pweight -- compute weights for use by the splice task
pweight intable outtable
This is an IRAF task to compute weights and include them as a column in spectral data, for use with the splice task. The fweight task also computes weights for splice, making use of the spectral data and a continuum fit to the spectrum. By contrast, pweight uses primarily the throughput table PHOTTAB to compute the weights.
The weight should be the signal-to-noise squared, (S/N)**2, and if N = sqrt (S), the weight is just S. The basis for determining the weight from the reference files alone is that the weight should be proportional to the number of electrons that you expect to have detected. It's only proportional and not equal because we don't take into account the actual flux of the target, and we don't use the NET column. This task would be appropriate for the case that the flux is so low that almost no counts were detected in at least one of the tables to be spliced together. For almost all other situations, the fweight task can be used instead of this one.
Because we don't know the normalization, the weights written by this task will typically be quite different from those written by fweight. It is therefore essential, when running splice, to consistently use weights written either by pweight or by fweight, but not to mix them.
The weight at each pixel (each wavelength) is obtained by the following expression:
weight = throughput * (pct_h / pct_infinity) * Ta * HST_AREA * exptime / dispersion
HST_AREA is the area (cm**2) of the unobstructed telescope aperture, pi * radius**2, with an adopted value of 45238.93416 cm**2.
throughput is the quantum efficiency, electrons detected per incident photon. This includes the obstructions of the aperture, reflectivities of the optics, transmission of any filter, and sensitivity of the detector. The throughput of the slit is _not_ included in this term (see Ta). The width of the extraction box in the cross dispersion direction is taken to be infinite (see the pct factors).
Ta is the throughput of the slit, which is thought of as being narrower in the dispersion direction than in the cross dispersion direction.
(pct_h / pct_infinity) corrects for the fact that a box of height h pixels in the cross dispersion direction was used to extract the spectrum, rather than an infinite extraction box. "pct" refers to the "photometric correction table," which contains this information. By convention, pct_h will by one if h is the default extraction box height, and pct_infinity will be greater than one (this is actually taken from the row for h = 600 pixels).
exptime is the exposure time in seconds.
dispersion is the dispersion of the spectrograph in Angstroms per pixel.
Each of the above terms was included because it either increases or decreases the expected number of counts per unit wavelength interval due to the source. HST_AREA was included to allow splicing data from different telescopes. This task cannot assign weights for other telescopes, but the above description should be sufficient to allow the user with sufficient information to compute weights with the same constant of proportionality.
Note that when running splice on output from pweight, the parameter splice.sw_name for the scalar weight column name should be set to null (""), since the weight already includes the exposure time.
The WAVELENGTH column is required in the input tables; this column can contain either scalar values or arrays. SPORDER must be either a column or a table header keyword. The following keywords are required, and they can be either in the primary header or in the table header: PHOTTAB, OPT_ELEM, CENWAVE. The EXPTIME keyword is required; it will be looked for only in the table header, not in the primary header. The following keywords are optional; they can be either in the primary header or table header: PCTAB, APERTAB, APERTURE, HELCORR. If the HELCORR keyword is found and its value is "COMPLETE", then RA_TARG and DEC_TARG will be looked for (in either header), and EXPSTART and EXPEND will be looked for (just in the table header). If HELCORR is "COMPLETE" and all these keywords are found, the heliocentric correction will be applied to shift the wavelengths back to the observed values before interpolating the reference tables at those wavelengths. EXTRSIZE can be either a column or a table header keyword; if it is found, and if the value matches a row in the PCTAB for column EXTRHEIGHT, the value in the PCTAB for that extraction height will be used.
Several of the above keywords or columns are used for small corrections to the computed weights, or they are only relevant if tables with different parameters are to be spliced together. For example, the APERTAB gives the aperture throughput. If the data to be spliced together are all in one echelle spectrum, or if all observations used the same aperture, then the aperture throughput just results in a constant offset to the weights. The PCTAB correction for EXTRSIZE will only be relevant if a non-standard extraction height was used when running x1d and files with different extraction heights are to be spliced together. The heliocentric correction results in a small shift in the dispersion direction, of order a quarter Angstrom. The weights computed by pweight and fweight can be shifted relative to each other by more than this depending on the smoothed flux spectrum used for fweight.
- intable [file name template]
- A list of tables containing spectra
for which weights are to be assigned.
If an input FITS file contains more than one table, each such table needs to be listed individually.
- outtable [file name]
- A list of output tables. If outtable is blank, or if the file names are the same as intable, then intable will be modified in-place. If outtable contains just one file name, while intable contains more than one table name, all those in intable will be copied to outtable, one table per extension. Otherwise, the number of names in outtable must be the same as the number in intable.
- (wgt_name = "weight") [string]
- The name of the column to contain the weights. If a column by this name (case insensitive) doesn't exist, it will be created; if it does exist, the values will be overwritten.
- (verbose = yes) [boolean]
- The default is to print the names of the input and output tables as they are processed.
1. Assign weights to o52z24010_x1d.fits, writing the output to o52z24010.fits.
on> pweight o52z24010_x1d.fits o52z24010.fits
2. Assign weights to o52z24010_x1d.fits, as above, but create the weight column in-place in o52z24010_x1d.fits.
on> pweight o52z24010_x1d.fits
The weights written by pweight and fweight use different normalization, since one is based on the throughput table and the other on the data. When running splice, the weights must be set consistently either by pweight or by fweight.
This task was written by Phil Hodge.