2019 ESAC TRAINEES
Characterizing red stars in the Galactic Plane with Gaia, 2MASS, and GALANTE
Tutors: Jesús Maíz Apellániz , Pedro García Lario
The advent of large-scale photometric and astrometric surveys has opened the way to the study of the Galactic stellar population beyond 1 kpc, where photometry is dominated by the effect of extinction instead of the intrinsic spectral energy distribution of the stars, making all stars “red". At large distances the majority of the observed population in those surveys is composed of bright red giants but among them we also find obscured OB and AGB stars, among other interesting objects. Finding those needles in the haystack is the purpose of this traineeship. The trainee will combine the recent Gaia DR2 photometry and astrometry with ground-based photometry in the NIR (2MASS) and optical (GALANTE) to characterize the obscured population of the Galactic Plane, identify the different types of objects, and measure their extinction. This work is suitable to be considered as a research master project.
Diego Moral Pombo
Support for the CESAR Radio Telescope and Related Tools
Tutors: Michel G. Breitfellner, Manuel Castillo, Javier Ventura
CESAR (Cooperation through Education in Science and Astronomy Research) is a joint educational programme developed by the European Space Agency (ESA), the Spanish National Institute for Aerospace Technology (INTA) and INTA-owned company Isdefe. Its objective is to provide students from European secondary schools and universities with hands-on experience in Optical and Radio Astronomy. In addition, CESAR shall contribute with outreach activities to promote Space Science and to stimulate European students' interest in Science and Technology in general and Astronomy in particular.
The CESAR programme owns several telescopes and observatories which need improvements and upgrades. One of these observatories is the CESAR ESAC Radio Telescope (CERT). It is the result of the refurbishment as astronomical facility of the ESTRACK antennas located at ESAC, Spain. INTA contributes to CESAR sharing the use of a satellite tracking antenna (VIL-2) located in ESAC for education and science activities.
CERT consists of a Cassegrain antenna of 15m diameter, the added RF subsystems working in S-band and the operative control subsystems. The existing equipment together with a power meter and a Spectrum analyzer allow to perform radiometric measurement of signals emitted by natural radio sources. At present, after several research and development activities, it has been proven that several radio sources can be detected, measured and studied with CERT to show different radio emission mechanisms and its relation with different astronomical phenomena.
The project objective is to contribute to the CERT fine tuning for its scientific and educational operation in close collaboration with the CESAR/CERT team. His/her support will be required for the following activities:
- Participation in CERT observation and calibration campaigns;
- Carry out own and assigned Radio Astronomy observation plans;
- Evaluate and improve the CERT data analysis and processing tools;
- Evaluate the existing HW/SW identifying possible improvements or alternatives;
- Write new and update existing CERT documentation;
- Preparation of Radio Astronomy class room resources for both teachers and students.
The final goal of the project is to enable CERT for its systematic scientific and educational use.
Francisco Jesús Acién Pérez
Software development for space science education initiative
Tutors: David Cabezas, Michel G. Breitfellner, Javier Ventura-Traveset
CesaR (Cooperation through Education in Science and Astronomy Research) is a joint educational programme developed by the European Space Agency (ESA), the Spanish National Institute for Aerospace Technology (INTA) and INTA-owned company Isdefe. Its objective is to provide students from European secondary schools and universities with hands-on experience in Optical and Radio Astronomy. In addition, CESAR shall contribute with outreach activities to promote Space Science and to stimulate European students' interest in Science and Technology in general and Astronomy in particular.
The CesaR initiative has its own observing facilities which allow to observe the universe actively on-line through a dedicated Control Centre hosted at ESA’s European Space Astronomy Centre (ESAC) in Madrid, Spain.
ESA is offering a 6 months traineeship opportunity within the CesaR education initiative. The main tasks to be carried out by the trainee are the following:
- Create a Web Archive Viewer for the images produced by the CesaR observatories.
- Software integration and interconnection of subsystems.
- Improve the interactive scientific cases tools for schools and universities.
- Help creating new software for education using the latest technologies (mobile apps, vr glasses, 3D interfaces).
The final result of the project should be a set of software, apps and/or database applications with new and/or improved functionality.
