STScI Logo

Space Telescope Imaging Spectrograph
STAN - November 1997

STScI                        .  . __..___.                       Number 6 
Analysis                     |__|(__   |                         November 1997
Newsletter                   |  |.__)  |   
              __..__ .___ __ .___..__ .__. __. __ .__..__ .   ,
             (__ [__)[__ /  `  |  [__)|  |(__ /  `|  |[__) \./ 
             .__)|   [___\__.  |  |  \|__|.__)\__.|__||     |  
                             _   _   _   _        
                            / \ / \ / \ / \ 
               G H R S     ( S | T | I | S )      F O  S  
                            \_/ \_/ \_/ \_/          

       - HST SPECTROSCOPY STAN: Support for STIS, FOS and GHRS

       - STIS NEWS 
         + STIS on the WWW:  Data Handbook, Analysis Support, and more...
         + Cycle 7 GO observing and the Long Range Plan for STIS 
         + Technical News - On Orbit Performance and Advice for Observers
            * Cycle 7 Calibration Status
            * Scatter and Spectral Purity
            * CCD BIAS levels not stable
         + High S/N Test Results
         + Improved Target-Acq ETC and On-line Help on the Web
         + SLITLESS: A New STSDAS Task to Work with the Pure Parallel Data
         + STIS Pipeline Calibration News
            * New tasks in STSDAS
            * Recalibrating your STIS Spectroscopic data
            * Alerts to Potential Pipeline Processing Problems 
            * New Reference File Deliveries
            * CCD Darks and Hot Pixel Updates
         + ERO Data Recalibrated
         + New RPS2 
            * GO inserted Fringe Flats
         + Hubble Deep Field South Test Observations
            * ISRs (Instrument Science Reports)...
            * FAQs...
            * Advisories...
            * STIS Data Handbook and GO Primer

         + Volume II of The Archive Data Handbook and Reference File Updates
         + The IUE archive at STScI


With the closeout of the GHRS and FOS following the 2nd Servicing Mission, the
STIS Team at STscI has taken on the responsibility to provide continuing
support for those instruments, including responding to Help Desk questions,
data analysis issues, etc.  We will also report any related newsworthy items
specific to those SIs in our newly titled STAN, HST SPECTROSCOPY, which will
continue to contain the normal flood of the latest news and information about
STIS.  The distribution lists for all spectrograph STANs have been combined
into a single list allowing you to continue to receive information on HST
spectroscopic instrumentation without having to resubscribe.  If you do not
wish to receive this information, simply unsubscribe according to the
instructions at the bottom of this STAN. 

New STIS WWW Home Page Format/Updates

Check it out.  The STIS WWW Home Page has been revamped providing a simplified
and nicely presented format to link to more detailed information.  A number of
new pages have been added as well including, 
 On-Orbit Performance 
 Reference File History 
 Data Foibles 
 Calibration Accuracies 
 Pipeline Software History 

*DATA HANDBOOK - Volume 1 of the HST Data Handbook, which covers the current
 instruments on board HST (including STIS).
*CALIBRATION WORKSHOP PAPERS - (Papers from the STIS portion of the 1997 
 Calibration workshop)

 The STIS group has written a GO Primer, which is a brief introduction to 
 analysing STIS data.  This is an excellent resource for beginners, as well as 
 experts who aren't familiar with the new format of HST data.  There is also 
 information here about how to arrange a visit to the Institute.

Cycle 7 STIS GO Observing and the Long Range Plan
The STIS Cycle 7 GO and GTO programs began in earnest in late September; STIS has 
been observing for roughly 25-30% of the external prime orbits since then and that
rate is expected to continue during the next year.  The HST Long Range Plan
will be rebuilt during the month of December to incorporate the orbits from the
NICMOS Delta CP.  The expectation and hope is that the impact on STIS observers
will be small, but STIS observers can expect that there may be some movement in
the placement of their observations on the long range plan as shifting is done
to accomodate the new full Cycle 7 program.  By mid January, we hope the LRP
will be largely stabilized again, though even then there will be small future
adjustments for things such as the second NICMOS Camera 3 campaign. As always,
we ask for your patience with this difficult cycle and hope that you agree in
the end that your STIS data will have been well worth the wait! 

Technical News - On Orbit Performance and Advice for Observers:

Cycle 7 Calibration Status:

About 30% of the data for the STIS Cycle 7 calibration program
has now been obtained. Recent measurements include 

* Further monitoring of MAMA and CCD sensitivity 
	- Sensitivity in the low-order grating modes is stable to
	  better than ~2% or better. Systematic uncertainties in 
          dispersion solutions and vignetting corrections currently
	  prevent measurements to higher precision.

* Slit throughput measurements for MAMA observing modes
	- Slit throughputs measured on orbit meet or exceed pre-flight
	  expectations.  Reference files are in the process of
	  being updated.

