Application of Canadian weather radar data to plant disease management schemes in southern Ontario

Abstract

Management schemes based on the mean temperature and duration of wetness periods (dew or rain) have been developed to schedule fungicide applications rationally in a number of high‐value crops. Due to the spatial variability of convective rain events during the growing season in southern Ontario, data from an existing rain‐gauge network is inadequate for delivering advice to the farming community. Therefore, the use of both conventional and Doppler weather radar data to supplement rainfall data was investigated. Hourly measurements of leaf wetness duration and tipping bucket rainfall data were compared to data from the King City and Exeter radars in two disease warning schemes. When used in conjunction with the Penman‐Monteith model to estimate disease indices, the King City radar data were in better agreement with the leaf wetness sensor than the Exeter radar. There was no marked difference between conventional and Doppler data. When both conventional and Doppler radar data were used in combination, the success rate in detecting rainfall occurrence was 80–85%, with most errors due to false positives. Rainfall duration estimates from radar were sufficiently accurate in two disease management schemes that cumulate information from several successive wet periods.