VIEW ARTICLE

Ecology and Epidemiology

Nonrandom Patterns of Bacterial Brown Spot in Snap Bean Row Segments. B. D. Hudelson, Department of Plant Pathology, University of Wisconsin, Madison 53706; M. K. Clayton(2), K. P. Smith(3), D. I. Rouse(4), and C. D. Upper(5). (2)Department of Statistics, University of Wisconsin, Madison 53706; (3)(4)Department of Plant Pathology, University of Wisconsin, Madison 53706; (5)Agricultural Research Service, U.S. Department of Agriculture, University of Wisconsin, Madison 53706. Phytopathology 79:674-681. Accepted for publication 6 February 1989. This article is in the public domain and not copyrightable. It may be freely reprinted with customary crediting of the source. The American Phytopathological Society, 1989. DOI: 10.1094/Phyto-79-674.

Each leaflet on every plant in 37 5-m row segments and a single 12-m row segment from commercial snap bean fields was assessed for bacterial brown spot. Graphs of the proportion of diseased leaflets per plant (disease incidence values) versus plant position along the row suggested two types of nonrandom variability in disease: an extreme jaggedness superimposed on a slow, undulating change in disease. Arcsine square root-transformed disease incidence values were analyzed for spatial nonrandomness using three techniques: runs analysis, autocorrelation analysis, and auto-regressive integrated moving average (ARIMA) modeling. All three techniques detected the slow, undulating change in disease incidence values; however, only ARIMA modeling detected the jaggedness and could quantify both patterns. A “generalized ARIMA(1 0 1) model” was found to describe 35 of the 38 data sets. The biological mechanism generating these patterns is unknown. Knowledge of the existence of such patterns is important for developing effective sampling strategies for this disease. Theoretical characteristics of the generalized ARIMA(1 0 1) model indicate that random start systematic sampling will provide a better estimate of total or mean brown spot in a row than simple random sampling.