 respectfully. This is followed by three digits representing the Julian Day.

Unzip the downloaded tar.gz file two times within the LT50270352004251PAC01 folder. The first time you unzip the file the .gz (gnu zip) is removed and a single .tar file is created. Unzip the .tar file again to yield the individual bands.

Once the file has been fully unzipped you will see many individual files with _B1 through _B8 added on to the file name. These refer to the Bands. For example, _B1 represents Band 1.

Repeat the unzipping of the Landsat 5 2010 scene. Download the Tulsa_msa.shp zipped file to your Landcover folder. Unzipp the shapefile.

Step 2. Open ArcMap and look at the images. In ArcMap, you can use the ArcCatalog window to add bands 1 through 5 and band 7 of the Path 27, Row 35, 2004 Landsat scene. (If your ArcCatalog window is not available, click Windows> Catalog).

Navigate to the Landcover folder, expand it, and expand the LT50270352004251PAC01 folder.

Select and drag band 1 into your map. Repeat for bands 2 through 5 and band 7.

The bands should appear in your ArcMap Table of Contents. If your map takes a long time to redraw after the addition of each layer, click the Pause Drawing button in the bottom left corner of the map window. This will stop the map from redrawing.

Add the Tulsa_msa shapefile to your map document. This will give you a reference as to where the images are in relation to our study area.

Save your map document as “Landcover.”

Step 3. Create a geodatabase to hold your work. In ArcCatalog, right click on the Landcover folder and select New > File Geodatabase. Name

the new geodatabase “Landcover.”

Import the Tulsa_msa shape file into your geodatabase as a feature class. Call the output feature class “TulsaMSA.”

2

Remove the Tulsa_msa shapefile from the map document.

Display the TulsaMSA feature class as hollow, with an outline line color of Solar Yellow and an outline width of 1.

Step 4. Create composite images for the scenes.A composite image combines all the bands into one file called a layer stack to allow easy symbolization and band compositing.

Make sure the Spatial Analyst extension is turned on: Click Customize> Extensions. Ensure that the Spatial Analyst check box is checked and close the Extensions window.

Open the ArcToolbox window open the following tool: ArcToolbox> Data Management Tools> Raster> Raster Processing> Composite Bands. (You can also do a search for the Composite Bands tool.)

Drag all the bands (1 through 5 and Band 7) into the Input Raster field, and if necessary, re-order them. It is VERY important that the bands are in the correct order in the layer stack. Click the browse button and navigate to your Landcover folder. Name the output layer “L504comp.”

Once your L504comp composite raster has been added to ArcMap remove each of the LT50270352004251PAC01 bands from the Table of Contents. This should leave only the L504comp raster.

Step 5. Change the Raster Symbology Change the symbology (Layers > Properties> Symbology tab) for the composite image so that

Red displays Band 3, Green displays Band 2 and Blue displays Band 1.

Step 6. Copy the Raster to a New RasterDue to the path of the satellite, Landsat scenes appear to be rectangles that are rotated. This is fine except that ArcGIS requires the rasters to be drawn as upright rectangles. This results in areas outside of the image appearing black in order to create that upright rectangle. Those black areas actually are raster cells with 0 as a data value. Because digital numbers for a satellite image can actually be 0 and we will be doing

3

calculations using the value of the digital numbers, we don’t want those external cells to contain a value of 0. They should, instead, have a value of NoData. To get around this it is necessary to reclassify cells that have a 0 value to NoData. The easiest way to do this is to create a copy of the raster while specifying the border areas have a value of NoData.

Change 0 values to NoDatao In the ArcCatalog navigate to the Landcover geodatabase. Right click on the L504comp

composite image and select Export > Raster To Different Format.o In the Copy Raster dialog box save your Output Raster to the Landcover geodatabase

and call it “L504copy.”o Type a 0 for Ignore Background Value (Optional) o Select the OK box to launch the tool.

Step 7. Clip the Image to the Study Area.In this step you will clip the bands to include only our MSA study area. If you look at the MSA and compare it to the L504copy image you will quickly notice that it is greater than the satellite scene. For the sake of this exercise we will only analyze the areas of the image that fall within the MSA for this one scene. If we were to truly perform an analysis of the entire MSA for Tulsa we would need to include Row 27, Path 34 and Row 26, Path 35 scenes. [Those extra scenes would have to be clipped and the reflectance and radiance calculations performed on them separately (see Part II below). Once the calculations were done the three scenes could be mosaicked together to create one final raster. ]

Within the Search window type “Clip” and select the Clip (Data Management) option.

The Input Raster is the L504copy image. Navigate to and select the “Tulsa_msa” shapefile as the

Output Extent. Use Input Features for Clipping Geometry (optional)

should be checked. Save the output raster as “L504clip” to the Landcover

geodatabase. Click OK.

Step 8. Clean up the map document and table of contents.

Remove everything except the L504clip image and the Tulsa_msa shapefile in the table of contents. You now have a composite Landsat 5 Thematic Mapper image for only the area of interest: the Metropolitan Statistical Area for Tulsa.

Let’s now display the image so that it looks more natural. In the Table of Contents, right click on the new layer and choose Properties. Select the Symbology tab. Under "Draw raster as an RGB composite," click the down arrow for the Red channel and select band 3, the red Landsat band. The Green channel can remain as band 2, the green Landsat band. Change the Blue channel to band 1, the blue Landsat band. Click OK.

Take a moment to explore the image by zooming and panning. When you are finished, save your map.

4

Step 9. Try it yourself! Repeat steps 1 through 8 with the 2010 Landsat 5 scene. Below is a simple outline of the steps you just accomplished. When you are finished, save your map document.

Download and extract the LT50270352010123PAC01 zipped file in the Landcover folder. Create a composite image. Call it “L510comp.” Remove the black area surrounding the images by changing 0 values to NoData. Clip the mosaicked image to the MSA. Call the final clipped image “L510clip.”

Part 2: Convert Digital Number to Radiance

When examining change detection, it's important to normalize the images so that you can make comparisons between them. This involves two steps, taking the digital number (DN) values in each pixel and converting them to radiance and then to reflectance.

To do this, you will need to collect some information about your Landsat scenes to use as inputs for the equation to convert DN to radiance and then to convert radiance to reflectance. There are tables at the end of this module that you can use to find some of the values, and others are available in the metadata for your Landsat scene.

DN to RadianceIn this step, you will convert the DN in your scene to radiance, the amount of energy in watts at the satellite's sensor for each cell on the ground.

Here is the equation to convert DN to radiance.

Lλ = ((LMAXλ - LMINλ)/(QCALMAX-QCALMIN)) * (QCAL-QCALMIN) + LMINλ

L is the spectral radiance. So, LMAX and LMIN represent the highest and lowest possible values of radiance, which vary with gain state. This value is saved for each band in the MTL file saved with your Landsat scene. You can open the MTL file with WordPad or any other word processing program. It’s a good idea to note the values for all the bands in a table like the one below. At this time you do not need to worry about filling in the blanks for the Leap Year, Day of Year or below. We’ll get to that in Part 3.

5

MTL File for 2004 scene: This file comes in the Landsat package and contains information you will need to calibrate the Landsat scene.

QCALMAX and QCALMIN are the calibrated maximum and minimum cell values. These values are also listed for each band in the metadata.

QCAL = Is the digital number, or the cell value to be calibrated. So for that term in the equation, you will specify the target band, and the program will use the cell values in that band to calculate the radiance for that cell.

Using the MTL file and the tables at the end of this exercise fill in the table above for the 2004 scene.

