- Application Deadline
- Applicant notification
- Fellowship start date(s)
- Please send inquiries
about the program to firstname.lastname@example.org
The Center for Computational Astrophysics (CCA) at the Flatiron Institute is a vibrant research center in the heart of New York City with the mission of creating new computational frameworks that allow scientists to analyze big astronomical datasets and to understand complex, multi-scale physics in a cosmological context.
The CCA Pre-Doctoral Program will enable graduate student researchers from institutions around the world to participate in the CCA mission by collaborating with CCA scientists for a period of 5 months on site. With this opportunity, the selected group of researchers will be able to participate in the many events at the CCA and interact with CCA scientists working on a variety of topics in computational astrophysics (including both numerical simulations and sophisticated analyses of observational data), thereby deepening and broadening their skill sets.
CCA Pre-Doctoral Program participants will be employed for up to 5 months at the CCA as Research Analysts. More information about this paid position is available on the application page, which can be accessed by clicking ‘Apply Now‘.
Research Analysts will collaborate with one or more CCA scientists on a project of mutual interest. Potential applicants can find a list of projects proposed by possible CCA mentors by going to the mentors tab above. Alternatively, applicants may propose a project related to the interests of one or more of the CCA mentors listed. Before applying, applicants must contact one or more potential mentors to discuss the project of interest in detail and specify the selected mentor(s) in the research proposal. Applicants are encouraged to collaboratively develop their project proposals with their prospective mentor(s).
Applications for the Research Analyst position should be submitted here by April 15. Applicants will be notified about the status of their applications by May 15.
- Application Deadline
- Applicant notification
- Fellowship start date(s)
- Please send inquiries
about the program to email@example.com
|TITLE: Research Analyst||REPORTS TO: (Associate) Research Scientists and/or Group Leaders|
|DEPARTMENT: CCA||FLSA STATUS: Exempt|
|LOCATION: NY office (162)||EMPLOYMENT CLASSIFICATION: Fixed-term|
The Center for Computational Astrophysics (CCA) at the Flatiron Institute seeks temporary full-time Research Analysts as part of its Pre-Doctoral Program.
The aim of this program is to provide graduate students from institutions worldwide the opportunity to be employed at the CCA for 5 months for the purpose of working on a research project with one or more CCA staff mentors as well as expanding collaborative networks. The project should be distinct from but complementary to the student’s thesis. The program is open to individuals who are currently pursuing a PhD in a relevant field. Students who currently have limited access to computational astrophysics resources and mentorship are especially encouraged to apply.
Before applying, candidates for this position must contact one or more potential mentors, and the prospective mentor(s) must approve of the candidate’s possible participation in the program and commit to serving as mentor in order for the candidate’s application to be considered. Applicants are encouraged to collaboratively develop their project proposals with their prospective mentor(s). Candidates are encouraged to work with Flatiron Research Fellows, but at least one co-mentor must be at the Associate Research Scientist level or more senior. Mentors who are full-time CCA staff (and thus have limited access to students) will be prioritized. For further details about the program and a full list of mentors’ potential projects and descriptions of broad research interests, please visit: https://www.simonsfoundation.org/grant/flatiron-institute-center-for-computational-astrophysics-pre-doctoral-program/
- Currently enrolled in a PhD program in a field relevant to the proposed research
- 1-2 years of advanced coursework in a relevant field
- Demonstrated understanding of basic research skills. Please note that extensive research experience is not required for this position
Related Skills & Other Requirements
- Some basic technical skills (e.g. basic Linux, some experience with Python or similar)
- Technical and scientific curiosity
- Professional communication skills
REQUIRED APPLICATION MATERIALS
- CV and publication list
- Description of previous research experience (not to exceed two pages, including figures and references)
- Research proposal of not more than 2 pages (including figures and references) outlining planned work at Flatiron
- Two (2) letters of recommendation submitted confidentially by the letter writers to firstname.lastname@example.org. One letter must be from the applicant’s PhD supervisor and must explicitly approve the applicant’s possible participation in the Pre-Doctoral Program
- All applications must be submitted no later than April 15, 2022
THE SIMONS FOUNDATION’S DIVERSITY COMMITMENT
Many of the greatest ideas and discoveries come from a diverse mix of minds, backgrounds and experiences, and we are committed to cultivating an inclusive work environment. The Simons Foundation actively seeks a diverse applicant pool and encourages candidates of all backgrounds to apply. We provide equal opportunities to all employees and applicants for employment without regard to race, religion, color, age, sex, national origin, sexual orientation, gender identity, genetic disposition, neurodiversity, disability, veteran status, or any other protected category under federal, state and local law.