Social media and communication with emphasis on outreach and education
Tutors: Emmet Fletcher, Beatriz Arias, Arantxa Alonso, David Cabezas, Michel G. Breitfellner
ESAC Communication Office is responsible for official ESA communication activities in Spain and Portugal, organisation of launch and major events with national institutions (Museums, Planetariums, etc); all the Spanish and Portuguese media activities related to space; exhibitions; run and coordinate the Spanish and Portuguese ESA websites, social media activities, Intranet, as well as coordination of VIP visits to ESAC.
CesaR (Cooperation through Education in Science and Astronomy Research) is a joint educational programme developed by the European Space Agency (ESA), the Spanish National Institute for Aerospace Technology (INTA) and INTA-owned company Isdefe. Its objective is to provide students from European secondary schools and universities with hands-on experience in Optical and Radio Astronomy. In addition, CESAR shall contribute with outreach activities to promote Space Science and to stimulate European students' interest in Science and Technology in general and Astronomy in particular. It is based at ESA’s European Space Astronomy Centre (ESAC) in Villanueva de la Cañada, Madrid, Spain.
ESA is offering a 6 months traineeship opportunity at ESAC. The main tasks to be supported are:
- Audio visual support (recording videos and/or taking pictures of events and subsequent editing)
- Support in the organisation of events (press conferences, institutional events, VIP visits, social media events, CesaR workshops, videoconferencing, …)
- Support with the office inventory and ESA Communications Office monthly report
- Support with the design and implementation of strategy for social media for CesaR
- Manage and edit CesaR news for the web and other media
- Design and create multimedia contents for the CesaR web pages
- Help with the daily interaction related to CesaR’s Space Science Experience (SSE) programme with teachers and schools
- Carry out web and social network analysis and SEO monitoring
Eduardo Alonso Peleato
Cross Visualization ExoMars16 CASSiS and Mars Express HRSC Footprints
Tutors: Isa Barbarisi, Jaime Saiz, Tanya Lim, Emmanuel Grotheer, Bruno Merin
Within the Data and Engineering Division, the ESAC Science Data Centre (ESDC) is responsible for developing and operating the on-line science archives of all ESA space science missions (astronomy, planetary and solar heliophysics), in close cooperation with the Science Operations Centres at ESAC, the instrument teams and the scientific community. In particular, the Planetary Science Archive hosts all ESA planetary mission science data holdings and represent a unique platform for planetary science data exploitation.
The Trainee will be fully integrated into a small team of software engineers and planetary scientists at ESAC and will participate to the development of a prototype designed within the PSA (using the Agile/Scrum software development methodology), in particular in the context of Exomars 2016 CASSiS Footprints Visualization within a GIS Map Browser.
Duration: 6 months
Specific Tasks and Objectives: The student will work within the already existing infrastructure set in place in the ESAC Science Data Centre (Planetary department) in collaboration with Engineers and Scientists.
The goal of this work-package is to visualize both Mars Express footprints coming from HRSC instrument and ExoMars16 footprints from CASSiS instrument in a Map Browser Interface.
The student will be provided with the following inputs:
- EM16 Cassis PDS4 Products containing geometry
- MEX HRSC Datasets containing multiple products with well-established geometry
- Scientific Use Case
- User Interface (Map Browser) Prototype within Vaadin Framework
The student will deliver the following output:
- PDS4 Geometry Parser library in Java
- Definition and Implementation of a Data Model able to accommodate PDS4 geometry data and metadata
- Geoserver Layer able to access geometry data and metadata
- Set of Java classes able to display PDS4 Geometry in a GIS Map
- (Time allowing) Set of recommendations to PDS4 geometry working group for optimal display of PDS4 geometry in the PSA Map Browser
Output and Performance Monitoring:
The student will get experience with PDS3 and PDS4 standard, GIS standard, PDS3 and PDS4 geometry definition, Vaadin and OpenLayer. He will implement all the workflow, from ingestion to visualization of CASSiS products according to the PSA GIS architecture.
The output shall be measured and assessed on the basis of:
- The trainee will gain knowledge of various common data formats in planetary science and how/when to use each of them.
