Embed Size (px)
Citation preview
Master Thesis Proposals
2019-2020
Space Sciences
Faculty of SciencesLiège University
Important notice!
The present catalogue is especially relevant for master theses in the Research focus.
For the Professional focus, the master thesis must be an internship and there is no specific offerprepared in advance. Students are encouraged to search for opportunities out of the AGODepartment. To do so, contacting other institutes is highly recommended, including
- Liège Space Centre (Sart-Tilman) : http://www.csl.uliege.be (or via S. Habraken, C. Barbier, J.Loicq)- Belgian Institute for Space Aeronomy (Brussels) : http://www.aeronomie.be/en/- Royal Observatory of Belgium (Brussels) : https://www.astro.oma.be/en/- Royal Institute for Meteorology (Brussels) : https://www.meteo.be/en/- The AMOS company (Sart-Tilman) : https://www.amos.be/- The Aerospacelab company (Mont-Saint-Guibert): https://www.aerospacelab.be/
… or any other company involved in space activities.
About master theses out of ULiège...
Students involved in internships (abroad, in Belgium, and even at the Centre Spatial de Liège) haveto fill in an internship agreement and a risk analysis sheet. These documents must be completed inconcertation with the person responsible for the internship at the host institution, with the agreementof the teacher/academic supervisor in ULiège.
Link to the required documents :https://www.enseignement.uliege.be/cms/c_9413472/en/etape-4-fiche-d-analyse-de-risque-et-convention
For any question or request for assistance, the contact person for the Faculty of Sciences is MrsKristel Karremans: [email protected]
In addition, for a stay abroad it is mandatory to follow an on-line procedure to officially request theauthorization to the Rector of the University. This is necessary for the validation of the activityabroad in the student programme and for benefiting of an insurance coverage. The request should beintroduced at least one month (sooner is better!) before the expected date of departure.
Astroparticles, Dark matter and Gravitational waves
Contact person : J.R. Cudell
e-mail : [email protected]
Tel : 04 366 3654
Office: 4/44 (B5a)
Availability: most afternoons in May or June. Check via e-mail if you want to be sure,
Thematics : Cosmology and astropaticles
Description:
A number of possibilities exist (in particle physics, astroparticle physics, dark matter, gravitationalwaves,…), and I encourage interested students to come and see me.
Physical properties of gravitationally lensed AGNs
Contact person : Dominique Sluse
e-mail : [email protected]
Tel : 043669797
Office: B5c, +1/10
Availability: Interested students should contact me by e-mail to find the best time for an appointment.
Thematics : Extragalactic Astrophysics
Description:
The active galactic nucleus (AGN) in the center of M87 is the only supermassive black hole that wehave been able to image, and it will probably remain as such for a long time. The emission from an AGN is caused by accretion of gas onto the central supermassive black hole. It arises from a region no larger than 0.1 pc. For that reason, progresses in our understanding of AGNs require indirect methods. Gravitational mirages (aka gravitational lenses) can be used as natural telescopes to measure the size of the disc from which arises the AGN continuum emission, and the size and geometry of the region emitting the salient broad emission lines detected in AGN spectra. Those properties, as derived with the gravitational lensing method, need to be studied at the light of other properties of the AGN, such as the mass of the central supermassive black hole. Papers dedicated to the discovery of new gravitationally lensed systems focus on confirming the lensed nature of an object but generally provide no emphasis on the properties of the lensed source. Therefore, very little information is available for the latter. This project proposes to determine the physical properties of lensed AGNs. In particular, it will be asked to use existing spectra of lensed systems (obtained by our group and/or available in the literature) as well as multi-band photometry, to derive the main physical properties of known lensed systems. In practice, the student will:
Model the emission of (lensed) AGN spectra.
Perform a series of measurements based on the spectra to infer: the black hole mass, the Eddington ratio; but also search for the presence of intrinsic narrow/broad absorption lines, rare emission lines, ...
Search literature and databases for data of the scrutinized systems (using of virtual observatory tools): multi-band photometry, additional spectroscopy, …
Compare and discuss his/her results with literature (when available)
Other projects may be possible depending on the skills and interest of the student.
Having attended the classes on Extragalactic astrophysics (SPAT0011) and Programming techniques andnumerical methods (SPAT0002) is certainly helpful.
A detailed analysis of the spectroscopic variability of the massive Osupergiant HD152249
Contact person : Eric GOSSET
e-mail : [email protected]
Tel : 04/366.97.54
Office: B5c, office 2-16, Groupe d’Astrophysique des Hautes Energies
Availability: At any time, but please send an Email in advance, in order to fix the date.
Thematics : Astrophysics; Stellar physics; Observations; Spectroscopy; Pulsations versus Spots;Determination of the Physical Parameters.
Description:
The variability of massive O stars of the upper part of the Hertzsprung-Russell diagram is not yetfully understood. In particular, very few O supergiants have been studied. HD152249 is a star ofspectral type O9I which is most probably single. However, the lines present in its spectrum exhibitweak profile variations that are easily detected in our high S/N observed spectra. A firstobservational campaign took place and confirms that the star exhibits line profile variations with acharacteristic time scale lower than a day. This kind of variation could be associated to the presenceof pulsations or to a spotted surface of the star or to other phenomena still to be discovered. Threeother campaigns took place, one being particularly intensive. These data are not reduced yet, butshould contain a wealth of new information.
The work will consist to start from the raw new observational data and to properly reduce them. Theline profiles will then be inspected using various methods to investigate their probable variabilityand to further analyse and describe it. Once this analysis done, comparison of the line to linedifferential behaviour will be investigated in order to eliminate some explanations and to favourothers. The physical parameters to be associated with the star will also be investigated. They willserve as a basis of an extensive discussion about the origin of the variability and of thecompatibility of the possible origins with the canonical models of this kind of star.
Remarks: other similar subjects are available particularly on Wolf-Rayet binaries.
The impact of atmospheric eclipses on the lightcurves of close massivebinaries
Contact person : Gregor Rauw
e-mail : [email protected]
Tel : 04/366 9740
Office: B5c, 2/2
Availability: 7 & 14 May between 14h00 and 17h00. Interested students should ideally contact meby e-mail to arrange an alternative appointment if needed.
Thematics : Stellar astrophysics
Description:The vast majority of massive stars reside in binary or higher multiplicity systems. Eclipsing binariesare especially important as they allow to infer the fundamental properties (masses and radii) of thestars. The overwhelming majority of the studies of such systems assume that the stars are opaqueand that their shapes can be described by the Roche potential. However, massive stars feature dense,partially optically thick stellar winds. In close binary systems, these winds can absorb some of thelight of the companion, producing a so-called atmospheric eclipse and leading to situations wherethe surface corresponding to optical depth unity (τ=1) lies beyond the Roche lobe. Both effects(atmospheric eclipse and R(τ=1) larger than the radius of the Roche lobe) are not accounted for inthe codes based on the Roche model. The goal of this project is to implement a numerical code forthe calculation of lightcurves of close binaries which accounts for the shape of the stars (based onthe Roche potential), the mutual eclipses of the opaque stellar surfaces and the atmospheric eclipsesand to compare the results with existing data.
Recommended master courses: . Variable stars. Programming techniques, numerical methods and machine learning. Stellar atmospheres
Development of methods for simultaneous fitting of imaging, radialvelocity, and astrometric data of substellar companions
Contact person : Anne-Lise Maire
e-mail : [email protected]
Tel : +3243669764
Office: B5c office 210
Availability: Can be defined upon email request.
