Small Grain Disease Forecasting Models

How the leaf disease forecast models work

These models classify whether or not a given 24 hour period was favorable for infection. Infection happens when an infectious pathogen contacts a susceptible plant in a suitable environment. Multiply this event millions of times and you have the beginnings of a plant disease epidemic in a field.

The number of hours the weather is suitable for infection is known as an infection period. To find out if the previous day contained an infection period, the models use hourly weather data from the North Dakota Agricultural Weather Network (NDAWN). NDAWN data are sent to NDSU every morning and summarized from noon to noon. Noon, rather than midnight, was chosen as a cut-off so that the dew period, critical for disease development, is considered in its entirety.

Tan spot and Stagonospora (Septoria) blotch infection periods are modeled by a form of artificial intelligence called neural networks. Since the 1993 growing season, infection data and matching weather patterns have been collected at NDSU. The neural network was able to relate weather and infection data with an accuracy of 82% for tan spot and 84% for Stagonospora blotch.

Wet period duration and growing degree days were the most important factors for tan spot infection. From experiments in growth chambers, we know the critical minimum wet period is between 3 and 6 hours at a temperature of 60-75 F. Wet periods longer than 5 hours increase the chance for infection and promote the growth of existing lesions. The tan spot fungus tends to build up its population through the season, so a cumulative degree day value serves as an index to increasing risk of infection.

Precipitation and relative humidity were the most important factors for Stagonospora blotch. Again from studies in controlled environments, we know that some spores of Stagonospora are released under high humidity while others are splashed onto plants by rain. Rainfall amounts of less than 0.02 inch are apparently ineffective in spore dispersal.

The infection period for wheat leaf rust is determined by whether a minimum of six hours of wetness occurred. A minimum temperature of 59 F during the wet period helps insure that nights too cool for infection are excluded.

How the Fusarium head blight forecasts work

Fusarium head blight or scab infection periods are based primarily on two factors - spore counts and weather. Spore samples from ND and MN tell us if the fungus is present in the area. In the current growing season, 21 NDAWN stations are paired with nearby sampling devices.

Spore counts are difficult and expensive to obtain so they are not available from all locations. Therefore, two computer models that assess the risk of scab have been added to all forecast locations. The first computer model uses the number of hours of rainfall and hours of temperature between 60 and 85 F during the week before flowering to determine whether a severe epidemic is likely. Similar to spore counts, the first model assesses whether conditions are favorable for pathogen activity as the crop enters its most sensitive stage and can guide a fungicide spray decision. The second model is designed to be used at 10 days after flowering starts as a way to predict grain quality. This information may be helpful for marketing plans but is not intended as a head blight fungicide spray decision tool. The second model uses temperatures the week before flowering (as does the first model) and hours of relative humidity greater than 90% with temperatures between 60 and 85 F.

Small Grain Disease Forecasting Home
Plant Pathology Department,
North Dakota State University, 306 Walster Hall, Fargo, ND 58105-5012
Web Site:  http://www.ag.ndsu.nodak.edu   Email:  forecast@ndsuext.nodak.edu