Pest and disease forecasting in fruit orchards: model development, application and functionality testing


Abstract: This work considered the development, application and evaluation of customized pest and disease forecasting process models for extension systems and Integrated Pest Management in fruit orchards. The functionality of the pest forecasting models was based on the use of degree-days weather kernels which predict the phenology of the major Lepidoptera in fruit orchards including C. pomonella, G. molesta, A. orana and A. lineatella. Moreover, the development, application and field testing of multivariate disease models were also studied, including: Monilia fructigena, Taphrina deformans, Stigmina carpophila, Sphaerotheca pannosa. We applied a customized addVANTAGE Pro 6.4 software interface in conjunction with wireless weather station devices. The scheme was established on a customer/server architecture, and collected data from one or several Telemetry Gateways (receivers) and run the process models which had been prior installed on a server where all the actual processing took place. The functionality and forecasting performance of the algorithms was evaluated using pest and disease data that were registered during the growth season from experimental fruit orchards situated in the area of Naoussa located in Northern Greece. In most cases the forecasting algorithms generated warnings which matched over the observed population dynamics and disease epidemics. In some instances in which deviations were observed, at 2-5 days accuracy lag was registered depending on species specific thermal requirements in relation to local weather conditions (e.g. no wet conditions for spore germination). Regarding the increasing interest of biorational insecticides where precise timing of treatments is extremely significant, whether driven pest and disease forecasting algorithms could be a useful instrument for improving their efficacy in IPM and assuring fruit related residual levels safely intervals.

Cookie Consent with Real Cookie Banner