PMN Proeedings: 2011 Field Crops Rust Symposium

2011 Field Crops Rust Symposium:
Abstract

Poster #28

Predicting Stripe Rust Epidemics in Kansas Using Regional Drought Indices and Local Weather Conditions

Presenting Author: B. GRABOW

Coauthor: E. D. De Wolf

Affiliation: Kansas State University, Manhattan, KS, USA

Stripe rust (Puccinia striiformis) is an economically important plant pathogen in Kansas. Major epidemics of stripe rust occurred in 2001, 2003, 2005, and 2010, resulting in 7.3, 10.6, 8.0, and 10.3% statewide losses reported, respectively. In 2010, stripe rust caused yield losses of 42.5 million bushels of wheat valued at approximately 221 million dollars for farmers in Kansas. The objective of this research is to develop prediction models for stripe rust epidemics in Kansas using combinations of regional drought indices and local weather conditions. The prediction models were developed based on historical records of regional stripe rust epidemics occurring between 1999 and 2010. Results indicate that drought conditions in Texas during February and temperature in Kansas during March are strongly associated with stripe rust epidemics. Logistic regression models combining drought indices from Texas with different representations of March temperatures in Kansas correctly predicted stripe rust epidemics with greater than 80% accuracy. Adding variables describing May rainfall in Kansas further improved accuracy of the prediction but would be less useful for disease management. These results suggest that stripe rust epidemics can be predicted with a high degree of accuracy based on regional drought indices and local weather conditions. In the future, these models may help wheat farmers better evaluate the need for fungicide applications and minimize the risk for future yield losses caused by stripe rust in Kansas.