Computational quantum field theory and global fits of effective dark matter models
File(s)
Author(s)
Bloor, Sanjay
Type
Thesis
Abstract
The search for physics beyond the Standard Model is necessarily a multi-disciplinary field.
By including all data relevant to a particle physics model simultaneously in a ‘global fit’, it is
possible to make statistically meaningful statements about the viability of theories beyond the
Standard Model. The topics of this thesis are extending the global fitting software framework
GAMBIT, and performing global fits of effective dark matter models.
Firstly, I present GUM, the GAMBIT Universal Model Machine, a tool that interfaces sym-
bolic Lagrangian-level tools and GAMBIT to allow one to implement new physics models in
GAMBIT with minimal effort. I perform a fit of a simplified dark matter model using GUM and
GAMBIT.
Next, I present CosmoBit, the new GAMBIT module for cosmological observables and likeli-
hoods. I present an application of CosmoBit in which I perform a global analysis to place limits
on the lightest neutrino mass by consistently combining cosmological and terrestrial datasets.
I then consider global fits of effective dark matter models using GAMBIT. I consider models
in which the Standard Model is extended by either a fermionic or vector dark matter candidate
that interacts via the ‘Higgs portal’. I present comprehensive results in both frequentist and
Bayesian frameworks, combining constraints from direct detection, indirect detection from γ
rays and neutrinos, the invisible width of the Higgs, and the relic abundance of dark matter,
whilst ensuring that the effective model description does not break down.
Finally, I perform global fits of dark matter effective field theories defined at the partonic
level, in which a dark matter candidate interacts with quarks and gluons via effective contact
interactions. I consider cases in which the effective theory is generated by integrating out either
a scalar mediator, a vector mediator, or a heavy quark. In these fits, I combine constraints
from direct detection, indirect detection from γ rays, monojet searches for dark matter particles
from the LHC, and the relic abundance of dark matter.
By including all data relevant to a particle physics model simultaneously in a ‘global fit’, it is
possible to make statistically meaningful statements about the viability of theories beyond the
Standard Model. The topics of this thesis are extending the global fitting software framework
GAMBIT, and performing global fits of effective dark matter models.
Firstly, I present GUM, the GAMBIT Universal Model Machine, a tool that interfaces sym-
bolic Lagrangian-level tools and GAMBIT to allow one to implement new physics models in
GAMBIT with minimal effort. I perform a fit of a simplified dark matter model using GUM and
GAMBIT.
Next, I present CosmoBit, the new GAMBIT module for cosmological observables and likeli-
hoods. I present an application of CosmoBit in which I perform a global analysis to place limits
on the lightest neutrino mass by consistently combining cosmological and terrestrial datasets.
I then consider global fits of effective dark matter models using GAMBIT. I consider models
in which the Standard Model is extended by either a fermionic or vector dark matter candidate
that interacts via the ‘Higgs portal’. I present comprehensive results in both frequentist and
Bayesian frameworks, combining constraints from direct detection, indirect detection from γ
rays and neutrinos, the invisible width of the Higgs, and the relic abundance of dark matter,
whilst ensuring that the effective model description does not break down.
Finally, I perform global fits of dark matter effective field theories defined at the partonic
level, in which a dark matter candidate interacts with quarks and gluons via effective contact
interactions. I consider cases in which the effective theory is generated by integrating out either
a scalar mediator, a vector mediator, or a heavy quark. In these fits, I combine constraints
from direct detection, indirect detection from γ rays, monojet searches for dark matter particles
from the LHC, and the relic abundance of dark matter.
Version
Open Access
Date Issued
2020-09
Date Awarded
2021-04
Copyright Statement
Creative Commons Attribution NonCommercial Licence
License URL
Advisor
Scott, Patrick
Trotta, Roberto
Sponsor
Science and Technology Facilities Council (Great Britain)
Publisher Department
Physics
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)