- The trainees will gain knowledge of GIS tools and standards
- The trainee will gain knowledge of PDS4 Standard and Geometry working group.
- The trainee will learn how to research a problem in the domain of scientific computing, and implement their solution within an existing framework.
- The trainee will gain experience of the challenges associated with long term archiving of scientific data.
- The trainee will gain experience of day-to-day work in, and the culture of an International Organization.
- The trainee will learn about science operations and the variety of activities undertaken at ESAC.
- Daily Interaction with ESDC Engineers and Scientists within the framework of the Scrum methodology
- Weekly progress meetings with mentors and other scientists and engineers at ESAC. More informal discussions as required.
- Final assessment of the trainee’s work.
Specific Qualification Requirements:
- Applicants should be in their final years of a University course at Masters Level (or equivalent) in computer science or planetary science.
- Experience with Java programming languages is required. Knowledge of GIS, WebGL and PostgreSQL would be an asset.
- Applicants should have good interpersonal and communication skills and should be able to work in a multi-cultural environment, both independently and as part of a team.
- Applicants should be interested in planetary science, be open and curious minded. A good proficiency in English is required.
David Leonard Wenzel
Allowing users to google for ESA Space Science Datasets
Tutors: Jesús Salgado, Beatriz Martínez, Isa Barbarisi and Bruno Merín
Google has recently released a “Dataset Search” interface in beta to allow world-wide users to search for and discover public datasets for scientific or other interests (see search interface at https://toolbox.google.com/datasetsearch and an article about the interface in https://www.nature.com/articles/d41586-018-06201-x). While the future scientific use of this interface is to be determined, it is undeniable that a potentially large fraction of the world population might be using this interface to search for scientific data in the next couple of years.
At the present time, the datasets from all of ESA’s Astronomy, Planetary and Heliophysics extremely rich datasets cannot be found through this interface because they have not yet been indexed with the schema.org community-supported indexing schema. The work of the trainee consists in re-parsing the metadata tables from the ~ 20 ESA’s space data science repositories (see the list in https://cosmos.esa.int/web/esdc) from their original data model schemas into schema.org, publish those schemas online and get google to index them such that they can be found in Google’s Dataset Search engine. As an extra, the trainee could work together with the ESDC staff to enable automatic updates of newly upcoming data to the archives to the same system to keep it always up to date.
The trainee will work in the ESAC Science Data Center, supported by a team of very experienced software engineers and scientists, all specialists in ESA's datasets and will apply agile/scrum development methodology.
Optimizing two-point correlation statistics using Machine Learning techniques
Tutors: Ginevra Favole, Bruno Altieri, Roland Vavrek, Antonio D. Montero-Dorta (São Paulo University)
New-generation spectroscopic surveys as SDSS-IV/eBOSS, DESI, Euclid, 4MOST, WFIRST and Subaru PFS will target emission line galaxies (ELGs) out to redshift z~2 over enormous cosmological volumes. The clustering signal of these objects will be used as fundamental cosmological probe to trace the baryon acoustic oscillation feature, measure the growth rate of structure, study the distribution of matter and the star formation history of the Universe, and to unveil the nature of dark energy. In parallel, high-quality, multi-color imaging programmes as DECaLS, DES and Subaru HSC are providing estimates of galaxy shapes and photometric redshifts with unprecedented accuracy. These observations will critically improve our current understanding of the complex mechanisms that regulate the formation and evolution of galaxies. Therefore, understanding how to best measure and model the galaxy clustering properties of high-redshift emitters is a fundamental task we have to address now using the computing techniques currently available.
The main goal of this project is to build and deliver to the scientific community a software script optimized to compute two-point correlation functions (2PCFs), either from large observational data sets or large-volume highresolutioncosmological simulations. We will measure the 2PCFs of different SDSS-III/BOSS galaxy samples, selecting them by color, luminosity and morphology. In order to group galaxy populations with similar properties, we intend to use the scikit-learn clustering algorithm. We will interpret our results using the latest products of MultiDark-Galaxies, currently the cosmological simulation with best resolution/volume available run with different semi-analytic models of galaxy formation. The computational outcomes of this project will be of major interest for the new-generation of cosmological surveys, first of all the Euclid ESA mission.
