Investigating weather variables driving boxwood blight epidemics: Insights from field trials with Buxus sempervirens ‘Suffruticosa’ in northern Germany between 2006 and 2020

Boxwood blight is a highly invasive disease, but studies on host–pathogen–environment interactions are rare because the initial research emphasis has been on developing resistant cultivars and chemical control of the disease. We used generalized additive models to investigate weather variables driving boxwood blight epidemics in field trials conducted between 2006 and 2020. Briefly, three or four replicate rows with 10 boxwood plants per row were planted in 0.75 m2 plots. Plants were artificially inoculated in 2006, while those in subsequent years were naturally infected with inoculum left over from previous trials. Disease severity was assessed by estimating the percentage of leaves blighted, including fallen leaves. There was a significant positive main effect of mean rainfall per rainy day, daily minimum temperatures and daily minimum relative humidity on disease severity observed over individual field trial periods. There was a significant negative interaction effect of mean rainfall per rainy day and daily maximum wind speed, and daily minimum relative humidity and daily minimum temperature on disease severity. Higher disease severity was associated with higher mean rainfall per rainy day and lower daily maximum wind speed. Likewise, an increase in daily minimum relative humidity at lower daily minimum temperatures was associated with a greater increase in disease severity than at higher temperatures, suggesting that higher temperatures resulted in lower humidity that led, in turn, to less disease severity. The implications of our findings for forecasting models and conservation of boxwood are discussed.