Thematics : Astrophysics; Planetology and planetary systems
Description:
The monitoring of the orbital motion of young imaged low-mass companions around their star is apowerful tool to (i) measure their orbital parameters, (ii) analyze potential dynamical interactionswith other bodies in the systems and/or the circumstellar disks, and, (iii) for close companions,assess their dynamical mass when combining imaging with radial velocity (RV) and/or astrometry.Dynamical mass measurements of such objects have been very challenging due to their wideseparation to the stars (≥10 au) and the poor overlap with RV planet surveys (≤5 au, old stars). Yet,they are critical to test and calibrate their evolutionary models, which are commonly used to infer amodel-dependent mass of imaged companions from their luminosity and age.
The advent of the dedicated exoplanet imaging facility SPHERE (Beuzit et al. 2019) together withthe ongoing Gaia survey (Gaia Collaboration et al. 2016) and the commissioning of the high-resolution spectrometer CRIRES+ (Dorn et al. 2014) offer new prospects to measure the dynamicalmass of young imaged low-mass companions in the coming years and therefore calibrate theirevolutionary models. SPHERE allowed to measure the position of brown dwarfs and giant planetsclose to stars down to accuracies of ~1–2 mas. CRIRES+ will provide high-accuracy RVmeasurements of the companions (provided that their orientation on the sky is known), overcomingthe time limitations of previous surveys measuring the RV of the stars. Pioneer studies withCRIRES demonstrated the power of this technique on the giant planet β Pictoris b (Snellen et al.2014) and the brown dwarf GQ Lup B (Schwarz et al. 2016) by constraining their rotation periodand/or further refining their orbit. The value of stellar proper motion data from Gaia and Hipparcoswas outlined by, e.g., Snellen & Brown (2018), Brandt et al. (2019), and Dupuy et al. (2019). Thetwo last studies performed joint fits of imaging, RV, and astrometric data to constrain the dynamicalmass of a few brown dwarfs and of β Pictoris b. The inclusion of astrometric constraints allows toadd the stellar mass as a free parameter in the fits and to derive companion masses fullyindependent from models (assuming no other unknown companions in the system).
The goal of this Master thesis is to develop methods and tools for the simultaneous fit of imaging,RV, and astrometric data of brown dwarfs and giant exoplanets. Our group at the University ofLiège (Wertz et al. 2017) developed a Markov-chain Monte Carlo orbital fitting code for imagingdata written in Python that can be used as a starting point to include radial velocity and astrometricconstraints. To validate the code, it will be applied to literature measurements of known low-masscompanions that were used in previous studies to constrain their properties. Finally, the mass of thecompanions inferred from the fits will be used together with existing luminosity and age estimatesfor detailed comparisons to predictions from several formation and evolutionary models (e.g.,Baraffe et al. 2015, Marley et al. 2007).
Helium glitch analysis in solar-like stars
Contact person: Martin Farnir, Marc-Antoine Dupret
e-mail: [email protected], [email protected]
Tel: 04 366 9721
Office: B5c +1/1 or +1/12
Availability: M. Farnir : Anytime in May and June, M.-A. Dupret : Anytime in May and June
Thematics: Astrophysics: Stellar physics, Asteroseismology
Description:
As the search for habitable exoplanets goes on, our attention is drawn towards solar-like stars.Thanks to space missions such as Kepler, a wealth of data of unprecedented quality has yet to beanalysed. With such data, it becomes essential to exploit as much of the available information aspossible, and to do so in a reliable and accurate way. Solar-like stars are cool low mass starsexhibiting p-type stellar pulsations excited by stochastic mechanisms. Aside from the regularpattern observed in the oscillation spectrum (e.g. large and small separations) a very low amplitudeoscillating signal may be observed in the frequencies. This is the glitch which is caused by a sharpvariation in the stellar structure. The analysis of such a signal may then provide an inference on thesurface helium abundance, which is inaccessible by other means in these cool stars.
The aim of the thesis will be to study several targets from the Kepler LEGACY sample, whichrepresents the best data at disposal for asteroseismology of solar-like stars, to provide a preciseestimation of their fundamental parameters (mass, composition, age, etc). The exceptionnal qualityof the Kepler LEGACY data enables us to retrieve the faint signature of glitches. However, this alsocalls for very accurate oscillation spectrum analysis techniques. Appropriate tools have beendeveloped in the ASTA team such as the CLES stellar evolution code, LOSC oscillation code andthe WhoSGlAd method for the oscillation spectrum and glitches analysis. The best fit stellar modelsearch will combine these tools with a minimisation algorithm. The project may take severaldirections according to the student’s intents. Several solar-like targets may be treated (e.g. Sun as astar data, binaries or peculiar stars) but more theoretical work may be done also such as theexploration of the helium glitch signature, and associated seismic indicators, throughout the HRdiagram based on a grid of theoretical models. This project is well suited for a student that iscurious about the physics of stellar models and who has basic programming knowledge.
Recommended courses: Stellar structure and evolution I SPAT0044-1 (& II SPAT0045-1) & Stellarstability and asteroseismology SPAT0005-1 (Pr. M-A Dupret)
Modeling the Shear Layer Oscillation and its impact on the rotation of low-mass stars
Contact person : Charly Pinçon, Marc-Antoine Dupret
e-mail : [email protected], [email protected]
Tel : 04 366 9721
Office: B5c +1/1 or +1/12
Availability: CP: from May to July, MAD: From May to June
Thematics : Astrophysics: Stellar physics, Hydrodynamics, Rotation, Waves
Description:
Rotation is an important ingredient of the stellar structure. For example, it can induce transport ofchemical elements that is able to refresh nuclear burning cores with hydrogen and thus to increasethe star lifetime on the main sequence. However, current seismic observations show that the centralrotation of low-mass stars during evolution from the main-sequence to helium core burning stagesand even to white dwarfs, is lower than expected from the current stellar evolution codes. Anefficient mechanism therefore needs to be included in stellar models in order to reproduce theseobservations. Among different possibilities, internal gravity waves remain a serious candidate.
Internal gravity waves are excited by turbulent motions in the convective region and canpropagate in the underlying radiative region. During their travel throughout the radiatve core, thesewaves suffer radiative losses so that they can deposit a part of the angular momentum into themedium and thus locally modify the rotation rate. In particular, previous works showed that theydrive a small-scale oscillation of the rotation profile in a thin layer just below the base of theconvective region with a typical period of the order of the year, the so-called Shear LayerOscillation (SLO). This is equivalent to the Quasi-Biennal Oscillation (QBO) observed in theterrestrial atmosphere. The SLO filters the wave energy flux transmitted towards the center of thestar and thus can have an impact on the efficiency of the wave-induced core acceleration.Nevertheless, it is difficult to take this effect into account in stellar modeling since it deals withtimescales much smaller than the typical timescale of stellar evolution.
The aim of this master project is to quantify the impact of the SLO on the wave-induced angularmomentum transport through a simplified modeling. The work will consist in numerically solvingthe differential equations of the evolution of the mean flow in a thin layer just below the convectiveregion and exploring the associated parameters space. The initial configuration of the model will beone-dimensional, plane-parallel, with a constant stratification, and will consider only the effects ofmonochromatic waves and a constant turbulent diffusivity. Improvements in the modeling could beenvisaged given the advancement of the project. The final goal will be to propose simpleprescriptions that can be easily included in stellar evolution codes. This project is well suited forstudents interested in solving general physical problems and especially hydrodynamics, with goodanalytical and numerical skills.
Recommended courses: Stellar structure and evolution I SPAT0044-1 & II SPAT0045-1 (Pr. M-ADupret)
Deformation and break-up of fast rotating stars
Contact person : Marc-Antoine Dupret
e-mail : [email protected]
Tel : 04 366 97 32
Office: B5c, room 1/12
Availability: Anytime in May and June
Thematics : Astrophysics: stellar physics
Description:
Context: Many stars are rotating fast. A typical example is given by Be stars, among which some are close tothe rotation velocity break-up. As a consequence, they are strongly deformed by the centrifugalforce. However, this deformation is neglected in nearly all current stellar models. Taking it intoaccount would enable significant progress towards a more accurate modelling of fast rotating starsand their oscillations.
