Built-in Transients

Common supernova types can be simulated easily using built-in settings. The supported types are Ia, Ib/c, IIn, and IIP. Based on the type, the volumetric rate and luminosity functions (as well as distributions of color and stretch for type Ia SNe) are set to realistic values. Each individual setting can be adjusted to the user’s liking, see Custom Transients for details.

The minimum information required to create a TransientGenerator for a built-in transient type is the redshift range, a range of RA and Dec values that will cover the full survey footprint, and a range of MJD values (or any other times in units of days) that covers the whole duration of the survey. The range of MJDs is the range for the t0 value of the sncosmo.Model used to simulate the lightcurve, i.e. typically the time of explosion or the time of peak, thus the range should be set such that transients that would only be observed while they are fading when the survey starts or while they are rising and the survey is ending.

To create the TransientGenerator use the following command:

tr = simsurvey.get_transient_generator((0.0, 0.05),
                                       transient='Ia',
                                       template='salt2',
                                       ra_range=(0,360),
                                       dec_range=(-30,90),
                                       mjd_range=(58178, 58543))

The resulting TransientGenerator will simulate type Ia supernovae at redshifts between 0 and 0.05 using the SALT2 template. The survey area covers the whole sky down to a declination of 30 degrees and the time scale is one year. The full list of options for transient and template, as well as their associated distributions and parameters are discussed below.

sncosmo Sources

For built-in transients the models will be created using sources that are built into sncosmo and will be downloaded at the first attempt of loading them.

transient	template	sncosmo Source	Source Class	Notes
Ia	salt2	salt2	SALT2Source
Ia	hsiao	hsiao	StretchSource
Ibc	nugent	nugent-sn1bc	TimeSeriesSource
Ibc	snana	All ‘snana’ sources of types Ib and Ic	MultiSource
IIn	nugent	nugent-sn2n	TimeSeriesSource	Limited to first 150 days to avoid interpolations errors
IIn	snana	All ‘snana’ sources of type IIn	MultiSource
IIP	nugent	nugent-sn2p	TimeSeriesSource	Limited to first 130 days to avoid interpolations errors
IIP	snana	All ‘snana’ sources of type IIP	MultiSource

For references to the SEDs used, see the sncosmo documentation.

Rates and Luminosity Functions

Based on the supernova type the volumetric rate is set and is then is used to determine the number of SNe to be simulated as well as their redshift distribution. Furthermore the absolute B-band magnitude at peak of each individual supernova is set to a value drawn from a Gaussian distribution centered on \(M_B\) with a standard deviation \(\sigma_M\). Lastly all models except SALT2 have had an extinction effect at the transient redshift added, which simulates extinction by dust in the host galaxy (SALT2 has its on color law for that). For this effect \(R_V\) is fixed to 2and \(E(B-V)\) is drawn from an exponential distribution with a scale of 0.11.

SN type	Rate [\(\textrm{Mpc}^{-3}\textrm{yr}^{-1}\)]	\(M_B\) (peak)	\(\sigma_M\)
Ia	\(3 \times 10^{-5} (1+z)\)	\(-19.3\)	\(0.1^\dagger\)
Ib/c	\(2.25 \times 10^{-5} (1+z)\)	\(-17.5\)	\(1.2\)
IIn	\(7.5 \times 10^{-6} (1+z)\)	\(-18.5\)	\(1.4^\ddagger\)
IIP	\(1.2 \times 10^{-4} (1+z)\)	\(-16.75\)	\(1\)

\(\dagger\) In addition to the intrinsic scatter of SN Ia peak magnitudes, the Tripp relations [25] were used to simulate a realistic population. In case of SALT2, the parameters \(x_1\) and \(c\) are drawn from Gaussians centered around 0 with standard deviations 1 and 0.1, respectively. From the peak magnitude the code then subtracts \(\alpha x_1 - \beta c\) with \(\alpha=0.13\) and \(\beta=3\). In case of the Hsiao template, the “stretch” parameter \(s\) is drawn from a Gaussian centered around 1 with a standard deviation of 0.1. From the peak magnitude the code then subtracts \(\alpha (s-1)\) with \(\alpha=1.3\).

\(\ddagger\) To avoid simulating a large number of unrealistically bright SNe IIn, the Gaussian distribution of peak magnitudes was truncated at \(1\sigma\) on the brighter-than-average side.