STScI Logo

Hubble Space Telescope
NICMOS STScI Analysis Newsletter 4

NICMOS STScI Analysis Newsletter 4

August 1997


        - The 1997 HST Calibration Workshop

        - NICMOS NEWS
          + New NICMOS Information on the WWW
          + New Flight software => New Calibration Files
          + NICMOS Exposure Time Calculator has been Updated
	  + Brief History of CALNICA and CALNICB
          + STATUS of THE CORONOGRAPH for cycle 7-NICMOS

	- STSDAS/TABLES 2.0 Release



		The 1997 HST Calibration Workshop

               with a New Generation of Instruments

		     September 22-24, 1997

STScI will host the third HST Calibration Workshop on September 22-24,
1997.  Topics include the most recent calibration results for STIS and
NICMOS, updates for WFPC2, FOC, and FGS, closure reports for GHRS and
FOS, and an advance look at ACS and COS, the newly selected instrument
for the 2002 servicing mission.  To register, see the link at



New NICMOS Information on the WWW

Similar to all Space Telescope Instruments, the NICMOS home page on
the World Wide Web is updated with all instrument developments. The
NICMOS home page can be found by visiting the STScI WWW page
( and following Observing links to the NICMOS
Web pages.

Since last month, all NICMOS pages from the Top Page, have been updated.
Please visit them. In paritcular:

   * The Supplemental Cycle 7-NICMOS Call for Proposal.
   * The NICMOS Handbook version 2.0 (15-July-1997).
   * NICMOS Focus Update from the ADVISORIES page (22-August-1997).
   * The LIST of NICMOS Reference Files for Data Calibration (8-August-1997).
     This list contains a complete set of supporting calibration reference
     files, including synthetic darks based on an improved characterization
     of the MULTIACCUM dark behavior.
   * NICMOS Data available from the ADVISORIES page.
   * A CORONOGRAPH update from the ADVISORIES page.
   * The NICMOS Exposre Time Calculator is a new version (18-July-1997). 
Since last month, the following Instrument Science Reports have been made
available on the NICMOS documentation Web Page:
   * NICMOS Cycle 7 Calibration Plan (Newer version! 30-June-1997).

New Flight Software => New Calibration Files

On August 21st, new flight software was installed on NICMOS. This was 
intended to ameliorate the "pedestal effect". The pedestal is an effect seen in 
the NICMOS detectors after they have been idle for some period of time. When 
they are first used after such an idle period, a slowly changing bias is seen in 
the observations. This time dependent bias affects observations much like an 
excess dark current, adding as much as a few tens of ADUs to the observed 
signal. The excess signal appears in general to be more or less uniform across 
the detector field-of-view. The effect on calibration is to leave a uniform 
excess signal behind after dark current subtraction, so that when the data are 
then flat-fielded, an inverse flat-field pattern is imprinted on the calibrated 
image. This effect is very obvious in calibrated parallel exposures which are 
publically available from the HDA.

The new flight software changed the operation of the detectors in idle periods, 
in a way which was expected to reduce or eliminate the pedestal effect, 
following some tests conducted by the IDT at GSFC. The effects of the flight 
software change on the pedestal behaviour are still being characterized at 
STScI. However, we found very early after the change that the dark current 
behaviour has itself changed significantly as a result of the revised detector 
operation. As a result, the entire suite of dark calibration reference files, 
recently developed by us, are no longer applicable to the observations now being 
taken. All data taken from August 21 onwards, until such time as we obtain 
sufficient new dark observations to revise our calibration reference files, will 
have poor dark current subtraction, which will show up in the calibrated data as 
a combination of strong striations across the detector (the so-called "shading" 
pattern which all NICMOS detectors display) combined with an inverse flat-field 
imprint. We expect to be able to revise the dark current calibration reference 
files (albeit with a reduction in S/N compared with the files we are currently 
using) on a timescale of roughly 2-4 weeks from now.

We regret that in the meantime it will not be possible for observers to obtain 
well calibrated data.

NICMOS Exposure Time Calculator UPDATE!

The Exposure Time Calculator has been updated once again, so as to agree
with results of the NICMOS absolute photometry measurements during SMOV.
As a result, there are significant changes in predicted sensitivity AT
ALL wavelengths. The sensitivities implied by this version of the ETC are 
those included in the Supplemental Cycle 7-NICMOS Handbook.

