Application of Hyperspectral Techniques to Monitoring and Management of Invasive Plant Species Infestation

Abstract

This report summarizes and integrates the main findings, using several case studies to illustrate conclusions. Case studies include cheatgrass and Russian knapweed at Yakima Training Center, showing that multiple dates that combine different growing seasons improve map accuracy. Also included are maps for kudzu and Johnson lovegrass at Fort Benning, tamarisk at Yuma Proving Ground and phragmites at Aberdeen Proving Ground. Examples were chosen to illustrate a range of problems in correctly mapping invasive plant species under different environmental conditions. A detailed analysis of the CART results for Vandenberg Air Force Base show how it can be used to predict likely locations for spread and the most probable causes for invasion at specific locations. A CART analysis for kudzu at Ft. Benning is included and the same method was applied to tree decline at Ft. Benning. A User's Guide is provided in the appendix as is a hyperspectral training tutorial.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2008
Accession Number
ADA495506

Entities

People

  • Carlos M. Ramirez
  • Deanne Dipietro
  • Edward T. Tom
  • Emma Underwood
  • John Kefauver
  • Karen S. Olmstead
  • Maragaret E. Andrew
  • Mary Andrews
  • Nina Noujdina
  • Shawn Kefauver
  • Susan L. Ustin
  • Yen-ben Cheng

Organizations

  • University of California

Tags

Communities of Interest

  • Advanced Electronics
  • Sensors
  • Space
  • Weapons Technologies

DTIC Thesaurus Topics

  • Aerial Photography
  • Birds
  • Cameras
  • Chemistry
  • Data Mining
  • Dimensionality Reduction
  • Geographic Information Systems
  • Geography
  • Habitats
  • Image Processing
  • Information Science
  • Information Systems
  • Jet Propulsion
  • Optical Properties
  • Photographs
  • Supervised Machine Learning
  • Two Dimensional

Readers

  • Aquatic Ecology
  • Business Analytics
  • Neural Network Machine Learning.