Authors: Zachary Schwab*, Saginaw Valley State University, Rhett Mohler, Saginaw Valley State University
Topics: Remote Sensing, UAS / UAV, Natural Resources
Keywords: UAV Imagery, Invasive Species
Session Type: Poster
Start / End Time: 9:55 AM / 11:35 AM
Room: Lincoln 2, Marriott, Exhibition Level
Presentation File: No File Uploaded
Phragmites australis, also known as common reed, is an invasive wetland plant that can form thick, dense stands that overtake wetlands. Consequently, wetland managers like those on the Shiawassee National Wildlife Refuge (SNWR) spend a great deal of time and money controlling it. Currently, remote sensing methods exist to detect and map larger stands of P. australis, but the relatively coarse resolution of many sensors, even one-meter aerial photographs, does not allow for the detection of smaller phragmites stands or individual plants. Being able to detect these small stands is important, however, because personnel on the SNWR wish to treat phragmites patches at the smallest size possible, before they spread and become more difficult and more expensive to treat later on. One disadvantage of higher-resolution imager, however, is that it is more resource-intensive to collect and analyze. Therefore, we sought to identify a spatial resolution that would allow for the detection of small patches of P. australis, but not be of a fine spatial resolution than was necessary. To test this, we collected July drone imagery from an island in the SNWR that was known P. australis patches of different sizes. The imagery was processed to simulate pixel sizes of 2.5 cm, 5 cm, 7.5 cm, and 10 cm. These images were then classified to produce maps of P. australis, and these classifications were compared to ground-truthed patch data to determine their accuracy.