Periodic Reporting for period 1 - EDUCADO (Exploring the Deep Universe by Computational Analysis of Data from Observations)
Período documentado: 2024-01-01 hasta 2025-12-31
The formation and evolution of massive galaxies is reasonably well understood in the context of the successful standard ΛCDM formalism. Such simulations of cosmic evolution, however, lead to serious challenges in the regime of very faint galaxies, including the problems referred to as missing satellites, too big to fail, and planes of satellite galaxies. These discrepancies between what is expected from numerical modelling of the formation and evolution of galaxies, and what is observed with ground- and space-based telescopes, denote fundamental issues which call for enhanced precision and understanding on both the observational and the theoretical side. With the massive amounts of excellent data now being produced by modern astronomical surveys, we have a unique chance to solve these problems. To do this, we require innovative developments in information technology.
In EDUCADO (Exploring the Deep Universe by Computational Analysis of Data from Observations), an intensive collaboration at the intersection of astronomy and computer science, we bring together experts from different disciplines and sectors. We train 11 Doctoral Candidates in the development of a variety of high-quality methods, needed to address the formation of the faintest structures. We aim to reliably and reproducibly detect unprecedented numbers of the faintest observable galaxies from new large-area surveys. We will study the morphology, populations, and distribution of large samples of various classes of dwarf galaxies and compare dwarf galaxy populations and properties across different environments. We will confront the results with cosmological models of galaxy formation and evolution. Finally, we will perform detailed, principled, and robust simulations and observations of the Milky Way and the Local Group to compare with dwarf galaxies in other environments.
EDUCADO is delivering a comprehensive interdisciplinary, intersectoral, and international training programme including a secondment at one of our 11 associated partners for each DC. We are providing a fresh and sustainable way of training PhD scientists with interdisciplinary and intersectoral data science expertise, a requisite for future European competitiveness.
Further information on EDUCADO can be found on the project website, https://research.iac.es/proyecto/educado/(se abrirá en una nueva ventana).
In EDUCADO (Exploring the Deep Universe by Computational Analysis of Data from Observations), an intensive collaboration at the intersection of astronomy and computer science, we bring together experts from different disciplines and sectors. We train 11 Doctoral Candidates in the development of a variety of high-quality methods, needed to address the formation of the faintest structures. We aim to reliably and reproducibly detect unprecedented numbers of the faintest observable galaxies from new large-area surveys. We will study the morphology, populations, and distribution of large samples of various classes of dwarf galaxies and compare dwarf galaxy populations and properties across different environments. We will confront the results with cosmological models of galaxy formation and evolution. Finally, we will perform detailed, principled, and robust simulations and observations of the Milky Way and the Local Group to compare with dwarf galaxies in other environments.
EDUCADO is delivering a comprehensive interdisciplinary, intersectoral, and international training programme including a secondment at one of our 11 associated partners for each DC. We are providing a fresh and sustainable way of training PhD scientists with interdisciplinary and intersectoral data science expertise, a requisite for future European competitiveness.
Further information on EDUCADO can be found on the project website, https://research.iac.es/proyecto/educado/(se abrirá en una nueva ventana).
We have so far made excellent progress towards achieving the project objectives, and are fully on track to complete them by the end of the project. The scientific and technical progress is supported by our team of DCs, recruited on time during the first year of the project, and the training, supervision, and development practices which we have implemented as described in the DoA. In particular:
Objective 1: Here we have focused on the development of new algorithms to detect faint dwarf galaxies from new state-of-the-art imaging data sets. The relevant work is mostly carried out in WP2 by DC01 and DC02, with aspects related to the detection of faint structures in images in WP3 through the work done by DC02 on max-trees and by DC04 on extending the Locally Aligned Ant Technique (LAAT). LAAT forms part of the 1DREAM toolbox to find manifold structures in data sets, which DC08 in WP4 uses to detect faint features in imaging. Already in the first phase of the project, all involved DC’s have made significant progress, and several of them have advanced enough to prepare manuscripts which will be published during 2026.
