Web-Based System to True-Forecast Disease Epidemics

Abstract

One of the challenges for modelers, besides the development of mathematical equations integrating biology and climatology, is to provide a comprehensive model delivery system through the Internet. The use of near real-time and forecast weather data is key to true-forecast disease outbreaks at local and regional basis. A prototype of such a system is proposed here to predict and forecast infection risks for Fusarium head blight (FHB) using a model previously developed. The system was designed to access and retrieve weather data from an automatic weather station and from a remote database with 7-days weather forecast for the same local. The model is initiated through a web interface and the simulation starts by selecting heading date. Once the current day is entered, the model uses forecast information to warn disease outbreaks, by combining forecast and historical weather data. Model results are presented to the user in tabular, graphical and report, which shows interpretation for the results based on an expert assessment. Once a registered user set a heading date, daily simulation updates are sent daily to email and cell phones. The prototype proved functional, is flexible and has the potential to integrate a decision support system for disease management. A farmer cooperative is extensively testing the system during the current wheat season in the state of Rio Grande do Sul