Developing efficient and accurate numerical methods to simulate dynamics of physical systems has been an everlasting challenge in computational physics. Physics-Informed Neural Networks (PINNs) are neural networks that encode laws of physics into their structure. Utilizing auto-differentiation, they can efficiently solve partial differential equations (PDEs) by minimizing the loss function at certain points within the domain of interest. The remarkable efficiency exhibited by these networks when solving PDEs positions them as ideal solvers for simulating complex systems.  In this pioneering work, we take a first step towards simulating galaxy dynamics using PINNs by solving the gravitational Poisson equation. We initially substantiate the capacity of PINNs to solve the gravitational Poisson equation for the simple Hernquist (Hernquist, 1990) radial density profile, and for the parametric Dehnen (Dehnen, 1993) radial density profile. Following this, we extended our study to encompass a more complex axisymmetric density profile describing a Thick Exponentiel Disk.  The capacity of PINNs to generate comparatively accurate results has been validated with an average error of 1.71% and 3.75% respectively for the spherically symmetric Hernquist and Dehnen models. While for the axisymmetric thick exponentiel disk model the PINN demonstrated an average relative error of 0.36% with a maximum error of just 0.99% after fine-tuning the PINN’s hyperparameters. Although this model typically relies on the two coordinates R and z along with the ratio η of the model’s scale lengths, the PINN is here trained using a fixed, predetermined value of η.  Drawing upon the outcomes of the grid search implemented for the thick exponen- tial disk model, we provide a succinct examination of how the hyperparameters of the PINN impact the relative error. Given the limited quantity of datapoints, we refrain from formulating definitive conclusions, yet we do exhibit certain discernible patterns. Specifically, we demonstrate that the hyperbolic tangent (tanh) activation function con- sistently outperforms other activation functions in the context of our model. Addition- ally, it appears that augmenting the depth of the network offers superior error reduction in comparison to increasing its width, reinforcing the importance of architectural con- siderations in the optimization of Physics-Informed Neural Networks  Our results show clear advantages of PINNs over regular solvers in terms of efficiency. Despite the success of the two-parameter PINN for the thick exponential disk, further work is required to confirm its extension to three dimensions. This pioneering research offers a promising foundation for further developments in the field, and demonstrates the genuine practical utility of PINNs for simulating complex systems such as galaxies.

