Predicting Stripe Rust Severity in Wheat Using Meteorological Data with Environmental Response Modeling

Objective: The main objective of current investigation was to develop a predictive disease model based upon meteorological data, viz., maximum temperature, minimum temperature, rainfall, relative humidity, and wind speed to predict stripe rust severity (%). Methods: Five years' data of stripe rust severity on three wheat varieties, namely SA-42, Sandal-73, and Barani-70, continuously cultivated for five years (2013-2017), were collected from experimental trials of Deputy Director of Agriculture Extension Layyah to develop a predictive disease model. For validation of the model, a research trial was conducted in the Research Area of the Department of Plant Pathology, Bahadar Sub-Campus Layyah, during the crop seasons of 2018-2019, following procedures similar to those utilized in five years investigation. The data on epidemiological variables used in the present investigation was collected from the Pakistan Meteorological Observatory at Karor-Layyah. To evaluate the association between meteorological factors and disease severity correlation and regression analysis was performed. Results: All meteorological variables contributed significantly in disease development and showed 89 % variability in stripe rust severity (%). Root means square error (RMSE) and residual (%) were used to evaluate the model's predictions. Both indices were below 20, showing that the model could accurately predict the progression of disease. The regression equations of 5 years model (Y = -63.11 + 0.96x1 + 1.72x2 + 3.72x3 + 0.43x4) and 2 years model (Y = -40.2 + 1.80x1 + 1.18x2 + 2.29x3 + 0.39x4) validated each other. Scatter plots indicated that environmental factors such as maximum temperature (12.8-22.5 & DEG;C), minimum temperature (8.7-14.8 & DEG;C), relative humidity (50-85 %), and wind speed (1.3-4.5) influenced the progression of stripe rust epidemic.