IXO/CAT: Simulation: SN 1987A

[ back to IXO/CAT: Scientific Simulations page ]

Simulations of SN 1987A: shocked plasma velocities

The ability to accurately measure the ion Doppler velocities due to thermal and turbulent motions is a new capability provided by the IXO/CAT grating spectrometer.

(D. Dewey Nov 2008 dd@space.mit.edu)

The X-ray emission from SN 1987A is due to shocked material with a range of densities and shock velocities; modeled with two vpshock components it is a good representative of similar emission from other SNe and GRBs.

Because SN 1987A has a spatial extent of ~ 2 arc seconds it will appear to IXO essentially as an unresolved point souce. Hence it can be observed with the IXO/CAT at the gratings' nominal resolving power.

The simulations below use the Area = 3000 cm^2 and Resolving_power = 3000 Lorenzian resonse models of November 2008.

Simulation results -- overall spectrum

The 3-panel plots below show the spectra from a 50 ks observation of SN 1987A over the energy range most relevant to the IXO/CAT and SN 1987A: from 0.35 keV (just below the C Ly-alpha line) to 1.25 keV (just above the Ne Ly-beta line). The IXO X-ray microcalorimeter system (XMS) spectrum is shown in blue - these counts are collected along with the CAT grating spectrum, shown in black. For reference, the gray line shows the XMS response if no grating were included in the design.

The CAT spectrum, black, is the sum of several CAT dispersed orders, 1st through 5th, which are shown as well by the colored spectra: red, orange, green, light blue, and purple-ish, respectively.

The two simulations here show the SN 1987A spectrum as it would be seen without any velocity effects (at left) and when realistic velocity broadening based on current observations is included, see below for more details.

SN 1987A -- model with very narrow lines :
SN 1987A -- with as-expected line broadening :

Simulation results -- velocity broadening

The plots below are closeups of the O VIII (0.65 keV, ~19 A) region of the simulations for three values of the Gaussian velocity broadening. The FWHM velocity broadening increases from left to right, specifically:

  • left - 23 km/s FWHM - Essentially no broadening; this shows a comparison of the "raw" XMS and CAT resolving powers.
  • middle - 352 km/s FWHM - This approximates the broadening due to the radial motion ("v_ring" ~ 370 km/s; see Dewey et al. 2008) of the shocked emitting material from the whole SN 1987A.
  • right - 587 km/s FWHM - This is the broadening when ion thermal and turbulent motions of ~ 470 km/s FWHM are included along with the radial motion.

As these plots show, the IXO/CAT is very sensitive to the ion turbulent/thermal velocity: note the very clear changes in the CAT (black) observed line-width as the velocity increases. In contrast, the XMS response shape is only slightly changed and extracting a measure of the underlying line broadening will be very sensitive to the XMS response function calibration and variations.

The broadening values used here in the simulations are only very roughly determined by current HETG observations (Dewey et al. 2008): the ability to accurately measure these ion Doppler velocities due to thermal and turbulent motions is a new capability provided by the IXO/CAT grating spectrometer.

v_fwhm = 23 km/s :
v_fwhm = 352 km/s :
v_fwhm = 587 km/s :

Simulation script

The simulations and plots shown here were made using isis with the script: sn87a_ixo_sim.sl along with the utilities and responses downloaded from the IXO/CAT software page. The nominal SN 1987A spectral model corresponds to the Fall 2007 emission (during LETG observations; Zhekov et al. 2009); it is the sum of two XSPEC vpshock models using NEI APEC database files from Borkowski. Specifically, the simulation here reads in the isis par file: sn87a_MishMash_LEG07_NoNHrest.par to set the model parameters; note that NH and the redshifts are 0 in this file and are set appropriately in the simulation script. Finally, the velocity blur is included using the XSPEC gsmooth convolution model, e.g., setting this in isis via:

fit_fun("wabs(1)*gsmooth(1, vpshock(1)+vpshock(2))");
v_fwhm = 587.0; % km/s FWHM
set_par("gsmooth(1).Sig@6keV",6.0*(v_fwhm/2.35)/(Const_c*1.e-5));
set_par("gsmooth(1).Index",1.0);

dd @ space.mit.edu