Segmentation and tracking of solar eruptive events based on ground based and remote instrument data is essential in pursuit of better answers about the physical nature of these phenomena. A Data-driven approach appears as an obvious solution; nevertheless, engineering of training sets is a challenging task as obtaining labeled ground truth masks requires a significant
The quiet-time solar emission at metric-decimetric wavelengths results from a combination of three main emission mechanisms. The dominant emission mechanism is still the thermal bremsstrahlung from the thermal distribution of coronal electrons, and is ubiquitous throughout the solar atmosphere. Close to and above Active Regions (AR) with strong magnetic fields, gyroresonance emission is produced by
The MOSAIICS team organized the second conference on machine learning and computer vision in heliophysics in Bulgaria, after the one in 2023. This hybrid (online and in person) interdisciplinary conference was held in Sofia between April 07-09, 2025. It focused on applications of Computer Vision and Machine / Deep Learning techniques to heliophysics research and
Recently, we have extended our deep learning approach to tracking solar features and applied it to on-disk solar filaments in EUV and visible light observations. The first model we developed was trained on 1500 H-alpha solar images from the Kanzelhöhe Observatory of the University of Graz in Austria, obtained as part of an on-going collaboration.
After the redesign of the EPREM model, we have applied it to study inner heliospheric energetic particle transport. The source population of particles is a crucial yet very poorly constrained parameter in global solar energetic particle simulations. This is due to the lack of in situ measurements close to the Sun, where the suprathermal particle
Together with colleagues, we studied an impulsive X-class flare from May 10, 2022, where both a faint CME and EUV wave were observed (Rigney et al., 2024). We studied the characteristics of the radio emission, the propagation of the EUV wave and the relation between both phenomena. As the dynamic spectrum in Figure 1 shows,
A significant amount of work was done over the past two years to redesign and rebuild the Energetic Particle Radiation Environment Module (EPREM) global numerical model for energetic particle acceleration and transport in the heliosphere, which the PI Kozarev has helped develop and used extensively in the past. Figure 1 shows the non-standard Lagrangian grid
Solar type III radio bursts are quite common phenomena. They are the result of accelerated electron beams propagating through the solar corona. These bursts are of particular interest as they provide valuable information about the magnetic field and plasma conditions in the corona, which are difficult to measure directly. We have recently completed a study
Solar energetic particles (SEP) are mainly protons and originate from the Sun during solar flares or coronal shock waves. Forecasting the SEP flux is critical for several operational sectors, such as communication and navigation systems, space exploration missions, and aviation flights, as the hazardous radiation may endanger astronauts’, aviation crew and passengers’ health, as well
The MOSAIICS team organized the first of its kind conference on machine learning and computer vision in heliophysics in Bulgaria. This hybrid (online and in person) interdisciplinary workshop was held in Sofia between April 19-21, 2023. It focused on applications of Computer Vision and Machine / Deep Learning techniques to heliophysics research and forecasting frameworks,