Persieds 2013

[BillW]

One more spectrum graph....

This is of the fireball image. Areas "A" and "B" are regions of atmospheric lines. I've only marked the prominent lines.

Attachments:

[stewartw]

Hi Bill,

Many thanks for posting these and apologies for my delay in replying - am getting back on the air after replacing my laptop HDD.

I'm still trying to get my head around what I'm looking at ... you've pointed out that the spectra contains a mixture of lines (atmospheric and from the meteoroid itself) - am I right in assuming that the spectral plot is taken from the pixel values along the brightest part of the the image? Or are these often saturated (pixel value = 255) and hence you take it from a fainter region?

How do you differentiate between atmospheric and meteoroid lines?

Are you still going to IMC and if so, are you presenting?

Best regards

William

[BillW]

Hi,
It took me days to get my own head around what the lines were.
When you don't have the image of the meteor itself, the "zero order" image it is very difficult.

What I ended up doing was montaging some images to get a starting point. The other issue is measuring the actual dispersion. This is very easy in principle (using a light source with known lines) but not so easy in practice due to de focus and tilt. The low resolution also hinders things as there are several nm/ pixel so some fainter lines are just lost to bright lines. The 2nd order has twice the dispersion which is why some of the fainter lines are now discernible. It is just the fall of the meteors and what you catch. It gets worse if you go into the 3rd order as some of the blue lines are overlapped by the red lines in the 2nd order.


The atmospheric lines are well known. Its just a case of eliminating what you know and working from a laboratory list fitting what you don't.

It's all a trade off. A grating has almost linear dispersion (easier to measure) but generates multiple versions of the spectrum. A prism generates a single spectrum but has very non linear dispersion so measuring is more difficult, nothing is easy!

The graphs are just measurements of the intensity, pixel by pixel, along the spread of the spectrum. The 8 bit frame grabbers are very limiting. As I said on another forum, "it needs to be bright so we can get a good signal but it needs to be not too bright so it doesn't saturate, we're never happy!"
The axis are probably not helping as it's just an auto scaled form the image slice but the wavelength goes from ~350nm to ~850nm TWICE at different pixel scales.

I've only done this through the brightest parts at the moment but of course it can be done at sections through the image. This shows how the lines evolve of times during the meteors life.

Yes I'll be there, looking forward to it immensely and yes I'm presenting a short talk but it's not on spectra. That'll be for the next time (maybe!)


These are just demostrations as full photometric reduction to get line intensity ratios is currently beyond me at the moment. I'm working on it....

cheers,
Bill.

PS. The conference schedule has now been finalised and is on the IMO website in the conference section.