I am currently a wildlife biology major at the University of Nebraska at Kearney, working on various undergraduate research projects. I first learned about Platte Basin Timelapse (PBT) last spring and was thrilled to see how imagery was being used to convey important stories. After that meeting, I knew I wanted to contribute to this project in some way. I have a background in Geographic Information Systems (GIS) and a strong interest in conservation. I am working with Dr. Mary Harner to tell the story of how water changes over time in the Sandhills of Nebraska through the use of historical aerial imagery and GIS programs.
My project focused on the southeastern edge of the Nebraska Sandhills, west of Burwell, Nebraska, in and around the Switzer Ranch, an area of ongoing PBT activities. Already in place are the permanent PBT cameras that show us how these features change on a smaller scale at Latta Lake, Gracie Creek, and Shoemaker Lake; these cameras have been capturing change since 2012. My task was to tell the story at a larger scale of the entire landscape across several decades.
The process began by obtaining aerial imagery dating back to 1948. The images were captured via airplane and photographed by sections. The oldest imagery was originally taken on glass slides and later scanned on a computer. Images looked like this:
We obtained imagery from May of 1948, June of 1953, August of 1974, and March of 1982. Images for more recent years (2012-2016) were accessible online, already georeferenced. Georeferencing refers to the process of assigning images real-world coordinates in a GIS program. Each year was flown at a different elevation, had different coloration, and different resolution, which added to the already difficult job of comparing the images to one another. You can see differences in color and scale below:
I was assigned the task of image to image georeferencing the images and mosaicking them together using GIS programs so that I could compare how the landscapes changed over time. I began by georeferencing the aerial imagery, which involved working with as many as 45 images per year.
Starting with the images above, I georeferenced each image individually, meaning I matched each image to an image that already had coordinates and gave the image the coordinates from that image. Then, in a program called ERDAS Imagine, I was able to mosaic all the images together and create a single, combined image for each year. This process is almost like putting a puzzle together. Each image is a piece to the puzzle, and I made sure all the pieces fit together to form one, cohesive image.
I repeated this process for each year, and the result is a single image for the area.
Here are images of the same study area from more recent years that were already georeferenced and available online:
After completing the mosaics, I used ArcGIS to digitize the water features for each year. Digitizing in GIS is a tedious task that requires manually tracing the features. Here is an example of the 1953 mosaic after being digitized:
To display the changes in water over time, I created a video of the mosaics and traced water features for each year. The water features are consistently found in a ‘U’ shape, but the amount of water varies from year to year, depending on if it is a dry or wet year:
Working on a computer for eight hours, five days a week got a little monotonous at times, but being able to visit the place I had been looking at from the air made it all worth it. In early summer, I got to visit the Switzer Ranch for the first time.
Visiting the Sandhills was like an entirely different world. This trip made me further understand the importance of my work because I was able to see how essential water is to the ranch – whether it was in one of the lakes, creeks, or the windmills that are strategically placed to provide water for the cattle. While working on this project I have learned so much about GIS programs, met so many different people associated with the PBT project, and discovered one of the most unique areas in the world – the Sandhills of Nebraska.
This project was supported by the University of Nebraska Collaboration Initiative, with funding from Nebraska EPSCoR, as well as the University of Nebraska at Kearney Undergraduate Research Fellows Program.