Given the DN to radiance equation:

Lλ = ((LMAXλ - LMINλ)/(QCALMAX-QCALMIN)) * (QCAL-QCALMIN) + LMINλ

Your band math equation should look something like this, substituting the variables for their values (note that some values may be negative):

((LMAXλ – (LMINλ))/(QCALMAX – QCALMIN)) *((BAND LAYER – (QCALMIN)) + LMINλ)

For example, for Band 1 in the 2004 p27/ r35 scene, the equation would be:

((193.0-(-1.52))/(255-1))*((BAND1-1.0)+(-1.52))

In addition to the MTL file, these values can also be found in Table 1 at the end of this tutorial. For other Landsat missions, such as Landsat 7, consult this PDF document produced by NASA: Chander, Markham and Hedler. "Summary of current radiometric calibration coefficients for Landsat MSS, TM, ETM+, and EO-1 ALI sensors." Remote Sensing of Environment 113 (2009) 893–903http://landsathandbook.gsfc.nasa.gov/pdfs/Landsat_Calibration_Summary_RSE.pdf

Step 1. Add 2004 Band 1 to the Map Document.Now you will use the Raster Calculator to convert DN to radiance for each band. To do this, you will need to add the individual clipped bands to your map. If necessary, open ArcMap and open your Landcover.mxd file. In ArcMap, open the Catalog window, navigate to your Landcover geodatabase and expand the composite image for the 2004 scene that you made in Part 1 of this tutorial (L504clip). Drag Band 1 into your map.

Step 2. Calculate the Radiance.Open the ArcToolbox window and open the following tool: ArcToolbox> Spatial Analyst Tools> Map Algebra > Raster Calculator (You can also open the search box and search for Raster Calculator.).

Write the following expression in the Raster Calculator (You can also copy and

6

paste the expression if you have this document open in digital format.):((193.0-(-1.52))/(255-1))*((BAND1-1.0)+(-1.52))

Once the expression is pasted, highlight the term "Band1." Then find the Band 1 layer in the variables box and double click it. This will replace the "Band1" in the expression with the actual name of the layer.

Click the browse button, navigate to your Landcover geodatabase and name the output L504_B1_RAD.

Click Save and then click OK to execute the calculation. Wait for it to complete and for the new Band 1 radiance layer to be added to your map.

Remove the L504clip-Band 1 layer from the Table of Contents.

Step 3. Finish Calculating Radiance. Repeat this process for bands 2 through 5 and 7 in both the 2004 scene. * Repeat this process for bands 1 through 5 and 7 for the 2010 scenes. *

*See the Reminder below the table before you get to Band 7.

You will need to fill out the table below for the Landsat 5 2010 scene. At this time you do not need to worry about filling in the blanks for the Leap Year, Day of Year or below. We’ll get to that in Part 3.

REMEMBER! The values for LMAXλ and LMINλ will be different for each band, and you will need to look these up in the MTL file for each scene. Also, remember that band 6 in your composite is actually Landsat band 7. ArcMap has just renamed it to be Band 6 because there were only 6 bands that created the composite image. Name the output radiance rasters for Band 7 “L504_B7_RAD” and “L510_B7_RAD.”

Save your map. You will continue with the calibration process for reflectance in Part 3.

Part 3: Convert Radiance to Reflectance

Radiance to Reflectance:Top-of-atmosphere reflectance (ρλ) is a normalized, unitless measure of the ratio of the amount of light

7

The solar elevation and the solar zenith angle, vary with season and time of day. Images courtesy of NASA).

energy reaching the earth's surface to the amount of light bouncing off the surface and returning to the top of the atmosphere, to be detected by the satellite's sensors.

π * Lλ * d2

Reflectance: ρλ =

ESUNλ * cosθs

Lλ is, of course, the spectral radiance at the sensor's aperture, that is the radiance value calculated in the previous lab for each cell in each band. So the radiance bands you created in the last step will be the input for this term in the expression.

The variable d is the distance from the earth to the sun in astronomical units (AU). The earth's distance from the sun varies, depending on the date. To find the earth-sun distance, first use Table 2a or 2b at the end of this tutorial to determine the Julian day or "day of year" that the scene was taken (note that Table 2a should be used for scenes taken in non-leap years and Table 2b should be used scenes taken on leap years). Then use Table 3 to find earth/sun distance on that day. For example, for a scene taken on September 19, 2008 (a leap year) we use Table 2b to determine that this was day 263 of that year. So from Table 3, we see that the earth was 1.0043 AU from the sun on day 263. Therefore, d = 1.0043.

Determine the Earth/Sun distance (d) for the 2004 and 2010 scenes and record the values in the Inputs for Conversion of DN to Radiance and Reflectance tables in the lab exercise above.

ESUNλ is the mean solar exoatmospheric irradiance. In other words, it is the mean amount of light of a particular band that makes its way to the sensor from space, without passing through the atmosphere. You could think of it as ambient light around the satellite that is picked up by the sensor. This value

doesn't change over time and is constant for each band on the Landsat 5 sensor. These values can also be found in Table 1 at the end of this tutorial.

Determine ESUNλ for the 2004 and 2010 scenes and record the values in the Inputs for Conversion of DN to Radiance and Reflectance tables in the lab exercise above.

8

For other Landsat missions, such as Landsat 7, consult this PDF document produced by NASA: Chander, Markham and Hedler. "Summary of current radiometric calibration coefficients for Landsat MSS, TM, ETM+, and EO-1 ALI sensors." Remote Sensing of Environment 113 (2009) 893–903http://landsathandbook.gsfc.nasa.gov/pdfs/Landsat_Calibration_Summary_RSE.pdf

θs is the solar zenith angle. This is the angle between the sun and the satellite, which depends on how high the sun is above the horizon, i.e. the sun's elevation. The elevation of the sun over the horizon depends on both the time of day and the season, and it is recorded when the scene is taken. To find this value, find the sun's elevation in the MTL file for the scene, and subtract the sun's elevation from 90o. For example, if the sun's elevation is 42.9050062o, the solar zenith angle is 90o - 42.9050062o = 47.0949938o .

Determine the sun's elevation in the MTL file for the 2004 and 2010 scenes and record the values in the Inputs for Conversion of DN to Radiance and Reflectance tables in the lab exercise above.

Calculate the solar zenith angle for each of the 2004 and 2010 scenes and record the values in the Inputs for Conversion of DN to Radiance and Reflectance tables in the lab exercise above.

So, for Band 1 in the 2004 p27/ r35 scene, the entire process would look like this:

The equation for converting radiance to reflectance is... π * Lλ * d2

ρλ =

ESUNλ * cosθs

Here are all the variables for Band 1 of the 2004 p27/ r35: θs = 90 - 52.31588317 = 37.68411683, and this will be converted to radians by multiplying it by

π/180 in the expression below Day of Year (leap year) = 251, so d = 1.0075 ESUNλ for band 1 from Table 1 below is 1983 Lλ is radiance band 1: L504_B1_RAD

To put it all together, you would enter the following expression in the Raster Calculator:

(3.14159 * "L504_B1_RAD " * 1.0075* 1.0075) / (1983) * Cos(37.68411683* 3.14159 / 180)

Step 1. Convert each radiance band to reflectance. To begin, open your Landcover.mxd file if you have previously

closed it. Make sure that radiance bands 1 through 5 and 7 are in the Table of Contents for both the 2004 and 2010 scenes.

Step 2. Calculate Reflectance. Open ArcToolbox> Spatial Analyst Tools> Map Algebra > Raster

Calculator. For raster calculations using trigonometric functions, you will need to set environment variables.

