VIEW ARTICLE

Ecology and Epidemiology

Detecting Patterns of Wheat Stripe Rust Pandemics in Time and Space. X. B. Yang, Department of Plant Pathology, University of Arkansas, Fayetteville 72701 and Department of Plant Protection, Beijing Agricultural University, Beijing, China; S. M. Zeng, Department of Plant Pathology, University of Arkansas, Fayetteville 72701 and Department of Plant Protection, Beijing Agricultural University, Beijing, China. Phytopathology 82:571-576. Accepted for publication 4 February 1992. Copyright 1992 The American Phytopathological Society. DOI: 10.1094/Phyto-82-571.

Pandemics of wheat stripe rust from 1950 to 1990 were analyzed with data from five key regions in different geographical areas in northern China. Epidemics in one region were correlated with epidemics in other regions. Spatial correlation coefficients increased as spatial distance decreased. The epidemic time series of the source region, where rust occurs year round, was not correlated with lagged epidemic series of regions in the dispersion area. Average epidemic indices of one region were associated with distance from the source region and latitude of the region. Time series analysis also was used to determine temporal pattern. An epidemic time series of each region was identified as the autoregressive integrated moving average (ARIMA) (3,1,0) for the source region and ARIMA (2,2,0) for regions in the dispersion area, which indicate that the epidemic value in a given year is related to epidemics of the previous 4 yr. There was a downward trend in the 41-yr pandemic series, except in Gangu. Mean epidemic indices from northern China were greater for the period of 1950–1969 than for the period of 1970–1990. Less frequent and less destructive pandemics were noted in the second period; this may be due to improved disease management by advanced agricultural techniques.