Developing a Cloud service to detect Verticillium-induced stress on olive trees by Sentinel-2 Imagery

There is a variety of approaches for plant disease management, but they all converge in early and precise detection of stress as a major factor, in order to make applications more effective. In this work, a remote sensing approach for timely detection of stress caused by the fungal pathogen Verticillium dahliae in olive trees is presented. A detection model that calculates a vegetation index based on satellite spectral information (Sentinel-2) was developed and was used to estimate the stress levels in olive trees in the Halkidiki region, which are directly related to Verticillium infestations. Index values are calculated only for pixels within the boundaries of the parcels of interest. ArcGIS 10.5 software and Python 3 were used to process the data and to produce the resulting model together with the libraries concerned. This model has been implemented in an e-infrastructure as a cloud service that offers the end user the ability to create an account and store searches and results. The resulting map provides an overview of the situation regarding the levels of water stress on olive trees over a large area and in little time. Implementing this model as a cloud service ensures its automated and near real-time application and its viability, while webGIS facilitates the user's experience and icreases its usability by interested parties, such as agricultural consultants and scientists.