ARTIFICIAL NEURAL-NETWORK MODELS OF WHEAT LEAF WETNESS

Most bacterial and fungal plant pathogens require free moisture for in fection and reproduction so a disease forecasting model needs to estim ate wetness duration on plant surfaces if direct measurement is unavai lable. Artificial neural network (ANN) models with a backpropagation a rchitecture were developed to predict wetness on wheat flag leaves usi ng environmental variables recorded at 0.5 h intervals by an electroni c datalogger. For comparison, multivariate discriminant and stepwise l ogistic models also predicted wetness, and dew periods were predicted with a classification tree-discrimination approach, a relative humidit y indicator, and a physical model. ANN and logistic models correctly c lassified 93% and 90% of the validation cases and were more accurate t han discriminant models. Many of the ANN errors occurred during times of dew onset or evaporation and were less than or equal to 1 h per eve nt. Model accuracy for all methods improved 3-4% when input data from a wetness sensor positioned above vegetation were included. The averag e absolute error per night for prediction of dew period was 0.6 h for the classification-discrimination model, 0.8-1.1 h for ANN models, 1.2 h for a logistic model, 1.4 h for the physical model, and 2.1 h for t he relative humidity index. ANN and logistic models thus performed wel l compared to previously developed models for prediction of dew durati on and these models predicted leaf wetness from both dew and rain, an advantage for disease forecasting. Further improvements in ANN perform ance are possible, making the technique a viable research tool. (C) 19 97 Elsevier Science B.V.