- Application Deadline
- Applicant notification
- Fellowship start date(s)
- Please send inquiries
about the program to email@example.com
Megan Bedell (CCA, ARS)
Open-source Tool for RV Observation Scheduling (With Lily Zhao)
Using an information-theory approach, design algorithms to calculate optimal future observation times for further follow-up of potential planet host stars in radial velocity (RV) surveys. Make the results available as an open-source tool for use in planning RV surveys including HARPS3/Terra Hunting Experiment, EXPRES, NEID, KPF, etc.
Simulating the Sun As a Star (With Lily Zhao)
Modern extreme precision radial velocity (EPRV) instruments often observe disk-integrated sunlight, and some of these data sets are publicly available (e.g. 3-yr HARPS-N, NEID). This project will use solar observations to validate software for simulating magnetically active stars and test the accuracy of common assumptions about the effects of stellar activity on the observed spectrum.
Matteo Cantiello (CCA, RS) (with Mathieu Renzo)
Stars are likely formed in, or captured by, the disks of active galactic nuclei (AGN). The disk conditions profoundly change the star’s evolution, with AGN stars accreting large amounts of mass and becoming massive / very massive. See Cantiello, Jermyn & Lin 2020
– Modeling the accretion stream on AGN stars with radiation hydrodynamics software instruments like Athena++
– Calculate rate of visible explosive transients in AGN disks from a population of massive AGN stars. Use order of magnitude estimates from Jermyn et al. 2022, plus AGN disk models neglecting migration effects (e.g. Sirko & Goodman 2003). Results from Perna et al. 2021 can help determine transient visibility in different parts of the disk. Connections to e.g. VRO/LSST, LIGO/VIRGO
– WD accretion rate and nucleosynthesis to explain \alpha-enrichment of AGNs, radial rate of SNIa vs. Novae
Late Phases of Massive Stars Evolution
The spin rate of black holes and neutron stars, and its relation to the pre-explosion core structure and the physics of stellar explosion is an as-yet insufficiently explored topic. Present measurements in X-ray binaries and GW sources show very different black hole spin-distributions, and many theoretical processes might influence the final core spin and the explosion of massive stars. These include stochastic spin up/down of the inner core by late shell burning and accretion of high angular momentum material from convective shells in red supergiants. These mechanisms can introduce a stochastic component in the spin evolution of stars, and their importance for the sample of observed compact object spins is as of yet poorly explored.
Study the spin rate of compact remnants due to
a) Stochastic spin up from IGWs (extending the work of Fuller et al. 2015)
b) Stochastic angular momentum accretion from convective shells fallback (extending the work of Quatert et al. 2019) In particular the goal will be to characterize amplitude and relative orientation of the spin vector, consider different types of objects (e.g. He stars) for applications to GW progenitors from isolated binaries.
The majority of massive stars live their life with companions and will exchange matter through their Roche lobes or possibly merge before exploding. These binary interactions have a strong impact on the appearance and structure of both stars, and observational advancements (e.g., Gaia, VRO/LSST, LIGO/VIRGO) require detailed understanding of common and rare evolutionary paths.
