This study develops and tests a novel approach for including regional risk factors in operational disease risk warnings against potato late blight. The central premise is that fungicide inputs can be reduced by omitting applications on days when conditions are unsuitable for the atmospheric transport of viable sporangia. The decision support system first decides whether a specific day is `high risk¿ (suitable for disease development in planta). Simulation studies revealed that on such high risk days, the capacity of the atmosphere to transport sporangia viably over relevant distances varies widely. An additional rule assesses this capacity, which is high when weather conditions allow a large number of spores to be released from the canopy and transported viably over long distances. When this capacity is high the original spray advice is followed, and when it is low a no-spray advice is given. The concept is implemented using the published decision support system SIMCAST complemented with models for spore release from sporangiophores, spore escape from the canopy, a newly developed model for spore dispersal and dry deposition of spores, spore survival during transportation, and weather forecast data from the mesoscale meteorological model MM5. Cultivar resistance was also incorporated into spray advice. The concept was tested in a field experiment in 2007 with three cultivars, representing a range in resistance to potato late blight from susceptible to highly resistant, and compared to a `stand-alone¿ version of SIMCAST. In a period with normal `infection pressure¿ (risk of disease) one third of the spray recommendations made by SIMCAST alone were modified and negated by the new system for the highly resistant cultivar. These savings came on top of a reduced, resistance-level dependent dose rate of Shirlan (a.i. Fluazinam). The results demonstrate the feasibility of including dispersal modeling and forecasted meteorology in disease warnings against Phytophthora infestans, even if the whereabouts of sources is unknown. The principles can be used in many decision contexts, but further work is needed to test and refine the method before it can be used in practice.
- phytophthora-infestans sporangia
- atmospheric surface-layer
- gaussian plume model
- maize pollen
- soybean rust