A core-collapse supernova (CCSN) is an astronomical explosion that indicates the death of a massive star. The iron core of the star collapses into either a neutron star or a black hole while the rest of the material is expelled at high velocities. Supernovae (SNe) are important for the chemical evolution of the Universe because a large fraction of the heavier elements such as oxygen, silicon, and iron are liberated by CCSN explosions. Another important role of SNe is that the ejected material seed the next generation of stars and planets. From observations, it is clear that a large fraction of all massive stars undergoes SN explosions, but describing how SNe explode has remained a challenge for many decades. The attached papers focus on comparing theoretical predictions with observations, primarily observations of SN 1987A. The compact remnant in SN 1987A has not yet been detected and we have investigated how a compact object can remain hidden in the ejecta (Paper I and II). Because of the high opacity of the metal-rich ejecta, the direct X-ray observations are not very constraining even for potentially favorable viewing angles. However, the combined observations still strongly constrain fallback accretion and put a limit on possible pulsar wind activity. The thermal surface emission from a neutron star is consistent with the observations if our line of sight is dust-obscured, and only marginally consistent otherwise. Future observations provide promising opportunities for detecting the compact object. We have also compared the most recent three-dimensional neutrino-driven SN models that are based on explosion simulations with early X-ray and gamma-ray observations of SN 1987A (Paper III). The models that are designed to match SN 1987A fit the data well, but not all tensions can be explained by choosing a suitable viewing angle. More generally, the asymmetries do not affect the early emission qualitatively and different progenitors of the same class result in similar early emission. We also find that the progenitor metallicity is important for the low-energy X-ray cuto↵. Current instruments should be able to detect this emission from SNe at distances of 3–10 Mpc, which correspond to distances slightly beyond the Local Group. / En kärnkollapssupernova (CCSN) är en astronomisk explosion som indikerar slutet av en massiv stjärnas liv. Stjärnans järnkärna kollapsar antingen till en neutronstjärna eller ett svart hål medan resten av materialet slungas iväg med höga hastigheter. Supernovor (SNe) är viktiga för Universums kemiska utveckling eftersom en stor andel av alla tyngre element såsom syre, kisel, och järn frigörs i CCSN-explosioner. Ytterligare en viktig roll för SNe är att nästa generations stjärnor och planeter bildas av det utkastade materialet. Från observationer är det tydligt att en stor andel av alla massiva stjärnor genomgår SN-explosioner, men att förklara hur SNe exploderar har kvarstått som en utmaning under flera decennier. De bifogade artiklarna fokuserar på att jämföra teoretiska förutsägelser med observationer, primärt observationer av SN 1987A. Det kompakta objektet i SN 1987A har ännu inte blivit detekterat och vi har undersökt hur ett kompakt objekt kan förbli dolt i ejektat (Paper I och II). De direkta röntgenobservationerna är inte så begränsande även längs potentiellt gynsamma siktlinjer på grund av det metallrika ejektats höga opacitet. Däremot begränsar kombinationen av alla observationer starkt ackretion och sätter en gräns för möjlig pulsarvindsaktivitet. Den termiska ytstrålningen från en neutronstjärna är konsistent med observationerna om vår siktlinje är skymd av stoft, och bara marginellt konsistent annars. Framtida observationer utgör lovande möjligheter för att detektera det kompakta objektet. Vi har också jämfört de senaste tredimensionella neutrinodrivna SN-modellerna, som är baserade på explosionssimuleringar, med tidiga röntgen- och gamma-observationer av SN 1987A (Paper III). SN 1987A-modellerna passar datan väl, men alla diskrepanser kan inte förklaras av ett lämpligt val av observationsvinkel. Generellt så påverkar inte asymmetrierna den tidiga emissionen kvalitativt och olika föregångarstjärnor av samma kategori resulterar i likartad strålning. Vi finner också att föregångarstjärnans metallisitet är viktig för egenskaperna av lågenergiröntgenstrålningen. Befintliga instrument borde kunna detektera denna emission på 3--10 Mpc, vilket motsvarar avstånd lite bortom den Lokala galaxhopen. / <p>Examintor: Professor Mark PearceQC 20190121</p>

Recent detections of massive quasars at z &gt; 6 has proven to be an interesting challengefor cosmology given the short time frame that these objects have to reach a certainmass via conventional methods (e.g., galaxy mergers). Direct collapse black holes(DCBHs) are theorized black hole seeds that propose a neat solution to the formationof supermassive black holes (SMBHs) by z ≳ 6. During this report, we followed thefirst photometric method developed to identify these black hole seeds, adapting it tothe James Webb Space Telescope (JWST ) photometric system to facilitate its usefor upcoming data. Two DCBH models were used to discard nearby objects (e.g.,cool stars, brown dwarfs and giant exoplanets) as possible interlopers, and allowed usto identify a possible confusion between DCBH signatures and high-redshift galaxies(z = 6) within an age range of 0.8-0.9 Gyr. This confusion could grow larger if wereconsider the analysis adding dust to the galaxy models we used