Proposed work:The main task will be to implement a code modelling accurately the centrifugal deformation of arigidly (or cylindrically) rotating star. This problem has the appropriate level for a master thesisthanks to a significant simplification: for rigid (or cylindrical) rotation, the fluid is barotrope and thepressure and density are constant on equi-potentials. Spherically symmetric models computed withthe Code Liégeois d’Evolution Stellaire (CLES) will be taken as initial models. Next, solvingiteratively the Poisson equation will provide the final centrifugally deformed model. Once the code works correctly, comparison with the approximate Roche models will be considered(centrifugal deformation, gravity-darkening) for models encompassing a wide range of masses andevolution stages. Finally, a very efficient student could consider the generalization to the non-conservative, non-barotropic case.This subject is ideal for a student who likes solving physical problems by himself and interpret theresults. It could also constitute a first step before a PhD thesis, which would focus on the 3Dmodelling of close binary stars deformed by both rotation and tidal effects.
Recommended course: SPAT0045-1 Stellar structure and evolution II (Pr. M.A. Dupret)
Assessing the probing potential of mixed modes in Red Giants
Contact person : Charly Pinçon, Marc-Antoine Dupret
e-mail : [email protected], [email protected]
Tel : 04 366 9721
Office: B5c +1/1 or +1/12
Availability: CP: from May to July, MAD: From May to June
Thematics : Astrophysics: Stellar physics, Asteroseismology
Description:
Once their central hydrogen is exhausted, low-mass stars leave the main sequence and startevolving on the red giant branch. At this evolutionary stage, stars are composed of a small-sizedhelium core in contraction located below a thin hydrogen-burning shell, all surrounded by a dilutedexpanding envelope. Due to their core-envelope structure, red giant stars exhibit a peculiar kind ofstochastically-excited oscillations called mixed modes. These modes can propagate in the centralradiative region, where they behave as gravity modes, and in the convective envelope, where theybehave as acoustic modes. Unlike pure acoustic modes in main-sequence stars, the frequencypattern of mixed modes in red giant stars thus gives us the opportunity of probing the properties notonly of their outer envelope, but also of their inner layers.
Theoretical analyses of mixed modes demonstrate that their frequency pattern depends on fourmain parameters. Due to the high-quality seismic data provided by the CoRoT and Kepler satellites,these indicators have already been measured in thousands of evolved stars and have already givenus stringent constraints, for instance, on the contraction state of their helium cores. Nevertheless, therelation existing between these four indicators and the internal properties of red giant stars needs tobe further scrutinized in order to fully exploit their ability of constraining not only theirevolutionary state, but also their whole evolution in general. Indeed, the structure of red giant starsmay conserve the signature of past events and may also supply the initial conditions for themodeling of later phases such as white dwarfs. Moreover, due to their high luminosity and theirwide range of mass, red giant stars represent key targets for Galactic archaeology.
In this context, the aim of this project will be to assess the link between the seismic parametersassociated with mixed modes and the stellar internal properties. The work of the student will consistin determining the considered seismic indicators throughout a grid of stellar models and in studyingtheir dependence with the stellar parameters such as the age, the mass, the metallicity, the mixing-length, the efficiency of the convective boundary mixing, as well as in understanding the physicalorigins of the observed trends. To do so, the student will use the stellar evolution code CLEScoupled with the oscillation code LOSC that are developed in the team ASTA. The results of theinvestigation will be compared to the current large-scale observations and discussed accordingly.This subject is ideal for a student interested in stellar evolution and stellar oscillations, with basicanalytical and programming skills.
Recommended courses: Stellar structure and evolution I SPAT0044-1 (& II SPAT0045-1) & Stellarstability and asteroseismology SPAT0005-1 (Pr. M-A Dupret)
Analysing internal rotation from oscillations of γ Dor stars observed byKepler
Contact person : Sébastien Salmon & Marc-Antoine Dupret
e-mail : [email protected]; [email protected]
Tel : +32 4366 9765/9732
Office: B5c, room 1/14 or 1/12
Availability: Anytime in May or June, but please contact us before for convenience
Thematics : Astrophysics: stellar physics, asteroseismology
Description:
Gamma Doradus stars are main-sequence A and F-type stars presenting high-order gravity modes ofpulsations. Some of them present an hybrid-pulsator character by also showing pressure modes,known as δ Scuti variability. For a while, the periods (a few hours) and amplitudes (a few mmag)of their pulsations made them hardly detectable from ground-based observatories. Yet, the Keplerspace telescope, with its long and uninterrupted observing campaigns, revealed rich and complexstellar pulsation spectra in hundreds of γ Dor stars. Since a large fraction of these stars are moderateto fast rotators, the analysis of their pulsations has revealed a huge potential for characterising theirinternal rotation profile and thus internal transport processes of chemical species and angularmomentum. In particular, recent studies reported detection of rotational signatures both in theirgravity and pressure frequency spectra of oscillations. Furthermore, convincing evidence of thedetection of Rossby waves was reported. These gravito-inertial waves only appear with rotation asCoriolis force acts as part of their restoring force. The imprint of rotation in oscillations of differentnature paves the way for characterising rotation in different parts of the star.
Within the ASTA team, we are developing for many years numerical tools alowing us to indepentlymodel stars from their structure to the computation of their pulsations. With help of these uniquetools, the student will analyse and do the detailed modelling of some of the most promising hybridγ Dor stars detected by Kepler. He/she will focus on the constraints that these stars can bring on theinternal stellar rotation, and the link with their past stellar evolution.
Recommended courses: Stellar structure and evolution I SPAT0044-1 (& II SPAT0045-1) & Stellarstability and asteroseismology SPAT0005-1 (Pr. M-A Dupret)
An Analysis of the Red Giant Bump
Contact persons : M.-A. Dupret et A. Noels-Grotsch
e-mail : [email protected], [email protected]
Tel : 04 366 97 32
Office: B5c 1/12 - 1/4
Availability: Anytime in May or June. For A. Noels, please contact by e-mail.
Thematics : Stellar Evolution
Description:
During its ascension of the red giant branch (RGB) low mass stars encounter a sort of zigzag, whichis known as the RGB bump. This bump is observed as an accumulation of stars in HR or Color-Magnitude diagram in globular clusters as well as in dSph galaxies.
The physical reason for the presence of this bump is related to the interaction between the hydro-gen-burning shell (H-shell) and the convective envelope. After the vanishing of the convective coreat the end of the main sequence, hydrogen burns in the H-shell, which progressively moves outwardas the helium core mass grows. This is accompanied by a deepening of the cooler and cooler con-vective envelope. The base of the envelope eventually penetrates layers, which have been suffi-ciently affected by earlier nuclear burning. This is the first dredge-up, which brings to the surfacematter processed by hydrogen burning with a decrease of the C12/N14 and C12/C13 ratios.
This deepening stops when the base of this envelope reaches layers close to the H-shell. The heatingof these layers to ensure hydrogen combustion in the H-shell keeps the opacity low enough to pre-vent convection and the convective envelope recedes. However, while receding, the convective en-velope leaves a discontinuity in chemical composition located at its maximum extent. Once the H-shell comes into contact with this discontinuity, it suddenly operates in a matter richer in hydrogen.To prevent an unruly increase in the nuclear energy production, the H-shell expands and coolsdown. This is followed by a global contraction and a decrease in luminosity before resuming the as-cension of the RGB. However, the theoretical RGB bump is more luminous (0.2-0.4 mag) than ob-served in globular clusters.
The proposed research work would be to analyse this aspect of the RGB and to propose leads to de-crease the RGB bump down to the observed value. Among these possible remedies the student willexamine (1) undershooting below the convective envelope, (2) diffusion and/or (3) increasedabundances of α-elements.
