Follow-on Science Instrument

Contract NAS8-01129

Monthly Status Report Numbers 018

August 2003

Science Theme: Supernova Remnants II

Prepared in accordance with DR 972MA-002; DPD #972

Prepared for

National Aeronautics and Space Administration

Marshall Space Flight Center, Alabama 35812


Center for Space Research; Massachusetts Institute of Technology; Cambridge, MA 02139

Supernova Remnant Progress, Part II (June’02 – August’03)


Summary of Supernova Remnant Observations and Activities

In the last 14 months progress has continued in the analysis and understanding of our GTO observations of Supernova Remnants. A major accomplishment was the recent submittal to Astrophysical Journal of a comprehensive paper on E0102: “Chandra High-Resolution X-Ray Spectrum of Supernova Remnant 1E0102.2-7219” by Flanagan et al. A contribution to the High Resolution Spectroscopy workshop, “Extended Source Analysis for Grating Spectrometers” by D. Dewey (available online), summarized some techniques and results for HETG SNR analysis. The determination of masses and abundances was a focus this period and has lead to an understanding of the importance of spectral model assumptions and geometry assumptions in these calculations. The Table below lists our SNR GTO observations and for reference the other 5 known “Oxygen-rich” remnants as well; these 5 have all been observed with ACIS and no grating. Specific progress is described in the following pages.





















1045, 2410,




LMC, 50




Strong Si and Fe lines.

Type 1a or Core collapse?









Oxygen-rich SNRs w/HETGS:




Cas A

Gal., 3.4






121, 1828



LMC, 50




Bright Filaments, “O blob”



120, 968





SMC, 60




Weak Fe lines

90 deg roll from Cycle 1 obs.









Other Known O-rich SNRs:




Puppis A

Gal., 2.0




~ 5,000






Gal., 4.8



< 1,600






LMC, 50



~ 1,700

PWN; 8” dia. O III ring.





SMC, 60




Older version of E0102 ?

Park et al. Astro-ph/0309271




NGC 4449



< 0.6

~ 100

Patnaude and Fesen, ApJ 2003



We have devoted quite a bit of energy to analysing and understanding SNR E0102 during the past year or so. During this time a comprehensive paper was completed and submitted to ApJ and includes a range of our initial results and more recently estimates of the mass of the Oxygen and Neon in the ejecta and a discussion of possible mechanisms to produce an axial/cylinderical symmetry of the remnant. Look for Flanagan et al 2003 in ApJ very soon!


Research extending beyond the paper has been on-going and some items are described here below.



A second E0102 HETGS observation, obsid 3828, was taken in December 2002 at a roll 90 degress from the earlier data; the MEG m=-1 order image is shown above. This 135 ks exposure has equal or greater counts than the earlier obsid 120 data (87 ks) and simultaneous analysis of them is planned especially for the Oxygen lines (red above.)


Count rate ratios of these Oxygen lines do show the effect of the ACIS contamination layer and their quantitative values provide support for the External Cal Source measured absorption model – indicating that there is some small additional absorption not currently in the edges-only derived model.


A first look spatial-spectral analysis of these new E0102 data give a spatial-velocity image for the Ne X line at right very similar to that derived from obsid 120; this inspite of the 90 degree rotation is a good confirmation of the reality of the result.



The HETGS data for E0102 can be roughly fit by a single, global vnpshock model, above, which allows relative abundances of the elements to be determined under this model assumption. In the plot at right, these abundances (red) are compared with expectations from SNR models, specifically Core-collapse models (purple) and Type Ia models (green); for reference SMC/ISM abundances are also shown (blue.) As expected E0102 is unambiguously a core-collapse remnant from a progenitor of ~30 solar masses.


Using Chandra’s high spatial resolution we have started to study localized regions of the E0102 ejecta ring, specifically the Northern “shelf” and Southeast “arc” bright features. The plots below show contours in temperature-tau model space for Oxygen line ratios from these regions. Although not identical, the similarity of temp and tau for ejecta in these two regions is a little puzzling given other indications of overall temperature and tau differences between north and sourth parts of the remnant, e.g. the intensity of flux above 2keV shown at left. These results therefore add constraints on the overall 3D model of the ejecta-plus-blastwave that we hope to create, see “3D Modeling” section below. (Plots from Fredericks et al., poster at 201st AAS meeting.)