Many stellar phenomena have been linked with accretion in a binary system (e.g., Be stars, LBVs, etc.), and detailed modeling of the internal structure of both stars and their reactions to binary interactions is possible with the MESA software instrument, while rapid population synthesis simulations (e.g., COSMIC) can be used to explore the parameter space and make predictions for event rates, and (magneto-)hydrodynamical codes (e.g., Athena++) might allow to run direct numerical simulations of rapid binary interactions in the very high mass regime.
– Mass transfer with MESA: Study stellar structure reaction to varying mass and angular momentum accretion rates. Compare to full binary run, possibly find analytic fits.
– Model grid of accretors to be used for unpublished galactic eclipsing binaries, Be stars in LB1 and HR6819 right after RLOF, Be-XRB, Be vs. B[e] stars
– Modeling of late CCSNe from RSG+CO WD (cf. Zapartas et al. 2017), to check if it leads to lifting the WD degeneracy or to a type 1.5 SN
– ATHENA++ modeling of RLOF accretion on a ~50+Msun star: relevant to GW formation scenarios and might be possible if thermal and dynamical timescale are similar.
Computational stellar physics. Angular momentum transport in stars. Massive Stars. Stellar explosions. Stellar Evolution in AGN Disks. Stellar magnetism. Binary Stars.
James Cho (CCA, RS)
Neutron Star Thermonuclear Storm Dynamics
The Coulomb-liquid oceans and plasma atmospheres on neutron stars can enhouse Nature’s most extreme storms that are powered by thermonuclear burning. We will work together with the student to analyze the latest computational fluid dynamic simulations modeling this phenomena. The student will also collaborate with us to implement a more sophisticated thermonuclear burning reaction networks to the simulation code. The results will help us advance our understanding of thermonuclear X-ray bursts on accreting neutron stars. (This project is jointly supervised with Dr. Joonas Nättilä)
Solar Convection Zone’s Differential Rotation
Helioseismology shows that the convection zone inside the Sun is differentially rotating. In this project, the student will perform and analyze three-dimensional, global hydrodynamics and magnetohydrodynamics simulations to understand the generation and stability of the differential rotation . The student will become acquainted with the following for convection in rotating-magnetized environments: analytical method for analyzing the governing differential equations and numerical method based on spectral transforms.
William Coulton (CCA, FRF)
Cosmology beyond the two-point function
Upcoming cosmology surveys will provide exquisitely precise measurements of our universe, however performing unbiased inferences that maximally exploit the potential of these data-sets is an ongoing challenge. I am interested in a broad range of areas from developing methods to efficiently simulate observables, implementing novel non-Gaussian statistics that complement existing analysis methods, and exploring likelihood approximation schemes / LFI. There is a range of potential projects, primarily focused on CMB and weak lensing observables, but I would also love to collaborate on topics within the broader areas. If you’d like to discuss this further please contact Will Coulton.
Shy Genel (CCA, RS)
My interests are in modeling galaxy formation using cosmological simulations such as IllustrisTNG or CAMELS, with a particular focus on galaxy morphology and dynamics, galaxy growth through accretion and mergers, and mock observations of simulated galaxies. I welcome candidates’ ideas for specific areas of study and will be happy to develop a project vision together.
Sultan Hassan (CCA, FRF)
Self-consistent models of AGN and photon escape fraction durning reionization (with Shy Genel)
There is currently debate on the role of AGN during reionization. The student will explore all possible models of AGN and photon escape fractions to constrain the contribution to reionization. As a result, a new self-consistent model of AGN and photon escape fraction will be designed. Parameter inference will be performed using different techniques such as MCMC and likelihood free inference.
Machine learning applications to emulate reionization and galaxy formation models (with Shy Genel)
Extracting the maximum amount of information from future surveys remains a challenge. Depending on the interest, the student will explore different generative models to emulate high dimensional datasets to enable rapid parameter inference at the field level, in order to maximize the scientific return of future surveys. Several directions can be discussed using datasets either from CAMELS or reionization simulations.