Recommended course: SPAT0045-1 Stellar structure and evolution II (Pr. M.A. Dupret)
Revisiting the instability strips of sdB stars
Contact person : Valérie Van Grootel & Marc-Antoine Dupret
e-mail : [email protected] ; [email protected]
Tel : +32 4366 9730/9732
Office: B5c, room 1/13 or 1/12
Availability: V. Van Grootel and M-A Dupret: anytime in May or June, but please make a firstcontact by email to agree on an apointment.
Thematics : Astrophysics: stellar physics, asteroseismology
Description:
The stars studied in this master project are extreme horizontal branch stars, also known as subdwarfB (sdB) stars. They represent an advanced stage of stellar evolution, after the main sequence wherethe stars spend most of their lives and after the first red giant phase. These hot (Teff=20,000-40,000K) and compact (log g=5.2-6.2) objects burn helium in their cores into carbon and oxygen and aresurrounded by an extremely thin H-rich envelope. Understanding the formation of sdB stars is oneof last big mysteries of stellar evolution. Some sdB stars present pulsations, which open thepossibility to apply asteroseismology (the study of stellar pulsations) to model these stars.
The proposed master thesis concerns the instability strip of sdB stars: we aim to determine when(i.e. at which masses, effective temperature and surface gravity) these stars present excited pulsationmodes and when they are not predicted to pulsate. Three groups of sdB pulsators are known: theshort-period (80-500 sec) sdB pulsators with temperatures Teff around 34,000K which were the firstto be discovered; the long-period (1-3h) sdB pulsators with cooler Teff around 28,000 K and lowersurface gravities; and finally very hot (50,000-60,000 K) and very compact post-sdB pulsators withvery short periods (less than one minute), recently found. Our classical models of sdB starscorrectly predict and reproduce the properties of the pulsations of the first group (the short-periodpulsators), but not at all those of the second (long-period) and third (very hot) group. The culprit isidentified since a couple of years: the absence of Ni in our sdB models. Unfortunately, no trustableopacity tables for Ni were available to us until very recently. This is now the case, and new sdBmodels including Ni have now been built.
The aim of this Master thesis is to revisit the instability strips of the three groups of sdB pulsatorswith these new sdB models. The goal is to determine if we can still correctly reproduce thepulsation properties of the first group of sdB pulsators, and if we can better reproduce the observedinstability strips of the second and the third groups with these new models. The non-adiabatic MADpulsation code (developed by Pr. Marc-Antoine Dupret) will be used to pulsate the sdB models. Inparticular, the MAD code includes various prescriptions for the interaction between pulsations andconvection, a potentially important effect to account for pulsations in sdB stars, but that has neverbeen tested.
Recommended courses: SPAT0005-1 Stellar Stability and asteroseismology, SPAT0045-1 Stellarstructure and evolution II (Pr. M.A. Dupret)
Modeling TESS data of extreme horizontal branch stars byasteroseismology
Contact person: Valérie Van Grootel
e-mail: [email protected]
Tel: +32 4366 9730
Office: B5c, room 1/13
Availability: Any time in May or June, but please contact me by email to agree on an appointment.
Thematics: Astrophysics: stellar physics, asteroseismology
Description:
The TESS satellite from NASA gathers since December 2018 high-quality photometric data onvarious stars, for searching transiting exoplanets but also for asteroseismology. Asteroseismology isthe study of stellar oscillations in order to tightly constrain the physics inside stars and hence, torefine the models of the structure and the evolution of stars.
Among these stars, TESS observes each month (one Sector of the sky each 27 days) dozens ofextreme horizontal branch stars, and discovers/confirms pulsations in a few of them. Extremehorizontal branch stars, also known as subdwarf B (sdB) stars, represent an advanced stage ofstellar evolution. These hot (Teff=20,000-40,000 K) and compact (log g=5.2-6.2) objects burnhelium in their cores into carbon and oxygen and are surrounded by an extremely thin H-richenvelope. Understanding the formation of sdB stars is one of last big mysteries of stellar evolution.
The proposed master thesis concerns the asteroseismic modeling of sdB stars observed by TESS.First step will consist in selecting the most promising targets for asteroseismic modeling: presenceof a rich pulsation spectrum, availability of good spectroscopic constraints. The second step ispreliminary asteroseismic analyses on the most promising targets, in order to select one that will bestudied in depth during this master thesis in the third step. The asteroseismic modeling consists inquantitatively comparing the computed oscillation periods for large sets of stellar models to theobserved periods. By optimizing this comparison (through genetic algorithms that have beendeveloped for this purpose) to find the best-fitting model to the observations, the seismic modelingwill yield the global parameters (e.g. stellar mass and radius) and internal structure and composition(e.g. envelope layering, core composition) of the star. Results will then be exploited, by comparingthem to those of other sdB stars modeled by asteroseismology and by interpreting them in a contextof sdB formation. All the tools are available and ready for a direct application to these TESS data.
This subject is well-suited for a student who like to work on concrete applications ofasteroseismology and space-based observations.
Recommended courses: SPAT0005-1 Stellar Stability and asteroseismology, SPAT0045-1 Stellarstructure and evolution II (Pr. M.A. Dupret)
Simulations of photometric performance of a CubeSat
Contact person : Valérie Van Grootel & Sébastien Salmon
e-mail : [email protected] ; [email protected]
Tel : +32 4366 9730/9765
Office: B5c, room 1/13 or 1/14
Availability: Anytime in May or June, but please first contact us by email to agree on anappointment.
Thematics : Instrumentation and methods
Description:
CubeSats are a class of standardized nanosatellites (spacecrafts that weigh 1-10 kg) by unit “U” of10x10x11 cm3. They are an expanding space activity, with several hundreds of CubeSats forecastedfor a launch in the coming years, including for scientific purposes. In particular, thanks to hugetechnical progress in satellite pointing stability and attitude control, high-precision photometry forastrophysics is now achievable.
3U/6U CubeSats are promising to achieve high-precision photometry for astrophysics, for examplefor transiting exoplanets, stellar oscillations, stellar activity, etc. The most interesting low-Earthorbits to ensure optimal visibility of stars are sun-synchronous dawn/dusk (6h/18h) orbits, where asatellite rides the terminator between day and night.
The goal of this master thesis project is to characterize the photometric performance of a 3U/6UCubeSat in a dusk/dawn orbit. Several solutions for the design of the payload and the platform willbe quantified and tested for their photometric performance. This will allow us to dimension theoptical design in order to meet astrophysical objectives. The performance simulator for theCHEOPS space mission (http://sci.esa.int/cheops/), for the preparation of which the promoters ofthis master thesis are implied, will be used as the tool to carry out the simulations. Photometricperformance of ASTERIA, a 6U CubeSat currently flying from ISS-like orbit, will be studied as aconcrete example.
Recommended courses: SPAT0035-1 Space Exploration (Pr. G. Rauw), AERO0018-3 Spaceexperiment development (J. Loicq)
Development of a program for automatic detection of solar system objectsin TRAPPIST telescopes archive images
Contact Person: Jehin Emmanuël
e-mail: [email protected]
Tél: (0)4 3669726
Office: B5c 1/9
Availability: please contact me
Thematics: Astrophysics (small bodies of the solar system)
Description:
The TRAPPIST telescopes installed at the Silla observatory in Chile in 2010 and in Morocco in2016 by our team are dedicated to the research and the study of exoplanets in transit and the studyof the small bodies of the system solar (comets and asteroids). Each night we collect hundreds ofimages of a few fields, both using short and long poses. In this work we propose the development ofa program to detect the objects of the solar system (which move relative to the fixed stars). Amongthese objects are mainly known asteroids moving several arcseconds per hour but sometimes alsoobjects still not listed thanks to the limiting magnitude (Rc=21) reached by TRAPPIST during longexposures.
As some fields are observed several nights in a row and for many hours, it is also possible to findobjects that move more slowly like comets and distant Transneptunian objects (less than 1arcsecond per hour).