Isabel Poveda Galvez
Building Information Management (BIM) tool
Tutors: Panos Machairas
The EFM Service at ESAC is developing a new method for the construction and follow up management of the ESA infrastructure. The method is based on the creation of three-dimensional object-based building models that contain all the information needed for the construction and the consequent Life Cycle Management of the asset.
The aim of the present Trainee Project, is to put in operation and test in practice the work performed in the course of a previous young graduate trainee project, during which the complete model of a building was created; it is moreover our aim to extend this work to additional buildings at ESAC.
The selected candidate has to be able to work autonomously in a multi-faceted development and application project, particularly in the coordination, application and integration phases.
Gaia, a micrometeoroid hit detector?
Tutors: Cian Crowley, Alex Bombrun, Juanma Fleitas, Nicolas Altobelli, Uwe Lammers
Description: ESAC is responsible for developing and running AGIS, the software that computes the astrometric parameters for the Gaia mission. As a side product of our data reduction, we identify features in our scanning rate data that originate from major micrometeoroid hits on the spacecraft. It turns out that Gaia seems to be a promising micrometeoroid hit detector. In order to better understand the limits on Gaia’s micrometeoroid hit detection we are looking for a candidate to improve our open source hit simulator (see https://github.com/bombrun/GaiaLab), as well as to continue the analysis of the real Gaia data. The results of this work will be then compared to the distribution of detected hits from simulations that are based on an up-to-date model of the solar system.
GaiaLab: simulation of a global astrometric solution
Tutors: Alex Bombrun, Jose Hernandez, Alfonso De Torres
ESAC is responsible for developing and running AGIS, the software that computes the astrometric parameters for the Gaia mission. The design and validation of Gaia global astrometric solution requires to be able to run simulations that include complex calibration issues. The current state of the art is AgisLab. This code is proprietary of DPAC, the scientific consortium processing the Gaia data and responsible for the publication of the final star catalogue. This code requires complex configuration and is very tied to Gaia data model.
Last year we have started the GaiaLab open source project, a simplified version of AgisLab in Python available on https://github.com/bombrun/GaiaLab. It is developed by students for students. The first part of the project has been carried on to simulate observations with a nominal scanning law and one has started to investigate some parts of the underlying least square problem. In order to simulate a global astrometric solution further developments are needed.
The ideal candidate should demonstrate basic knowledge of mathematics (linear algebra, polynomial, trigonometry) and more important the motivation to promote an open source project around global astrometry. The trainee will receive support regarding astrometry and Python.
Fast Outflows in Protoplanetary Nebulae and young Planetary Nebula observed by Herschel/HIFI
Tutors: Pedro Garcia-Lario, David Teyssier, Valentin Bujarrabal
The HIFI instrument (Heterodyne Instrument for the Far-Infrared) was the high-resolution spectrometer on-board Herschel. After almost 4 years of activity in orbit, a unique scientific data-base has been collected, offering a mine of information for scientific investigation. Among the unique data obtained by this instrument were a set of observations in mid- to high-excitation transitions of 12CO in the envelopes of Protoplanetary Nebulae (PPNe) and young Planetary Nebula (PNe). A sub-set of those data was analysed and published by Bujarrabal et al. 2012, and was later augmented by companion project (Danilovich et al. 2015) allowing to collect data from various transitions ranging from J=1-0 to J=14-13, evidencing a large variety of velocity profile and of physicochemical conditions at play in the environments of those objets.
The goal of this project is to study the physical conditions in the fast outflows observed in some of those evolved stars by means of the comparison between the line intensity observed at different locations of the envelopes and winds, and in different transitions of the 12CO line. In addition, observations obtained in SiO in some of the targets will be used in order to probe different temperature and density conditions. The project involves analysis of high-resolution spectra obtained in about a dozen of sources, and comparison with simple model of line excitations in order to derive parameters such as density and temperatures. If time allows, modelling using more sophisticated tools (e.g. SHAPEMOL, Santander-Garcia et al. 2015) could be considered in selected targets.
