A methodology for predicting barley β-glucan levels during pre-harvest stages

This study looks at the development of a farm-level baseline model to simulate the impact of various genotypic, environmental and agronomic factors influencing the level of β-glucan content in both hulled (H_B) and hull-less barley (H_LB) varieties. Monte Carlo simulation techniques were employed to model various stages in pre-harvest processes and to simulate the factors impacting the β-glucan content of barley cultivars. Probability density functions were fitted to relevant data sets collated from the scientific literature to capture the inherent uncertainty and variability in the factors. The model was developed in Microsoft Excel using the @Risk add-on package and the model run for 10,000 iterations. The mean simulated level of β-glucan was estimated to be 4.53 g/100g and 6.27 g/100g for hulled (H _B) and hull-less barley (H_LB) varieties, respectively. A sensitivity analysis illustrated that barley cultivar, harvest delay, germination time, storage days and precipitation are factors significantly influencing the β-glucan content (correlation coefficients of 0.62, 0.45, -0.33, -0.24 and -0.17 for hulled and 0.74, 0.38, -0.29, -0.22 and -0.14 for hull-less barley) This study highlights the influence of pre-harvest operations on β-glucan content and thus allows strategies to be identified to influence its content.