STScI Logo

rcombine stsdas.hst_calib.ctools



rcombine -- Combine HRS or FOS rapid-readout data with the option of propagating errors and data quality flags.


rcombine input output operation


The rcombine task combines data groups within HRS or FOS rapid-readout images. It will sum or average the groups of a group format image into an image with one or more groups and will optionally propagate the maximum value of the associated data quality (cqh) image. For calibrated (c1h) images, the user also has the option to combine, in quadrature, the groups in the associated error (c2h) image. The input can be a single file name or a file name template. The output can be a file name root or a directory name.

You must specify if the data are to be summed or averaged, and whether or not to also process the associated data quality and/or error images. Additionally, you may specify which groups to combine (the default is all groups in the image) and how to bin the input groups into the resulting output (the default is all input groups into one output group).

Each pixel in the input file is added to the same pixel in the other groups of the file. If a subset of groups is specified then only those groups will be combined (e.g., setting the parameter groups = 1-x2 will use every other group, i.e., step by 2). If nbin is set to a value less than the total number of groups in the input image, then nbin input groups will form one output group. If nbin does not divide evenly into the total number of input groups, then the last output group will contain the remaining groups.

If the data quality (cqh) image is included in the operation, then the maximum value of a given pixel amongst the groups being combined is propagated to the output data quality file.

If the error (c2h) image is included in the operation, then each pixel in the error file is added in quadrature to the corresponding pixels in the other groups of the error file. After nbin groups have been added, the square root of the summed errors is taken and, if operation=average, the result is divided by nbin.


input [file name]
Name of the input image (this can be a file name template). The image extension, e.g., .d0h, .c1h, .c4h, must be included in the name.
output [file name]
File name or directory in which the output image will be stored. It is not necessary to specify the image extension as the output image extension will be the same as the input image extension.

If a single input file is used, then the output can be a single file name or directory name. If an input image template is used, then the output must be a directory name. When a directory name is specified, the output file names will be the same as the input file names but will be written to the specified directory.

operation="average" [string, allowed values: average | sum]
Operation to be performed, either average or sum.

If averaging and error propagation are chosen, groups from the "c2h" file are added in quadrature, the square root is taken, and then divided by the number of groups processed. If sum is chosen, then there is no division by the number of groups.

(groups = "-") [string]
Range or list of groups to be processed. Using a dash (the default) will process all groups in an image.

You can specify that only a subset of groups be used, according to the syntax described in xtools.ranges (type "help ranges"). For For example: 1-50 uses only the first fifty groups; 1-50x2 uses every other group of the first fifty groups. You will probably want to use "-" most of the time. Note the interaction between this parameter and the nbin parameter.

(nbin=INDEF) [integer]
Number of input groups that will be binned into each output group.

The default is to process all groups in the image into a single output group. However, you might want to process every 10 groups into a single ouput group, then you would set nbin=10. If your input image has 123 groups then your output image will have 13 groups. The first 12 groups will result from processing sets of 10 consecutive input groups while the 13th output group will result from processing the last three groups. Note carefully the interaction between this parameter and the groups parameter. If your input image has 123 groups and you specify nbin=10, groups=x2, you will get 62 output groups (ceil(123/2)) and only every other input group will be processed.

(dataqual=no) [boolean]
Process data quality files along with input spectral files?

The data quality file must have the same root file name as the input file and a file extension of ".cqh".

(errors=no) [boolean]
Process error files along with input spectra files?

The error file must have the same root file name as the input file and a file extension of ".c2h". If the input spectral file is not type "c1h", this parameter is ignored.


1. Suppose the files testim.c1h, testim.c2h, and testim.cqh are all in your current directory and have 144 groups. To average the groups, including error and data quality propagation, and produce an output file called newim containing a single group you would type:

   fo> rcombine testim.c1h newim average dataq+ errors+

This would produce three files newim.c1h, newim.c2h and newim.cqh.

2. Process the same images as above, but this time average every 10 groups into a single output group and put the output into a subdirectory called new:

   fo> rcombine testim.c1h new/ average nbin=10 dataq+ errors+

The last group would be the average of the final four input groups.

3. Average all groups of the count rate image y0nt0303t.c4h into the image newim.c4h containing a single group:

   fo> rcombine y0nt0303t.c4h newim average


None known.


imsum, imcombine, scombine, ranges

Source Code · Package Help · Search Form · STSDAS