Although echomop was written with the UCLES Echelle Spectrograph in mind. It is in no way limited to processing data from this particular instrument, or even, to data from échelle spectrographs. The package can equally well be used on multi-object spectra obtained, for example, using popular fibre based instruments. If you are processing this type of data the following point needs to be remembered:
echomop is capable of reducing longslit spectra taken with a variety of instruments. In order to tune echomop for this task it will usually be necessary to increase the two parameters:
TUNE_MXSKYPIX
the maximum number of pixels in the spatial direction.
TUNE_PFLSSAMP
the number of sub-samples in spatial direction (e.g., 10 x TUNE_MXSKYPIX).
Although explicit support for extraction of extended objects is not provided in this release, it is possible to use echomop to extract an increment at a time. To do this the normal extraction should first be performed to obtain a reliable sky model. Then the ech_spatial task/ECHMENU Option 4 should be used to mask out all object increments except one. The extraction should then be re-done and the resulting spectrum copied into a section of a user-created data array using KAPPA facilities. This procedure needs repeating for each increment required and could be automated using a command file.
echomop has special facilities for particular detector dependent problems. In particular, for CCDs and similar detectors, the data may become obscured by cosmic-ray events over a long exposure. The package provides 2 different methods of identifying and flagging such pixels. CCDs also necessitate the provision of a number of extra pieces of information to the reduction process. Care must be taken to ensure these values are accurate as they can have a severe impact upon the resultant spectrum.
For IPCS detectors, frames can be highly distorted in both X- and Y-directions. echomop provides for a detailed 2-D polynomial fit to be made to the distortions present in each order.
echomop provides three major methods of cosmic-ray identification:
Iterations continue until the overall significance of all outliers in the order is below a user-supplied threshold. Each order is processed independently in two parts, first the sky pixels, and then the object pixels.
The significance of a fit is calculated by assuming a Gaussian distribution of deviations from expected intensities and using the Kolmogorov-Smirnov statistic (Ref. Numerical Recipes. Cambridge Press Sec. 13.5)
A post processor is used to counteract the tendency of cosmic-ray identifiers to be fooled by sky line pixels. This task examines the geometry of all ‘connected’ cosmic-ray pixels and restores any which it adjudges are actually due to bright sky emission lines.
provides this facility and should be the preferred method when multiple frames are available. The procedure calculates a median image and then rejects pixels which deviate by more than n sigma from the median value (note that this means that all the frames must be of equal exposure time). Bad pixels are flagged by placing values in the quality array of the frame concerned.
CCD detectors are subject to an error due to the final amplification immediately prior to the data leaving the chip. This is called readout noise and is due to uncertainty in the operation of the on-chip amplifier.
Recent developments have led to very significant improvements in this area (giving readout noise of less than 1 count per-pixel), but many older chips are still in use.
echomop looks for a value for readout noise specified in the data frame itself (for CCD frames). If no value can be found then you will be prompted for a value by the READOUT_NOISE parameter.
Some of the more modern CCDs use multiple on-chip amplifiers in order to provide a rapid readout facility. Each of these will have its own independent readout noise value. Currently echomop assumes that a single readout noise value is applicable for the whole frame so each sub-section of a multiple readout CCD image must be processed separately.
This release of echomop provides no special support for flux calibration of the extracted spectra. fiGARO provides extensive support for this procedure, including many standard star data tables.
You should note that due to the extremely high resolution of the spectra produced by UCLES, that there may not be sufficiently high resolution data available for the flux standards.
A common situation is that the fiGARO table for the standard contains (at most) one sample (flux) point per order, and that such points are often 40 Angstrom averages. In such cases it is difficult to ascribe much confidence to the resulting flux values.
A program of high resolution observations (of commonly used flux standards) to combat this problem is being planned.
A limited set of high resolution tables are provided. Type:
to see which ones. These tables may be used with the fiGARO flux calibration programs to flux calibrate spectra output from echomop.
If flux calibration is not being performed it is sometimes desirable to remove the ‘blaze’ function from the extracted spectrum to assist in fitting line profiles etc. during data analysis.
A task is provided for this purpose which operates by fitting polynomials to the flat-field orders. The fits may be automatically or interactively clipped and the resulting blaze spectrum is normalised such that its median intensity is unity. The normalised blaze is then divided into the extracted spectrum.
After a blaze function has been applied to the extracted order all its values may be reset to unity to ensure that the order(s) cannot be re-flattened in error. If the blaze is to be re-applied then the correct procedure is to first re-extract the order(s) concerned and then re-fit the blaze.
In some cases you may want to subtract the contribution of scattered light from the object instead of using the sky pixels (which implicitly also contain scattered light information). The single function task ech_mdlbck performs this scattered light background fitting and places the resulting fitted values in the ‘sky fit’ arrays used by the extraction routines. The background fitter uses a similar set of parameters to the the normal sky modeller but applies its fits to the sets of inter-order pixels at each X-position in the frame. Note that this involves a lot of I/O and can be quite a slow operation for large frames.
The wavelength calibration process may be performed using either a single ARC frame, or by using two frames (before and after the object exposure) and having the wavelength scale interpolated.
The advantage of using a single frame (possibly generated by co-adding the before and after arc frames) is that this allows the resolution of the final spectrum to be easily investigated. This is because the arc spectrum is always extracted in an identical manner to that of the object; using the first (and in this case only) supplied arc frame.
Using two frames and having to interpolate the wavelength scale will be most useful when there is a significant shift during the exposure such that co-adding the two arcs would produce a frame with ‘doubled’ lines. You should note that in this situation the generated wavelength scale is only a guess based on the assumption that the difference between the two arcs was due only to a stable linear shift between the exposures. You should ensure that the fractional weights assigned to the two arcs (parameter TUNE_SCFRACT ) are appropriate to the specific situation by investigating the times of the exposures.
The effect of the 2-D distortion correction can be examined by using the ech_scrn2d task to create a
‘corrected’ image of the orders. This should be displayed using the regular imaging commands.
It is also possible to use software intended for processing single spectra to process such
corrected images. You should take care however to note that the pixels in such images are
no longer independent as they represent the scrunching of two or more pixels in most
cases.
It is often useful when reducing ‘difficult’ data, to be able to examine various intermediate ‘spectra’
generated during the reduction process. echomop provides a very flexible method of examining any
intermediate arrays produced by it or other tasks. The ech_plot task/ECHMENU Option 27
provides zooming, re-binning, re-scaling etc. on any array in the reduction database. Arrays
may be specified by name or by using a set of reference names by which echomop tasks
access them. To obtain a directory of known reference names use the D(irectory) option in
ech_plot.
echomop can apply heliocentric corrections to the output wavelength scales during ‘Write results’ and ‘Multi-merge’ operations. In order for this to be possible it is necessary for to provide the package with the date and time of the observations involved. This is done by entering the values in the following manner (SETOBJ is a fiGARO command):
If these values are not provided a warning message will be printed, otherwise the applied correction will be shown.