Pictures and Press from the MMT

• NGC 7479 (data taken by D. McCarthy and J. Hinz; image processed by Craig Kulesa)
• color image of NGC 7479 (data taken by D. McCarthy and J. Hinz; image processed by Phil Hinz)
• UofA Press release

Instrument Description

• For a complete description of the PISCES instrument, check out our PASP paper.
• PISCES is a wide-field infrared camera, employing a 1024x1024 Hawaii (HgCdTe) array.
• On the Steward 90", PISCES has an inscribed field with an 8.5' diameter, 0.5"/pixel.
• At the MMT, PISCES has a plate scale of 0.18"/pixel and a field size of 3.1".
• Click here to see a picture of the dewar. Check out Joannah's page. She has some other info and pictures.

Detector Characteristics

• Gain = 4.35 e-/ADU (from C. Kulesa)
• Readnoise = 3.3 - 4 ADU for double correlated sampling (C. Kulesa)
• Some plots illustrating:
  • Counts vs. time
  • Linearity
  • Gain

Data Reduction Procedure

• Check Pixel Types
  • All frames should have pixel type set to "short" - Note: this is done automatically with corquad.

• Make sure dark frames have values for RA and DEC (If you are reducing old PISCES data, you will have to add RA and DEC to the header for all images.)
  • cl> hedit dark*.fit RA 0 ver-
  • cl> hedit dark*.fit DEC 0 ver-

• Correct for Cross-Talk (a.k.a. dark comets)
  • Roelof de Jong de Jong has written an IRAF task to correct for cross-talk.
  • The task is called corquad, and you can find a tar file here.
  • Gunzip and untar as usual in a directory where you keep your IRAF *.cl scripts.
  • A detailed description of the routine and how to install it can be found in the included README file.
  • Note the platform-dependent compile instructions in the README file.
  • Linux user might want to: (1) insert "status=0;" before fits_open_file is called in the file corquad.c, and (2) in the Makefile make sure "LIBS = " has no arguments.
    • compiled version for linux - corquadnew.e

• Combine Dark Frames

• Dark Subtraction

• Apply Badpixel Mask
  • Recent version by Jackie M. is here and inverse is here. Rename to .pl files after download (for security reasons you can't download a *.pl file).
  • Here is a mask I made for the old array, good for data up through 1/99. After masking images with ccdproc, I also used this file as input to fixpix to get rid of the Nike swish.
  • Here is a mask for the new array. This mask has good values of zero.
  • Here is another mask for the new array that is more generous with the field size. This mask was made from a dark frame, so pixels with high dark current are flagged. If you don't like this method for making the mask, you might prefer mask4.fits listed below. This mask has good values of zero.
  • mask4.fits is also for the new array and has a more generous field size. No pixels are flagged within the useable field. This mask has good values of zero.
  • mask6.fits is also for the new array and has a more generous field size. Bad pixels within the useable field were determined by doing an 8-sigma clip on the ratio of two flats using the IRAF task ccdmask. This mask has good values of zero.
  • Here is a iraf batch file that Kevin Luhman wrote to mask out the part of the array outside the circular field of view. Note that this cuts off some useable parts of the field.
  • This IRAF script, mask.cl, replaces the masked out parts of the image with the mode.
    • You need to declare mask.cl a task in your loginuser.cl file.
      • Ex: task mask = home$scripts/mask.cl
    • The instructions on how to use mask.cl can be found by listing the task parameters.
      • cl> lpar mask

• Make Flat
  • Suggested imcombine parameters are
    • imcom(images,output=out,scal="median",stats="[200:800,200:800]", reject="ccdclip",lsigma=5.5,hsig=4.5, rdnoise="RDNOISE",gain="GAIN", maskt="goodval", lthres=500,hthres=32000,nkeep=2,blank=-999,grow=8)

• Flatfield Data

• Sky Subtraction (optional)

• Correct for Geometric Distortion
  • Roelof de Jong wrote an IRAF script to automatically correct for geometric distortions.
    • The script is called matchusno.cl, and all the necessary files can be found here .
    • Gunzip and untar as usual in a directory where you keep your IRAF *.cl scripts.
    • A detailed description of the routine and how to install it can be found in the included ReadMe file.
  • You can also correct for distortion manually using the IRAF geotran routine. When running geotran, the database is the name of the file below, and the transform is whatever name comes after "begin" in the files below. These files can be fed into geotran to undistort images.
    • Steward 2.3m - Here are distortion corrections generated from geomap for the April 99 run. (The distortion may vary run to run...) Pat Hall did the J and K; I did H.
      • J_9904_geomapout
      • H_9904_geomapout
      • K_9904_geomapout
    • MMT - Distortion corrections generated from Oct 13, 2002 data of NGC 1193 for J and narrow-band filters. Fit w/4th order Legendre polynomial. RMS is 2 pixels (0.36") for J and <3 pixels (0.54") for narrow-band filters.
      • geodb4

• Align and Stack
  • Roelof de Jong has written a task to mosaic images. Don't have a link yet!
  • autoalign.cl is another alternative for mosaicing. You need to run Roelof's matchusno.cl routine first so that (1) the images are undistorted, (2) North is up and East is to the left, and (3) the plate-scale is 0.495 arcsec/pixel. Download
    • autoalign.cl
    • getshifts.cl
    Add the following to your loginuser.cl file,
    • task autoalign = /path where you keep scripts/autoalign.cl
    • task getshifts = /path where you keep scripts/getshifts.cl
    and off you go! I haven't made a README file, but there are instructions at the beginning of the autoalign.cl file. Let me know if you have questions.

Other Reduction related stuff

• If you are not running Roelof de Jong's corquad task, make sure all files except darks have pixel type = ushort. Darks should have pixel type = short. If you need to change the pixel type of an image, you can use the IRAF task images.imutil.chpixtype.
  • Examples:
    • cl> chpixtype dark_in dark_out short
    • cl> chpixtype image_in image_out ushort
  • pisces.cl is an IRAF task written by Don that converts file pixel types to ushort. This is useful to run in the background while observing. Note: you do NOT want to run this on darks because IRAF will then interpret negative counts as large numbers. For darks you should leave the pixel type as short.
  In general, your in and out names should be the same. There's no need to keep copies with the wrong pixel type.
• This is a perl script that will take the output from hedit and create a huge IRAF batch file called batch.cl. See notes at the beginning of the file for more details. To run the perl script (in case you haven't used perl before) type "chmod +x reduce_ir.perl". Then "reduce_ir.perl" will run the script. Let me know if you have problems.
  • reduce_ir.perl perl script

Observing Tools

• Here is a new observing log. PISCES is not yet equipped to receive header info from the MMT TCS, so this log helps keep track of important info that is usually in an image header.
• Catalog of Persson et al. (1998, AJ, 116, 2475) standard stars.