Scientific work in SWATNet is divided in three Work Packages (WPs) consisting of 3-5 individual research projects. Each project forms the basis of a PhD work for one student. The WPs are briefly described below

WP1: Modelling and forecasting solar activity

WP1 focuses on solar eruptions and their solar cycle variations. The core premise of this WP is that understanding and providing context to the physical conditions of the magnetic field and plasma prior to major eruptions is key to enhancing our eruption prediction capability. WP1 studies both non-eruptive (i.e., confined eruptions) and eruptive events originating from active regions of the Sun and investigates pre-eruption dynamics, deciphers and tracks the evolution of magnetic topology both at the photospheric boundary and above active regions. Advanced dynamo modelling will yield new insight on longer term variations in solar activity.

List of WP1 projects

  • Project 1 Pre-eruption magnetic configuration and eruption forecasting
  • Project 2 Assessment of the Near-Sun CME Magnetic Field
  • Project 3 Three-dimensional solar flare forecasting
  • Project 4 Modelling periodic and quasi-periodic variations in solar activity

WP2: Coronal and heliospheric modelling and forecasting

WP2 aims at breakthroughs in modelling the solar wind and the propagation of solar eruptions and charged particles they accelerate in the corona and heliosphere. This WP explores the formation and propagation of solar eruptions and their role they have in the self-consistent acceleration of energetic particles in realistic and complex solar-wind backgrounds in a time-dependent manner through novel advanced simulations. Arrival time predictions will be improved using probabilistic drag-based modelling along sophisticated three-dimensional magnetohydrodynamic simulations.

List of WP2 projects

  • Project 5 Global MHD coronal model
  • Project 6 CME evoluation in the corona
  • Project 7 Particle acceleration at coronal shocks
  • Project 8 Particle transport in interplanetary medium
  • Project 9 Forecasting CME arriva in the whole heliosphere

WP3 Forecasting Space Weather with Artificial Intelligence

WP3 uses latest approaches in Artificial Intelligence (AI) and Machine Learning (ML) to advance the classification and forecasting of space weather. The use of Neural Networks to forecast the solar wind velocity or the onset of geomagnetic storm dates back to the late 90s, but those works relied on a restricting small number of parameters. The unique opportunity for this WP arises from the wealth of solar observations now available and the recent huge leaps in AI and Machine Vision knowledge opening new horizons for using these approaches also for space weather purposes.

List of WP3 projects

  • Project 10 Forecasting solar activity with deep learning
  • Project 11 CME arrival modelling with Machine Learning
  • Project 12 Development of mathemathical morphology algorithms to characterize the solar activity