Research Interests: Reionization, Galaxy Formation, Machine Learning
Chris Hayward (CCA, RS)
Simulating the birth and adolescence of galaxy clusters (with Doug Rennehan)
Recent observations suggest that there are highly overdense, merging groups of galaxies in the early Universe that are undergoing extreme star formation. These unprecedented systems are protocluster cores — the progenitors of the cores of galaxy clusters. They should quickly merge, leading to the most massive galaxies in the Universe forming only a few billion years after the Big Bang (seemingly contrary to the hierarchical assembly picture). Incoming FRF Doug Rennehan and I have been running high-resolution, state-of-the-art numerical simulations of these systems. We are entering the analysis phase of our project, and there are therefore several projects that would be possible with the data we are producing from our unique simulations. Some exciting projects involve using post-processing tools that simulate existing and future instruments in order to compare to real observed galaxies.
Possible projects include:
– The history of the intracluster light: Use mock observations for JWST to make predictions for the light profiles within the nascent clusters.
– The history of the CGM: When does the CGM form? When do its properties get set? Use the pyxsim Python package to make X-ray predictions for future instruments.
– Field vs. cluster: Are the properties of galaxies forming in such an overdense region significantly different from those outside?
– Protoclusters in the IR/submm: Use the SKIRT radiative transfer code to compute IR/submm emission from the simulated protoclusters and compare with observed systems discovered with the South Pole Telescope, Planck, and Herschel.
Projects about the circumgalactic medium, galactic winds, or multiphase mixing (with Drummond Fielding)
Drummond and I would potentially be interested in co-mentoring a student on a project about the circumgalactic medium, galactic winds, or multiphase mixing. Please reach out to discuss ideas!
– Galaxy formation, with a particular focus on “bridging the gap” between simulations and observations. Feedback in Realistic Environments (FIRE) simulation collaboration, and projects analyzing existing FIRE simulations or running new ones using the FIRE-3 code are possible.
– Stellar feedback and outflows – Infrared/submm-detected galaxies
– Obscured AGN
– Galaxy protoclusters – Dust radiative transfer/synthetic observations – Using machine learning techniques to emulate radiative transfer (with Rachel Cochrane)
– Spectral energy distribution modeling.
– Reach out so that we can find out what interests we share and collaboratively come up with a project idea.
David W Hogg (CCA/NYU, Group Leader)
Equivariant Machine Learning for Cosmology
We have developed new methods for making machine-learning models obey exact symmetries, such as rotation, translation, and permutation; and also dimensional scalings. At the same time but separately, there have been great results from CCA on applying machine learning to cosmology and cosmological simulations to make emulators, translators, and classifiers. The family of projects I have in mind would involve mashing up these two lines of research to make symmetry-respecting machine-learning methods targeted at cosmological applications. There is a range of possible projects that could be pursued here; we could start by finding the right fit, given any applicant’s interests.
Yan-Fei Jiang (CCA, ARS)
Accretion Onto Stellar Mass Black holes in a Disk
This project is to study how accretion happens when a stellar mass black hole is embedded in a disk. We will perform 2D global radiation MHD simulations from the Hill radius down to the region near the black hole. We will study the properties of the accretion flow for gas with different thermal properties and angular momentum fed from the Hill radius.
Effects of Rotation On Massive Star Structures
The project is to study the 3D structures of massive star envelopes with rotation. The first step is to analyze existing simulation data to understand how convection in the radiation pressure dominated envelope is modified by different rotation speeds, and how the angular momentum distribution is modified by convection. The second step will require running a few 3D radiation hydrodynamic simulations to explore different locations in the HR diagram.
Radiation MHD simulations of various astrophysical systems and development of new numerical techniques. In particular, I am interested in accretion flows onto compact objects and stars, multi-dimensional simulations of massive stars, cosmic ray and radiation driven winds.