In the study of habitability of terrestrial exoplanets, both life-supporting conditions and the prevalence of transient life-threatening events need to be considered. One type of hazardous effect that has so far not received much attention is the thermal effect of a nearby active galactic nucleus (AGN), or in this particular case, the class of the AGN known as a quasar. In this work we investigate the thermal effect from a quasar by calculating the number of habitable terrestrial planets (HTP) in an elliptical or bulge-dominated galaxy, that goes extinct when exposed to the quasar radiation in a limited wavelength range. This is done by approximations and modelling along with pre-existing formulas and data from earlier publications. As a result, the influence by a quasar during the time span of quasar activity will have a less significant impact on the habitability in solar-type stellar systems than expected. Assuming tQSO = 108 yrs of quasar activity, results in the number of affected HTP, ≈ 1 × 105, 9 × 105 and 4 × 108 for isotropic spherical radiation and ≈ 1 × 106, 8 × 106 and 3 × 109 for a double-conical radiation. In terms of stellar mass fraction, ≈ 1.3%, 1.0%, 0.4% for isotropic radiation and ≈ 12.8%, 9.5%, 3.8% for conical, is affected. The results of this work are hoped to provide a rough estimation of the thermal impacts of a quasar on the habitability as well as to point out the most important parameters when considering this model. / I studier om beboeligheten på jordlika exoplaneter övervägs både förutsätningar för liv på planeten men även livshotande händelser i planetens närhet. En typ av farlig effekt som hit- intills inte fått mycket uppmärksamhet, är det termiska effekterna från en aktiv galaxkärna (AGN) eller som i detta fall, AGN-typen kvasar. I detta arbete studeras de termiska effekterna från en kvasar genom att beräkna antalet beboeliga jordlika exoplaneter (HTP) i en elliptisk eller bulge-dominerad galax, (bulge-centralförtätning), som blir obeboeliga då de utsätts för kvasarens strålning i ett begränsat våglängdsområde. Detta görs genom antaganden och modellering av redan befintliga formler och data från tidigare publikationer. Detta resulterar i en mindre inverkan av kvasaren på system kring sollika stjärnor än förväntat. Antaget tQSO =108 år av kvasar-aktivitet ger antal påverkade HTP, ≈ 1 × 105, 9 × 105 och 4 × 108 vid isotropisk strålning och 1 × 106, 8 × 106 och 3 × 109 vid dubbel-konisk formad strålning. Uttryckt i andel stjärnmassa motsvarar detta ≈ 1.3%, 1.0%, 0.4% för sfäriskt fall och ≈ 12.8%, 9.5%, 3.8% vid koniskt. Detta arbete hoppas kunna ge on grov uppfattning om kvasarens termiska effekter på beboligheten men även identifiera det mest betydande parametrarna i denna modell.

Some stars with similar properties to our sun (solar twins) have differ- ent chemical composition than the rest of the solar twins. One explanation might be planet engulfment. Therefore we did a large number of simu- lations where a disturbing star passed a sun and a planet at a distance closer than 100 AU to see how often the planet was engulfed. The result was that the planet in most cases was thrown out of the system, but it was engulfed in about 10 − 30% of the simulations when the planet was close to its star. The conclusion was that planet engulfment indeed can be a good explanation for the different chemical compositions of solar twins, at least in dense stellar clusters where such close passages should be quite common. / Vissa solliknande stja ̈rnor (s ̊a kallade soltvillingar) har en annorlunda kemisk sammansa ̈ttning ja ̈mfo ̈rt med resten av soltvillingarna. En mo ̈jlig fo ̈rklaring kan vara att dessa stja ̈rnor har slukat planeter. Da ̈rf ̈or gjorde vi ett stort antal simuleringar d ̈ar en sto ̈rande stja ̈rna passerade ett sys- tem best ̊aende av solen och en planet. Stj ̈arnan passerade p ̊a ett avst ̊and under 100 AU och vi observerade hur ofta planeten slukades. Resultatet var att planeten oftast slungades ut i rymden, men den slukades i 10-30% av simuleringarna da ̈r planeten kretsade na ̈ra sin stja ̈rna. Slutsatsen vara att planetslukning kan vara en rimlig f ̈orklaring till de annorlunda kemiska sammansa ̈ttningarna hos soltvillingar, ̊atminstone i ta ̈ta stja ̈rnhopar da ̈r na ̈ra stja ̈rnpassager bo ̈r vara ganska vanliga.

The object of this thesis is to extend the current tables for two useful quantities when calculatingcollisional broadening, the cross-section of interaction and the velocity parameter.These quantities, which have hitherto been tabulated for lower states, have now been calculatedfor higher lying states and are used in the ABO - model for spectral line broadening.Having larger tables for these values enables broadening calculations for more spectral linesand this thesis shows examples of calculations which beneted from this. During the calculationsof the tables some unexpected behaviour in the distribution of values arose whichhas not been seen previously. This could point to an unknown underlying mechanism.

New constraints on the scattering cross sections of the Kaluza Klein photon as a darkmatter candidate, its annihilation rate in the Sun and the resulting muon flux on Earth are derived.For this purpose, data collected in the IceCube Neutrino Observatory during 532 days of exposurein the austral winters between 2011 and 2014 have been analyzed with Poisson confidence intervals (J. Conrad et al., 2003) and compared to the simulated prediction achieved with the WimpSimsoftware (J. Edsjö et al., 2003). The results do not allow for any detection claim, but they improveby one order of magnitude the constraints formerly presented in R. Abbasi et al. (2010). Despitethe recent results from LHC experiment which discard lower masses for the Kaluza Klein photon (N. Deutschmann et al., 2017), the new constraints are still relevant for masses above 1500 GeV.