9

Click the Environments button, expand the Processing Extent section and set the Extent to Same as layer "L504_B1_RAD."

Expand the Raster Analysis Section, and set the cell size and mask to the "L504_B1_RAD" layer.

Click OK to apply the settings. Enter the expression below (or copy and paste it from this document) for band 1 of the 2004 p27/ r35 image.

(3.14159 * "L504_B1_RAD " * 1.0075* 1.0075) / (1983) * Cos(37.68411683* 3.14159 / 180)

Click the browse button, navigate to your project folder, name the output file “L504_B1_REF.” Click Save and then OK to execute the expression. Wait for the process to complete. The output

layer will be added to your map.

Step 3. Finish Calculating Reflectance. Repeat this process for bands 2 through 5 and 7 for the 2004 scene. * Repeat this process for bands 1 through 5 and 7 for the 2010 scene. **See the Reminder below before you get to Band 7.

REMEMBER! The value for ESUNλ will be different for each band, and you will need to look these up in Table 1 for each calculation. Remember to replace the input band with the correct one. Also, remember that band 6 in your clip composite is actually Landsat band 7. Name the output reflectance rasters for Band 7 “L504_B7_REF” and “L510_B7_REF.”

Step 4. Create Composite Reflectance Image.Now combine the reflectance bands for the 2004 scene into a new composite layer to allow easy analysis and band compositing. Make sure the Spatial Analyst extension is turned on.

If needed, click Customize> Extensions. Ensure that the Spatial Analyst check box is checked and close the Extensions window.

Open the ArcToolbox window open the following tool: ArcToolbox> Data Management Tools> Raster> Raster Processing> Composite Bands (you can also do a search for the Composite Bands tool).

Drag all the reflectance bands into the Input Raster field, and if necessary, re-order them. It is VERY important that the bands are in the correct order in the composite. Click the browse button and navigate to your Landcover geodatabase. Name the output layer "L504_REFcomp."

When you are finished, remove the original bands from your map, leaving only the composite reflectance image in your Table of Contents.

Repeat the process to create a 2010 reflectance composite called “L510_REFcomp.” Once both the composite reflectance images have loaded into ArcMap, change the symbology

so that Band 3 is Red, Band 2 is green and Band 3 is blue. Save your map.

10

Landsat 5, 2004 and 2010 images with pixel errors in Wagoner County.

Part 4: Create a Cloud Mask

Generally at this point you would have one more important step when it comes to preparing the data for analysis. It isn't always possible to get Landsat imagery that is cloud-free for the time frames needed for a particular study. While 0% cloud cover is ideal, sometimes it is necessary to use images with clouds and the shadows cast by clouds. If a study is looking at land cover, however, it is necessary to exclude the clouds and shadows from any analysis. To do this, a cloud mask can be created and applied to the scene. The cloud mask assigns the pixels marred by clouds and shadows a value of NoData. The NoData values are then not processed in later analysis. The same process can be used to mask scan line gaps in Landsat 7 data collected after May 2003 when the scan line corrector on the Landsat 7 satellite failed. For more information about the L7 scan line corrector failure, visit this NASA webpage: http://landsat.usgs.gov/products_slcoffbackground.php.

In this exercise you are working with two cloudless Landsat 5 scenes. A rare feat indeed! It would seem that you do not have to create a cloud mask. However, in the lower right hand corner of Wagoner County there is some unnatural striping of the Landsat data. These pixels should not be included in our analysis and we will use the cloud mask process to remove them.

Step 1. Prepare the Data. Insert a new Data Frame and rename it

“Mask.” Copy and paste the 2004 reflectance

composite: L504_REFcomp to the it. Make sure the reflectance composite is symbolized so that Band 3 is Red, Band 2 is green and

Band 3 is blue. Note: Since we are not looking for clouds in our images we are going to keep the symbology as a natural color band combination. This makes it easy to see those pixels that obviously have a problem. However, if you were doing this step because you were trying to mask out clouds and cloud shadows, a false color band combination (4,3,2) would be more appropriate.

To begin, you will create a polygon shapefile and digitize polygons over the pixels with errors (or clouds and their shadows if that was the goal).

In the Catalog window, navigate to your Landcover geodatabase, right click on it and choose New > Feature Class. Name the new Feature Class “pixel_mask.”

Under Types of features stored in this feature class, choose Polygon Features. Click the Next button. For the coordinate system select Projected Coordinate System> UTM> WGS1984> Northern

Hemisphere> WGS 1984 UTM Zone 14N. Click the Next button two times to keep the default settings for the next 2 dialog boxes and the

Finish button to generate the new feature class. The new empty feature class called “pixel_mask” will be added to your table of contents.

Make the pixel_mask polygons symbolized as hollow with an outline color of Solar Yellow.

11

The create features window with the pixel_mask template acitve.

Sample polygon created for the pixel_mask feature class.

Step 2. Digitize.Now you will digitize polygons over the areas where there are errors (Or in the case of an actual cloud mask you would digitize the clouds and their shadows.).

Add the Editor toolbar. Right click the pixel_mask feature class layer in your table of

contents. Choose Edit Features > Start Editing. The Create Features window will appear.

In the Create Features window, click the pixel_mask template. Under Construction Tools on the bottom of the Create Features

window, click Polygon. Zoom in, as needed, and

digitize polygons over each of the problem pixels.To finish a polygon, double click. You don't need to be very precise, just make sure to completely mask the problem pixels.

When you are finished digitizing, click Editor > Stop Editing. Click

Yes to save your edits.

Step 3. Convert Polygon Feature Class to Raster.Now you will convert the pixel_mask layer to a raster.

Open ArcToolbox and open the following tool: Conversion Tools > To Raster > Polygon to Raster.

Drag and drop the pixel_mask layer to the Input field. ArcGIS should fill in the Value Field as ObjectID.

Save the output layer to your Landcover geodatabase folder and name it “pixel_mask_ras.” Set the Cell Size to 30. Click the Environments button. Expand the Processing Extent item, and under Extent choose

"Same as layer L504_REFcomp." Expand the Raster Analysis item and set the Cell Size and Analysis Mask to Same as Layer

L504_REFcomp and L504_REFcomp respectfully. These environment variables ensure that the output raster will have the same cell size and extent as the reflectance composite layer.

Click OK. Wait for the process to complete. The pixel_mask raster layer will be added to your table of contents.

Remove the pixel_mask feature class from your map.

Note that the cells corresponding to the cloud and shadow polygons have a value of 0. By default, cells with a value of NoData are displayed as transparent.

12

To assure yourself that this new grid has the same cell size and extent as the reflectance stack, open the Properties window for the pixel_mask raster layer. In the Symbology tab, under Display NoData as, select any color. If necessary, zoom out to see the entire grid. The NoData cells should completely cover the reflectance composite layer.

Step 4. Reclassify.Now you will reclassify the mask.

Open ArcToolbox, and open the following tool: Spatial Analyst Tools > Reclass > Reclassify.

Under Input Raster, choose pixel_mask. Make sure the Reclass Field is set to Value. Under the New Values column type "NoData" for Values of 1, 2

and 3. Under the New Values column for the row that states NoData

for the Old Values, type 1. Name the output pixel_mask_null and save it to your

Landcover geodatabase. Click OK and wait for the process to complete. The new layer will be added to your map. Note

that now the error pixels have a value of NoData, so they are displayed as transparent. The remainder of the cells now have a value of 1.

Step 5. Apply the Mask.In this final step, you will apply the mask to the 2004 scene. To do this, you will need to add the individual reflectance bands you created in the previous section.