ESA Human Resources – Procedure for Newcomers
Tutors: Javier Delgado, Patricia Donat
ESAC constantly receives new staff (Young Graduate Trainees, Research Fellows and permanent staff). Their arrival is known some months in advance and some procedures need to be implemented before their arrival.
The project would consist in giving administrative support to establish a formal procedure before the arrival of the newcomers. This work would be done together with the Facility Management team and the Division Assistants. The post-holder will have to understand ESA’s structure, attend to the internal meetings and prepare a document to be distributed to the people involved.
Bert Vander Meulen
Multi-purpose Geometric Light curve model
Tutors: Jan-Uwe Ness
From large distances, small geometric structures such as binary stars cannot be resolved directly, but under certain circumstances such as viewing from the side, the evolution of the total brightness allows reconstruction of the geometric configuration from eclipses, shadowing, reflection etc. Light curves (brightness evolution with time) can be simulated assuming certain geometric configurations. Good agreement with an observed light curve can be interpreted as the assumed configuration to correspond to the unresolved system configuration.
Publicly available light curve models are limited to special situations without being flexible enough to be expanded to address new questions. The objective of this project is to develop a public multi-purpose light curve simulation tool. While such a tool will always be limited in scope, a modular design allows newly arising questions to be addressed by adding new modules.
The final tool should be documented for both, users and developers, such that it can be further expanded by others (e.g. future trainees).
Héctor Rueda Ricarte
Rosetta data tutorials using Jupyter Notebooks
Tutors: Mark S. Bentley, Dave Heather
The successful Rosetta mission generated a wealth of data from a variety of instruments onboard the orbiter and lander – from cameras to spectrometers, microscopes to magnetometers. Many of the instruments teams have recently been working on delivering enhanced data products to the Planetary Science Archive. As such there is now a treasure trove of data ready for analysis, but the barrier to getting started with this analysis is often high. This project proposes that a trainee pick one or more Rosetta instruments and digs into their data format to understand and, if possible, try to reproduce the results of a publishes paper. The result would ideally be a series of Jupyter Notebooks with both complete code and a description of the end-to-end process. To facilitate this, the EPN-TAP (EuroPlanet Table Access Protocol) interface to the PSA could be used. This allows data to be queried and retrieved directly into a Notebook, meaning that anyone could run the notebook to retrieve, process and display the data “live”. An example of such a notebook, written by one of the proposers for the Rosetta MIDAS instrument, is available on github and described in this blog post. The results of such Notebooks could be made available, for example, on BitBucket and linked from the Rosetta mission Cosmos pages to helpfuture users start to use the data.
Oriol Cervelló i Nogués
Universitat Politècnica de Catalunya - BarcelonaTech (UPC)
Earth Tides Measurement with GNSS
Tutors: Manuel Castillo, Fernando Martín-Porqueras, Javier Ventura-Traveset
The GNSS Science Service Centre (GSSC) works with GNSS data and products for scientific GNSS applications and research. They have many scientific receivers, GNSS data processors and expert software tools. Moreover, they manage external GNSS scientific projects.
Since long time, it was suspected that a relation exists between the tidal stress produced by the Moon and the Sun on the Earth crust (Earth tides) and the occurrence of big earthquakes and volcanic events. Few years ago, it was shown that a significant statistical correlation exists between earthquake and the tidal Sun-Moon influence. An Earth Tides model is considered within the Precise Point Positioning (PPP) Techniques as one of the corrections to be performed for higher accuracy in the GNSS positioning.
The proposed project I carried was a first data exploration to have a glimpse of the extent of Earth tides in the GNSS signals. Using the data provided by the GSSC acting as a Global Data Center of the International GNSS Service (IGS) we made a first analysis of the Earth tides from different IGS stations. The selected stations were close ones but positioned over different geological ground, to check if the Earth tide have different behaviors (amplitudes and frequencies). Using GNSS data, a new type of Earth tides model may be created to have a better characterization of the ground type beneath the station and its behavior to the tide. As future extend of the project, once having a better characterization of the Earth tides it could be possible to correlate them with seismic and volcanic events, as demonstrated, and use GNSS stations as a monitoring network for these events.
