Unruh, Sandra: Weak lensing magnification & baryon acoustic oscillations in galaxy-galaxy lensing. - Bonn, 2021. - Dissertation, Rheinische Friedrich-Wilhelms-Universität Bonn.

Online-Ausgabe in bonndoc: https://nbn-resolving.org/urn:nbn:de:hbz:5-61772

Online-Ausgabe in bonndoc: https://nbn-resolving.org/urn:nbn:de:hbz:5-61772

@phdthesis{handle:20.500.11811/9296,

urn: https://nbn-resolving.org/urn:nbn:de:hbz:5-61772,

author = {{Sandra Unruh}},

title = {Weak lensing magnification & baryon acoustic oscillations in galaxy-galaxy lensing},

school = {Rheinische Friedrich-Wilhelms-Universität Bonn},

year = 2021,

month = sep,

note = {Cosmology is the science that aims to explain the Universe in its entirety. While the standard cosmological model has had tremendous success in explaining independent astrophysical observations, we still lack understanding of the very nature of its main constituents, namely dark matter and dark energy. To better understand their origin, we can map matter structures throughout the Universe and their evolution using the distortion of light rays as they travel through the inhomogeneous Universe. This approach is called weak gravitational lensing, and this thesis focusses on galaxy-galaxy lensing (GGL), which directly reveals the relation of the visible ‘normal’ matter to underlying dark matter structures. Typically, GGL is measured in terms of tangential shear, i.e., the distortion of the observed galaxy shapes with respect to foreground galaxy positions.

Shear estimates from weak lensing surveys will soon enable us to determine cosmological parameters with sub-percent accuracy. The necessary analyses require excellent control over detector systematics, a sound theoretical model, and capable numerical tools. Therefore, we first developed an open-source numerical tool to extract GGL signals efficiently and then used it to pursue the following science cases.

Weak lensing magnification describes the change of a galaxy’s observed flux. It consequently changes the observed number density of galaxies on the sky, which in return affects the observed tangential shear. In this thesis, we provide leading-order analytical descriptions for the magnification effects. Further, we present numerical methods to select samples of foreground (lens) and background (source) galaxies that are unbiased by magnification. Currently, the combination of the three surveys KiDS+VIKING+GAMA provides one of the best constraints on cosmological parameters. We analysed the impact which neglecting magnification effects has on such a survey and find that, for lens galaxies at redshift z_d = 0.36 and source galaxies with mean redshift z_s = 0.79, the shear profile is changed by 2% and the mass of the lens is biased by 8%. We conclude, magnification effects by source and lens galaxies must be carefully taken into account even for ongoing surveys, while the statistical power of future weak lensing surveys certainly warrants correction for this effect.

The shear-ratio test (SRT) is a null-test that probes for systematics in galaxy shape and redshift estimates simultaneously. It is a purely geometrical probe that relies on shear and distance measurements for one foreground and two background galaxy populations. In this thesis, we show that the test is heavily biased if weak lensing magnification is not accounted for. The bias is stronger for increasing redshift of lenses and therefore, affects future surveys more severely. Using simulations, we find that an SRT with flux-limited lens galaxies at redshift z_d = 0.8 deviates up to 9sigma from zero. To retain the useful properties of the SRT, we provide a mitigation strategy that solely relies on already present observational data. The mitigation reduces the bias by a factor of ~100 and, at the same time, reduces the total uncertainties. This results in a deviation of typically <1sigma.