Add the following to your new Data Frame: L504_B1_REF L504_B2_REF L504_B3_REF L504_B4_REF L504_B5_REF L5_B7_REF

(Alternatively, you can add the individual bands from the reflectance composite.)

Open the Raster Calculator (Spatial Analyst Tools > Map Algebra > Raster Calculator).

To build your expression, under Layers and Variables, double click the "pixel_mask_null" layer. Click the * button once. Then under Layers and Variables, double click reflectance band 1.

Your expression should look like this:

"pixel_mask_null" * "L504_B1_REF"

Name the output L504_B1_REF_CF (the CF indicates that the image is now cloud-free—or pixel error free).

13

With the NoData cells displayed in yellow and pixel error cells shown in green, blue, and red the mask grid completely covers the 2004 reflectance composite.

Save it to your Landcover geodatabase. Click OK and wait for the process to complete. The new layer will be added to your map. The

cells corresponding to clouds and shadows now have a value of NoData. Use the Identify tool to make sure.

Step 6. Repeat the Raster Calculation For All Landsat 2004 Reflectance Bands. Repeat the raster calculation with the rest of the 2004 reflectance bands.

Step 7. Create a Pixel Error (or Cloud) Free Composite. When you have conducted the raster calculation for all of the 2004 reflectance bands, create a

composite image from them called “L504 _CFcomp.” Save it to your Landcover geodatabase. Remove all layers from your map document except L504 _CFcomp. Change the symbology to

reflect the natural band combination (i.e. 3, 2, 1) Save your map document.

Step 8. Repeat the Create a Pixel Error (or Cloud) Free Composite.Because the 2010 Landsat 5 image also has pixel errors in the same general area of Wagoner County you will need to repeat steps 1 through 7 with that scene. Below I’ve repeated the steps and given you naming conventions to help you keep it straight.

o Step 1. Prepare the Data. Use the Mask Data Frame and copy and paste the 2010 reflectance composite:

L510_REFcomp it. Symbolize it as a natural band image. Create a new feature class in the Landcover geodatabase called “pixel_mask2.” Symbolize the pixel_mask2 polygons as hollow with an outline color of Solar

Yellow.o Step 2. Digitize.

Digitize a polygon (or polygons) over the areas where there are errors.o Step 3. Convert Polygon Feature Class to Raster.

Save the output layer to your Landcover geodatabase folder and name it “pixel_mask2_ras.”

o Step 4. Reclassify. Name the output pixel_mask2_null and save it to your Landcover geodatabase.

o Step 5. Apply the Mask to all of the 2010 Reflectance bands.o Step 7. Create a Pixel Error (or Cloud) Free Composite.

Call the 2010 pixel error free composite “ L510_CFcomp.” Display it as a natural band combination.

Part 5: Use band combinations to symbolize and then compare your scenes.

In this part of the tutorial, you will use ArcMaps RGB compositing capabilities to symbolize the Landsat bands in several different combinations. This allows you to highlight different land cover types of interest using bands in both the visible and infrared Landsat bands. There are many different ways to symbolize these images through the use of band combinations. A band is assigned to each of the red, green, and blue channels of the image, and the color composite of these channels shows details in the

14

landscape pertaining to the three bands that were chosen.

The band combination 3-2-1 reveals the image in “natural color.” This combination shows the land much as the human eye would see it from above, using visible wavelengths of light--red, green and blue—each assigned to the appropriate channel. In Part 4 you were already asked to symbolize the 2010 and 2004 pixel error free composites as a natural color. However, we are going to improve upon that just a little bit.

Step 1. Modify Image Properties. Open the Layer Properties dialog box for the L504_CFcomp image. In the Properties window, double check that the bands associated with each channel are Red =

3, Green = 2, and Blue = 1. For the Stretch Type select Standard Deviation with n = 2.5. Click OK to apply the changes and dismiss the Properties window. Notice how this changes the

image.

Step 2. Repeat the Process. Repeat the symbology property changes for the L510_CFcomp image.

Step 3. Look at the Images. Turn the top layer on and off by clicking its check box in the table of contents to view the

changes between 2004 and 2010. Use the magnification tools to zoom in and inspect the images more closely.

Check it out! Can you see areas of the image that have changed during the time period? What might have caused the changes you observe between the two time frames? Were the changes natural or caused by humans?

Another common band combination is known as “false color,” where the band order is 4-3-2. This combination uses the near-infrared band (band 4) in the red channel. Since green vegetation strongly reflects near-infrared light, lush and healthy vegetation appears as a dark red instead of green; the darker the red the color, the healthier the vegetation. Now you will change the band combination for one of the image files to show the image in false color.

Step 4. False Color Composite. Right click on the 2010 image in the Table of Contents

and select Properties. Choose the Symbology tab and use the Band drop-

down menus to set the red channel to band 4, the near-infrared band. For the green channel, choose band 3, and for the blue channel, choose band 2. Click OK.

Repeat this step for the 2004 image, then examine the images to see the result. Remember, vegetation will appear red in a false color image.

Hint: Learn more about band combinations. This NASA site

15False color band combination with Landsat Band 4 (near-IR) assigned to the red channel, Band 3 (red) assigned to the green channel, and Band 2 (green) assigned to the blue channel. Note that vegetation appears

False Color Composite of downtown Tulsa.

Natural Color Composite of downtown Tulsa.

has a clear explanation: http://landsat.gsfc.nasa.gov/education/compositor/. You may also wish to consult a remote sensing and image analysis textbook for a more in-depth discussion of band combinations. Be sure you can explain the advantages of the false color combination. Why do we use band 4, the near infrared, in the false color band combination?

Try it yourself! Use some of the band combinations listed in the NASA reading above. Try your own combinations and see which features pop out as a result of your choices. Use the USGS Spectral Characteristics Viewer (http://landsat.usgs.gov/tools_viewer.php) to try to figure out why certain features are more prominent as a result of your chosen band combination. Discuss your findings with a classmate.http://web.pdx.edu/~emch/ip1/bandcombinations.html

Part 6: Supervised Classification Now you will use the supervised classification toolset to classify land cover types in both scenes to de-termine if the amount of urban growth in the MSA has changed from 2004 to 2010. Supervised classifi-cation requires you to first create a training data set identifying specific cover types of interest. ArcMap can use training samples to conduct supervised classification. The software conducts a statistical analy-sis of all the bands in the composite stack and then checks each of the other cells in the scene to deter-mine if it is similar to the training samples. It does this by analyzing the spectral characteristics of the cover types in the training data set and identify other places in the scene where those cover types occur.

Step 1. Set up a Training Set.First you will open the Image Classification Toolbar.

Click Customize > Toolbars > Image Classification to add the Image Classification Toolbar.

Use the drop-down menu to select L504_CFcomp in the Image Classification Layer menu.

Since we are interested in urban areas, in particular, we will zoom to a location where there is a large amount of urban space.

Center your image on the Tulsa’s downtown area. You should be able to clearly see the urban area within the Interdispersal Loop (IDL).

On the Standard toolbar, click the scale pull-down and choose “1:24,000.”

Inspect the image. According to James Quinn (2009) [http://web.pdx.edu/~emch/ip1/bandcombinations.html] a False Color composite image illustrates vegetation as shades of red. Urban areas are cyan and shades of light blue. In general a dark red indicates healthy vegetation. Lighter shades of red and pink indicate grassland or sparsely vegetated areas. Note that the urban area appears in various shades of light blue to cyan. Notice too that there are pink and red pixels in the False Color image indicating that they are covered with vegetation.

16

B

The Merge button can be used to combine several training samples into one class.