Objectives: first develop the detection strategies according to the type of objects to detect (close byNear Earth Asteroids (NEA), or very distant Trans-Neptunian Objects (TNO) and make a script forautomatic detection. Then perform the astrometric and photometric measurement of the targets andpresent the results in the Minor Planet Center (MPC) format.
If time permits, identification of the targets with MPC catalogs.The tools used will include IRAF, SExtractor etc.Thousands of data are already available to test the program. The ultimate goal is to integrate thisprogram into the TRAPPIST online reduction pipeline.
http://www.orca.ulg.ac.be/TRAPPIST/Trappist_Main_Fr/Home.html
Study of the chemical composition of comets atmospheres using the TRAPPIST telescopes
Contact person: Jehin Emmanuël (ULiège)
e-mail: [email protected]
Tél: (0)4 3669726
Office: B5c 1/9
Availability: please contact me
Thematic: Astrophysics (small bodies of the solar system)
Description:
Comets are among the best preserved specimens of the primitive solar nebula. This status of“fossils” gives them a unique role in understanding the origins of the solar system. The success ofthe Rosetta space mission was very important and is changing our knowledge about comets. But itshowed also that observations from the ground continue to be important: they make it possible tosupplement the data in situ by obtaining information on larger scales of the coma and tails, and for amuch larger number of comets, which is necessary to extrapolate the results to the entire cometarypopulation. The link between the composition of comets and their dynamic history must still beclarified and a complete comet classification is still missing.
In this context, we propose the observation and analysis of the coma of two or three bright cometswith the TRAPPIST telescopes network. These robotic telescopes installed by our team in Chile (in2010) and in Morocco (in 2016) are equipped with narrow band filters to isolate the emissions ofdifferent gases and dust contained in the atmosphere of comets. The student will have to prepare theobservations, calibrate the data and calculate the production rates of the different gases using the so-called Haser model (1957). The necessary tools for this kind of measures have already beendeveloped in our team. The student will have to become familiar with the various techniques, adaptand improve if necessary the reduction procedures and scripts and run the models. The results mightlead to the publication of an article.
The work will be done in the comet group of the OrCa Service (+1) and the TRAPPIST team http://www.trappist.uliege.be/cms/c_3300885/en/trappist-portail
Uncovering terrestrial exoplanets from direct imaging data
Contact person : D. Defrère, O. Absil
e-mail : [email protected], [email protected]
Tel : 043724701 (D. Defrère), 043669724 (O. Absil)
Office: B5c (second floor)
Availability: any day (please contact us to agree on a time slot)
Thematics : Planetology and planetary systems, image processing
Description:
The detection and characterization of exoplanets is one of the major science cases of modernastronomy. In particular, the University of Liège is currently involved in two major projects tosearch for habitable exoplanets around nearby sun-like stars. The first project, called NEAR, hasjust started and is focused on the alpha Centauri system using the Very Large Telescope installed inChile. The second project, called METIS, is an instrument for the future Extremely Large Telescope(ELT) and will search for habitable exoplanets around a sample of nearby Sun-like stars (2025+).One of the major issues to reach the sensitivity required to detect habitable exoplanets is theoverwhelming thermal emission from our Earth's atmosphere, which is at least one million timesbrighter than an Earth-like exoplanet located around our nearest stellar neighbour. In order to enablesuch detection, the thermal background from the atmosphere has to be carefully removed usingdedicated image processing techniques.
Under the supervision of Denis Defrère and Olivier Absil at the department of Astrophysics,Geophysics, and Oceanography (AGO), the goal of the master thesis will be to develop, test, andcompare different innovative background subtraction techniques. These techniques will be tested onexisting data sets obtained with similar instruments such as ESO’s VLT/VISIR installed in Chileand NASA’s LBTI/NOMIC installed in Arizona. The optimized background subtraction techniquewill then be applied to existing exoplanet data sets. The results of the master thesis are expected tocontribute to the scientific preparation of METIS.
Recommended master courses :
SPAT0063: Introduction to exoplanetology
SPAT0067: Atmospheric and adaptive optics
SPAT0056: Planetary and exoplanetary atmospheres
SPAT0002: Programming techniques, numerical methods
Constraining the dynamics and masses of the TRAPPIST-1 planets withthe transit timing variations method
Contact person: Michaël Gillon
e-mail: [email protected]
Tel: 043669743
Office: Institute of Astrophysics, B5c, -1/1
Availability: contact by e-mail
Thematics: Planetology and planetary systems
Description:
The nearby ultracool dwarf star TRAPPIST-1 hosts a compact system of seven temperate Earth-sized planets, three of them orbiting in its habitable zone.
This Master thesis will focus on the analysis of transits of the planets to be observed between Juneand Dec 2019 by the SPECULOOS-South and SPECULOOS-North Observatories. The measuredtransit timings will then be analyzed with a N-body integration code to constrain further the massesand orbital parameters of the planets.
Jupiter's aurora (several subjects)
Contact persons : Denis GRODENT, Bertrand BONFOND
e-mail : [email protected] , [email protected]
Tel : 04 366 97 73
Office: B5c - 0/17
Availability: Please, set an appointment by e-mail
Thematics : Planetology and planetary systems
Description:
The overall theme consists in using image processing tools to analyse Jupiter’s permanentauroral emissions, the brightest in the solar system. Caused by the impact of charged particles intothe planetary neutral atmosphere, these emissions of light form a synthesis of the many processestaking place in the whole magnetosphere.
Jupiter and its aurorae are currently under great scrutiny, as the NASA Juno spacecraft orbits theplanet since July 2016. Among its instruments is a UV imaging spectrograph, named Juno-UVS,dedicated at giving global and close-up spectrally resolved pictures of the atomic and molecularhydrogen in the ultraviolet wavelength range. In order to complement these observations and get themost of this unique opportunity, we executed one of the most comprehensive planetary programwith the Hubble Space Telescope (HST), covering the first four years of operation of Juno.
These ever-growing HST and Juno datasets already contain an enormous amount of information,which we need to thoroughly analyze and interpret. In that regard, several master degree thesismay be defined on the basis of this vast dataset and may cover various aspects of the study ofJupiter's aurora and magnetosphere. It is possible to address rather technical or more theoreticalresearch themes. applied or theoretical sides A first example of master thesis project is not to look directly at the aurorae, but to look for theregions devoid of them. Indeed, in the region located inside the main auroral oval lies a so-called“Dark-region”, supposedly related to return currents accelerating electrons away from Jupiter ratherthan towards its poles. The student will create an automated method to identify and measure thecharacteristics (size, location, brightness, etc.) of this region on Hubble Space Telescope and Juno-UVS auroral observations.
Another topic would be to use a motion tracking software to determine the motion of some auroralfeatures in order to recognize and classify distinctive patterns. Indeed, some of the auroralemissions in the so-called “active polar region” appear to be spreading, while others seem toliterally move across the pole. The velocity and amplitude of this mysterious dynamics suggests thatthey correspond to ionospheric processes rather than magnetospheric processes.
A third topic would involve the automated recognition of auroral arcs at Jupiter. The student willadapt a method used for similar purposes in order to track the location, shape, brightness and size ofthe main auroral oval on images obtained either by the Hubble Space Telescope or the UVSinstrument on board Juno.
Additional subjects may also be ‘custom-made’ to match students skills and preferences.
More information on Jupiter’s UV aurora may be obtained here (previous Master thesis on Jupiter’saurora): http://hdl.handle.net/2268/154269
Preview movies and additional information about the Jupiter-Hubble-Juno campaign are availablehere: http://www.lpap.uliege.be/cms/c_3478756/en/lpap-hst-campaign.
