Brock University – Department of Geography and Tourism Studies
GEOG 2P07: Introduction to Geospatial Technologies
Lab Assignment #5: Remote Sensing Analyses
Date Assigned: The week beginning November 20, 2017
Date Due: The week beginning December 4, 2017
Special Note for Students:
Lab assignment 5 is due by 4:30 p.m. on the weekday of your lab session the week beginning December 4. Please submit
your lab to your TA’s mailbox, located in the Department of Geography and Tourism Studies by 4:30 p.m. on that date.
You will not meet in lab the week of December 4.
The main objective of this lab is to provide you with practical experience extracting information from remote-sensing
data using ENVI 5.3 (32-bit version). Subsequent interpretation of the results of data analyses is also an important part
of this process. In this lab, you will make use of GeoEye-1 data acquired over St. Catharines on July 31, 2009.
By the end of this lab you will be able to:
1. assess the quality of the image data;
2. apply a spectral transformation (NDVI) to satellite image data to extract meaningful information about
vegetation from those data;
3. interpret the results of a spectral transformation of satellite image data; and
4. export the results to ArcMap 10.5.1 to produce a cartographically sound map.
Data Use License Agreement (GeoEye-1 Data)
Important Student Reminder:
RESTRICTION/COPYRIGHT: These data are licensed to Brock University faculty, staff and students for teaching and research
purposes. At the end of this lab assignment, you need to attach a signed copy of the Data Use License Agreement provided to you
as a PDF on Sakai. Failure to include this agreement with your lab assignment will result in a grade deduction.
You must properly cite all geospatial datasets used in this assignment. The sample citation format for these data is provided to you
St. Catharines sections: po_552588 [GeoEye-1 Scene 2009073116134721603031606123] [electronic resource]. (2009). Lethbridge,
AB: Alberta Terrestrial Imaging Centre. Available: Brock University Map Library Controlled Access
po_552588_grn_0000000.tif, po_552588_red_0000000.tif, and po_552588_nir_0000000.tif (Accessed current date).
During the lab, your TA will first introduce this assignment to you. He/she will then demonstrate some of the tools you
will need to complete this assignment. Please note that you can access help with the ENVI software by going to the Help
pull-down menu and clicking on Contents (you can also select the Index tab and search for a tool by name). Note that
under the Help menu you will also find a number of tutorials that are available for your use.
Lab Assignment Instructions
Please carefully follow the instructions provided below. You will need to spend time working with the software, course
text, lecture materials, and other resources available to you to answer the questions in this lab assignment.
To access these digital data files, you must carefully follow the instruction provided below:
First, navigate to the P:/GEOG 2P07/ drive and copy the folder (labelled Lab5_Data) to your space (based on your
username) on the W: drive (navigate to W:/GEOG_2P07) or USB. You can copy these files from any of the PC computer
labs on campus (Note: you cannot access these files or the ENVI 5.3 software from the James Gibson Library or the Map,
Data and GIS Library). To do this, you can drag and drop (or copy) the file folder called Lab5_Data from the P: drive (i.e.,
P:/GEOG 2P07/Lab5_Data) to your space on W:/GEOG_2P07/username/. Once you have copied this folder into your W:
drive space or USB, start working on the assignment [the file is called GEOG_2P07_F2017_Lab_Assignment_5.pdf and is
posted on Sakai].
***NOTE: You will have to import the TIFF files into ENV 5.3 and then create an ENVI file prior to any image
processing. The instructions for this were provided in lab assignment #4.
Remember to save a copy of all your files on a separate device (i.e., USB key). You must also properly cite your
information sources and provide your answers in complete sentence format or marks will be deducted from your final
Question 1: Data Quality [12/60 marks]
a) Are there any clearly visible anomalies in the image data itself (something unexpected, different, or peculiar)? If so,
what are the geographic coordinates of the anomaly/anomalies? What might cause this/these anomaly/anomalies?
Explain. [5/12 marks]
b) Do data quality problems (such as data anomalies caused by atmospheric effects or instrument errors) have
implications on further image processing and analysis? Explain. Use at least two credible sources of information to
support your claims. (Reminder: Do not use online dictionaries, software manuals, Wikipedia, or lecture notes to
support your claims.) [7/12 marks]
Question 2: Transformations of Image Data: Vegetation Indices [48/60 marks]
Using bands 3 (red) and 4 (NIR) of the GeoEye-1 data, produce an NDVI image of St. Catharines, Ontario, using ENVI 5.3
Creating a Spatial Subset and NDVI image in ENVI:
First, launch the 32-bit version of ENVI 5.3 and load the GeoEye-1 image file by going to File > Open…(navigate to the
location where your lab 5 data file is stored). Next, you must subset (resize) the data to a size of 4,321 by 4,741 pixels.
The spatial subset tool in ENVI allows you to focus on a specific geographic area within an image scene. To do this, you
must go to the Toolbox panel and select Raster Management > Resize Data. From the Resize Data Input File dialogue
box, select the GeoEye-1 data file and click on the Spatial Subset button. In the Select Spatial Subset dialogue box, copy
the dimensions of the spatial subset as shown below (shown in the rows labeled Samples and Lines) and click OK.
Click OK again. Next, a Resize Data Parameters dialogue box will appear. You must provide a new filename (no spaces)
for this newly created ENVI subset image file. When you are finished creating a new filename and assigning a location,
click OK in the Resize Data Parameters dialogue box. This newly created ENVI file, along with the bands of data
associated with that file, will appear under Layer Manager. [Note: If you right click on the band names and select
Rename Item… you can assign a new name (or modify the existing name) to each of these bands.]