Others appear green indicating that they are bare. For more information on band combinations please see James Quinn’s website http://web.pdx.edu/~emch/ip1/bandcombinations.html.

Change the symbology to reflect the Natural Color Composite (3, 2, 1) combination. Go back and forth between the False and Natural Color Composite combinations. As you

inspect your 2004 False Color and Natural color composite images you will begin to get an idea as to how we can differentiate between landcover types. In order to keep this lab simple we are going to only be interested in 4 basic landcover types: water, bare ground, vegetation, and urban areas.

When doing a supervised classification the image analyst must determine which color composite best suites his/her purpose. Because we are only looking for 4 simple categories we could use either the False Color composite or the Natural Color composite for our purposes. This time, we will use the Natural Color Composite combination.

If needed, change the symbology to reflect a Natural Color Composite.

On the Image Classification toolbar, click the Training Sample Manager button . The Training Sample Manager window will open.

Click the Draw Training Sample with Polygon button on the Image Classification Toolbar.

Now click on the image to outline the urban area within the IDL of Tulsa. Zoom in, if necessary, to be sure you don't include other cover types (i.e. water or vegetation) in your sample.

Double click to finish the polygon. Note that a row is populated in the Training Sample Manager window corresponding to the polygon you've just drawn.

Draw another polygon around another urban area, and continue to do so until you've drawn polygons indicating four or five urban areas. You may need to zoom in and out or use the pan buttons to find other urban areas. You can change the color of a polygon by clicking the color patch in the Training Sample Manager window. If you make a mistake while digitizing, you can delete a polygon by clicking its number in the Training Sample Manager window and clicking the X button.

Step 2. Modify the Training Set. Since these polygons all indicate the same land cover type, we will combine them into the same training sample layer.

In the Training Sample manager window under ID, click the number “1” to highlight that row. Hold down your Shift key and click the last row in the list. This will select all the polygons you have drawn—note that each selected polygon is hatched in ArcMap.

Click the Merge button on the Training Sample Manager window toolbar. The polygons are merged into a single layer and assigned the same color.

17

Under class name, change the name of the newly merged layer to “Urban.”

Step 3. Create the Bare, Urban, Water and Vegetation Training Classes.For the following steps, you may need to refer to other imagery to be sure of your classifications. Ideally, you would have access to aerial photos or imagery for 2004 or a close year. However, for Oklahoma the best we can do is to use the current imagery provided under Basemaps from ESRI.

Click the drop-down button next to the Add Data button. Select Add Basemap and click on Imagery and the Add button. The

basemap will be added to your table of contents. To see the basemap, turn off the overlaying false color image layer.

Repeat Step 2 with polygons representing water areas. Merge them, and name the new layer “Water.”

Repeat the training process with polygons for various vegetative areas. Make sure you get dark reds and light pinks as they both represent vegetation.

Repeat the training for the greener areas of the image. Merge those classes and call them “bare.”

Hints on Digitizing Training Samples: There is no need to outline a sample feature perfectly

when you are drawing sample polygons. It is more important to include a number of adjacent cells of the same cover type.

If you are unsure of the classification, don't include it in the training sample.

Try to include 500 or more cells in each training sample class. You can reorder the samples in the Training Sample Manager window by selecting a row and

clicking the arrows on the toolbar. This allows you to group similar cover types. You may find it fastest to create many samples of a single cover type then merge them together. If you merge some polygons in error, you can unmerge them with the Split button next to the

Merge button.Step 4. Clean up the Training Sample Manger.When you are finished creating training samples and have merged, renamed and colored the samples, use the arrow buttons in the Training Sample Manager window to reorder the classes as they appear in the graphic above.

Click the Reset Class Values button: . Make sure the class values match those in the graphic to the right.

Click the save button on the Training Sample Manager toolbar to save the sample data set as a shapefile. Name the shapefile “training_samples_2004.shp” and save it in your Landcover folder (or the folder where the Landcover geodatabase is located).

Step 5. Classify the 2004 Composite Image.Now you will use the Interactive Supervised Classification tool to classify the 2004 scene. In this step, ArcMap will use a maximum likelihood classification method to analyze the spectral signatures of the

18

Example of the sample training manager with the 4 classes. Note that the Count column includes the number of cells in each sample type.

samples and classify the entire scene by comparing the spectral characteristics of each cell with the spectral statistics for each class. It assigns the cell to the class with the highest probability of a match.

If you would like to learn more, see “How Maximum Likelihood Classification works” on the ArcGIS Desktop 10 Resource Center website (http://bit.ly/max_likelihood_help).

On the Image Classification toolbar, click Classification > Interactive Supervised Classification. A new layer will appear in your map.

To make this layer permanent (it is currently saved in your temp folder), right click on it in the table of contents and choose Data > Export Data.

Save the output in your Landcover geodatabase and name it MLClass_L504. When asked, add the exported layer to the map. You may remove the temporary classification layer.

Step 6. Clean Up the New Raster. Once the MLClass_L504 layer has been added, change the symbology so that the Urban areas

are black, the water areas are blue, the bare areas are beige, and the vegetation areas are green. Step 7. Examine the results. Use transparency or the swipe tool to compare the classification layer with the satellite image.

Did ArcMap correctly classify the cover types? Why or why not? It is common to do image classification many times to achieve satisfactory results. If you were to do this classification again, what would you do differently and why?

Note: Urban areas are sometimes misclassified with bare soil. Did this happen in your classification

scene? If so, where did it happen? Why might that have happened? How might you solve the problem in the future? How would this type of problem impact your results?

There are many small, isolated patches of land cover in the classified image leading to a "noisy," speckled appearance. We will clean up the image with post-processing later.

Step 8. Clean up and save your map. You may wish to collect the classification layer and the cloud free composite layer into a group

layer called 2004. Turn off the basemap and all of the 2004 layers and save your map.

Step 9. Classify the 2010 Pixel Error Free Composite. Add the L510_CFcomp image to your map and symbolize with a false color display (3,2,1) if

you need to do so. Change the target layer in the Image Classification toolbar to the L510_CFcomp, and if

necessary, clear the old training samples from the Training Sample Manager window. Build a training data set for the 4 classes. Use the imagery basemap if needed to help you

determine what is likely on the ground at particular locations.

Things to remember as you create a 2010 training data set: Don't forget to change the target layer on the Image Classification toolbar to the 2010

reflectance composite image.

19

Be persistent and try to include at least 500 cells for each cover type. Reorder the classes in the Training Sample Manager window to match the order we used in

classifying the 2004 scene. Transparency settings or the Swipe tool may be useful in creating the training data set.

When you are finished building a training data set, save it as “training_samples_2010” and run an Interactive Supervised Classification.

If you are satisfied with the results, make the classification permanent by exporting it to your Landcover geodatabase. Name it MLClass_L510.

Examine the results and compare them with the classified 2004 scene. What differences, if any, do you see?

Group your MLClass_L510 layer with the L510_CFcomp layer. Save your map doc.

Part 7: Post-Processing Image Classification

Now we will begin post-processing of the classified image. Post-processing will remove small isolated patches from our classified image. Many of these smaller patches are erroneously classified or are too small to concern us in our analysis. This will also smooth the boundaries between patches. Post-processing involves a series of geoprocessing tools in the Spatial Analyst tools in ArcToolbox. This process is most easily done using Model Builder.

To learn more about post-processing classified imagery, visit the “Processing Classified Output” page on the ArcGIS 10 Desktop Help website.

Step 1. Set up the Workspace and the Create Toolbox.First, you will create a new toolbox to hold your new model.