Chirag Modi (CCA/CCM, FRF)
Differentiable Halo Finder (in collaboration with Shirley Ho and other ARS/FRF at CCA)
Forward modeling approaches for cosmological analysis such as reconstruction of initial density field or likelihood free inference require developing differentiable simulations to model the observables. In this regard, connecting underlying dark matter field with dark matter halos and galaxies in a differentiable way is an open problem. The finer details of the project will be developed in collaboration with the student but the overall goal of this project will be two fold- i) to develop new machine learning approaches to model dark matter halo and galaxy field in a differentiable fashion and ii) compare them with traditional, as well as other more recently proposed methods, both in terms of the accuracy of the forward model and robustness of corresponding inference.
Learning Subgrid Physics for Cosmological Simulations (in collaboration with Shirley Ho and other ARS/FRF at CCA)
Efficacy of the forward modeling approaches for cosmological analysis, such as reconstruction of initial density field or likelihood free inference, relies on fast and approximate simulations. However these do not simulate accurate physics, especially on subgrid scales, since they miss the small scale gravitational forces as well as baryonic effects. Broadly the aim of the project is to learn this missing subgrid physics. While the details will be developed in discussion with the interested students, one of the ideas is to investigate a hybrid approach to learn the subgrid physics by combining correct simulations on the large scales with machine learning approaches on small scales. Moreover, in addition to testing the accuracy of the forward model, the idea is to also focus on testing the accuracy and robustness of inference when using these subgrid models on realistic data.
Developing forward modeling approaches for cosmological analysis as well as Bayesian inference. Under this umbrella – Build differentiable simulations and models for cosmological observables which sometimes use machine learning. Methodologies for high dimensional inference with statistical methods like monte carlo and variational inference or use machine learning tools like normalizing flows, learnt optimization and likelihood free inference.
Joonas Nättilä (CCA, FRF)
Kinetic modeling of turbulent accretion disk outflows
Non-thermal plasma dynamics of accretion flows around neutron stars and black holes are still largely unknown. We work together to set up large-scale 2D and 3D particle-in-cell simulations to model the turbulent plasmas in accretion disk outflows. We will then use these simulations to study the energization of the plasma. The results will help us understand the formation of non-thermal radiation from accretion disks from first principles.
Magnetic reconnection or collisionless shocks in ultra-strong magnetic fields of magnetars
Magnetic field around magnetars is so extreme that it can influence many of the fundamental plasma processes via quantum electrodynamic (QED) effects. We will work together to set up kinetic plasma simulations to study how QED effects change the evolution and plasma energization processes of i) magnetic reconnection or ii) collisionless shocks. The project is based on recently developed QED interaction algorithms and previously performed reconnection/shock simulations; we will work together to combine these two in order to understand their feedback on each other. The results will help us build physical understanding of the most extreme plasma processes in the Universe.
Research Interests: Extreme plasma astrophysics around neutron stars and black holes. Use first-principles supercomputer simulations and theoretical pen & paper work. Collaborate with students e.g., in understanding (plasma) physics of pulsar and magnetar magnetospheres, accretion disks, or neutron star mergers. Additionally, Development of new state-of-the-art numerical algorithms to better model these systems; this can, for example, include collaboration with the students to develop new machine-learning-accelerated simulation algorithms.
Rachel Somerville (CCA, Group Leader)
Planting Better Merger Trees
“Merger trees” represent how dark matter halos are built up via mergers of smaller halos, and encode a key set of information about galaxy formation and assembly. They are commonly used in semi-analytic models of galaxy formation. Existing semi-analytic methods to generate merger trees are not very accurate, and do not correctly build in the correlations between larger scale environment, halo structure, and halo mass accretion history. In this project we will use machine learning (Recurrent Neural Networks and/or Graph Neural Networks) to develop a fast and accurate method for generating merger trees that correctly builds in these correlations. co-mentor: Chirag Modi
Developing Next Generation Semi-analytic Flow Models for Fast, Interpretatable Emulation of Numerical Hydrodynamic Simulations
Classical semi-analytic models represent the main physical processes of galaxy formation using simple phenomenological recipes. They have been very successful at predicting and interpreting a vast range of galaxy properties, and are much more computationally efficient than numerical hydrodynamic simulations. However, many of the recipes lack rigorous physical grounding. Semi-analytic models are a form of “flow” model, which track the flow of various mass (and energy) components through different reservoirs, such as the IGM, CGM, ISM, and stellar body of a galaxy. This framework also provides useful insights into the baryon cycle in numerical simulations. This project will focus on one of the physical processes in a SAM, and develop a new model that attempts to “emulate” the physics in the IllustrisTNG simulation. Physical processes that could be chosen include heating and cooling, star formation, black hole accretion and feedback, and satellite stripping. One approach to this project could be developing a hierarchical Bayesian model to fit the flow quantities that we have extracted from the numerical hydrodynamical simulations FIRE and/or IllustrisTNG. Depending on the project, we may work with one or more CCA co-mentors.