Analysis of the generation of cloud bands on Jupiter or Saturn
Contact person : Jean-Marie Beckers, Denis Grodent
e-mail : [email protected], [email protected]
Tel : +3243663358
Office: B5a 2/5
Availability: Preferentially monday’s afternoon.
Thematics : Planetology, Geophysical Fluid Dynamics
Description:
Cloud bands are visible both on Jupiter and Saturn with a strong zonal flow direction alternatingfrom the equator to the poles. The proposed works aims at analysing the dynamics leading to suchbehaviours, focussing on one of the two planets.
The student will start with an analysis of the different explanations provided up to now, includingtheir hypothesis on thermodynamics and geometrical constraints (barotropic or baroclinic modes,vertical extensions and dynamics, small-scale turbulence effects etc). Based on the currentknowledge of Jupiter’s or Saturn’s atmosphere, a model providing a good compromise betweenrealism and simplicity will be chosen for further investigation. In particular, the idealized modelwill then be used to simulate how coherent structures as a big vortex or the bands can emerge andunder which conditions involving geostrophic turbulence and planetary waves. Testing hypothesiswill typically include activating or deactivating certain processes during the simulations. The focusof the work is thus the understanding of the underlying dynamics and less the best reproduction ofobservations.
If the analysis turns out to be strongly dependant on a poorly known parameter, the study couldactually be turned into an inverse problem. In that case, using actual observations on the banddistributions, the parameter could be calibrated and the thesis provide an indirect estimate for it.
The work will rely on a good understanding and analysis of fundamental geophysical fluiddynamics and the use of numerical models.
Study of fireballs with BRAMS and FRIPON
Contact person : Emmanuel Jehin (ULiège), Hervé Lamy (IASB)
e-mail : [email protected], [email protected]
Tel : +32 (0)4 3669726, +32 (0) 2 3730418
Office: B5c 1/9
Availability: E. Jehin will be available most afternoons in May and June.
Thematics : Planetology and planetary systems
Description:
Every day, the Earth’s atmosphere is hit by tiny dust particles collectively called meteoroids with asmall part of them bigger enough (size ~ 1 cm) to produce a visible phenomenon called meteor (orshooting star). Fireballs are very bright meteors due to much larger objects entering the atmosphere.These objects penetrate much deeper in the atmosphere and might not be fully ablated leaving astrongly slowed down small part which falls down freely to the ground as a meteorite.
Meteors can be observed from the ground either with optical or radio techniques. With the latter, aradio wave is sent from a transmitter on the ground and is reflected on the ionized trail created bythe meteoroid along its path when hitting atoms and molecules of the upper atmosphere. Forfireballs, a reflection can also be obtained on the ionized region in front of the meteoroid.
In this master thesis, we propose to study fireballs observed above Belgium with both optical andradio techniques. For optical observations we will use data from FRIPON (Fireball Recovery andInterPlanetary Observation Network, www.fripon.org) which is a French network of all-skycameras separated by typically 100 km. A detection of a fireball by several cameras allows areconstruction of the fireball trajectory and speed. FRIPON has lately been extended to othercountries including Belgium. One camera is on the roof of IASB-BIRA in Uccle and is runningsince 2016. Another one is installed since summer 2018 at ULiège on the roof of the Physicsbuilding. For radio observations we will use data from the BRAMS (Belgian RAdio MeteorStations, brams.aeronomie.be) which is a network run by IASB-BIRA and made of a dedicatedtransmitter located in Dourbes (South-East of Belgium) and 26 receiving stations located all overthe Belgian territory, one of them being at the University of Liège and running for a few years.
The following tasks will at least be carried out by the candidate: Understand how both BRAMS and FRIPON experiments works and what data are produced Study of overdense trail meteor echoes observed by the BRAMS network : timing, power
variation, relation to mass? Study of head echoes observed by BRAMS : determination of the speed of the meteoroid
from multi-station observations, comparison with FRIPON estimate of speed. Comparison of trajectories of fireballs retrieved from BRAMS and FRIPON data
This work will be done in collaboration with the BRAMS team (PI : Hervé Lamy) working at BISA(Belgian Institute for Space Aeronomy) and with the FRIPON team (PI : François Colas) working atObservatoire de Paris.
Effect of charged particles on Doppler tracking
Contact person: Promoteur : Véronique Dehant ; Co-Promoteur : S. Le Maistre and N. Bergeot
e-mail : [email protected], [email protected], et [email protected]
Tel: 02 373 0266 (V. Dehant)
Office: Royal Observatory of Belgium
Availability: any time by skype; please email to decide when
Thematics: Planetology
Description:
The aim of this study is to analysis the effects of plasma turbulence in the solar wind on Doppler radiomeasurements from interplanetary spacecraft.A spacecraft receives an uplink signal at X-band (7.3 GHz), and transmits data (telemetry or science data) toEarth at X-band (8.4 GHz). In addition, a small amount of downlink power is transmitted at S-band (2.3GHz) for evaluating the link performance. The two-way communication involves a 34-m ground station onthe Earth that is equipped to emit an X-band signal and to receive both X- and S-band signals. At the groundstation a loop tracker records carrier phase at both downlink carriers and this data is analyzed to examine theeffect of charged particles, from solar plasma and the Earth’s ionosphere, on Doppler passes typically usedfor tracking and navigation of interplanetary spacecraft. The Doppler tracking data allows, for example,calculating the gravitation field of the planet around which the spacecraft is orbiting. The experimental dataset that will be used in this study are dual-frequency tracking of the European Space Agency’s (ESA) MarsExpress (MEX) spacecraft.
Task Description:
Based on the power spectral analysis of the dual-band data, the objective of this work is to determine anappropriate model for treating charged particles in single-band Doppler tracking. The statistic of phasevariations will be described in terms of Allan variances (root-mean-square differenced-phase variations) andclassical frequency variance. A realistic modelling of the charged particle variations should be adopted ingood agreement with current models developed to estimate the solar plasma variations as a function of SEPangle.
This work can lead to a publication and can be continued in a PhD thesis.
Whitening of Doppler data in Radio Science
Contact person: Promoteur : Véronique Dehant ; Co-Promoteur : S. Le Maistre
e-mail : [email protected] et [email protected]
Tel: 02 373 0266 (V. Dehant)
Availability: any time by skype; please email to decide when
Thematics: Planetology and planetary systems
Description:
The Deep Space Network (NASA) and ESTRACK (European Space Tracking - ESA) ground stationscommunicate directly with space probes using uplink radio signals (Earth → Transponder) and downlinksignals (Transponder → Earth). The frequency band used is often the X band or the S band.
When the Earth is in the field of the antennas of the probes and if the programmatic allows it, the probestransmit for a limited period of time telemetry data (navigation or sensors data) and / or scientific data(binary data or pure modulated tones). The scientific data of interest to us in this case are the Doppler dataproduced by the DSN and ESTRACK antennas after receiving the descending signals (One-Way path: Probe→ Earth or Two-Way path: Earth → Probe → Earth).
The data stored in flat files are often temporally correlated and therefore require pre-processing so as not tooverestimate the quality of the Doppler data with artificially small formal errors.
The objective of this master thesis is to apply a suitable data-whitening algorithm that will de-correlate andnormalize the components of the Doppler signal embedded in background noise (Gaussian white noise).
Description of tasks:
Examine the different whitening algorithms that allow to process highly noisy data and comparetheir respective performances.
Implement under Matlab the algorithm of whitening that will minimize the amplification of the noisepresent in the Doppler data (files from DSN and ESTRACK ground stations).
This work can lead to a publication and cab be continued in a PhD thesis (application to real data).