If necessary, open the ArcMap document you saved at the end of Part 6. Open the catalog window, navigate to the Landcover folder. Right click on the folder and select

New> Folder. Name the new folder “Scratch.” In the catalog window right click on your geodatabase and choose New > Toolbox. Name the new toolbox “Post” and click Enter to apply the name. Right click on the Post toolbox you have just made and choose New > Model. A new model

window will appear. In the model window, click Model > Model Properties. Select the Environments tab. Scroll

down and check the box next to Workspace. Click on the Values button. If needed click on the double chevron next to Workspace to expand

the options. Set the Current Workspace to your geodatabase. Set the Scratch folder to the Scratch folder you just created in the Landcover folder.

Click OK twice.

Step 2. Build the Model for the Majority Filter. Drag and drop the Interactive Supervised Classification layer for 2004 (MLClass_2004) from

20

your table of contents into your model.

Next you will add the Majority Filter tool to your model. This tool will change the classification of small, isolated patches in the classified layer to that of the majority of neighboring cells. For example, if a single cell classified as bare is surrounded by six forest cells and two water cells, that cell will be classified as forest in the output.

Open ArcToolbox and expand Spatial Analyst Tools > Generalization. Most of the tools you will need for this model are in the Generalization toolset.

Drag and drop the Majority Filter tool into your model. When the Majority Filter tool appears in your model, use the Connect tool to connect the MLClass_L510F oval to the Majority Filter tool and select Input Raster. The layers and tools will be filled with color.

In your model, double click on the Majority Filter tool to open it. Name the Output Raster maj_filter and make sure it is saved to your Scratch folder. Leave the defaults for all other fields and click OK. Save your model by clicking Model > Save, and run it by clicking the run button. When the model is finished running, dismiss the results window, if necessary. Right click on the green maj_filter oval, and choose Add to Display to add the maj_filter layer to your map. Examine the new layer and compare it with the MLClass_L510 layer. Note that some of the speckling in the image is gone. Save your map.

Step 3. Build the Model for the Boundary Clean Tool. We'll use the Boundary Clean tool next, to smooth the edges between patches of different classes.

Drag and drop the Boundary Clean tool from the Generalization tool set into your model. Use the Connect tool to connect the maj_filter

oval to the Boundary Clean tool and select Input Raster.

In your model, double click the Boundary Clean tool to open it. Name the output raster bndry_clean and make sure it is saved in your scratch folder. Under Sorting Technique, choose Ascend.

Leave the default for the Output Raster. Uncheck the optional Run expansion and shrinking

twice. Click OK. Save your model and run it. Close the Results box.

21

Add bndry_clean to the map display by right clicking on the bndry_cln oval. Examine the results. Note that the boundaries between patches are smoother, but there are still

many isolated, small patches. We will take care of them in the next step.

Step 4. Develop the Model to Remove Small Patches.We will use three tools to eliminate all patches with 4 or fewer contiguous cells. This works out to be 38,750 square feet which is just under an acre (1 acre = 43, 560 square feet). The Region Group tool will count the number of cells in all patches. Then the Set Null tool will select patches with fewer than 4 cells and give them a null value. Then the Nibble tool will use the Set Null output to identify the small patches and will assign them values of their nearest neighbors. So, for a patch consisting of 4 cells classified as bare and surrounded by a large vegetation area, the process will reclassify the 4 cells of the small patch as vegetation.

Hint: If you close Model Builder in the middle of the tutorial, or if Model Builder crashes, do not double click the model to reopen it. Instead, right click on the model in ArcCatalog and choose Edit.

Drag and drop the Region Group and Nibble tools from the Generalization tool set into your model.

In ArcToolbox, navigate to Spatial Analyst Tools > Conditional, and drag the Set Null tool into your model.

Use the Connect tool to connect the bndry_clean oval to the Region Group tool and select Input Raster.

Double click the Region Group tool to open it. Save the output to your scratch folder and name it reg_grp. Uncheck the box next to Add link field to output. Keep the default values for all other fields and click OK.

Save your model and right click the Region Group tool and choose Run. Wait for it to execute, and if necessary, dismiss the results window.

Use the Connect tool to connect the reg_grp oval to the Set Null tool and select Input Conditional Raster.

Double click the Set Null tool to open it. Use the SQL button to open the expression builder and build

the following expression: "Count" < 4. This indicates that region groups with fewer than 4 cells will be set to null in the output.

Select the reg_grp as the Input false raster or constant value. Save the output to your scratch folder and name it

set_null_4. Click OK. Double click the Nibble tool to open it. Under Input raster,

choose the bndry_clean layer. Under Input Raster Mask, choose set_null_50.TIF. Name the Output raster class_pp.

22

Save your model and run it. When the process completes, dismiss the results window, if necessary, and add the class_pp to the display.

Examine the results.

Step 5. Reclassify.For the last step in the model, we will conduct a reclass. Recall that we are specifically interested in identifying urban areas in our study area. We will use reclassification to create a binary layer in which cells classified as urban have a value of 1. All other classes will have a value of zero.

Look

at the

pp_class layer and determine what classes (i.e., urban, vegetation, bare, water) are associated with what numbers. Make a note of this.

In ArcToolbox, navigate to Spatial Analyst Tools > Reclass, and drag the Reclassify tool into your model.

Use the Connect tool to connect the class_pp oval to the Reclassify tool and select Input Raster.

Double click the Reclassify tool to open it. Under Reclass field, choose Value. Click the Unique button if it is not grayed out, so that individual cell values appear under Old

Values in the Reclassification table. Under New Values for the bare,

vegetation, and water classes enter the new value as 0. For the urban class old value enter the new value as 1.

Name the output urban_rcls_04, and save it in your Landcover geodatabase, not your scratch folder (this is the final output of the model).

Click OK. Save your model, run it, and add the

final output to your display. Examine the results. Note that the urban class has a value of 1, while all other classes have a

value of 0.

If you had a lot of pixels along the river system erroneously assigned to the urban category it will be very apparent in this final layer. What might you do to correct this problem? Would it be best to have more classes during the classification process? Is there a way to mask out the water or rivers before you perform a classification? Would working with the False Color

23

Composite image while classification help with this problem?

Save your map.

Step 6. Rerun the Model for 2010.Now, you will run the model again using the 2010 classification layer, MLClass_L510. First, we must delete the intermediate files created when we ran the model to post-process the 2004 classification layer.

In the Model Builder window, click Model > Delete Intermediate Data. The urban_rcls_04 layer should remain in your table of contents, while the other layers produced by the model are removed.

In your model, double click the Majority Filter tool to open it. Under Input Raster, select the MLClass_L510 layer. Click OK.

In your model, double click the Reclassify tool to open it. Change the name of the Output Raster to urban_rcls_10.

Save the model and run it. It may take a few minutes to execute all the processes in the model. When it is finished, dismiss the results window, if necessary, and add the urban_rcls_10 layer to your model.

Compare it to the urban_rcls_04 layer.

Do you see any areas of change between the two time frames? Have urban areas expanded?

Save your map document.

Part 8: Change Detection

Now you will use the ArcGIS Minus tool to identify changes in urban development between 2004 and 2010. Using simple subtraction (later image minus initial image) ArcGIS will compare the classified, scenes to determine whether the amount of urban land has increased, decreased or remained the same.

Step 1. Open ArcMap. If necessary, open the Landcover ArcMap document. Add the urban_rcls_2010 and urban_rcls_04 layers you created in the previous section.

Also, add the Imagery basemap if needed.