It should be pointed out that the SMOV measurements were made using only
a few filters per camera. In general the results did not agree with the
predictions of the ETC, or those of the SYNPHOT code in the IRAF STSDAS
package. (Synphot has traditionally been used in the calibration of the
other HST instruments, and has been used at STScI for NICMOS calibration.)

On average, the absolute photometry measurements showed that the total
throughput for NICMOS was of order 80% of that predicted by the ETC and
Synphot (which agree closely with one another). There was a significant
scatter about this mean, with some filters showing more, and some less,
than this throughput. No obvious wavelength dependence could be ascertained
from the results. Therefore we have modified the calibration reference
files used by Synphot and the ETC to reduce the total throughput to
76%, 77% and 83% of what is theoretically expected for Cameras 1, 2 and
3 respectively. 

As a result, the predictions of the ETC now agree with observations in
terms of countrate. The predictions for deep exposures are also close to
the results actually obtained in some early Cycle 7 and ERO observations. 

IMPORTANT: The following caveats should be borne in mind when using the
ETC. First, for one filter, the F090M in Camera 1, the ETC overestimates
the countrate by about 20%. This is because the reflectance of the 
fore-optics and OTA drop shortwards of 1.0 micron. This deficiency in
the ETC will be corrected soon. Second, although for all other filters
Synphot and the ETC agree very closely with one another, the observations
indicate a scatter in throughputs about the mean adopted here. We do not
yet understand the cause, but note that the sensitivities implied by
the ETC do agree *on average* with observations, and so can be used
to constrain orbit requirements for new proposals.

Brief History of CALNICA and CALNICB

CALNICA: Instrumental calibration of single datasets

Version 2.1 of CALNICA was in use in OPUS at the time NICMOS was launched.
On 19 May 1997, version 2.2 went into use in OPUS, which fixed a couple of
sporadic arithmetic bugs in the non-linearity correction (NLINCORR) and
MultiAccum zero-read subtraction (ZOFFCORR) steps, and properly updated the
"FILENAME" keyword in all output files produced by CALNICA.
Version 2.3 was installed in OPUS on 19 June 1997. This version contains a
major rewrite of the cosmic ray identification (CRIDCALC) routine for
MultiAccum observations and populates the new "CAL_VER" keyword in output
files. The "CAL_VER" keyword indicates which version of the calibration
software was used to process the data.

The new CRIDCALC routine incorporates two major changes which correct for
deficiencies in the previous version of the routine. First, the process of
rejecting outliers from the stack of readouts for a given pixel was modified
so that instead of simultaneously rejecting all samples that lie more than
5 sigma from the mean countrate and then iterating the computation, only the
largest outlier is rejected. After rejecting the largest outlier, the
computation of the mean countrate and search for outliers is repeated until
there are no samples above the rejection threshold. The second change is that
instead of using the weighted mean countrate of the individual samples as the
final countrate for a given pixel, the final countrate is computed using a
linear regression routine to fit the relation between accumulated counts
and exposure time for the non-rejected samples. The combination of these two
changes has resulted in a more robust routine and yields a better estimate of
the true signal.

We are currently working to resolve several known deficiencies that still
exist in CALNICA v2.3. First, when processed on some systems, the HISTORY
keywords that are written to the science (SCI) image extension headers in the
output ima and cal files will erroneously indicate that the CRIDCALC step was
skipped because the readnoise reference file (NOISFILE) was dummy, when in
fact this file was not dummy and the CRIDCALC step was performed. The
processing log contained in the trailer (trl) file, as well as the calibration
indicator keyword for the CRIDCALC step ("CRIDDONE") correctly indicate that
the step was performed. Second, in the event that the CRIDCALC step rejects all
but one sample for a given pixel, the computation of the final countrate can
sometimes result in unphysically large (e.g. +-10**8 DN/sec) values. This
usually only occurs when processing heavily saturated data. Finally,
observations of very bright targets in MultiAccum mode can lead to a
significant amount of charge accumulated during the 0.203 seconds that it
takes to perform the initial (or "zeroth") readout of the detector. Once the
subtraction of the zeroth read (ZOFFCORR) has been performed, the total amount
of charge in the pixel is no longer correctly represented when the linearity
correction (NLINCORR) is performed, which can lead to either an inaccurate
linearity correction or allowing saturated values to pass through unflagged.
Corrections for these problems have already been devised and will be
incorporated into the next version of CALNICA.

