Innovation
The methods outlined are representative of the most emerging analysis techniques in the scientific domains covered by the project. The development and application of predictive models based on geographically enabled machine-learning techniques is a frontier topic in GIS science as well as the most promising method for improving predictions accuracies by exploiting computing resources achieved by modern IT facilities.
Applications into the local analysis of farming-related PM are poorly treated in the literature, especially to what concerns the inclusion of multivariate spatial observations including satellite-based PM estimates. The availability of multiple PM estimates from space as open-data has been accomplished only recently and the Sentinel-5P is one of the few Earth Observation satellite missions providing them operatively. This opens newsworthy opportunities for satellite-based PM and air quality monitoring. To that end, research activities will also include state-of-art on-field PM sampling and characterization that will empower enhanced source apportionment and exposure modelling combined with the satellite data assets. In parallel, collected data will provide a valuable source of information for locally validating the predictive models and therefore a better awareness of method performances which are of primary interest to future practical applications. The extensive adoption of open data, as well as the use of free and open-source software, provides the analysis with a potential to be empowered, replicated, and improved; thus favouring the scientific debate and benchmarking on the addressed topics. From a more general perspective, the project aims at contributing to the baseline research on ground and satellite sensors interplay for air quality monitoring that is advised by the international scientific community for contributing to next-generation air quality geostationary satellite missions.