Step 2. Open the Minus Tool. Open ArcToolbox and open the following tool: Spatial

Analyst Tools > Math > Minus. Minus is a very simple

24

tool. It simply subtracts the values of overlapping cells in a pair of grids. Under Input raster of constant value 1, choose urban_rcls_10 (the later image). Under Input raster of constant value 2, choose urban_rcls_04 (the initial image). Name the output raster urban_difference. Make sure it saves to your Landcover geodatabase. Click OK and wait for the process to run.

Step 3. Examine the Urban_difference Raster.When the new layer is added to your map, take a moment to examine it. Note that there are three pixel values: 1, 0 and -1. Recall that land classified as urban in each layer had a value of 1, while anything else had a value of 0. So, the possible combinations are...

Step 4. Calculate Area.Now you will add a field to the attribute table and use it to calculate the area of the various classes.

Open the attribute table for the urban_difference layer. In the Table Options pull-down menu, select Add Field. In the Add Field window, name the new field Area_Acres. Under Type, and select Long Integer. Click OK. When the new field appears in the attribute table, right click on the field header and choose

Field Calculator. In the Field Calculator window, build the following expression: ([COUNT] * 900) / 4047.

Each cell in the grid is 30m X 30m = 900 m2, so we multiply the number of cells, the count, by 900 to get the total area in square meters. Dividing by 4047 converts the area to acres, a common unit used in agricultural statistics.

Step 5. Consider the results. In the period between 2004 and 2010, according to your analysis, was there a net gain or loss of urban land? When you examine the difference layer and the Landsat images, does this result make sense? Why or why not? How much area indicated a loss in urban area? Does this seem realistic to you? Why or why not? Look over the urban difference layer and compare it to the World Imagery Basemap. What primarily accounts for those negative 1 values? How might you fix this problem if you were to redo this lab?

Step 6. Reclassify. Open the Reclassify tool. Under Input Raster select urban_rcls_2004. Under New Values, type "NoData" in the first row, and

"2004" in the second row, as shown here.

25

Name the output urban_2004 and save it in your geodatabase. Click OK and wait for the process to run. Repeat the reclass with the urban_rcls_2010 layer. For New Values enter 2010 in the second

row as the new value for 1.

Step 7. Create Vector Datasets. Open the following tool: Conversion Tools > From

Raster > Raster to Polygon. Under Input Raster, select urban_2004. Make sure the Field is set to Value. Name the output urb_poly_2004.shp, and save it to your

geodatabase. Check the box for the option to Simplify polygons. Click OK and wait for the process to run.

A polygon layer will appear in your map with polygons indicating the urban land cover class.

Open the attribute table for the new polygon layer, and add a long integer field named Area_Acres.

When the new field appears, right click the field header and choose Calculate Geometry. Click yes, and the Calculate Geometry window will open. Under Property, choose Area. Under Units, choose Acres. Click OK. ArcMap will calculate the acreage for each of the polygons. Repeat this process for the 2004 layer.

Step 8. Explore the results. You can right click on the header of the Area_Acres fields and select Statistics to learn more about the data. Look at the Sum and the Mean for the urban areas and each time frame. What do these numbers tell you? Has there been a lot of urban growth within the time frame? Do you see evidence of such change when you examine the original Landsat image?

Step 9. Generate a Map Layout.Create a layout showing changes you have identified in urban growth. Your layout should include the urban difference layer, as well as, the urban areas defined for 2004 and 2010. Include a table derived from the attribute table of your difference layer and text discussing your findings. Also, include a brief statement about the potential sources of error in your analysis and recommended next steps.

Save your layout, and export a PDF.

Optional follow-up activities: 1) If you live in the Tulsa MSA, consider a ground truthing expedition with GPS and spectrometer. 2) Try refining the classification process so that bare and water pixels are not classified as urban.3) Try classifying an image using the False Color Composite as your guide for classification.4) Analyze an earlier or more recent Landsat image of the same area and look for additional

changes.

26

Table 1 from Chander, Markham and Hedler. "Summary of current radiometric calibration coefficients for Landsat MSS, TM, ETM+, and EO-1 ALI sensors." Remote Sensing of Environment 113 (2009) p. 896, Table 3 http://landsathandbook.gsfc.nasa.gov/pdfs/Landsat_Calibration_Summary_RSE.pdf

27

DATE JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC1 1 32 60 91 121 152 182 213 244 274 305 3352 2 33 61 92 122 153 183 214 245 275 306 3363 3 34 62 93 123 154 184 215 246 276 307 3374 4 35 63 94 124 155 185 216 247 277 308 3385 5 36 64 95 125 156 186 217 248 278 309 3396 6 37 65 96 126 157 187 218 249 279 310 3407 7 38 66 97 127 158 188 219 250 280 311 3418 8 39 67 98 128 159 189 220 251 281 312 3429 9 40 68 99 129 160 190 221 252 282 313 34310 10 41 69 100 130 161 191 222 253 283 314 34411 11 42 70 101 131 162 192 223 254 284 315 34512 12 43 71 102 132 163 193 224 255 285 316 34613 13 44 72 103 133 164 194 225 256 286 317 34714 14 45 73 104 134 165 195 226 257 287 318 34815 15 46 74 105 135 166 196 227 258 288 319 34916 16 47 75 106 136 167 197 228 259 289 320 35017 17 48 76 107 137 168 198 229 260 290 321 35118 18 49 77 108 138 169 199 230 261 291 322 35219 19 50 78 109 139 170 200 231 262 292 323 35320 20 51 79 110 140 171 201 232 263 293 324 35421 21 52 80 111 141 172 202 233 264 294 325 35522 22 53 81 112 142 173 203 234 265 295 326 35623 23 54 82 113 143 174 204 235 266 296 327 35724 24 55 83 114 144 175 205 236 267 297 328 35825 25 56 84 115 145 176 206 237 268 298 329 35926 26 57 85 116 146 177 207 238 269 299 330 36027 27 58 86 117 147 178 208 239 270 300 331 36128 28 59 87 118 148 179 209 240 271 301 332 36229 29 88 119 149 180 210 241 272 302 333 36330 30 89 120 150 181 211 242 273 303 334 36431 31 90 151 212 243 304 365

Table 2a: JULIAN DAY TABLE, NON-LEAP YEAR (for leap year, see 2b below)

From http://amsu.cira.colostate.edu/julian.html

28

DATE JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC1 1 32 61 92 122 153 183 214 245 275 306 3362 2 33 62 93 123 154 184 215 246 276 307 3373 3 34 63 94 124 155 185 216 247 277 308 3384 4 35 64 95 125 156 186 217 248 278 309 3395 5 36 65 96 126 157 187 218 249 279 310 3406 6 37 66 97 127 158 188 219 250 280 311 3417 7 38 67 98 128 159 189 220 251 281 312 3428 8 39 68 99 129 160 190 221 252 282 313 3439 9 40 69 100 130 161 191 222 253 283 314 34410 10 41 70 101 131 162 192 223 254 284 315 34511 11 42 71 102 132 163 193 224 255 285 316 34612 12 43 72 103 133 164 194 225 256 286 317 34713 13 44 73 104 134 165 195 226 257 287 318 34814 14 45 74 105 135 166 196 227 258 288 319 34915 15 46 75 106 136 167 197 228 259 289 320 35016 16 47 76 107 137 168 198 229 260 290 321 35117 17 48 77 108 138 169 199 230 261 291 322 35218 18 49 78 109 139 170 200 231 262 292 323 35319 19 50 79 110 140 171 201 232 263 293 324 35420 20 51 80 111 141 172 202 233 264 294 325 35521 21 52 81 112 142 173 203 234 265 295 326 35622 22 53 82 113 143 174 204 235 266 296 327 35723 23 54 83 114 144 175 205 236 267 297 328 35824 24 55 84 115 145 176 206 237 268 298 329 35925 25 56 85 116 146 177 207 238 269 299 330 36026 26 57 86 117 147 178 208 239 270 300 331 36127 27 58 87 118 148 179 209 240 271 301 332 36228 28 59 88 119 149 180 210 241 272 302 333 36329 29 60 89 120 150 181 211 242 273 303 334 36430 30 90 121 151 182 212 243 274 304 335 36531 31 91 152 213 244 305 366

