A model-based approach for the prediction of banana rust thrips incidence from atmospheric variables
(2022)
This work focuses on the development of a mathematical model for the population growth of banana red rust thrips (Chaetanaphothrips signipennis) based on a modified temperature-based growth rate with the addition of climatic variables, such as relative humidity, wind speed and rainfall rate. The aim is to enable better prediction of the pest incidence and improve decision making, productivity, as well as quantifying the influence of these variables on the development of red rust thrips. The developed model is then compared with current solutions for predicting the pest incidence, showing improved accuracy (higher than 67%) versus experimental data, for which the state-of-the-art models indicate extremely poor fits.
This contribution introduces a data acquisition and modelling framework for the prediction of banana pests’ incidence. An IoT sensors-based system collects weather and micro-climate variables, such as temperature, relative humidity, and wind speed, which are uploaded in real time to a cloud storage space. The incidence of the red rust thrips (Chaetanaphothrips signipennis) is collected “manually” by periodic inspection. The mathematical model is adapted from population growth functions and a model of insect species development and allows predictions to be made at various time intervals with an accuracy greater than 80%, improving decision-making capacity for agro-producers and enabling the improvement of pest management actions.
This paper examines the most important approaches that could be applied for the introduction of circular economy strategies for the banana production process in the region of Piura (Peru). Based on this, a framework for an optimized economic cycle that is able to conserve resources and minimize the capital investments of farmers, while simultaneously increasing their production, can be created. Challenges and potential solutions are discussed for the production stages.