Validating the Dark Matter Origin of a LDMX Signal with Direct Detection Experiments
Ladda ner
Publicerad
Författare
Typ
Examensarbete för masterexamen
Master's Thesis
Master's Thesis
Program
Modellbyggare
Tidskriftstitel
ISSN
Volymtitel
Utgivare
Sammanfattning
The search for a particle candidate explaining the origin of dark matter is one of the central
goals in modern astro-particle physics. Many experiments based on different measurement
strategies are being built in an effort to detect such a candidate. In the event that a dark
matter signal is seen, it is crucial to verify the signal with other experiments. In this thesis
we explore the possibility of using a direct detection experiment to validate a hypothetical
dark matter signal seen at the Light Dark Matter eXperiment (LDMX). We consider a rep resentative light dark matter model with masses ranging between ∼ 1 − 1000 MeV, which
are assumed to be mediated by a massive dark photon. Hypothetical dark matter signals
are simulated, both for an LDMX-type experiment and for a semiconductor based direct
detection experiment. These signals are generated for varying values of the “true" dark
matter mass mχ,true and dark photon kinetic mixing εtrue parameters, which are chosen
under thermal relic abundance constraints. Validation is achieved by using the MultiNest
Monte Carlo algorithm to estimate the marginal posterior of the mχ and ε parameters
given the hypothetical direct detection signal. The parameter estimate is input into the
LDMX-type simulation software and compared with that of the hypothetical LDMX re sult. The chi-square hypothesis test is used to conclude whether the two distributions are
consistent with a single distribution function. A conclusion is drawn regarding the amount
of exposure required for a direct detection experiment to validate a hypothetical LDMX
signal for some choices of mχ,true. For instance, when mχ,true = 10 MeV the threshold
exposure for validation was determined to be 0.05 kg-year.
Beskrivning
Ämne/nyckelord
dark matter, light dark matter, LDMX, direct detection, exclusion limit, experiment validation, Bayesian inference, Madgraph, DarkELF, MultiNest