Author
![]() |
ROOT, RALPH - USGS |
![]() |
KOKALY, RAY - USGS |
![]() |
BROWN, KARL - USGS |
![]() |
DUDEK, KAY - USGS |
![]() |
USTIN, SUSAN - UC DAVIS |
![]() |
ZARCO-TEJADA, PABLO - UC DAVIS |
![]() |
PINILLA, CARLOS - UC DAVIS |
![]() |
Anderson, Gerald |
![]() |
HOLROYD, EDMOND - COLORADO STATE UNIV |
![]() |
HAGER, STEVE - NATIONAL PARK SERVICE |
Submitted to: Airborne Science Workshop
Publication Type: Proceedings Publication Acceptance Date: 12/11/2002 Publication Date: 12/11/2002 Citation: Root, R., Kokaly, R., Brown, K., Dudek, K., Ustin, S., Zarco-Tejada, P., Pinilla, C., Anderson, G.L., Holroyd, E., Hager, S. 2002. Comparison of AVIRIS and EO-1 Hyperion for classification and mapping of invasive leafy spurge in Theodore Roosevelt National Park. In: Proceedings of the 11th Jet Propulsion Laboratory (JPL) Airborne Earth Science Workshop, JPL Publication 03-4, Jet Propulsion Laboratory, Pasadena, California. p. 297-305. Interpretive Summary: Effective weed control depends on efficient and economical methods of identifying and mapping invasive populations. Three imaging devices were used to map leafy spurge in Theodore Roosevelt National Park. Each imager was designed to acquire images differently. The CASI (flown in a small aircraft) was designed to acquire high resolution images of relatively small areas, AVIRIS (flown in a large military aircraft) was designed to acquire images of larger areas and the Hyperion satellite for acquiring images of even larger areas. This study evaluated the capabilities of each instrument to identify and map leafy spurge and provide information useful for controlling weed populations. Preliminary results indicate that leafy spurge is generally separable from its associated vegetation types using the three imaging devices. Further research is expected to quantify classification results and explore the potential for generating more regional, multi-flightline maps of leafy spurge infestations that will provide valuable contributions to monitoring and controlling efforts. Technical Abstract: Invasive species are rapidly becoming a threat to the world's biodiversity. In the United States alone, non-native species are causing environmental damage and economic losses estimated to exceed $100 billion per year (Pimentel et al., 2000). Morse et al. (1995) estimate that approximately 5,000 plant species that have escaped from cultivation have subsequently invaded natural ecosystems in the United States. One of these species, leafy spurge (Euphorbia esula L.), entered North America from Eurasia in 1829 (Council for Agricultural Science and Technology (CAST), 2000). Some years later, during the Homestead Period, it found its way into North Dakota and subsequently has spread throughout the upper Midwest and northern Rocky Mountain States, causing estimated annual losses of revenue in excess of $200 million. Leafy spurge causes severe ecosystem alteration due to its aggressive growth relative to that of native flora, its ability to invade unaltered habitats, and its persistence once established. It forms nearly monotypic stands and has the capacity to alter ecological processes and visitor perceptions of Theodore Roosevelt National Park and the surrounding region (Trammell, 1994). This paper outlines tests of state-of-the art remote sensing technology for detection and mapping of leafy spurge at Theodore Roosevelt National Park began in 1998 as part of a cooperative demonstration project between the National Aeronautics and Space Administration (NASA) and the United States Department of Interior. (DOI). Within the NASA/DOI Hyperspectral Technical Transfer Project (Root and Wickland, 2001), high altitude Airborne Visible and Infrared Imaging Spectrometer (AVIRIS) data were collected over the South Unit of the park in July, 1999. Preliminary analysis of these data demonstrated that leafy spurge could be identified and mapped (O'Neill et al., 2000). AVIRIS provided an excellent reference point for comparison with the Hyperion imager on the orbital side, and CASI on the low altitude aircraft side, for effectiveness of mapping leafy spurge with imaging spectrometers spanning a wide range of spatial resolution. Preliminary results of three different classification techniques indicate that leafy spurge is generally separable from its associated vegetation types. Further research is expected to quantify classification results and explore the potential for generating more regional, multi-flightline maps of leafy spurge infestations that will provide valuable contributions to monitoring and controlling efforts. |