11/03/2025
This week in lab, the class did something slightly different. Instead of working with a scan, there was a pre-planned dataset used in order to better understand the Arc GIS Pro software. This dataset was used from flights conducted in Doak Forest, an area of the Midwest that Professor Hupy had flown in order to see the difference between areas before and after a controlled burn. While we were looking at these differences in the imagery before and after the burn, we were also conducting practice on how to use the software, which was very important as we would have to use it more and more toward the end of the semester. In this lab, the first thing completed was a review of color spectrum reflectance bands, which can be seen below in Figure 1.
Figure 1: Multi-spectral bands
After the data was downloaded from the class data folder, each color band was separated as an image layer on its own. Each color band required a screenshot to be put in the report, and each color band made different things stand out. For instance, shown below is the blue color band in Figure 2. This band made such things as the walkways shine bright and the tree shadows dim low. The affected areas changed with each color band as well as with infrared imagery. As an extreme example, as seen below in Figure 3, the long wave infrared (LWIR) band shows the wild difference between burned areas and non burned areas that the blue band does not in the post burn dataset.
Figure 2: Blue band pre burn
Figure 3: Thermal LWIR post burn
After this initial set of concepts was understood, pre and post brun images began to be directly compared. This was completed by downloading both file sets to the project and then switching back and forth between images. One way this was examined was through the use of false color infrared imagery. These images were created by reassigning the different color bands in an image to be a color they are not, which makes for very high contrast and attention grabbing images. Figure 4 below shows a false color infrared image of the study area before the burn, and this image can be directly compared to Figure 5, which is a false color infrared image taken after the burn. This color scheme highlight the differences between the two in an easy to spot way. Further on, in Figure 6, these two previous images can be seen laid over top of one another, with the swipe tool in use. This allows the mouse to act as a slider and pan back and forth between images to see differences in quick time.
Figure 4: Pre burn false color IR
Figure 5: Post burn false color IR
Figure 6: False color IR swipe tool
As a bonus set of images for this data set, normalized difference vegetation index (NDVI) imagery was examined, which is very useful in agriculture. This imagery, taken by a variety of aircraft across the country, does a very good job at showing vegetation health between images, and the two below images are examples of this. This is because it uses near infrared band color to spot subtle differences in vegetation life. In Figure 7, one can tell that there are not areas of red, or poor, vegetaion health except for some bottom left patches. However, in Figure 8, which is the post burn coutnerpart to the previous image, there are huge red squares in the center of the image. These areas indicate terrible vegetation health, which makes sense because that is where the controlled burns occured.
At the end of this lab, the color bands were experiemented with, and after switching the colors all around, it became very apparent that we leave the colors the way they are so that confusion does not occur. For instance, the NDVI vegetation health imagery is very easy to interpret because most people see green and imagine plants, which is also equivalent to good vegetation health in that imagery. However, when we start doing things like using blue to indicate health and yellow to indicate soil, the waters become muddied and the message may be lost. Overall, the operator shall do what makes the most sense to make a meaningful product for the task at hand. A full version of this report is available upon request.