Modeling the cosmic web with the Bisous method
Kuupäev
2023-10-04
Autorid
Ajakirja pealkiri
Ajakirja ISSN
Köite pealkiri
Kirjastaja
Abstrakt
Vaatluste põhjal kaardistatud galaktikad ja kosmiline gaas moodustavad keeruka suureskaalalise võrgustiku, mida kutsutakse Universumi kärgstruktuuriks. See kärgstruktuur on vormitud gravitatsiooni ja Universumi paisumise poolt. Suurte peaaegu tühjade tühimike ümber paiknevad kosmilised filamendid, mis ühendavad galaktikate parvi. Enamus Universumi massist asub just filamentides ja neil on arvestatav mõju galaktikate arengule. See doktoritöö uurib Bisous mudelit, mis on matemaatiline meetod kärgstruktuuri mudelleerimiseks vaatlusandmestikest. Bisous mudel otsib tumeaine filamentide tuvastamiseks galaktikate ahelaid, mudelleerides neid ahelaid ühtse ühendatud võrgustikuna.
Mudeli usaldusväärsuse hindamiseks uurisin selle varieeruvust rakendatuna samadel andmetel. Tulemused näitasid, et mudel on praktiliselt koondunud ja usaldusväärne. Kuna valgustugevus väheneb kaugusega, siis see mõjutab ka galaktikate arvtihedust vaatlusandmestikes erinevatel kaugustel. Doktoritöö raames uurisin kuidas sõltub tuvastatud filamentaarne võrgustik sisendandmete tihedusest ja selgus, et madala arvtiheduse korral tuvastab mudel küll vähem filamente, aga samas ei väljasta Bisous peaaegu üldse valepositiivseid filamente, mis tõstab mudeli usaldusväärsust.
Galaktikate arvtihedust on võimalik tõsta kasutades fotomeetriliste punanihete andmeid, mida saab mõõta korraga tuhandete galaktikate jaoks, aga kauguse hinnang on märgatavalt ebatäpsem kui spektroskoopiliste vaatluste korral. Doktoritöö esitleb meetodit, mis kasutab nii fotomeetrilisi kui ka spektroskoopilisi punanihkeid ja seeläbi tõstab filamentide tuvastamise efektiivsust. See meetod on kasulik, kui spektroskoopilisi andmeid on vähe.
Bisous mudeli karakteriseerimine on tähtis tulemuste korrektseks tõlgendamiseks ja fotomeetriliste andmete kasutamine võimaldab mudelit rakendada rohkematele andmetele. Pidev metoodika areng ja vaatluste võimekusega kaasas käimine on tähtis, et täiendada meie teadmisi kärgstruktuurist ja galaktikate evolutsioonist.
The observations have shown that galaxies and cosmic gas form intricate web-like large-scale structure of the Universe. This cosmic web is sculpted by gravity and cosmic expansion. The filaments that connect the galaxy clusters and surround the huge empty voids are the most remarkable of the cosmic web elements, hold most of the Universe’s mass, and profoundly influence galaxy evolution. This thesis explores the Bisous model, a framework developed to model the filaments using observational data. The Bisous model tackles the challenge of studying the unobservable dark matter structure by connecting galaxies into chains and optimizing the distribution of these chains to form an interconnected cosmic web. To assess the Bisous model’s reliability, we examined its variance when applied to the same data. The results confirmed the model’s convergence and showed its robustness. For flux-limited surveys, the observed galaxy number density depends on the distance due to diminishing light. This presents a challenge for detecting the cosmic web at greater distances. Our analysis revealed and quantified a strong relationship between galaxy number density and the completeness of detected filaments. Despite lower densities, the model rarely produced erroneous filaments, adding to its robustness. As a way to boost the galaxy number densities, we also explored using photometric redshift. Photometric data can be measured in bulk for thousands of galaxies, but the distance estimates have significantly larger uncertainties than for spectroscopic redshift data. Our analysis revealed that mixing photometric and spectroscopic redshift data improved filament detection. This approach proves valuable when spectroscopic data are scarce. The main results of the thesis are the quantitative characterization of the model to help interpret the results and a method to boost filament detection. This will pave the way for a more comprehensive understanding of the cosmic web and galaxy evolution.
The observations have shown that galaxies and cosmic gas form intricate web-like large-scale structure of the Universe. This cosmic web is sculpted by gravity and cosmic expansion. The filaments that connect the galaxy clusters and surround the huge empty voids are the most remarkable of the cosmic web elements, hold most of the Universe’s mass, and profoundly influence galaxy evolution. This thesis explores the Bisous model, a framework developed to model the filaments using observational data. The Bisous model tackles the challenge of studying the unobservable dark matter structure by connecting galaxies into chains and optimizing the distribution of these chains to form an interconnected cosmic web. To assess the Bisous model’s reliability, we examined its variance when applied to the same data. The results confirmed the model’s convergence and showed its robustness. For flux-limited surveys, the observed galaxy number density depends on the distance due to diminishing light. This presents a challenge for detecting the cosmic web at greater distances. Our analysis revealed and quantified a strong relationship between galaxy number density and the completeness of detected filaments. Despite lower densities, the model rarely produced erroneous filaments, adding to its robustness. As a way to boost the galaxy number densities, we also explored using photometric redshift. Photometric data can be measured in bulk for thousands of galaxies, but the distance estimates have significantly larger uncertainties than for spectroscopic redshift data. Our analysis revealed that mixing photometric and spectroscopic redshift data improved filament detection. This approach proves valuable when spectroscopic data are scarce. The main results of the thesis are the quantitative characterization of the model to help interpret the results and a method to boost filament detection. This will pave the way for a more comprehensive understanding of the cosmic web and galaxy evolution.
Kirjeldus
Väitekirja elektrooniline versioon ei sisalda publikatsioone
Märksõnad
universe, evolution of galaxies, large-scale structures, research methods, mathematical models