Prediction of weather damage of mungbean seed in tropical Australia. II. Model application

To demonstrate a model to simulate the risk of weather damage of mungbean, application studies were undertaken using 27 years of climatic data collect ed at Katherine, Northern Territory, Australia. In terms of the risk of wea ther damage, the transition from high risk to low risk occurred after mid-F ebruary but before 20 March. High quality seed could be expected in 70% of seasons for a crop that matured after 20 March. For planting dates prior to 25 January, the chance of producing premium quality seed was enhanced to 4 0-70% of seasons by sowing a cultivar that matured 2 weeks later and by har vesting promptly (4 days after maturity). There was no benefit from later m aturity or harvest promptness where sowing was made after 25 January, becau se maturity occurred after the wet season was complete. In contrast, yield was optimised at early January sowing dates. Calculating gross margins by c ombining yield and weather damage simulations identified an optimum sowing date between the optimum for yield and seed quality. It was shown that late r maturity combined with photoperiod sensitivity increased the sowing windo w from 10 to 29 days compared with a short duration variety that was insens itive to photoperiod. The relative merits of modelling and field experiment ation in assessing the cropping potential for mungbean in a new region are discussed. The need to be able to simulate the yield of the second flush of flowers was acknowledged as a future research requirement.