Wheat disease survey of the Texas panhandle using Landsat Thematic Mapper imagery Conference Paper uri icon

abstract

  • Viral pathogens commonly infect hard red winter wheat in the Texas panhandle generally causing chlorosis and stunted growth. Landsat Thematic Mapper (TM) images were acquired throughout the growing season and used to classify the vegetative status of wheat in the region. Normalized difference vegetative index (NDVI) was calculated during dormancy to identify wheat fields and create a mask on subsequent imagery. A supervised classification was applied to the wheat fields using the maximum likelihood algorithm with five initial classes. The classification was based on ground truth data ranging from lush vegetation to severe chlorosis. Classification error was measured with confusion matrix and the area of each class was measured using geographic information system software. In the 2004 growing season, the western portion of the panhandle included 3.2 million hectares (ha) of which 281,218 ha were classified as wheat. The area that was severely infected in February, March and May was 21,103 ha, 33,959 ha, and 39,541 ha, respectively. The overall accuracy for each date is 73, 70, and 40 percent, respectively. This suggest that Landsat TM imagery can be used to detect disease, however more training classes need to be used as the season progresses to compensate for other factors which may affect vegetation late in the season such as cattle grazing.

published proceedings

  • American Society for Photogrammetry and Remote Sensing - 20th Biennial Workshop on Aerial Photography, Videography, and High Resolution Digital Imagery for Resource Assessment 2005

author list (cited authors)

  • Jones, D. C., & Rush, C. M.

complete list of authors

  • Jones, DC||Rush, CM

publication date

  • December 2006