Once you have resized the image to 4,321 by 4,741 pixels (rather than 4,321 by 8,753 pixels), apply a spectral
transformation to these data. To do this, you must go to the Toolbox panel and select Spectral > Vegetation > NDVI. In
the NDVI Calculation Input File dialogue box, select your new subset file of GeoEye-1 data. In the NDVI Calculation
Input File dialogue box, click OK. In the NDVI Calculations Parameters dialogue box, leave the default Input File Type as
Landsat TM (since the red reflected energy is contained in band 3 and the NIR is contained in band 4 in this GeoEye-1
subset file). Choose an Output Filename for the results (saved on your W: drive space or USB) by clicking the Choose
button. In the newly-opened Output Filename dialogue box, navigate to the W: drive or USB in the left column and
provide an intuitive name for your file in the field File name: Click Open. When you return to the NDVI Calculations
Parameters dialogue box, click OK. The NDVI result will appear as a layer in the Layer Manager panel. It is customary to
display an NDVI image as a gray scale image.
Note: Like ArcGIS, ENVI can display raster data in layers. To view the spatial subset image that you created and the NDVI
image alone, you may have to turn off other layers. Use the checkboxes in the Layer Manager panel to turn on and off
a) Why does the normalized difference vegetation index (NDVI) make use of the red and near-infrared (NIR) bands to
assess vegetation health (i.e., why does it not use green and red or blue and red or some other combination of
bands)? Please provide credible information sources for your answer. [6/48 marks]
b) In a properly formatted table (i.e., one that is produced in MS Word or in an MS Excel spreadsheet as opposed to an
ENVI screen capture), list the minimum, maximum, mean, and standard deviation for the NDVI image (please round
numeric values to four decimal places). What does this information tell you about this dataset? Discuss. (Be
specific!) [6/48 marks]
How to Generate Statistics on Different Layers of Data in ENVI (e.g., the layer could represent a band of image data or a
vegetation index image):
From the Toolbox panel, select Statistics > Compute Statistics. When the Compute Statistics Input File dialogue box
appears, select the file name containing your NDVI image and click OK. In the Compute Statistics Parameters dialogue
box, you must make sure you place a check mark in the empty box to the left of the word Histograms. In the bottom
part of this same box, you should also place a check mark in the empty box to the left of the words Output to a Statistics
File and Output to Text Report File. Click Choose and save both of these files to your W: drive space (or to a USB key).
Enter file names (without spaces) for each of these two files. [Note that if you want to view the statistics later, you do
not need to compute them again. Simply go to the Toolbox panel and select Statistics > View Statistics File and navigate
to the statistics file name [the file will have a *.sta file extension] that you are about to create here. The *.txt file you are
about to create can also be imported into MS Excel for proper formatting.] To generate these statistics files, click OK. A
Statistics View window should appear.
To view a histogram of your data (onscreen), click on the Select Plot tab and select the histogram number you
would like to view (e.g., Histogram Band 1). Please note that the histogram should be saved in a *.jpg file format. To
save the histogram for the NDVI image, click on the Export tab and select Image… In the Please Select a File for Writing
dialogue box, change the Save as type to *.jpg; *.jpeg. Next, enter a file name (do not add spaces to this filename. The
file will be assigned a *.jpg extension), and click OK. Please note that you can also import these histograms into MS Excel
by saving them in ASCII file format.
c) Using the Cursor/location tool, complete the following table (you will need to recreate this table in your lab report):
Land-use/Land-cover Type Typical NDVI value(s) Tone in NDVI Image
(Light, Medium, or Dark)
Water (Lake Ontario)
Agricultural Fields (Non-vegetated/bare)
Agricultural Fields (vegetated/planted)
d) How many pixels does this NDVI image contain? [1/48 Marks]
e) How many pixels contain NDVI values lower than -0.75? [1/48 Marks]
f) How many pixels contain NDVI values higher than 0.50? [1/48 Marks]
g) What proportion (as a %) of the image has NDVI values greater than 0.60? [1/48 Marks]
h) What proportion (as a %) of the image has NDVI values less than -0.52? [1/48 Marks]
i) Do you think your answers to questions 2e) to 2h) above would change if this image were
acquired in October rather than July? If so, would they be higher/lower? Explain. [3/48 Marks]
j) Produce a properly formatted histogram of the NDVI results and include it in your lab report. Clearly describe the
shape and main characteristics of this histogram. Do these data appear to be normally distributed? Explain. (Be
specific) [5/48 marks]
k) Produce a cartographically sound map of your NDVI results in ArcMap. [10/48 marks]
How to Export an ENVI Image (or Data Layer) to ArcGIS:
Launch ArcMap 10.5.1 from the Start Menu, and open a blank map (you may need to click on New Maps > Blank Map
and click OK). You must also Launch the 32-bit version of ENVI 5.3. This is found in the Start Menu > ENVI 5.3 > ENVI for
ArcGIS > ENVI 5.3 (32-bit). Open the NDVI file in ENVI. Once your NDVI file is loaded, go to File > Data Manager. In the
Data Manager dialogue box select the image and click the ArcMap icon in the top toolbar. Once clicked, go to ArcMap. A
Create Pyramid dialogue box will show. Click Yes and the image will load. Save the ArcMap project in your W: drive or
on a USB key. Using the cartographic knowledge and skills developed from previous lab assignments produce your map
for question 2k) with this image.
l) How could you make use of the results from this NDVI analysis? (You must give credit to your information sources in
answering this question).