Willingthon Pavan

Web-Based System to True-Forecast Disease Epidemics

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

Web-Based System to True-Forecast Disease Epidemics – Sisalert

One important step-in developing plant disease and pest forecasting systems is to provide an easy and comprehensive way to either run the models or deliver results to the users. The use of near realtime and, in some cases, forecast weather data is crucial to forecast outbreaks of plant diseases or pests at both local and regional basis. In this work, we present a web-based information system designed to retrieve weather data from both weather stations and remote database with 7-days weather forecast and run the models. The system has been adjusted to predict outbreaks of two diseases of wheat and four diseases of apple. The models are run in a daily basis and the results are real-time generated and presented in tabular, graphical and text format - an interpretation of results based on an expert assessment. Alternatively, alerts are sent to emails and cell phones. The system proved to be functional, flexible and has the potential to be integrated into a decision support system for pest and disease management. The system has been tested by growers and preliminary results showed a significant reduction of pesticide applications compared to conventional systems.

Web-Based System to True-Forecast Disease Epidemics — Case Study for Fusarium Head Blight of Wheat

Disease forecasting has become an established component of quantitative epidemiology. The mathematics of disease dynamics is the core of several disease forecast models that have been developed in the last four decades. However, many models have not lived up to the expectations that they would play a major role and lead to a better disease management. Amongst the reasons, the presumption of a disease forecast model is that it makes projections of major events in disease development and most present forecast models do not (Seem 2001). An exciting development in this area is the possibility to use weather forecasts as input into disease models and consequently output true disease forecasts. As weather forecasts improve together with more accurate estimations of micro environmental variables useful for plant disease models, as such precipitation and leaf wetness duration, it will be possible to provide seasonal estimates of disease likelihood and forecast outbreaks. This is especially interesting for field crops for the reason that unnecessary sprays has a significant impact on production costs, and no timely applications may result in inadequate control.

A Virtual Environment for Drone Pilot Training Using VR Devices

This paper presents the development of a Virtual Reality environment for drone pilot training using interaction devices. We built an immersive interface in order to enhance the user experience in UAV training tasks comparing to the traditional control interfaces. To reach this aim, we chosen the Microsoft Kinect to control the drone and the Oculus Rift to visualize the scene, as well as keyboard, mouse and joystick support. As a result, we created a virtual environment in which users can train the drone pilot safely. Despite a good impression on preliminary tests, the solution still needs an evaluation by users and improvements in Physics simulations.

Statistical Correction of the Result of Weather Forecast by Applying the Model Output Calibration

This work describes the simulation models calibration method called Model Output Calibration. In order to verify its effectiveness, presents the application of the MOC in weather forecast correction generated by the Eta 15Km model at CPTEC/INPE. Eta is a regional model for numerical weather prediction. The results of the statistical correction of Eta forecast were positive, with satisfactory improvements in the variables tested (temperature and relative humidity). The use of this approach shows the possibility of gains in the results of simulation models of crops and diseases that use as predictive variables the variables generated by weather forecast models.

Development of a disease forecasting system for strawberries as a tool on AgroClimate

Strawberries are one of the most valuable crops in Florida. The state produces around 16 million flats of strawberries every year, which represents 15% of nation’s berries and virtually all the berries grown during the winter. The high value of the crop often compels growers to protect their profits by making numerous applications of fungicides, insecticides, and acaricides on a strict calendar schedule. In Florida, fungicides are applied on a weekly schedule, mainly for control of anthracnose and Botrytis fruit rot, from December through March. These are the most important diseases for production of annual strawberries in central Florida and worldwide. Different predictive models for anthracnose fruit rot and Botrytis fruit rot were evaluated for timing fungicide applications for control of those diseases under Florida conditions. The most effective models were embedded in a web-based tool developed for use by growers to schedule their fungicide applications. The implementation of this internet-based forecasting system to predict anthracnose and Botrytis enable the growers to easily access the information necessary for them to decide on the need for a fungicide application. The benefits of such a tool is that growers can apply fungicides only when conditions are favorable for disease development, thus reducing the number of applications and production costs without compromising disease control.

Um sistema web para acompanhar o crescimento e desenvolvimento do algodão

This work presents the development of a Web-Based simulator, designed to track the cotton growth, making use of the relationship between the phenological stages and the degree-days accumulation. In this manner, are presented a research on the cotton growth, in order to determine how is the development of culture and its relationship with the heat accumulation and a survey on current Web technologies. The tool proposed operates in three different ways: past, present and future, supplying the user with relevant information in the cotton crop.