Effects of the meteorite and comet impacts on the degassing of volatilecontained in the icy soil
Contact person: Promotor: Véronique Dehant; Co-Promotor: E. Gloesener
e-mail: [email protected] et [email protected]
Tel: 02 373 0266 (V. Dehant)
Office: Royal Observatory of Belgium
Availability: any time by skype; please email to decide when
Thematics: Planetology
Description:
Mars past: Geological evidences suggest that Mars underwent a period in the beginning of its history wherethere was liquid water. Early in its history, Mars had a denser atmosphere with sufficient greenhouse gases tosustain the presence of stable liquid water at the surface. Impacts by asteroids and comets have played asignificant role in the evolution of the martian atmosphere, not only by causing atmospheric erosion but alsoby delivering material and volatiles to the planet. A period with intense bombardment of meteorites couldhave increased the atmospheric loss, but to explain the loss of a speculative massive atmosphere in the LateNoachian (< 3.5 Ga), other factors of atmospheric erosion also need to be taken into account. CO 2 andmethane (CH4) may well be part of this process, with potential storage in the martian regolith, in clathrates.Clathrates or gas hydrates are crystalline compounds formed by the inclusion of gas molecules in the cavitiesof water molecules network and are typically stable at high pressure and low temperature. These clathratescould have remained stable at depth over geologic time.
Mars present: The discovery of methane on Mars is a very fascinating subject because it could possibly berelated to past or present life... However, this is not the only explanation as it could be related to degassing ofclathrates. Nowadays, martian methane remains surrounded by a lot of mystery as its source and sink areunknown. Its observed spatial and temporal variations challenge our current understanding of atmosphericchemistry and physics and, as a result, its presence on the Red Planet is still debated today. The ExoMarsTrace Gas Orbiter (TGO), currently in orbit around Mars, provides new CH 4 data. This mission, launched in2016 is specially designed to detect methane and other trace gases. The scientific team of TGO recentlyannounced that they did not find methane while the presence of methane has been detected at the surface ofMars by Mars Science Laboratory (MSL). The destabilization of clathrates in regions where methane hasbeen locally reported requires considering processes such as seismic activity or impacts.
This work: Meteorite and comet impacts still exist on Mars and may play a role in the degassing of methanetrapped in clathrates as the impact may increase the surface and subsurface temperatures. However, this hasnever been demonstrated. The idea of the work is to investigate temperature and pressure changes followinga small impact in order to compute methane degassing in the atmosphere of Mars from the regolith whendestabilizing the clathrates.
Task description:
Get familiar with existing codes for computing methane degassing Compute the temperature associated with an impact Verify if this allows degassing of methane.
This work can lead to a publication and can be continued in a PhD thesis.
Remote sensing of forest vegetation biomass from drone flights towards thevalidation of dynamic vegetation models
Contact person : Louis FRANCOIS
e-mail : [email protected]
Tel : 04/3669776
Office: B5c 0/4
Availability: May : 21-24, 27, 29 ; June : 11-14 (preferred dates for ISSeP colleagues), 17-19, 21, 24-28
Thematics : Climate, environment and oceanography
Description:This master thesis draws on an earlier work undertaken by a space science student within the framework ofthe BELAIR programme of the Belgian Science Policy (BELSPO). The objective of BELAIR is to developremote sensing test sites within various ecosystems in Belgium. A new site has been created recently withinBELAIR in connection with forest ecosystems. This site is named BELAIR SILVA. It is located in theBelgian Ardenne, spanning the region comprised between Vielsalm and Eupen, and also extending towardsthe Sart Tilman domain and Liège. In 2018, BELSPO organised airplane and drone flight campaigns over theBELAIR SILVA area. These flights were equiped with multispectral and/or hyperspectral imaginginstruments.
In particular a drone flight was organised over the Vielsam FLUXNET site (https://fluxnet.fluxdata.org),where an eddy covariance station is installed since 1996 to monitor the exchange fluxes of water vapour andcarbon dioxide between the forest and the atmosphere. The forest at this site is a mixed temperate forest withseveral tree species, including European beech, Douglas fir, silver fir, Norway spruce, etc. In the earlierstudy, techniques have been developed to identify the species and map their distribution from themultispectral images taken during the drone flight. The height and crown area of each tree has also beenretrieved. These parameters can then be used to evaluate the biomass of the tree by the use of allometricrelationships used by foresters and available for most common tree species. In this way, it will be possible toget high resolution biomass data per tree species in an automatic way and for relatively large territories. Suchdata are important to validate forest vegetation models and refine the estimates of forest carbon storage atregional scale.
The objective of the master thesis will be to extend the methods developed in the earlier work, by testingthem on other sites and for other tree species. Drone flights will be organised in collaboration with ISSeP.Ground-based field works will also be performed to evaluate biomass and net primary productivity byconventional forestry methods (ground-truth measurements). Possible sites include oak forests near Eupenand the Hautes-Fagnes area already ground-based sampled and modelled in a recent study (Paillet et al.,submitted), as well as some woodlands within the Sart Tilman university campus. Besides biomass,possibilities of retrieving some trait data (such as leaf nitrogen content or specific leaf area) necessary formodelling the tree species will also be investigated. The work will also include the use of the CARAIBdynamic vegetation model, which will be run over the studied forest sites and will be validated with theacquired ground-based and remote sensing data.
Collaboration : Benjamin BEAUMONT and Eric HALLOT, Remote Sensing and Geodata Unit, ISSeP, ruedu Chéra 200, 4000 Liège
Alain Hambuckers, UR SPHERES, University of Liège, BelgiumBELAIR SILVA consortium, Belgian Science Policy (BELSPO)
Future ranges of tropical tree species under climate change : improving thepredicting capacity of a dynamic vegetation model
Contact person : Louis FRANCOIS
e-mail : [email protected]
Tel : 04/3669776
Office: B5c 0/4
Availability: May : 21-24, 27, 29 ; June : 11-14, 17-19, 21, 24-28
Thematics : Climate, environment and oceanography
Description:Climate change is expected to severely impact the distribution of plant species in the future on all continents.This may have huge consequences for the ecosystems and the services they provide to human populations.
Species range projections for the future are often made with niche-based models (NBM). These are statisticalmodels that establish empirical relationships between the distribution of a species and the climate variables.Another type of model is also commonly used to project the impact of climate change on vegetation: thedynamic vegetation models (DVM). These models are process-based. They are most often used for largegroups of plant species (the so-called plant functional types, PFT, such as, the needle leaf trees or thebroadleaf deciduous trees in temperate regions), for instance to project the carbon budget of ecosystems intothe future. However, a few DVM can be run at the species level. They can then become very powerful toolsto project the current ranges of plant species into the future and to assess the risk of reduction of theirdistribution, or even the risk of species extinction, in the face of climate change. Indeed, these modelsintegrate ecological and physiological processes governing the growth of the species. However, they need acareful calibration and validation at the species level, to increase their predictive capacity, since up to nowthese models have mostly been used at the PFT level.
Our laboratory has developed such a species-based DVM: the CARAIB model. A recent comparison ofCARAIB with a NBM (MaxENT) for more than 50 tropical tree species of South America showed twophenomena.
First, the DVM projected clearly larger present-day distributions than the NBM. Now, NBMs are recognizedfor their good ability to project the realized niche (ecosystem position taking into account the bioticinteractions) and the current distributions. This raised the hypothesis that the DVM is able to predict thefundamental niche of the species, i.e. the distributions when the species are not limited by their parasites orpredators. The natural enemies constrain species distributions to their most favourable climatic regions,where also the species are the most efficient. Indeed, in these conditions, the plants are able to grow rapidlyand to replace their injured parts and they produce more secondary metabolites allowing them to poison theirpredators. The first objective of the master thesis would be to test the efficiency of one or several climateconstraints in the model on its predictive power.Secondly, the predictive power of the DVM shrank when the number of presence coordinates was low (<20). The coordinates allow the determination of bioclimatic thresholds. It is probable that the informationwith small numbers of presence data did not encompass enough the thresholds. This raised the hypothesisthat the use of the density distributions of the climatic factors would allow to determine with more precisionthe bioclimatic thresholds. The second objective of the master thesis would be to test this hypothesis.