One item of note in N132D analysis is the overplotting of the grid of shock model predictions on the ratio-ratio plot shown above. The ratio of O VIII Lyman alpha to total O VII triplet (r+I+f) flux is plotted on the X-axis. The Y-axis gives the O VII forbidden to resonance ratio. Values of these ratios for a variety of features/filaments in N132D are plotted by red symbols. The overlayed grid shows the region spanned by the model for various T and tau values. The NE-corner region has an unexpectedly high f/r ratio. The raw data for NE-corner are shown at right with the model fit and decomposition into the r, i, and f contributions; the bottom panel shows the nominal feature profile and specific region of interest. (Plots from Fredericks et al., Poster at Con-X meeting, May 2003.)





We fit the global HETGS spectrum of N103B with a model, shown above left, that is the sum of high-T (green) and low-T (red) “vnpshock”s with variable abundances plus a small synchrotron “srcut” component (blue.) The total spectrum agrees well with our data.

The plot in the upper right shows the contribution of Fe emission (green) to a shock model (other elements’ emission shown in grey.) The “Iron forest” between 7A and 18A is a major contributor to the N103B model in that range making it hard to clearly identify Ne and Mg emission in that range.

Shown at right are the abundances from the 2T model (red) compared with Ia and core-collapse models (green, purple) and the LMC ISM abundance (blue dotted.) Our model abundance ratios are tantalizingly in between the Ia and core-collapse predictions; mass estimates are ongoing. (From M.S. Thesis by J. Migliazzo, 2003.)


3D Modeling


1:, 2:, 3:, 4:.


The red-green spatial-velocity image of derived for the Ne X line in E0102 suggests a cylinderical ejecta distribution, 1 above, viewed nearly on-axis. In order to qualitatively and quantitatively test this and related ideas on E0102’s structure we’ve started to explore the creation of 3D plasma models and their X-ray emission. As an example 2 above, shows the X-ray color-coded intensity image cylinderical ejecta surrounded by a spherical blastwave shell.

We’re also looking at available 3D software and how it could be used in 3D astrophysical modeling. For example, 3 above, shows a view from the modeling window of the Blender software in which a cylinder (green mesh), an observer (pink) and light source (yellow dot) have been defined. When this model is “rendered” the image shown in 4 above is generated. Ideally the same model definition software could be used as input to an “X-ray renderer” for X-ray modeling.


Supernova Remnant Plans, Part II


Our plans for the next year or so are given here. A general additional topic for all of our SNR research would be to include and understand obseravtions made at other wavelengths like radio, optical and IR.



·      Compare fluxes with models, e.g., power-law shock model values.

·      Investigate 3-D models for their agreement with the Ne X spatial-velocity results and the variation of ionization stage with position.

·      Create s/w to jointly analyze the Cycle 1 and Cycle 4 E0102 observations, especially for O VIII, O VII lines.



·      Try including an Fe "pseudo continuum" to better measure Ne and Mg line fluxes in the various regions.

·      Understand the enhanced O forbidden emission in the NE-corner region.

·      Find the abundance ratios for a global fit to a single vnpshock model, compare w/Ia,II yields.



·      Complete Paper 1 on N013B.

·      Carry out spatial-spectral analysis of Si lines by combining MEG and HEG data sets.


Cas A

·      For the Si feature combine grating line flux values with ACIS zeroth-order continuum values to get n_H/n_Si and feature masses.

·      Analyze additional features in Si and S for flux, temp, mass, and doppler velocities.

·      Find the abundance ratios for a global fit to a single vnpshock model, compare w/Ia,II yields.


Advanced Analysis

·      Investigate use of 3D s/w and tools for SNR modeling, specifically for E0102 modeling.

·      Work with ISIS s/w team to determine useful spatial-spectral s/w functions to be added to ISIS.