Identifying spectral patterns of Eurasian watermilfoil for mapping and monitoring

Session: Invasive Species (5)

Colin Brooks, Michigan Tech Research Institute, [email protected]
Amy Marcarelli, Michigan Tech. University, [email protected]
Amanda Grimm, Michigan Tech Research Inst., [email protected]
Richard Dobson, Michigan Technological University - MTRI, [email protected]

Abstract

Being able to reliably identify the location and extent of invasive submerged aquatic plants would help with efforts to understand the effectiveness of different treatment methods. A Michigan Tech research team has developed methods to collect and compare spectral profile data at multiple scales and identify when a particular species can be separated from other submerged aquatic vegetation (SAV). Through an US EPA-funded GLRI project, we have been demonstrating how unmanned aerial system (UAS) enabled sensing can help map the extent of Eurasian watermilfoil (Myriophyllum spicatum or EWM) and track the impacts of different treatment methods. Spectral profiles were collected for EWM and other SAV species for plants taken out of the water, immediately over the plants still in the water, and from a UAS platform approximately 15m above the water. This enabled evaluating the strength of the SAV species spectral signature as three different scales, as well as demonstrating a UAS-enabled monitoring technique. Results show that the normalized difference vegetation index (NDVI) is useful for identifying EWM, and that a hyperspectral level of bands more reliably differentiates EWM from other SAV species.

Twitter handle of presenter
@cnbinaa