Collaboration : Alain Hambuckers, UR SPHERES, University of Liège, Belgium
Hydrodynamic ensemble simulations of the North Sea
Contact person : Alexander Barth (Uliege), Sébastien Legrand (RBINS)
e-mail : [email protected], [email protected]
Tel : 043663664
Office: B5 2/56
Availability: 29 May 2019, 3 June 2019 (please confirm dates by email)
Thematics : environment and oceanography
Description:
Operated by RBINS, the Belgian Marine Forecasting Centre (http://www.marineforecasts.be/) provides, twice a day, five-day marine forecasts (currents, waves, storm surges, sea water temperature and salinity, etc.) on the English Channel (from 5°W) and the southern North Sea (upto 57°N) with a special focus on the Belgian part of the North Sea. RBINS marine forecasts are based on the model COHERENS and are used daily in support to numerous human activities at sea, for professional, recreative or scientific purposes. The range of applications spans various areas of benefit such as civil protection, maritime safety, marine and coastal environment protection and scientific research.
High resolution models do however not necessarily provide an accurate forecast. Initial and forcing errors, and the chaotic nature of ocean flows can degrade the forecast accuracy when making predictions. The aim of this master thesis is to quantify the error variance of the model by making ensemble simulations (also called Monte Carlo simulations) by perturbing uncertain aspects of the models. In essence, the master student will run the model of RBINS with slightly perturbed initial conditions and atmospheric forcing fields (like wind stress, air temperature, cloud coverage). The magnitude of these perturbations should match the uncertainty of the input fields. Based on an ensemble of input fields an ensemble of model results will be generated. From this ensemble of model simulations, one can for example derive the error variance of the temperature forecast or the probability that the currents at a given location exceed a given threshold value. Some perturbations will have the tendency to make the model error grow in time and thus defining over which time scale a useful model forecast can be made.
This master thesis will give the master student hands-on experience in computational fluid dynamics and will benefit from a strong collaboration between RBINS and ULiege.
Recommended lectures: Structure and application of numerical ocean models (OCEA0036-1) Data assimilation and inverse methods (OCEA0097-1)
Neural networks to reconstruct missing data in ocean satellite observations
Contact person : Aida Alvera-Azcárate, Alexander Barth (ULiege)
e-mail : a.alvera@ ulg.ac.be; [email protected],
Tel : 043663664
Office: B5 2/56
Availability: 29 May 2019, 3 June 2019 (please confirm dates by email)
Thematics : environment and oceanography
Description:
The GHER is currently developing a method to reconstruct missing data in satellite images using a convolutional neural network. Satellite observations working in the optical and infrared bands are affected by clouds which obscure part of the ocean underneath. This limits the amount of available data for subsequent studies. Neural networks are specially well positioned to detect the presence of nonlinear, stochastic features like small scales in the ocean surface that are measured by satellite sensors, and will be used in this Master Thesis to reconstruct missing data in satellite images. This method will be applied to a series of sea surface temperature (SST) data. More specifically, the influence of multivariate input data in the quality of the reconstructed SST will be assessed. Variables like ocean surface winds and air temperature have a direct influence in the SST and will likely provide useful information to reconstruct ocean variables. The method will be applied to a time series of 25 years of data in the Ligurian Sea (northwestern Mediterranean Sea), and results will be used to analyse interannual variability and to assess the influence of atmospheric and oceanic indexes like the North Atlantic Oscillation.
Recommended lectures: Data acquisition and analysis, complements (OCEA0035-1) Data assimilation and inverse methods (OCEA0097-1) Satellite oceanography (OCEA0087-1)
Spatio-temporal variability of eddies in the vicinity of the Gulf Stream
Contact persons: Sylvain Watelet, Jean-Marie Beckers (GHER)
E-mails: [email protected], [email protected]
Office: B5a, 2/12
Availability: Preferentially monday’s afternoon.
Thematics : Climatology, Oceanography
Description:
Ocean currents are often represented as straight flux lines between large scale gyres. For instance,the mean Gulf Stream path separates from the US east coast at Cape Hatteras and then flowsbetween the subtropical and subpolar gyres towards the central North Atlantic from where itsextension to the European and Arctic seas is known as the North Atlantic Drift. Although the meanpath of the Gulf Stream has been studied for centuries (already observed in the 16th century), itsspatio-temporal variability and their causes are still subject to some debate. In particular, it wasdiscovered in 1947 that rings typically smaller than 100 km are generated from the cut-off of GulfStream meanders. On its northern side, the Gulf Stream releases warm-core eddies rotatingclockwise while cold-core eddies rotating anti-clockwise are generated on its southern side. Theseeddies have a large impact on Gulf Stream heat transport and more generally on marine biology oreven hurricanes strength. Although the recent arrival of satellites greatly improved our knowledgeof their structures, the studies on Gulf Stream eddies are often limited to short time periods.Besides, satellite data only allow the study of the ocean surface while most ocean climate modelshave a coarse resolution (~1°) and are thus not « eddy-permitting ». Here, in the frame of anongoing PhD thesis at GHER, very high resolution fields of temperature, salinity, surface height andspeed from DRAKKAR simulations at 1/12° will be made available to the master student. Thesefields available for the period 1970-2015 will be used to detect the eddies, characterize their spatialvariability and highlight possible trends. The teleconnections between the atmospheric forcings(NAO,…) and these ocean turbulences might be studied as well.
Interpolation of SWOT altimetry data using variational methods
Contact persons: Charles Troupin
E-mails: [email protected]
Office: B5a, 2/11
Availability: Afternoon except Wednesday.
Thematics : Oceanography (remote-sensing)
Description:
SWOT stands for Surface Water and Ocean Topography (https://swot.jpl.nasa.gov/) is a missionfocused the world's oceans and its terrestrial surface waters. The SWOT satellite will provide oceansurface measurements with an unprecedented resolution, over a 120 km wide swath with a ~20 kmgap along nadir. This particular spatio-temporal data distribution makes it necessary to use specificinterpolation techniques to create gridded fields of sea surface height.
While the satellite is expected to be launched in April 2021, the SWOT simulator(https://github.com/SWOTsimulator) already provides artificial SWOT measurements on seasurface height with simulated errors based on numerical ocean simulations. Such a dataset could beused to perform some tests in order to prepare for the real data coming after the satellitedeployment.
The proposed work consists in:1. Making an inventory of the interpolation techniques usually applied to altimetry data and to simulated SWOT measurements.2. Apply the DIVAnd (https://github.com/gher-ulg/divand.jl) tools on simulated SWOT data in a selected region and assess:
the computation time, the quality of the gridded fields and the estimation of the analysis parameters.
Simulating Tsunami propagations
Contact person : Jean-Marie Beckers,
e-mail : [email protected],
Tel : +3243663358
Office: B5a 2/5
Availability: Preferentially monday’s afternoon.
Thematics : Oceanography
Description:
Since the Sumatra 2004 disaster, Tsunami alert systems have been implemented in a series ofplaces, including models for Tsunami-wave propagation. The proposed work will provide anassessment of the associated uncertainties.
The work will start with an analysis of the different mathematical models used to simulate Tsunamipropagation (linear models, Boussinesq models, ray-tracing models). One model will be selected toprovide a realistic simulation of the Sumatra case. A baseline simulation, reasonably close theSumatra case, will then serve to analysis how uncertainties on the model formulation (including ornot the non-linearity of the Boussinesq model), topographic uncertainties or the initial shape of theperturbation lead to different predictions. To do so, one has on the one hand side to define a strategyto create perturbed simulations and on the other hand to defined adequate metrics (moment of wavearrival, amplitude, …).
The outcome will then be an estimate of errors in the Tsunami propagation model.
The work will rely on a good understanding and analysis of fundamental geophysical fluiddynamics and the use of a numerical model.