Lastly, we explore the influence of baryon acoustic oscillations (BAO) in the GGL signal. BAO are frozen-in density fluctuations in the large-scale structure that were generated by sound waves in the early Universe. The detection of the BAO signal as a function of redshift is an excellent probe for the time evolution of dark energy. Upcoming surveys will enable us to constrain the BAO signal from GGL measurements for the first time. In this thesis, the BAO signal is first modelled analytically. Then, we aimed to compare the model to the signal estimated from weak lensing simulations. However, various problems with the weak lensing simulations were discovered that prohibit detecting BAO. Nonetheless, the analysis pipeline has been set-up successfully and improved data catalogues can be analysed on the time-scale of an hour.},

url = {https://hdl.handle.net/20.500.11811/9296}

}

urn: https://nbn-resolving.org/urn:nbn:de:hbz:5-61772,

author = {{Sandra Unruh}},

title = {Weak lensing magnification & baryon acoustic oscillations in galaxy-galaxy lensing},

school = {Rheinische Friedrich-Wilhelms-Universität Bonn},

year = 2021,

month = sep,

note = {Cosmology is the science that aims to explain the Universe in its entirety. While the standard cosmological model has had tremendous success in explaining independent astrophysical observations, we still lack understanding of the very nature of its main constituents, namely dark matter and dark energy. To better understand their origin, we can map matter structures throughout the Universe and their evolution using the distortion of light rays as they travel through the inhomogeneous Universe. This approach is called weak gravitational lensing, and this thesis focusses on galaxy-galaxy lensing (GGL), which directly reveals the relation of the visible ‘normal’ matter to underlying dark matter structures. Typically, GGL is measured in terms of tangential shear, i.e., the distortion of the observed galaxy shapes with respect to foreground galaxy positions.

Shear estimates from weak lensing surveys will soon enable us to determine cosmological parameters with sub-percent accuracy. The necessary analyses require excellent control over detector systematics, a sound theoretical model, and capable numerical tools. Therefore, we first developed an open-source numerical tool to extract GGL signals efficiently and then used it to pursue the following science cases.

Weak lensing magnification describes the change of a galaxy’s observed flux. It consequently changes the observed number density of galaxies on the sky, which in return affects the observed tangential shear. In this thesis, we provide leading-order analytical descriptions for the magnification effects. Further, we present numerical methods to select samples of foreground (lens) and background (source) galaxies that are unbiased by magnification. Currently, the combination of the three surveys KiDS+VIKING+GAMA provides one of the best constraints on cosmological parameters. We analysed the impact which neglecting magnification effects has on such a survey and find that, for lens galaxies at redshift z_d = 0.36 and source galaxies with mean redshift z_s = 0.79, the shear profile is changed by 2% and the mass of the lens is biased by 8%. We conclude, magnification effects by source and lens galaxies must be carefully taken into account even for ongoing surveys, while the statistical power of future weak lensing surveys certainly warrants correction for this effect.

The shear-ratio test (SRT) is a null-test that probes for systematics in galaxy shape and redshift estimates simultaneously. It is a purely geometrical probe that relies on shear and distance measurements for one foreground and two background galaxy populations. In this thesis, we show that the test is heavily biased if weak lensing magnification is not accounted for. The bias is stronger for increasing redshift of lenses and therefore, affects future surveys more severely. Using simulations, we find that an SRT with flux-limited lens galaxies at redshift z_d = 0.8 deviates up to 9sigma from zero. To retain the useful properties of the SRT, we provide a mitigation strategy that solely relies on already present observational data. The mitigation reduces the bias by a factor of ~100 and, at the same time, reduces the total uncertainties. This results in a deviation of typically <1sigma.

Lastly, we explore the influence of baryon acoustic oscillations (BAO) in the GGL signal. BAO are frozen-in density fluctuations in the large-scale structure that were generated by sound waves in the early Universe. The detection of the BAO signal as a function of redshift is an excellent probe for the time evolution of dark energy. Upcoming surveys will enable us to constrain the BAO signal from GGL measurements for the first time. In this thesis, the BAO signal is first modelled analytically. Then, we aimed to compare the model to the signal estimated from weak lensing simulations. However, various problems with the weak lensing simulations were discovered that prohibit detecting BAO. Nonetheless, the analysis pipeline has been set-up successfully and improved data catalogues can be analysed on the time-scale of an hour.},

url = {https://hdl.handle.net/20.500.11811/9296}

}