Table 2b: JULIAN DAY TABLE, LEAP YEAR (for non-leap year, see 2a above)

From http://amsu.cira.colostate.edu/julian.html

29

Table 3: Earth/ Sun Distanced DOY d DOY d DOY d DOY d DOY d

1 0.98331 61 0.99108 121 1.00756 181 1.01665 241 1.00992 301 0.993592 0.98330 62 0.99133 122 1.00781 182 1.01667 242 1.00969 302 0.993323 0.98330 63 0.99158 123 1.00806 183 1.01668 243 1.00946 303 0.993064 0.98330 64 0.99183 124 1.00831 184 1.01670 244 1.00922 304 0.992795 0.98330 65 0.99208 125 1.00856 185 1.01670 245 1.00898 305 0.992536 0.98332 66 0.99234 126 1.00880 186 1.01670 246 1.00874 306 0.992287 0.98333 67 0.99260 127 1.00904 187 1.01670 247 1.00850 307 0.992028 0.98335 68 0.99286 128 1.00928 188 1.01669 248 1.00825 308 0.991779 0.98338 69 0.99312 129 1.00952 189 1.01668 249 1.00800 309 0.9915210 0.98341 70 0.99339 130 1.00975 190 1.01666 250 1.00775 310 0.9912711 0.98345 71 0.99365 131 1.00998 191 1.01664 251 1.00750 311 0.9910212 0.98349 72 0.99392 132 1.01020 192 1.01661 252 1.00724 312 0.9907813 0.98354 73 0.99419 133 1.01043 193 1.01658 253 1.00698 313 0.9905414 0.98359 74 0.99446 134 1.01065 194 1.01655 254 1.00672 314 0.9903015 0.98365 75 0.99474 135 1.01087 195 1.01650 255 1.00646 315 0.9900716 0.98371 76 0.99501 136 1.01108 196 1.01646 256 1.00620 316 0.9898317 0.98378 77 0.99529 137 1.01129 197 1.01641 257 1.00593 317 0.9896118 0.98385 78 0.99556 138 1.01150 198 1.01635 258 1.00566 318 0.9893819 0.98393 79 0.99584 139 1.01170 199 1.01629 259 1.00539 319 0.9891620 0.98401 80 0.99612 140 1.01191 200 1.01623 260 1.00512 320 0.9889421 0.98410 81 0.99640 141 1.01210 201 1.01616 261 1.00485 321 0.9887222 0.98419 82 0.99669 142 1.01230 202 1.01609 262 1.00457 322 0.9885123 0.98428 83 0.99697 143 1.01249 203 1.01601 263 1.00430 323 0.9883024 0.98439 84 0.99725 144 1.01267 204 1.01592 264 1.00402 324 0.9880925 0.98449 85 0.99754 145 1.01286 205 1.01584 265 1.00374 325 0.9878926 0.98460 86 0.99782 146 1.01304 206 1.01575 266 1.00346 326 0.9876927 0.98472 87 0.99811 147 1.01321 207 1.01565 267 1.00318 327 0.9875028 0.98484 88 0.99840 148 1.01338 208 1.01555 268 1.00290 328 0.9873129 0.98496 89 0.99868 149 1.01355 209 1.01544 269 1.00262 329 0.9871230 0.98509 90 0.99897 150 1.01371 210 1.01533 270 1.00234 330 0.9869431 0.98523 91 0.99926 151 1.01387 211 1.01522 271 1.00205 331 0.9867632 0.98536 92 0.99954 152 1.01403 212 1.01510 272 1.00177 332 0.9865833 0.98551 93 0.99983 153 1.01418 213 1.01497 273 1.00148 333 0.9864134 0.98565 94 1.00012 154 1.01433 214 1.01485 274 1.00119 334 0.9862435 0.98580 95 1.00041 155 1.01447 215 1.01471 275 1.00091 335 0.9860836 0.98596 96 1.00069 156 1.01461 216 1.01458 276 1.00062 336 0.9859237 0.98612 97 1.00098 157 1.01475 217 1.01444 277 1.00033 337 0.9857738 0.98628 98 1.00127 158 1.01488 218 1.01429 278 1.00005 338 0.9856239 0.98645 99 1.00155 159 1.01500 219 1.01414 279 0.99976 339 0.9854740 0.98662 100 1.00184 160 1.01513 220 1.01399 280 0.99947 340 0.9853341 0.98680 101 1.00212 161 1.01524 221 1.01383 281 0.99918 341 0.9851942 0.98698 102 1.00240 162 1.01536 222 1.01367 282 0.99890 342 0.9850643 0.98717 103 1.00269 163 1.01547 223 1.01351 283 0.99861 343 0.9849344 0.98735 104 1.00297 164 1.01557 224 1.01334 284 0.99832 344 0.9848145 0.98755 105 1.00325 165 1.01567 225 1.01317 285 0.99804 345 0.9846946 0.98774 106 1.00353 166 1.01577 226 1.01299 286 0.99775 346 0.9845747 0.98794 107 1.00381 167 1.01586 227 1.01281 287 0.99747 347 0.9844648 0.98814 108 1.00409 168 1.01595 228 1.01263 288 0.99718 348 0.9843649 0.98835 109 1.00437 169 1.01603 229 1.01244 289 0.99690 349 0.9842650 0.98856 110 1.00464 170 1.01610 230 1.01225 290 0.99662 350 0.9841651 0.98877 111 1.00492 171 1.01618 231 1.01205 291 0.99634 351 0.9840752 0.98899 112 1.00519 172 1.01625 232 1.01186 292 0.99605 352 0.9839953 0.98921 113 1.00546 173 1.01631 233 1.01165 293 0.99577 353 0.9839154 0.98944 114 1.00573 174 1.01637 234 1.01145 294 0.99550 354 0.9838355 0.98966 115 1.00600 175 1.01642 235 1.01124 295 0.99522 355 0.9837656 0.98989 116 1.00626 176 1.01647 236 1.01103 296 0.99494 356 0.9837057 0.99012 117 1.00653 177 1.01652 237 1.01081 297 0.99467 357 0.9836358 0.99036 118 1.00679 178 1.01656 238 1.01060 298 0.99440 358 0.9835859 0.99060 119 1.00705 179 1.01659 239 1.01037 299 0.99412 359 0.9835360 0.99084 120 1.00731 180 1.01662 240 1.01015 300 0.99385 360 0.98348

361 0.98344362 0.98340363 0.98337364 0.98335365 0.98333366 0.98331

DOY

DOY = Day of Year or Julian Day; d = Earth/ Sun Distance in Astronomical Units

Table 3 from Chander, Markham and Hedler. "Summary of current radiometric calibration coefficients for Landsat MSS, TM, ETM+, and EO-1 ALI sensors." Remote Sensing of Environment 113 (2009) p. 896, Table 6Http://landsathandbook.gsfc.nasa.gov/pdfs/Landsat_Calibration_Summary_RSE.pdf