On the RL/CEL discrepancy

Grazyna Stasinska

Grazyna Stasinska (LUTH, Observatoire de Paris, Meudon, France)

The discrepancy between abundances derived from collisionally excited lines (CEL) and recombination lines (RL) in  H II regions and planetary nebulae is still not fully understood.

As I understand, the effective recombination rates for the lines have been computed for electron temperatures “typical” of photoionized nebulae, i.e. Te < 20 000 K. What if inside the nebulae there are zones of much higher temperatures? Such zones can be produced inside dusty filamentary nebulae, as shown by Stasinska & Szczerba (2001), or in zones of shocked gas inside the nebulae. Could there not be a process occurring only at temperatures higher that 20 000 K, that would enhance the production of recombination lines (something similar to dielectronic recombination at high temperatures) ?  Of course, CELs would also be affected by such high temperature zones, but differently and this could perhaps resolve the RL/CEL discrepancy.

I know that, presently, temperature diagnostics from recombination lines always indicate low temperatures. But those could be flawed as well, if the high temperature zones give a noticeable contribution.

This entry was posted in Atomic data applications and tagged , , , . Bookmark the permalink.

6 Responses to On the RL/CEL discrepancy

  1. wmuphysics says:

    Interesting idea! Are you planning to quantify the predictions of such a model?

  2. Hi Grazyna,

    I just read your comment on the discrepancy.
    The RL-CEL discrepany is due to atomic physics treatment as well. We
    have been studying oxygen, particularly O II. REL (same as RL) has been
    predicting higher oxygen abundance in nebular gas than CEL. The main reason
    is lower recombination rate at low temperature. Our recent calculations on
    photoionization show low energy near threshold resonances, forming due to
    relativistic effect but not allowed in LS coupling. This will increase the
    recombination rate at low temperature and thus will reduce the oxygen
    abundance in agreement with that by CEL. Please see “Low Energy Fine
    Structure Resonances in Photoionization of O~II”, Nahar et al. Phys. Rev.
    A 82, 065401 (2010). We have also written up a paper, under review, on
    the O III line ratios using relativistic collision strengths. The results
    are consistent with relativistic fine structure effects.

    With best regards,
    – Sultana

  3. grazyna stasinska says:

    Dear Sultana,
    I had not seen your paper. Very interesting indeed!

  4. RW says:

    Surely the only way to test the hypothesis that high temperature processes are affecting the abundances is to calculate the recombination rates at those higher temperatures. Are such calculations available, or likely to be carried out?

    I’m very interested to read about the low temperature effects. When you say they “will reduce the oxygen abundance in agreement with that by CEL”, I guess that refers to the typical factor of 2-3 discrepancy. Would it be fair to say, then, that a separate mechanism still needs to be found for the nebulae which show much higher discrepancies of x5 or more?

  5. grazyna stasinska says:

    hello RW ! (who are you, by the way?)

    I think that the pb of the CEL/RL discrepancy is not the same in HII regions (with their typical discrepancy factor of 2-3) and in planetary nebulae, where it can be much higher. Note that, in PN where the ADF is very large, the temperature from the Balmer jump is very low . There is no question about that, it is very clear on the spectra around the Balmer jump that the discontinuity is enormous. This goes well with the hypothesis of cold hydrogen-poor clumps (see a the paper by Henney & Stasinska 2010, where we discuss such a scenario in depth).

    of course, if recombination rates change at low temperature by a factor 2-3, the exact values of the inferred abundances in and outside clumps will change, if one does a detailed model, butthe nature of the problem in PNe with extremely large ADFs will not be changed.


  6. Roger Wesson says:

    Very belatedly returning to this discussion… my previous comment was not very clear thinking. The way to test the hypothesis would be to obtain spectra in which thermal broadening can be observed. Then you would see directly whether RLs were emitted, on average, by regions with higher, lower or similar temperatures to the CELs. If these high temperature effects exist and affect RLs more than CELs, then RLs should be broader.

    I know of some measurements like this for PNe, eg Barlow et al. 2006 who found RLs to be narrower than CELs in NGC 7009 and NGC 6153, and on the other hand Peimbert et al. 2004 who measured very similar line widths for both types of line in NGC 5315. I don’t know of any measurements which show RLs broader than CELs though.

    (I signed as “RW” before)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s