My research focuses on stellar astrophysics, particularly the distinct evolutionary phases of massive stars across diverse galactic environments. I have extensive observational experience, utilizing both small and large facilities equipped with a wide array of state-of-the-art instruments. I apply statistical analysis and machine/deep-learning techniques to address complex astrophysical challenges. As a strong advocate for professional-amateur collaborations, I am committed to fostering openness and inclusivity within the scientific community.
Full list of publications: NASA/ADS and Google Scholar
Research IDs: ORCID 0000-0002-0891-7564 and Web of Science ResearcherID S-5054-2017
…Machine-learning applications
The rise in computational power and the continuous development of machine/deep learning techniques have transformed their application from scientific contexts to everyday life. Simultaneously, the increasing availability of data from modern astronomical surveys and facilities underscores the imperative nature of employing these automated approaches.
Throughout my career I have consistently used statistical tools and have applied machine/deep learning approaches in solving complex astrophysical problems (including data mining and cleaning, limited datasets, missing values, feature engineering optimization). Some example cases:
- During my current work for the ASSESS project I have been responsible to develop a machine-learning algorithm (using an ensemble approach combining Support Vector Machines, Random Forest, Multilayer Perceptron). The purpose of this classifier is to provide a classification prediction using multi-wavelength all-sky photometric surveys, such as Pan-STARRS, GAIA, 2MASS, Spitzer.
(Maravelias+2022a, 2022b) - During my PhD studies I worked on the development of an automated classification method for the optical counterparts of X-ray binaries, using a Naive Bayes classifier. This work was further expanded by implementing a Random Forest classifier by the MSc. student Elias Kyritsis (who I co-supervised, with Prof. A. Zezas, Univ. of Crete).
(Maravelias 2014; Kyritsis+2022) - Application of Bayesian approaches and Gaussian mixture processes to identify cluster ages and cluster formation episodes in the Magellanic Clouds.
(Bitsakis+2017, 2018)
…Massive stars in different galaxies
Massive stars (defined as those with masses above 8 Msolar) play a critical role in the Universe, contributing significantly through their intense UV radiation, powerful stellar winds, and eventual supernova explosions. Their impact spans from influencing the re-ionization of the early Universe (Haiman+1997) to enriching the interstellar medium with heavy elements, shaping galaxy formation and evolution (Aoki+2014), and serving as a primary source of momentum in the Universe (Bresolin+2008). The death of massive stars, observable across cosmic distances, leads to some of the most energetic phenomena, including gamma-ray bursts (Yoon+2006). Understanding the evolution of massive stars, particularly up to their supernova deaths, is crucial for comprehending their profound influence
Massive stars have a wide range of manifestations: Wolf-Rayet (WRs), Luminous Blue Variables (LBVs), Blue Supergiants (BSGs), B[e] Supergiants (B[e]SGs), Red Supergiants (RSGs), Yellow Supergiants (YSGs), O/B main-sequence stars, Be stars. Although all these acronyms may sound bizarre there is a common physical underline mechanism: all stars lose mass from the start of their lives up to their spectacular end as supernovae. This continuous mass loss transfers energy and momentum to the interstellar medium, and chemically processed material as the stars evolve, contributing significantly in the galactic environment of their host galaxies. The major factors determining the evolution and final stages of a single massive star are metallicity, stellar rotation, and mass-loss (Ekström+2012; Georgy+2013; Smith 2014). Additionally, the presence of a companion (most typical among massive stars, with binary fractions of ~50-70%; Sana+2012, 2013; Dunstall+2015) can significantly alter the evolution of a star through strong interactions. Although, all these factors critically determine the future evolution of the star and how it will end its life, they are, unfortunately, not well-constrained in many cases. Although stellar evolution models had great successes we have not yet complete the puzzle.