During my trainee I also had the opportunity to work with and install some GNSS receivers and antennas from the GSSC lab. Participate in progress meetings of projects the GSSC carry with external organizations. And enjoy other ESAC activities, such as organized seminars in which it was possible to get to know projects from many areas carried by the ESA.
Noiseless Image Enhancement in Astronomy
Tutors: Maggie Lieu, Lyndsay Old, Bruno Altieri, Ivan Valtchanov
In terms of observation time, astronomy is expensive. In such low light conditions, our data suffers from low photon counts and low signal-to-noise, in particular if our observation times are short. By using longer exposure times we can improve the noise levels, but at the cost of blurring. Often in astronomy we tend to stack multiple images to simultaneously reduce the noise and increase the signal. But more recently, Chen et al 2018 have developed a convolutional network that is able to enhance low light images to qualities better than any traditional signal processing technique, removing both noise and colour biases. Such a technique if successfully applied to astronomy could have big implications for astronomers, including significantly reducing observation times and cost. In this project we will train the convolutional network on our astronomical data and compare it to more traditional methods such as stacking, and testing the scientific viability of machine learning
Marta Donate Fuente
CubeSat: cubesat communications subsystem
Tutors: Julio Gallegos, Xavier Dupac, Fernando Martín Porqueras
Our group has been working on the development of different subsystems for a CubeSat with the focus on an astronomical application. We would like to continue in the refinement of what we already have and advance the project to a full system in the near future.
If you decide to join us, you will work on the communication system to expand the present UHF/VHF to include a S-Band channel and to study the feasibility of an X-Band system. The ground station is fully operation in UHF/VHF and it will be part of the work to operate and decode the signals received. In addition, you will work on the link between the ground station and the AOCS demonstrator and, if time permits, a link to a simulated rover vehicle.
The roadmap for this project would include the design of the S-band addition to the present ground station, integration of the AOCS with UHF/VHF communication system and the power system. In parallel, the operation of the ground station. This project has a strong hardware component and you will need to work with electronics and antennas.
Victoria Skiba Maldonado
Universitat Politècnica de Catalunya (UPC)
Improving the quality of Science Ground Segment
Tutors: Maria Garcia-Reinaldos, Julio Gallegos, Jose Marcos, Silvia de Castro, Luis Martin
The prime objective of the Science Operations Product Assurance and Quality unit (SCI-OQ) is to ensure that science operations projects accomplish their defined mission objectives in a safe, available and reliable way. The management of Product Assurance is fully embedded in the management of the project. The early identification of aspects potentially detrimental for safety and mission success, and the cost-effective prevention of any adverse consequence of such aspects are the basic principles for the ECSS Product Assurance requirements. The disciplines covered by the unit are: product assurance programme implementation, processes assurance and product assurance.
Under the scope of the product and processes assurance, the SCI-OQ unit is interested in having objective evidences of the quality of the processes and the products in the different Science Ground Segment projects. PA/QA support is provided to those projects individually but up to now there are no quality measures of all projects as a whole. Having global quality metrics would allow SCI-OQ to have a global view of the SCI-O projects quality, to comparatively analyse the different projects and to draw conclusions about the quality of the Science Ground Segment projects as a whole. These conclusions will be later on used to define concrete quality objectives for the future.
SCI-OQ has recently started the preparation of a Product Assurance Environment that can be used to perform tests and analysis of the project processes and products using tools replicated from the projects and being able to simulate alternative solutions to improve quality.
During my traineeship I helped to develop a tool for the QA/PA team to extract objective data that measured the quality of the missions. I worked in the definition, setup, configuration and maintenance of the Product Assurance Environment while requested to define, collect, analyse and report on global metrics of applicability to all science projects. I was required to interact with the different projects, to extract results and to critically analyse them and extract conclusions that can be implemented to improve the quality of the processes and the products.
