Faint Object Extraction Issues with Spextool

[Nathaniel Ross]

Short version:
I have had success in extracting standard stars following the Triplespec Spextool Point Source 'Cookbook' PDF, but have encountered endless problems in extracting very faint sources. Does anyone know of a faint source 'cookbook'?

Long Version:
I've been following the PDF 'Cookbook' for extracting my standard stars and everything works well in Spextool in the 'A' beam mode for bright sources. I have then been trying to extract my very-faint-object spectra in the 'A-B' mode and have encountered a myriad of crashes.

Strategy number 1:
In 'Point Source' tab, select reduction mode 'A-B'
Using fixed aperture locations and 'optimal' extraction, Spextool sometimes (but not always!?) crashes with the following message:

% Attempt to subscript DISP with Z is out of range.
% Execution halted at: XSPEXTOOL_WRITESPEC 5060 /data/Spextool3/Pro/xspextool.pro
%                      XSPEXTOOL_PSEXTRACTSPEC 2863 /data/Spextool3/Pro/xspextool.pro
%                      XSPEXTOOL_EVENT  2057 /data/Spextool3/Pro/xspextool.pro
%                      WIDGET_PROCESS_EVENTS
%                      $MAIN$         

Strategy number 2:
Attempt to combine A-B image in the 'Combine Images' tab. I select 'A-B' beam switching mode. I specify the images and then click 'Combine Images' which crashes with the following message:

% TOTAL: For input argument MASK, Dimension must be 1 or 2.
% Execution halted at: XSPEXTOOL_COMBINEIMGS 2548 /data/Spextool3/Pro/xspextool.pro
%                      XSPEXTOOL_EVENT  1924 /data/Spextool3/Pro/xspextool.pro
%                      WIDGET_PROCESS_EVENTS
%                      $MAIN$         

Strategy number 3:
In the 'Combine Images' tab, coadd each ABBA dither quad, then in the 'Point Source' tab, extract spectra in fixed apertures, with 'Optimal' extraction mode selected.
This usually runs to completion, although, it may fail to extract over large swaths of wavelengths for which it spits out warnings in the following format:

Warning:  Optimal extraction failed at wavelength 1.2053233.
Warning:  Optimal extraction failed at wavelength 1.2054674.
Warning:  Optimal extraction failed at wavelength 1.2056116.
Warning:  Optimal extraction failed at wavelength 1.2057558.
Warning:  Optimal extraction failed at wavelength 1.2058999.
Warning:  Optimal extraction failed at wavelength 1.2060441.
etc.

Strategy number 4:
In 'Combine Images' tab, coadd ALL images, then in 'Point Source' tab, extract spectra in defined apertures with 'Optimal' extraction mode. This successfully extracts a spectrum, but may not be best practices. It is not, for example, what was done by Belli et al. (http://adsabs.harvard.edu/abs/2013arXiv1302.3614B),
my reading of which suggests following strategy number 1 or 2 above.

As a side note, what statistic do people prefer for combining the images and/or spectra (there are a lot of options in the GUI): Robust Weighted Mean, Robust Mean, Mean, Median, etc.?

[Kelle Cruz]

I've never encountered this problem, but gonna be honest, this is exactly why I integrate on faint sources longer than recommended. (up to 10 mins.) I'd rather have sky lines to deal with rather than not be able to extract my spectrum!

kelle

--
You received this message because you are subscribed to the Google Groups "P200 TripleSpec Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to p200-triplespec-...@googlegroups.com.
To post to this group, send email to p200-tripl...@googlegroups.com.
Visit this group at http://groups.google.com/group/p200-triplespec-users.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

--
Kelle Cruz, PhD — http://kellecruz.com/

917.725.1334 — Hunter: x16486 — AMNH: x3404

[Babar Ali]

Hello Nathaniel:

I'll add the Strategy I know about, in case you have not tried it yet.

* Use bright source data taken as close in time and configuration (important) to your faint source as possible.
* Trace the apertures on the bright source and save that aperture trace.
* On the faint source, don't ask it to trace the aperture but load the traced apertures from your bright source.

The rest should be the same.
Stick with the default setting: Robust weighted mean.

cheers,
Babar

[Nathaniel Ross]

I did try this. Every time I specify the trace file to import from and then click "Find/Store Ap Positions", it fails to import and prints a message to the command line: "Fix this."

Nate

> To unsubscribe from this group and stop receiving emails from it, send an email to p200-triplespec-users+unsub...@googlegroups.com.

[Michael Cushing]

On Jun 25, 2013, at 8:34 PM, Nathaniel Ross <ros...@gmail.com> wrote:

> I did try this. Every time I specify the trace file to import from and then click "Find/Store Ap Positions", it fails to import and prints a message to the command line: "Fix this."

:). Yeah this has to do with the fact that with TSpec I have to do the tracing slightly differently. If you do the median combine for the image, you should see the object and be able to trace it (and not have to store the standard position).

-Mike

> > To unsubscribe from this group and stop receiving emails from it, send an email to p200-triplespec-...@googlegroups.com.

> > To post to this group, send email to p200-tripl...@googlegroups.com.
> > Visit this group at http://groups.google.com/group/p200-triplespec-users.
> > For more options, visit https://groups.google.com/groups/opt_out.
> >
> >
>
>

> --
> You received this message because you are subscribed to the Google Groups "P200 TripleSpec Users" group.

> To unsubscribe from this group and stop receiving emails from it, send an email to p200-triplespec-...@googlegroups.com.

[Nathaniel Ross]

Hi Mike,

A couple more crashes I have encountered:

In the combine images tab, with A-B beam switching mode selected and a median combine, I specify two images (an A-B pair) and click 'Combine Images'. I get the following crash:

% TOTAL: For input argument MASK, Dimension must be 1 or 2.
% Execution halted at: XSPEXTOOL_COMBINEIMGS 2548 /data/Spextool3/Pro/xspextool.pro
%                      XSPEXTOOL_EVENT  1924 /data/Spextool3/Pro/xspextool.pro
%                      WIDGET_PROCESS_EVENTS
%                      $MAIN$         

This does not crash when I select 'ABBA to ABAB' and then specify an ABBA quad.

Sometimes (but not always) when I click the 'Scale Lines' button in xtellcor, I get this crash:
% Attempt to subscript <PtrHeapVar20> with STDIDX is out of range.
% Execution halted at: XTELLCOR_GETSCALES  532 /data/Spextool3/Pro/xtellcor.pro
%                      XTELLCOR_EVENT    141 /data/Spextool3/Pro/xtellcor.pro
%                      WIDGET_PROCESS_EVENTS
%                      $MAIN$         

Nate