Link to home

Aerial Photography Used for Spatial Pattern Analysis of Late Blight Infection in Irrigated Potato Circles

July 2003 , Volume 93 , Number  7
Pages  805 - 812

Dennis A. Johnson , J. Richard Alldredge , Philip B. Hamm , and Bruce E. Frazier

First author: Department of Plant Pathology, Washington State University, P.O. Box 646430, Pullman 99164; second author: Program in Statistics, Washington State University; third author: Department of Botany and Plant Pathology, Hermiston Agricultural Research and Extension Center, Oregon State University, P.O. Box 105, Hermiston, OR; and fourth author: Crop and Soil Sciences Department, Washington State University

Go to article:
Accepted for publication 17 February 2003.

Spatial and temporal dynamics of late blight were investigated from color, infrared aerial photographs of five commercial potato fields in the Columbia Basin during epidemics in 1993, 1995, and 1998. Aerial photographs were taken one to four times at 6- to 21-day intervals. Photographs were scanned and pixels, representing approximately 1 m2 in the field, were used in the analysis. Late blight-infected plants were aggregated as indicated by runs analysis. Significant z-tests were computed for four directions during each sampling date in each of the five fields. Absolute z-values for runs analysis increased, indicating increasing aggregation in the four directions, as disease incidence increased in the early and midphases of the epidemics in each field. Variograms indicated the existence of autocorrelation among infected plants in four directions; the range of influence increased as disease incidence increased except at the highest levels of disease. Late blight was observed to spread in fields as foci. Late blight foci enlarged in size, produced distinct daughter foci, and coalesced. A field where initial inoculum likely originated from infected seed tubers exhibited less initial aggregation than the other fields, perhaps due to a different source of primary inoculum. Aerial photography coupled with spatial analyses of late blight-infected plants was an effective technique to quantitatively assess disease patterns in relatively large fields and was useful in quantifying an intensification of aggregation during the epidemic process on a large scale.

© 2003 The American Phytopathological Society