The rapid spread of non-native invasive plant species, including plants regulated as noxious weeds, is causing irreparable damage to the natural resources on military installations. This research moves beyond current remotely sensed vegetation indices and classification algorithms to take advantage of information generated from high-resolution spectra.

Spectral angle mapper classification of common reed performed on the physiological indexes NDVI, NDWI, and PRI.

The objective of this project was to develop and demonstrate a new remote sensing methodology using hyperspectral imaging (HSI) for mapping invasive plants. The outcome was a predictive model of potential plant invasion that integrated HSI information with a geographic information system (GIS). This methodology has broad applicability to military installations.

Six military bases from the southeastern, southwestern, and northwestern regions of the United States, each with different non-native invasive plant taxa and intensities and patterns of environmental disturbances, were selected to demonstrate, refine, and validate the methodology. These case studies demonstrated the portability of the methods under various types of military activities. Airborne flightlines were identified to provide data for mapping various species of invasive plants under the diverse conditions existing at these sites. These data provided a basis for demonstrating and assessing the benefits of HSI. The combination of HSI tools and images provided a robust protocol for monitoring plant invasions. The HSI information was integrated into a GIS database of other site characteristics to develop a predictive model of the potential for plant invasion.

In the short-term, this project will improve understanding of the distribution of key non-native invasive plants on military bases and the environmental conditions associated with their distributions and spread. The long-term benefit will be the development of a cost-effective method for mapping invasive plants that can be used to monitor their spread to new locations. With these new tools, installation managers can develop an accurate distribution of invasive species coverage within their boundaries and natural resource managers can evaluate their invasive species control approaches. Initial estimates indicate that remote sensing surveys can dramatically increase the survey area while reducing associated costs by more than 90%. (Project Completed - 2008)