CALNICB: Processing associated datasets

CALNICB version 0.1.4 was in use in OPUS at the time of launch. On 11 April
1997, version 0.1.5 was put into use, which fixed a bug in the assignment of
input images to output mosaics for FOUR-CHOP and EIGHT-CHOP patterns that
have enough images to fill the pattern more than once. Version 0.1.6 was
put into OPUS operations on 19 May 1997 and included modifications to update
the "FILENAME" keyword in output mos, asc, and spt files, and to update the
"EXTNAME" keyword in the output asc file.

CALNICB v2.1.1 was put into use in OPUS on 14 July 1997 (versions 2.0 and 2.1
were never put into operations). This was a major upgrade of the task which
offers several new capabilites. These include: the identification of pixels
that are suspected to contain flux from a source and the subsequent exclusion
of these pixels when computing the scalar background level in each image;
the use of a cross-correlation routine to refine the image offsets computed
from World Coordinate System (WCS) header keyword values; special processing
for Darks and Flats which eliminates background subtraction and mosaicing.
CALNICB v2.1.2 was put into OPUS on 23 July 1997, which fixed a sporadic
arithmetic error in the cross-correlation routine which would cause the task
to crash.


Tests of the NICMOS coronograph have shown that it performs well and
can be expected to give substantial reductions in scattered light
compared to direct images.  Changes to the NICMOS onboard software and
to ground software will allow for automatic location of the
coronographic hole and will readjust the focus to optimize the
performance of the coronograph.  Proposers whose scientific goals would
benefit from use of the coronograph should consider requesting this
observing mode in response to the cycle 7-NICMOS call for proposals,
although some caveats apply. Relevant details of the coronograph
performance are described in this document.

Recommendations for Proposers

Proposers for cycle 7-NICMOS who want to observe faint objects 0.3 - 3
arcsec from a bright point source should consider using the coronograph
to increase the contrast for their target object relative to the wings
of the PSF of the brighter companion object.

As discussed below, the most effective reduction in the scattered
background is obtained when images at two different roll angles are
obtained and subtracted.  Dithering is not an option for coronographic
observations, thus identification and removal of artifacts from images
also depend on having images at two different roll angles.  The
spacecraft can be rolled at approximately 6 degrees per minute.  After
a significant roll of the spacecraft, new guide stars must be acquired
and a new coronographic target acquisition must be done.  Thus, while
it is possible to perform all these activities in a single orbit, the
inefficient use of telescope time and the difficulty of scheduling lead
us to recommend separate orbits/visits for different roll angles for most

Observers whose target is bright enough to saturate NICMOS in a narrow
band filter in less than the shortest available integration time will
need to add one full additional orbit per target for acquisition.  This
is done through a somewhat cumbersome real-time procedure and can only
be supported for a limited number of targets.


	by Rick White & Perry Greenfield

We are pleased to announce the availability of the beta release of TABLES
and STSDAS, versions 2.0.  This version includes new tasks for accessing
NICMOS and STIS datasets along with a number of tools for manipulating
these and other data stored in FITS image extension format.

This beta release is "load and go" for Sun SPARCstations running SunOS
4.1.3 or Solaris 2.5 with the IRAF 2.11 beta release, which means it has
already been compiled for execution under these versions of the Sun
operating systems.  Once the tar file has been read, you can use the
package.  Please note that IRAF 2.11 is REQUIRED to run STSDAS/TABLES
2.0.  They will NOT run correctly under 2.10.

The final (non-beta) release of STSDAS/TABLES 2.0 is expected on about
September 7th.  The exact timing depends on the release by NOAO of the
final version of IRAF 2.11 (which should be available any day now.)  For
details on how to obtain and install this beta release and for
information on the status of our final release, check the release status
web page at .


                          APPENDIX: NICMOS Contacts
Any questions about the scheduling of your observations should be
addressed to your Program Coordinator. Post-Observation questions can
be addressed to your Contact Scientist. If you do not know who these
persons are, you can find the information on the WWW at

Analysis, STSDAS or any other HST-related questions can also be
addressed to
To subscribe or unsubscribe send a message to with
the Subject: line blank and the following in the body:
                     [un]subscribe nicmos_news YOUR NAME
Comments, questions, suggestions, etc. can be e-mailed to
The Space Telescope Science Institute is operated by the Association of
Universities for Research in Astronomy, Inc., under NASA contract