Skip to main content

Mission Overview

Roman Strongly Lensed Supernova Simulations


Primary Investigator: Justin Pierel

HLSP Authors: Justin Pierel, Steven Rodney, Giorgos Vernardos, Masamune Oguri, Richard Kessler, Timo Anguita

Released: 2022-06-29

Updated: 2022-06-29

Primary Reference(s): Pierel et al. 2021

DOI: 10.17909/t9-k8w7-zk32

Citations: See ADS Statistics

Read Me


Left has plot showing J band flux as a function of time of SN1a with macro-lensing. Upper right has same SN1a light curve but with microlensing from Pierel et al. 2021 applied. Lower right shows microlensing curve in magnitudes as a function of time.
Simulated, Lensed SNIa

Left panel: Simulated lensed SN Ia with macro-lensing parameters applied. Upper right panel: Simulated lensed SN from left panel, now with microlensing applied from Section 3.2 of Pierel et al. 2021. Lower right panel: The microlensing curve applied to image 1 (black), and image 2 (red), in magnitudes. In both light curve plots, the solid green curve denotes the underlying (un-lensed) model used to create the simulation before the application of microlensing.

Left is a scatter plot showing max. time delay as a function of max. separation between multiple images of the source in arseconds. Right is a scatter plot of the source redshift as a function of the lens redshift.
Rubin Highz Lens Catalog Compared To RSLSS Catalog

Comparison of properties between the Oguri & Marshall 2010 Vera C. Rubin Observatory LSST mock lens catalog and the Highz lens catalog created in this work for the Roman Space Telescope. We use 1000 random lenses from each catalog for comparison, though the actual projected yields for Rubin are much higher than for Roman (Goldstein et al. 2019; Huber et al. 2019). In both figures, Rubin data are shown in blue and Roman data are shown in red. (Left) The maximum time delay for each mock vs. the maximum separation between multiple images of the source in arcseconds. The grey dashed line represents the likely point spread function (PSF) full width at half maximum (FWHM) for Rubin; images are “resolved” when their separation is ≥ 2/3 θP SF . (Right) The source redshift vs. the lens redshift for each lensing system. The Rubin mock lens catalog uses a 23.3 mag (i-Band) detection threshold, while we use 26.2 mag (H-Band) for Roman (see Table 1 of Pierel et al. 2021).


Using anticipated characteristics of the Roman Space Telescope Type Ia supernova (SN Ia) survey, the team has constructed mock catalogs of expected resolved lensing systems and strongly lensed Type Ia and core-collapse (CC) SN light curves, including microlensing effects. The data products include mock catalogs for four Roman SN Ia survey variants and the High Latitude Survey (HLS), as well as 2.4 million simulated light curves (600k for each of 4 SN classes: Ia, Ibc, IIn, IIP) based on the Hounsell et al. 2018 "Allz" strategy. These products were used to predict that Roman will find ~11 lensed SNIa and ~20 CCSN for the fiducial SN Ia surveys, with a time delay precision of ~2 and 3 days (measured with the SNTD package:, respectively. The predicted parameters were then used to construct Fisher matrices, also included, and derive projected constraints on cosmological parameters for each SN Ia survey strategy.

Data Products

Simulated light curves and tables of fit parameters are grouped and named with the following convention:



  • <sn_type> is the type of supernova (e.g., Ia = "sn-ia", Ib/c = "sn-ibc", IIn = "sn-iin", IIP = "sn-iip").
  • <file_type> is the file type ("lcs" for light curves, "fitparameters" for fit parameters).
  • <ext> is the extension (".tar.gz" for light curves, ".fits" for fit parameters).

Mock catalogs of lensing systems are named with the following convention:



  • <strategy> is the Roman survey strategy, one of "allz", "highz", "lowz", "hls", or "ultradeep".

Fisher matrices constructed from the predicted parameters are named with the following convention:



  • <strategy> is the Roman survey strategy, one of "allz", "highz", "lowz", "hls", or "ultradeep".
  • <cosmology> is the assumed cosmology, either "flatwcdm" or "flatw0wacdm".

Data file types:

lcs.tar.gz Bundles of simulated light curves

Tables of fit information for the simulated light curves

mocks.txt Mock catalogs of lensing systems
fisher.csv Fisher matrices based on predicted parameters

Data Access

Simulated Lightcurves by Supernova Type
Supernova Type   Lightcurves File Size
SN Ia Fit Parameters Lightcurves 8.7 GB
SN Ib/c Fit Parameters Lightcurves 7.7 GB
SN IIn Fit Parameters Lightcurves 7.9 GB
SN IIP Fit Parameters Lightcurves 13 GB


Mock Catalogs and Fisher Matrices by Roman Survey Strategy
Strategy Mock Lens Catalogs Fisher Matrices 
Allz Mock Lens Catalogs

FlatwCDM Cosmology

Flatw0waCDM Cosmology

Highz Mock Lens Catalogs

FlatwCDM Cosmology

Flatw0waCDM Cosmology

Lowz Mock Lens Catalogs

FlatwCDM Cosmology

Flatw0waCDM Cosmology

Ultradeep Mock Lens Catalogs

FlatwCDM Cosmology

Flatw0waCDM Cosmology

HLS Mock Lens Catalogs


Please remember to cite the appropriate paper(s) below and the DOI if you use these data in a published work. 

Note: These HLSP data products are licensed for use under CC BY 4.0.