PREDICTION OF DISEASE DAMAGE, DETERMINATION OF PATHOGEN SURVIVAL REGIONS, AND CHARACTERIZATION OF INTERNATIONAL COLLECTIONS OF WHEAT STRIPE RUST

Thesis (Ph.D.), Department of Plant Pathology, Washington State UniversityStripe rust of wheat (Triticum aestivum L.), caused by Puccinia striiformis Westend. f. sp. tritici Erikss., is an economically important disease worldwide. Three studies were conducted in regional, national, and international scopes, with a focus on the epidemiology of the disease.A series of models for predicting potential yield loss for the U.S. Pacific Northwest were developed using climatic variables and historical disease data. Simple and multiple linear regression models were developed to estimate yield loss using winter climatic variables that were significantly correlated with yield loss. These models allow forecasting of potential yield loss and improve management of stripe rust in the major wheat growing areas in the Pacific Northwest. Regions for overseasoning of the stripe rust pathogen were determined in the mainland U.S. using long-term means for temperature, relative humidity, rainfall, dew point, snow depth, and availability of plant hosts. The pathogen can oversummer in most regions north of 40oN and in the highlands of southern states either in the Rocky or Appalachian Mountains. Winter survival can occur in most regions south of 40oN and the Pacific rims. The fungus cannot survive both summer and winter in most of wheat growing regions. It can oversummer and overwinter in the Pacific rims, highlands of southern states, and in the Appalachian Mountains. A total of 235 P. striiformis f. sp. tritici isolates from 13 countries were tested on 20 single Yr-gene lines and 20 wheat genotypes that are used to differentiate races of the pathogen in the U.S. Virulences to 13 Yr genes and 14 U.S. differentials were detected in all countries. At least 80% of the isolates were virulent on 15 wheat differentials. All isolates were avirulent to Yr5 and Yr15. Molecular characterization of 292 isolates from 18 countries was conducted using 17 simple sequence repeat markers, which separated the isolates into two genetic groups with some of admix genotypes. The most genetic variation was among the isolates within countries. This information helps us understand virulence and genetic variations of the pathogen populations and should be useful for control of stripe rust using disease resistance in the world.Department of Plant Pathology, Washington State Universit