Ulrich Steinwandel (CCA, FRF)
The Impact of Magnetic Fields on Ram Pressure (with Stephanie Tonnesen)
Ram Pressure Stripping is a gas-removal mechanism that acts on satellite galaxies in which an interaction between the interstellar medium (ISM) and surrounding intracluster medium (ICM) removes the ISM. Observations and simulations indicate that ICM magnetic fields may drape over the satellite galaxy and impact gas stripping. This may also effect the magnetic field orientation near satellite galaxies. However, there are many unknowns about the importance of magnetic fields in ram pressure stripping. Does magnetic draping have an effect on the gas removal rate via ram pressure? Does draping depend on the surrounding magnetic field strength or the galaxy size or velocity? How far behind satellite galaxies are magnetic fields aligned? In this project we will compare a cluster simulation run with HD to one run with MHD. We will determine if the magnetic field changes the stripping rate of galaxies. We will also study the magnetic alignment near galaxies to determine whether magnetic draping is occurring and how it impacts the cluster magnetic field.
Structure of Galactic Outflows Driven by Resolved SN-feedback (with Chris Hayward)
Recent simulations of galaxy formation and evolution have reached a resolution that allows us to simulate the multiphase ISM at solar mass and sub parsec resolution, at least for galaxies on the low mass end with halo masses of up to 1e11 solar masses and stellar masses that can reach up to a few times 1e8 solar masses for gas rich dwarf galaxies. However, these simulations allow for a detailed investigation of the structure of the outflowing gas beyond the typical outflow statistics that is currently used to quantify galaxy outflows in n numerical simulations, such as the mass outflow rate or the mass loading factors. In this project we will investigate in detail, the outflow phase structure of galactic winds based on a sample of ultra high resolution samples of dwarf galaxies that are part of the GRIFFIN (Galaxy Realizations Including Feedback From INdividual massive stars) project (https://wwwmpa.mpa-garching.mpg.de/~naab/griffin-project/), where the multiphase ISM is modeled via non-equilibrium cooling and heating processes. The formation and destruction of molecular hydrogen is modeled directly in the simulation over non-equilibrium rate equations. In all the simulations, the cold phase of the ISM forms single massive stars and we follow their UV and photo-ionising photon budget. Once the massive stars reach the end of their lifetime we explode them as core collapse supernovae for which we explicitly resolve the cooling radius for up to 99 per cent of the SN-explosions and drive large scale (resolved) galaxy outflows.
Modeling the Growth of Magnetic Fields with Semi-analytic Models
In this project we will combine the well established analytic dynamo theory, which has been very successful at describing magnetic field amplification, with semi-analytic models of galaxy formation and evolution, to provide a new path forward for predicting the properties of magnetic fields in galaxies. Many previous works applying analytic dynamo theory have adopted overly simplistic or static assumptions for galaxy properties. State of the art semi-analytic models can efficiently provide predictions for the relevant properties of galaxies. In particular, the velocity dispersion of the disk and its dependence on galaxy properties and redshift is a critical ingredient in the turbulent dynamo model. Recent work by Forbes and collaborators provides a comprehensive model for how cosmological accretion, disk inflows, and other processes determine the velocity dispersion of the ISM.
