Beatrice Maddalena Scotto

Particle trajectories are generally calculated using current fields that commonly come from Ocean General Circulation Models (OGCMs). The first studies about these models date back to the 1980s and dealt mostly with small-scale examples, only in the 90s are obtained models of trajectories of particles on a global scale related to ocean circulation. In recent years many articles have been written on this subject and include several studies on 'Lagrangian ocean models' including those of interest for the study of this thesis, those relating to oil spills . The Particle Tracking Model (PTM) is created to track particle dispersion in ocean currents from two different initial release configurations, which can be point release or area release respectively, depending on the type of study being conducted. As for the temporal configuration, the release can be instantaneous, where all particles are released at the same time, or periodic, where particles are released at different times. The PTM to be operational uses the Python computer language through the OceanParcels toolbox, ""Probably A Really Computationally Efficient Lagrangian Simulator"" developed to simulate in a customizable way the trace of both passive and active virtual particles. Hereinafter is presented a briefly outline what has been done in previous studies regarding the topic of oil spill. The analysis is made to consider multiple ocean models that present data of surface currents. The choice of the different models is due to the great complexity in the prediction and simulation of the circulatory currents, therefore there is a significant difference between the distinct models. A dispersion was created through the Lagrangian method of Particle Tracking Model, that uses virtual particles to properly simulate the real phenomenon. Lagrangian models are very advantageous for this type of analysis as they solve the trajectory of the particle. What highlighted is a probabilistic distribution of how the oil slick generated evolves according to the different models, to see in which area there is the presence of greater concentration. To have a more complete vision, the barycentres of the slick in evolution in the time have been calculated leaving the trace of the positions, to have a graphic visualization of the route followed by the various particles and to have a measure of the distance covered. In addition, the extent of oil slick dispersion in each time and in x, y directions have been evaluated.

The goal is to develop an application that can manage and analyze data related to spillage events in the Mediterranean Sea. The system will be able to predict the spread of spills over time and study past events to provide detailed analysis. The project will involve research on different ocean models, simulation of oil spills, and analysis of satellite images. The portal will be user-friendly, allowing for easy access to data sheets and graphs with spill characteristics. The system will evaluate the extent of cloud spreading and enable efficient tracking of pollutants back to their source. The aim is to build a solid and efficient program to manage and solve spillage events in the Mediterranean Sea.

The methods used are summarized in the following points: - Collection of series of data from ocean models. - An in-depth study of JavaScript and a detailed study of the Python language, to manage a web application. In particular, use of OceanParcels toolbox to simulate and study the dispersion of virtual particles. - Use of GIS tools for visualise e analyse geospatial information.

Currently, my research has yielded a method for calculating the dispersion of inertial particles that incorporates various contributions, each scaled with specific coefficients. These contributions include surface currents, wind, and waves, resulting in a simulation that closely reflects real-world scenarios. Moving forward, I plan to further enhance the accuracy of this method by automating the calculation of the coefficients from observational data. This will allow for even more precise simulations, providing a valuable tool for studying the dispersions.