Towards this direction I have worked to understand the properties of specific phases and their possible connecting mechanisms in a wide range of galactic environments, including the Milky Way, the Magellanic Clouds, and other galaxies. Example publications include:
- Studies of the circumstellar environments of B[e] Supergiants and Luminous Blue Variables.
(Maravelias+2015; Kraus+2016, 2017; Maravelias+2018; Torres+2018, Maravelias+2023) - Similarities and possible physical links between B[e] Supergiants with other phases such as Luminous Blue Variables, and Yellow Hypergiants.
(Maravelias+2014; Aret+2017a, 2017b; Kraus+2022; Maravelias+2023) - Studying the properties and evolutionary status of Yellow Hypergiants.
(Kraus+2019; Kourniotis+2022; Maravelias & Kraus 2022) - Properties of Red Supergiants and mass loss.
(Yang+2021; de Wit+2023; Yang+2023) - Massive stars populations in different galactic environments and low metallicity:
(Maravelias+2017, 2019; Yang+2019, 2020, 2021, 2023; Kourniotis+2022; Vink+2023; Bonanos+2023).
…X-ray binaries
High mass stars end their life with a spectacular explosion (supernova) and the outcome will be either a compact object or a black hole. If this happens within a binary system then the system will enter the X-ray binary phase (e.g. Langer+2020). In this case the massive star loses material either through a decretion equatorial disk or through strong stellar winds and results into accretion onto the compact object, followed by X-ray emission (Reig 2011). Hence, these systems are valuable to understand matter behavior under extreme physical conditions, disk creation and accretion, being in reality the low-mass equivalents of accretion onto massive black holes in galaxies. Moreover, the X-ray systems are indispensable tools to test stellar and binary evolution models, formation and evolution of compact objects, and are the channel to create compact object binaries, the progenitors of gravitational wave sources. (e.g. Bavera+2023).
I have been involved in studies of X-ray binaries with a focus on their optical counterparts. Some example works:
- Properties of the OB parent populations.
(Maravelias+2014, 2017, 2019; Kyritsis+2022) - Classification of X-ray binary counterparts with optical spectroscopy and imaging.
(Maravelias+2014, 2018b, 2018c; Kyritsis+2022) - Optical counterparts of gravitational wave events.
(Drout+2017, Shappee+2017)
…Professional-amateur collaborations & Outreach
Over the last three decades there has been a revolution relative to the availability of “hardware” (telescopes, CCD and video cameras) for amateur astronomers, which combined with software improvements and networking accessibility amateurs can contribute invaluable help. As an enthusiast of professional-amateur collaborations, I consistently work towards communicating with amateurs to convey the best practices for scientifically valuable observations.
This effort involves contributing to international projects and initiating campaigns and projects. Some examples:
- Stellar variability using amateur data.
(Maravelias+2012, Kraus+2019; Maravelias & Kraus 2022) - Contribution to professional-amateur projects in planetary atmospheres.
(Kardasis+2016, 2020, 2022) - Organization of multiple hands-on events and workshops on training amateurs in observing techniques for a wide range of topics, and subsequent publications of the experience gained.
( Voutyras+2013; Kardasis+2015; Maravelias+2018; Moutsouroufi+2019) - Contributing to the international campaign for the eclipse of epsilon Aurigae and leading its Greek part. I was responsible for: (i) communicating news to the public, (ii) training participants to observe, (iii) handling and analyzing the data, (iv) writing the final report (for Greece).
(Kloppenborg+2012; Maravelias+2012). - Coordinating the Greek campaign to observe the 2011 Draconid meteor outburst, for which I was responsible to: (i) communicating the campaign to the public, (ii) inform/train potential observers, (iii) coordinate observing parties.
- Meteor observations, including visual recording and analysis, as well as setting up a video station.
(Maravelias 2011, 2012) - From 2003-2023 i was appointed the positions of the position of Director of the Meteor and the Variable Stars Sections of the Hellenic Amateur Astronomy Association.
- I have participated and organized multiple outreach events for the public and schools.