The end goal of the project is to combine these well-established tools to build a new “magneto”-SAM, and to use it to make predictions of the detailed properties of the magnetic field in galactic disks over large cosmological volumes, and over cosmic time. These predictions would have many applications, including predictions for SKA and other next generation radio telescopes.
Magnetic fields play an important role across many scales in galaxy formation and evolution. Large scale numerical simulations have begun to include magnetic fields, but these simulations are computationally expensive and have limited resolution.
Research Interests: Physics of galaxy and galaxy cluster formation with a special focus on resolving physical processes that govern the respective scales (e.g. magnetic fields, cosmic rays and anisotropic diffusion and the structure of the multiphase ISM ). Numerical methods and parallel computing to treat the problems listed above.
Stephanie Tonnesen (CCA, ARS)
Spans galaxy evolution, range of projects, especially those related to:
• galaxy quenching
• what drives outside-in star formation and quenching
• satellite galaxy evolution
• mock observations of simulations of ram pressure stripped galaxies
• multiphase gas mixing (in the CGM and ICM)
Francisco Villaescusa-Navarro (CCA, RS)
Unveiling Hidden Connections Between Cosmology and Galaxy Formation with Machine Learning
Astrophysical processes such as supernova and AGN feedback affect cosmological observables on small scales. Do changes in cosmology also affect astrophysical objects such as galaxies? We will apply machine learning to the CAMELS simulations to investigate the links between cosmology and galaxy formation processes. Examples of such connections can be found in https://arxiv.org/abs/2201.02202 and https://arxiv.org/abs/2109.04484
Robust Likelihood-free Inference for Cosmological Inference
Neural networks are very powerful tools that can find complicated patterns in the data to perform regression and/or inference. However, they may also be learning spurious numerical effects from the simulations or systematics from the data. We will study this problem and develop machine learning models that will be robust against numerical artifacts and apply them to data from either the Quijote simulations or the CAMELS Multifield Dataset. Examples of this problem can be found in https://arxiv.org/abs/2109.09747
Developing tools to extract the maximum amount of information from the cosmological observations. Large set of N-body and hydrodynamic simulations (e.g. The Quijote and CAMELS simulations) with machine learning. Besides the projects listed above Discussing additional projects that involve cosmology, machine learning, and galaxy formation/evolution in broad terms. Potential students can reach out to me at firstname.lastname@example.org
Vladimir Zhdankin (CCA, FRF)
Microphysics of Turbulent Electron Energization
This project would involve using advanced numerical analysis tools (such as machine learning) to characterize the electron heating, acceleration, and trapping mechanisms in kinetic simulations of trans-relativistic plasma turbulence, in a bid to better inform analytical models of electron-to-ion heating rates. Such an analysis is critical for understanding the radiative properties (including the relatively low luminosity) of hot accretion flows around supermassive black holes, as imaged by the Event Horizon Telescope. The work will leverage big data including multi-scale magnetic field profiles and millions of tracked particles from particle-in-cell simulations.
Radiative Signatures of Plasma Turbulence
The radiative properties of a turbulent collisionless plasma are poorly understood, despite being critical to interpreting emission from a broad range of high-energy astrophysical systems. This project would involve measuring radiative signatures (synchrotron/inverse Compton lightcurves, nonthermal spectra, anisotropy, etc.) from a suite of the latest kinetic simulations of turbulence in relativistic plasma. The results will be compared to observations of compact object accretion flows, outflows, and/or magnetospheres.
Understanding the plasma physical origins of high-energy particles and radiation in astrophysical systems. Much of our knowledge of distant systems comes from observable radiation, but the processes responsible for energizing the emitting particles are often poorly understood. Recent kinetic simulations and analytical theories provide a first-principles description of the complex plasma processes (turbulence, magnetic reconnection, shocks, instabilities) that energize particles in high-energy astrophysical systems such as pulsar wind nebulae, accretion flows, jets, and compact object magnetospheres. However, applying these basic insights to model particular systems requires careful, hard work. Design student projects related to this topic.