In this paper we analyze the small C1.5 class solar flare observed on June 30th 2013 by the Swedish Solar Telescope. The evolution of asym- metries in the H-alpha line profile of the solar flare was studied where it could be seen how the number of red asymmetric regions had a maximum value near the beginning of the flare and then decreases rapidly in the first 4 minutes of the observations. This could be interpreted as a correlation with the HXR and microwave emissions of the impulsive phase of the flare as these emissions also typically have a similar rapid increase and decrease of emission intensity.

In this thesis project, we study the field of high-dispersion spectroscopy and methods for extracting the spectrum of circumstellar objects such as exoplanets from the combined signal of a stellar system. One of the only techniques for detecting absorption lines in exoplanetary atmospheres is to directly image a planet and record the reflected light. However, exoplanets are incredibly faint compared to the parent star and are often completely obscured in any images of the system. We utilize techniques such as high-dispersion spectroscopy (HDS) and high contrast imaging (HCI) in order to capture the planetary signal and develop methods for reducing only the stellar light while leaving the planet relatively untouched.   We investigate a method for removing the scattered starlight by utilizing the separate spectra of the star and the planet, where the signal from the objects will be spread out according to a point spread function (PSF) and laid on top of each other. By empirically determining the shape of the stellar PSF, reference profiles can be created for each wavelength and subtracted from the entire signal, revealing the planetary spectrum. To achieve this, we have constructed a spectrum extraction algorithm, written in Python 3.6, for use on the spectra of directly imaged exoplanetary systems. Additionally, we discuss many of the problems which may arise when reducing cross-dispersed echelle spectra and attempt to solve them with the algorithm.   To assess our algorithm, we utilize spectral images of the system  Pictoris, taken with the high-dispersion spectrograph CRIRES, and three model exoplanetary systems of varying brightness. When extracting the spectrum of the planets, we find that the method employed for constructing the reference stellar PSFs is partially flawed and leaves a substantial amount of residual stellar light in the reduced images. This leads to difficulties with identifying any spectral absorption lines and an alternative method is likely necessary. Nonetheless, the algorithm is found to successfully extract the spectrum and identify spectral lines of an exoplanetary atmosphere if the planet is sufficiently bright, although only for theoretically unrealistic luminosities. We expect that our algorithm can be improved upon with more well-researched methods for reducing the starlight and by using data recorded with spectrographs of even higher dispersive capabilities, such as CRIRES+, METIS, or HIRES.

Erosion has a long history in science and is used in many different fields today, for example in geology for coastal erosion and in the oil industry for pipe erosion. It is very difficult to study erosion both analytically. Numerically it is difficult due to moving and shape-changing boundaries. Here we develop a numerical model in 3D using the Lattice-Boltzmann method, which is good at simulating complex moving boundaries, and erosion capabilities are implemented. Both laminar and turbulent flow can be modelled with this program. Using an experimentally derived model for the mass change due to erosion in clay and mud-type objects, one can derive equations predicting that the volume of a sphere should, due to erosion, scale as V ∼ −t2. This is also observed with simulations. The shapes of a double sphere with different orientations and a cube in laminar flow we find to have similar power law exponent P, P = 2±0.1. But a cube eroding in Re = 800 had no power law behaviour, meaning that the current analytical framework is incomplete. The possibility of a more general framework is presented for future research. Different Reynolds number also affected the power law behaviour and the shape change over time for the different solids. Very little research has been made for erosion of planetesimals, but it has been argued that erosion can be relevant to their fate. Using the same erosion model, an equation of the erosion time is found for laminar flows and for a sphere. Simulation results find that the equation works within an order of magnitude for turbulent flows, a double sphere and a cube. This gives an estimate of the erosion time t∗ of planetesimals to be t∗ ∼ 1s, given a size of radius equal to 10cm and 1km, an orbital eccentricity e &gt; 10−2 and a distance at r = 1 a.u. Implying that orbits for planetesimals with low eccentricity might be favoured.