Project Overview
The origin of the astrophysical neutrinos detected at TeV-PeV energies remains one of the major open questions in astroparticle physics. This PhD project will address that question through a multi-messenger approach, combining neutrino event lists, gamma-ray source catalogues, variability information, and follow-up observations to investigate whether some active galaxies, especially blazars, are credible neutrino emitters. The project is conceived as a natural continuation of the work carried out by Dr. Enzo Oukacha in his PhD thesis (2025UNIP7179, 2025).
The project will develop along two complementary directions. The first is a statistical approach, based on the search for correlations between neutrino events and gamma-ray blazars using Fermi-LAT catalogues, Fermi light curves, IceCube public event lists, and, when available, selected KM3NeT samples. A key objective will be to move beyond simple positional coincidences by developing a more selective and physically informed association framework.
The second direction is a follow-up approach, aimed at identifying the most promising targets for Imaging Atmospheric Cherenkov Telescopes, especially H.E.S.S. The project will connect neutrino-triggered ranking strategies with very-high-energy observations and source interpretation in a coherent multi-messenger framework.
A central aspect of the thesis will be to combine neutrino-side information with source-side astrophysical knowledge. On the IceCube side, the analysis will exploit the event-level astrophysical probabilities already provided in published samples. On the KM3NeT side, the PhD candidate will investigate a post-reconstruction machine-learning framework to assign analogous astrophysical probabilities to selected high-energy through going muon events. These probabilities will then be used in an upgraded neutrino-blazar association framework.
In a broader perspective, the project may also explore whether the spectral energy distributions of the most promising candidate blazars are naturally described by standard leptonic scenarios or whether they suggest more complex physical interpretations. This additional source-side information could help reinforce the selection of the most physically credible neutrino-emitter candidates and improve the prioritization of targets for follow-up observations.
Scientific Objectives
The main objective of the project is to develop a new multi-messenger framework for identifying and ranking promising associations between high-energy neutrino events and gamma-ray blazars.
More specifically, the thesis will aim to:
- Improve the statistical characterization of local blazar environments around neutrino directions,
- Combine Fermi-LAT source properties with IceCube and KM3NeT event information,
- Construct weighted ranking strategies for candidate neutrino-blazar associations,
- Identify the most promising source populations and individual candidates for follow-up observations,
- And provide a more physically informed interpretation of candidate neutrino-emitting blazars.
through a combination of ranking methods, anomaly detection, and comparison with control or randomized samples in ordetor to identify statistically unusual and astrophysically-meaningful source environments.
Main Research Directions
- Search for neutrino-gamma-ray correlations using Fermi-LAT source catalogues, light curves, and IceCube event lists
- Develop advanced machine-learning methods for weighted neutrino-blazar association studies
- Investigate post-reconstruction event ranking and astrophysical-probability assignment for selected KM3NeT neutrino events
- Contribute to the prioritization of promising targets for H.E.S.S. follow-up observations
- Perform multi-wavelength and multi-messenger analyses of candidate sources
- Explore source-modelling strategies for the interpretation of the most promising blazar candidates
- Publish results in scientific journals and present them at international conferences
Data and Tools
The project will combine:
- Fermi-LAT catalogues and light curves
- IceCube public alerts and event lists
- Selected ANTARES/KM3NeT event samples
- H.E.S.S. data products and simulations
- Python-based scientific analysis and machine-learning tools