Recent ASCA spectra of Eta Car

100 ksec of SIS data were obtained on Jul 29 1996. The data were extracted by MFC and a 2 component RS fit was attempted, using enhanced N abundances similar to that of Tsuboi et al. (1995, PASJ 45, 35). A plot of the best fit using this model is here.

Details of the fit follow:

 Fit statistic in use is Chi-Squared
 Minimization technique is Lev-Marq
 Weighting method is standard
 Convergence criterion =   1.000000000000000E-002
 Querying enabled
 Prefit-renorming enabled
 Solar abundance table is angr

 Information for file   1
  belonging to plot group   1, data group   1
 telescope = ASCA , instrument = SIS0  channel type = PI
  Current data file: ascas0_ao5.pha
  Background file  :ascas0_ao5_bkg.pha
  No current correction
  Response (RMF) file    : ecs0b.rmf
  Auxiliary (ARF) file    : ascas0_ao5.arf
  XSPEC filter : none
  Noticed channels    17 to   278
  Spectral bins        1 to   262
  File integration time    8.1476E+04
     and effective area     1.000
  File observed count rate     1.216    +/-5.44224E-03 cts/s
  Model predicted rate :    1.215

  ---------------------------------------------------------------------------
  ---------------------------------------------------------------------------
  mo = wabs[4] (gaussian[1] + gaussian[2] + vraymond[3]) + wabs[6] (vraymond[5])
  Model Fit Model   Component     Parameter   Value
  par   par comp
    1    1    1       gaussian    LineE      2.44992     frozen
    2    2    1       gaussian    Sigma     0.000000E+00 frozen
    3    3    1       gaussian    norm      5.457980E-05 frozen
    4    4    2       gaussian    LineE      6.46177     frozen
    5    5    2       gaussian    Sigma     0.000000E+00 frozen
    6    6    2       gaussian    norm      1.534950E-04 frozen
    7    7    3       vraymond    kT(keV)   0.296819     +/- 0.00000E+00
    8    8    3       vraymond    He         1.00000     frozen
    9    9    3       vraymond    C          1.00000     frozen
   10   10    3       vraymond    N          50.9077     frozen
   11   11    3       vraymond    O          1.00000     frozen
   12   12    3       vraymond    Ne         1.00000     frozen
   13   13    3       vraymond    Mg        0.896769     frozen
   14   14    3       vraymond    Si         4.30533     frozen
   15   15    3       vraymond    S          2.03013     frozen
   16   16    3       vraymond    Ar         1.00000     frozen
   17   17    3       vraymond    Ca         1.00000     frozen
   18   18    3       vraymond    Fe        0.812778     +/- 0.00000E+00
   19   19    3       vraymond    Ni         1.00000     frozen
   20   20    3       vraymond    Redshift  0.000000E+00 frozen
   21   21    3       vraymond    norm      2.027430E-02 +/- 0.00000E+00
   22   22    4       wabs        nH 10^22  0.616438     frozen
   23   23    5       vraymond    kT(keV)    7.53189     frozen
   24   24    5       vraymond    He         1.00000     frozen
   25   25    5       vraymond    C          1.00000     frozen
   26   10    5       vraymond    N          50.9077     = par  10
   27   26    5       vraymond    O          1.00000     frozen
   28   27    5       vraymond    Ne         1.00000     frozen
   29   13    5       vraymond    Mg        0.896769     = par  13
   30   14    5       vraymond    Si         4.30533     = par  14
   31   15    5       vraymond    S          2.03013     = par  15
   32   28    5       vraymond    Ar         1.00000     frozen
   33   29    5       vraymond    Ca         1.00000     frozen
   34   18    5       vraymond    Fe        0.812778     = par  18
   35   30    5       vraymond    Ni         1.00000     frozen
   36   31    5       vraymond    Redshift  0.000000E+00 frozen
   37   32    5       vraymond    norm      7.030150E-02 +/- 0.00000E+00
   38   33    6       wabs        nH 10^22   3.98130     frozen
  ---------------------------------------------------------------------------
  ---------------------------------------------------------------------------
 Chi-Squared =      353.5     using   262 PHA bins.
 Reduced chi-squared =      1.370

Some points to note:

The N abundance derived here is 50, while Tsuboi et al. derived an overabundance of about 100. The actual N abundance is very sensitive to determination of low-energy (E~ 0.5 keV) background. Si & S are somewhat overabundant, while Fe is underabundant.
This model fits the PSPC data from 1992 & 1994, along with the ASCA PV data, but requires that the norm of the 7.53 keV component varies in each dataset. In addition, in order to fit the PSPC data, the temperature of the cooler component needs to decrease from 0.29 keV to 0.21.
We detect at high significance the presence of an Fe resonance line at 6.46 keV.

Comparison of the ASCA and Nearly-Simultaneous XTE spectrum

We also obtained an XTE PCA spectrum on Jul 31 1996, just 2 days after the ASCA observation. Bish Ishibashi has extracted the spectral data (using PCAs 0, 1 & 2) and used PCABACKEST 1.4c to generate an estimate of background. The background model includes the particle (q6) background, the activation background and an estimate of the cosmic background. Here's a plot comparing the best-fit ASCA spectrum (above) to the XTE PCA data.

Some points to note:

There's an obvious deficit at low energies (E< 5 keV). This is probably due to the presence of extended emission and unresolved sources which lie in the PCA field of view, and which are not well modeled by the standard cosmic X-ray background models available from the XTE GOF.
The spectrum above 7 keV is dominated by the 7.5 keV source from Eta Car. However,it appears that the q6+activation+cosmic model overestimates the background at E>7keV; as a result the predicted model falls above the observed data in this region.

Details of the data analysis are here

I also tried modeling the extracted data, using pcabackest 1.4 and the q6 and activation models only (i.e. I did not use the cosmic background model). A plot of the result is here. Details of the modeling are also available.

Some points to note:

Channel 1 is less than 0. (It's also less than 0 in the previous fit, but it's been ignored in that fit.) Thus the q6+qctiv model background overestimates the background contribution to the 1st channel.
If we ignore the cosmic background the net spectrum at E> 7 keV is now a PRETTY GOOD match to the ASCA model.

for more details, see

/rosat_stage/mfc/eta_car/COMPARE

on barnegat.

Back to the "X-ray Observations of Eta Car" main page.