Powdery mildew infection is affected by interacting factors in a polyploid goldenrod system

Monitoring disease incidence in wild plant populations is increasingly important as climate change expands pathogen ranges. Climate responses will be complex given that the prevalence and severity of plant disease is governed by both plant and pathogen genetics, and these interactions shift depending on environmental conditions, including water availability and flowering phenology. We assess 6 methods for quantifying powdery mildew infection on a complex genetic and environmental system. Diploid and hexaploid Soli dago altissima L. (Canada goldenrod [Asteraceae]) plants that were previously artificially selected for early or late flowering were exposed to 2 watering treatments and accrued powdery mildew infection in field. We conducted a comprehensive assessment of infection (percentage of infected leaves) and compared this metric to 5 other rapid assessment methods, 2 of which were quantitative and 3 qualitative, using 4 criteria: 1) time investment per measurement, 2) correlation of the results with those obtained from percentage infected leaves, 3) R² obtained from a standard analysis applied to each data set, and 4) differences in interpretation depending on the method employed. We found that the percentage of infected plant height was the most accurate but also the most time-consuming approach. A qualitative 4-level scale ranked second with respect to all metrics of time and accuracy. Other methods were relatively fast, but each suffered from a lack of accuracy. We discuss these results in the context of selecting the best method to assess powdery mildew infection in complex systems that include natural levels of genetic and environmental heterogeneity.