Objective 2: This objective covers two aspects, the first related directly to the properties of dwarf galaxies, and the second more generally analysing complex datasets. The first aspect of covered mostly by DC02 in WP2, and that part of the work has already led to a completed manuscript which will be submitted for publication in early 2026, on the evolution of intrinsic 3D shapes of star-forming galaxies. This work can be considered a proof of concept, which will in the second part of the project be applied to very large sets of imaged galaxies from the Euclid and LSST data sets. The second aspect is mostly developed within WP4, where DC09 and DC10 use advanced computational and analysis approaches to extract physical properties of galaxies and their components from multi-dimensional data sets, in our case integral field spectroscopy and radio interferometric imaging of spiral and dwarf galaxies.
Objective 3: This aspect of our overall scientific work focuses on the comparison of our own Milky Way galaxy and its dwarf companions with the similar galaxies studied primarily under Objective 1. Our WP3 is mostly centred on the study of our Milky Way, combining work in computer science (DC04) with the analysis of astronomical data, including those from Gaia (DC05 and DC06), and with numerical simulations of Milky Way analogues (DC07). The DCs have progressed well individually, but they have also worked very closely together. Through secondments and collaborative work, we expect this unique combination of computer science with observational and numerical astrophysics to lead to ground-breaking results in the second half of our project, and to reaching our Objective 3.
Overall, progress is excellent, and we are well on track to reach the main objectives set for the scientific work within the project.
Objective 1: Here we have focused on the development of new algorithms to detect faint dwarf galaxies from new state-of-the-art imaging data sets. The relevant work is mostly carried out in WP2 by DC01 and DC02, with aspects related to the detection of faint structures in images in WP3 through the work done by DC02 on max-trees and by DC04 on extending the Locally Aligned Ant Technique (LAAT). LAAT forms part of the 1DREAM toolbox to find manifold structures in data sets, which DC08 in WP4 uses to detect faint features in imaging. Already in the first phase of the project, all involved DC’s have made significant progress, and several of them have advanced enough to prepare manuscripts which will be published during 2026.
Objective 2: This objective covers two aspects, the first related directly to the properties of dwarf galaxies, and the second more generally analysing complex datasets. The first aspect of covered mostly by DC02 in WP2, and that part of the work has already led to a completed manuscript which will be submitted for publication in early 2026, on the evolution of intrinsic 3D shapes of star-forming galaxies. This work can be considered a proof of concept, which will in the second part of the project be applied to very large sets of imaged galaxies from the Euclid and LSST data sets. The second aspect is mostly developed within WP4, where DC09 and DC10 use advanced computational and analysis approaches to extract physical properties of galaxies and their components from multi-dimensional data sets, in our case integral field spectroscopy and radio interferometric imaging of spiral and dwarf galaxies.
Objective 3: This aspect of our overall scientific work focuses on the comparison of our own Milky Way galaxy and its dwarf companions with the similar galaxies studied primarily under Objective 1. Our WP3 is mostly centred on the study of our Milky Way, combining work in computer science (DC04) with the analysis of astronomical data, including those from Gaia (DC05 and DC06), and with numerical simulations of Milky Way analogues (DC07). The DCs have progressed well individually, but they have also worked very closely together. Through secondments and collaborative work, we expect this unique combination of computer science with observational and numerical astrophysics to lead to ground-breaking results in the second half of our project, and to reaching our Objective 3.
Overall, progress is excellent, and we are well on track to reach the main objectives set for the scientific work within the project.
All DCs in their individual research projects are performing cutting-edge research on the interface of astrophysics and computer science. Through aspects including the use of new data, the development of new algorithms and methods, and the application of those to observational data, all are extending the boundaries of knowledge and producing results that go beyond the state-of-the-art. For example, we are detecting dwarf galaxies in deeper images than ever before, and are extending the study of galaxy shapes from more traditional imaging to the newly available Euclid space telescope images. We are developing new computer science techniques, extending existing techniques by implementing the latest advances in AI. We are combining new data and analysis methods to increase our understanding of our Milky Way and its immediate surroundings. And we are developing and implementing new analysis techniques for integral field spectroscopic data sets, which are already leading to new insights into the structure and evolution of galaxies. Many of these findings have already been submitted for publication in top professional journals, with many other to follow soon.