My project was centered on measuring the expansion rate of the universe from independent, precise cosmological probes. For this purpose, I developed software for both the analysis of lensed supernova data and its discovery with wide-field surveys like the Zwicky Transient Facility. I also led (as principal investigator), photometric and spectroscopic programs for classification and confirmation of targets and detailed follow-up of lensed SNe. The work led to the discovery of SN Zwicky (Goobar et al. 2023, incl. SD) and the follow-up with the Liverpool telescope, which was crucial to infer the extinction properties and the lightcurve morphology in the redder filters (i and z band) for an intermediate redshift supernova. I led a successful JWST program to follow-up SN Zwicky in the nebular phase, to study the late-time properties of a Type Ia supernova (SNe Ia). The data have been obtained and are also a key part of modelling the stellar light of the lens using near infrared data. I am co-leading a project within the Dark Energy Science Collaboration of Rubin-LSST to determine, using a realistic survey strategy, the number and lightcurve characteristics of lensed SNe expected to be detected by LSST (Arendse, SD et al. internal review). I also worked on a hierarchical Bayesian model to combine information both from the time-delays and the lensing magnification of lensed SNe~Ia to forecast that LSST will be able to measure H0 to ~ 1.5% precision in the maximally conservative case (Birrer, SD, Shajib, 2022).
Independent of lensed SNe, I have developed new methods for using the local distance scale for measuring the Hubble constant. I led the first data release of Type Ia supernovae from the Zwicky Transient Facility (Dhawan et al. 2022a). I used a uniform distance ladder, with SNe Ia in both the calibrator and the Hubble flow samples, i.e. both the more nearby sample for which we can obtain independent distance constraints and the more distant sample that is sensitive to the expansion of the universe, observed with the same instrument and survey. This is a crucial step to reducing systematics from poorly understood selection effects and minimising the error from photometric cross-calibration between various different telescopes as has been done in the literature. Calibrating to the novel tip of the red giant branch method, I demonstrated that the single survey distance ladder yields an H0 value of 76.9 +/- 6.4 km/s/Mpc (Dhawan et al. 2022b). I also used a new approach to model the optical to the near infrared data of the Type Ia supernovae simultaneously with a hierarchical Bayesian lightcurve inference method, i.e. BayeSN and found an improvement of up to 15% in the uncertainty on H0 compared to the case with only the optical (Dhawan et al. 2023a). With my student, using a Bayesian model selection framework, we find a slight preference for using a student-t instead of a Gaussian distribution for modelling the SN Ia likelihood and it leads to a ~ 8% reduction in the uncertainty on H0 (Lovick, Dhawan, Handley in prep.).
A key new aspect in the search of new cosmology beyond the standard model is testing whether the expansion rate is the same in every direction. I used a model independent approach for the first time to constrain the quadrupole in the Hubble parameter (Cowell, SD, Macpherson 2023), finding that while isotropy is still favoured, anisotropies of upto ~ 10% in the Hubble parameter are still permitted by the data (SD et al. 2023).
Results from this project were disseminated via publications in peer reviewed journals and datasets were made public via the journal platforms and github. I presented the work via invited talks at several institutes (e.g. Duke, Pittsburgh, Oxford, LMU-Munich) and at the European Astronomical Society meeting. Moreover, I have also actively taken outreach roles like presenting talks for the public at the Institute of Astronomy and Lucy Cavendish College.
