Disease-yield relationship in barley.

The effect of disease on growth and yield of two barley cultivars sown at different times and in different years was investigated in New Zealand. The data were used to develop yield-loss models based on both disease severity (measured as green leaf area) and yield target (estimated by duration of crop growth). Disease influenced yield differently in Triumph and Sonya barley (spring and winter types, respectively), and yield components were affected to different degrees. Empirical yield-loss models for individual cultivars, sowing dates and seasons had different slope values, and the best models were based on measurements of disease at different growth stages. Combined models were less significant and explained less variation in yield than the individual models. Models which included the duration of crop growth as an estimate of yield target improved the fit to the data. Empirical models were specific to cultivar, sowing date and season, suggesting that they were not applicable in the varying conditions tested. The inclusion of estimated yield target improved the general applicability of models, and provided a method of using models in crops sown at different times and in different seasons, without measuring yield target directly.