Marina de Brito Barba
Advanced exploration of the Solar System through virtual reality
Tutors: Marc Costa, Bjoern Grieger, Vicente Navarro, Christophe Arviset
The ESA SPICE Service (ESS) provides ESA´s Solar System Exploration missions ancillary data and geometry information to the science community and to the science ground segments in the shape of SPICE data. Most of this data is three-dimensional, and its interpretation and visualisation is one of the challenges faced by the ground segments that operate the spacecrafts and the scientists that study its data.
Science Observations and contextual Data analysis of Planetary missions can be naturally accommodated into 3D visualizations. Virtual Reality (VR) brings an unprecedented level of interaction possible with these visualisations, and emerging VR platforms such as Oculus VR, Google Cardboard or Samsung Gear VR are making these technologies accessible to the wide public. VR might become in the near future a key part for spacecraft science operators and become a key element of the ground segments.
This project aimed to contribute as a first step of this milestone by delviering advanced functionalities for VR tools created to access Solar System geometry, with particular emphasis on visualisation of the science observations carried out by the ESA Planetary fleet on the Solar System and include advanced functionalities for navigation and data selection in a VR space through peripherals like Oculus Touch.
Can an artificial neural network help to understand X-ray spectra?
Tutors: Norbert Schartel, Maria Santos-Lleo, Richard Saxton
XMM-Newton is an ESA space observatory that collects X-rays from astronomical sources. It carries three high throughput X-ray telescopes with an unprecedented effective area. Each telescope has an X-ray CCD camera, comprising the European Photon Imaging Camera (EPIC). In addition XMM-Newton is also equipped with two Reflection Grating Spectrometers and an Optical Monitor for simultaneous X-ray imaging, spectroscopy and UV/optical measurements. The large collecting area and spectral capabilities combine with an ability to make long uninterrupted exposures making XMM-Newton an ideally suited observatory to provide highly sensitive, high spectral resolution, long and continuous X-ray observations.
At present, XMM-Newton has been observing the X-ray sky for more than 18 years and still continues. There exists a large data base of public X-ray spectra that populates the scientific XMM-Newton archive and which is growing.
In the astrophysical literature X-ray spectra are discussed in comparison to physical models that describe the radiation emitted under different scenarios. So far, none of the various attempts to classify X-ray spectra through machine learning techniques like the artificial neural networks was taken over by the astrophysical community. Therefore, comparison with physical models either remains a manual operation or can only be done in an automatic way for a few standard models. Large data-bases of X-ray spectra that try to provide the properties of the emitting material can only be populated with very limited results.
A group of Japanese scientists have recently used a neural network to analyse the observed X-ray spectrum of a well known astrophysical source. The interesting point of their approach is that the neural network provides the physical parameters of the X-ray emitting material and therefore bridges the gap to the discussions in the scientific literature. This was achieved by training the network with thousands of simulated (physical) emission spectra.
The aim of tis project was to explore the available public domain neutral network libraries and possible configurations for the analysis of X-ray data. The best suited combination to be installed, trained for XMM-Newton observations with a large number of simulated spectra and applied to several selected X-ray spectra of high and low resolution. The results allowed checking the performance of the constructed neural network and its re-usability in an automated way to perform specific tasks on real data and to infer the physical properties of the emitting material.
Searching for UFOs around black holes
Tutors: Michael Parker, Gabriele Matzeu, Norbert Schartel, Maria Santos-Lleo
The supermassive black holes in the centers of galaxies feed on the surrounding gas and dust, radiating huge quantities of energy in all wavelengths, particularly in the UV and X-ray bands. Because the accreting material around the black hole gets hotter further in, the corresponding emission shifts to higher energies, so using X-ray telescopes allows us to study extreme physics close to the black hole. One of the most interesting phenomena observed are the ultrafast outflows (UFOs), which are powerful winds launched from the accretion disk. These winds travel at speeds of 30000-100000 km/s, and slam into the interstellar medium of the host galaxy, driving gas out of the galaxy altogether and shutting off star formation. Because of their extreme temperatures, these winds are largely transparent and can only be detected by a handful of absorption lines in the X-ray spectrum.
The aim of this project was to use a new detection method developed and published in the astrophysics literature, based on the variable behaviour of these black holes, to take a fresh look at archival X-ray data and conduct a large search for new UFOs.