* Bias levels NOT stable

Analysis of all in-flight bias frames taken to date has shown that the behavior
of the bias level along rows of the sensitive region of the CCD is NOT stable;
the slope appears to change over time. The timescale of this change is of order
one week, but is not constant. Since overscan correction is currently performed
using columns in the left-hand serial overscan region of the CCD, the maximal
effect of the slope is found on the right-hand side of the CCD. The maximal
(peak-to-peak) difference in bias level on the right-side edge of the CCD
amounts to about 0.8 DN, which can be significant for some purposes, e.g. in
cases of short exposures in which the background intensity is negligible.  See
the On-Orbit performance page on the WWW for more details. 
Fortunately, the virtual overscan region (which is located at the bottom rows
of the raw images) turns out to track the behavior of the sensitive region very
well. Consequently, we are changing the bias level correction algorithm so as
to perform overscan subtraction using both the serial and parallel overscan
regions of the CCD. The revised algorithm will be available in calstis 1.6 (see
Pipeline NEWS). 
An implication of this change is that we have to re-create and re-deliver all
superbias and superdark reference files from March 1997 onwards. This effort is
currently being undertaken.  We expect the corrected superbiases and superdarks
to be delivered by mid-December. GOs and GTOs who are concerned about the
effect of this change to their data are advised to wait until the corrected
reference files have been delivered. 

* Scattering and Spectral Purity of STIS Spectroscopic data

By now, a number of observations have become available that are useful to study
the spectral purity of STIS data. If the PSF had an infinitely small FWHM and
no extended wings, the spectrum of a point source would produce a spectrum with
infinitesimal extension in the spatial direction. Furthermore, the spectral
resolution would be essentially the theoretical limit of the spectrograph,
independent of the entrance slit.  In practice, the OTA PSF is wider and more
complex, and there is scattered light from both the gratings and the detector
itself, leading to decreased spectral resolution and spectral purity which is
dependent both on the observing aperture and the extraction aperture used when
analyzing the data.  The properties of spectra produced by STIS, like those
from any other spectrograph, are a compromise between the requirements of
maximum S/N and spectral resolution. Often these two requirements are mutually
exclusive, or at least they cannot be maximized simultaneously.  A new posting
on the On-Orbit performance page on Spectral Purity describes the spectral
resolution and line profiles and their dependence on the slit size, both in
dispersion and in spatial directions. Here we provide a summary of the
recommendations for observers. You should look for more details on the WWW

Recommendations for Observers:

1) The maximum spectral purity is achieved with entrance slits of 0.2" width or
   smaller. Observers wishing to study spectral lines of continuum sources
   should always consider use of small entrance slits.

2) Echelle observations of continuum sources should be done with an entrance 
   slit height of 0.2", i.e. with the standard 0.2" x 0.2" aperture. Larger
   slit heights will make it difficult to subtract the background and lead to
   increased spectral impurity.

3) If allowed by S/N considerations, observers may want to experiment with
   smaller than the default extraction slit heights during data reduction 
   if the observations suffer from spectral impurity due to the use of a 
   >0.2" entrance slit (see STIS ISR 97-13).

4) Users of the exposure time calculator should be aware that the scattered
   light and spectral impurity results are not yet implemented in the ETC. 
   This, together with revised sensitivities leads to increased count rates 
   for a given incident spectrum. The revision is most significant for the 
   echelles.  While this does not immediately translate into higher S/N for a 
   given exposure time, it does affect the bright object screening limits 
   which have become more conservative by a factor of about 2 and 1.5 for the 
   E140H,M and E230H,M modes, respectively. We will update the ETC to reflect 
   these changes in spring 1998.  Meanwhile, observers should simply scale 
   their count rates by the appropriate factors. 

High S/N Test Results

Analysis of the data from SMOV proposal 7091, designed to test out the 
ability of STIS MAMA's to return very high signal to noise was, very 
successful.  Using the FPSPLIT slits, signal-to-noise of over 350:1 per 
resolution element integrated over the point source cross dispersion profile 
was achieved for both the E230M and E140M modes. 

Improved Target-Acq ETC and On-line Help on the Web

The revised and updated version of the target-acq ETC is now available on 
the web. Among other improvements, this improved version of the target-acq 
ETC has a 'CHECK-BOX' facility, with options for point and diffuse sources 
in keeping with the proposal instructions (check-box =3 for point sources, 
and 3 to 105 for diffuse sources). The default values of the Gain, CR-split, 
etc. are the actual values used for target acq.  Since only the CCD detector 
can currently be used for the target acquisition, this ETC does not have any 
MAMA options in the input form.  The ETC also includes a list of standard 
stars which can be used.
The ETC-help files have now been enhanced with more information which include 
more details on the background characteristics, some details on the 
uncertainties in the peak countrates, etc.  Links have been added from every 
(STIS)ETC to every other (STIS)ETC.

SLITLESS:  A New STSDAS Task to Work with the Pure Parallel Data

An IRAF program called SLITLESS has been written to allow the quick 
extraction of a spectrum from a direct image and slitless grating pair.
It is primarily aimed at quick reduction of STIS parallel data where 
the direct image is taken with the 50 CLEAR aperture and the spectrum 
image with the G750L grating and no slit. SLITLESS displays the direct 
image, then simply by selecting an object on that image, the matching 
spectrum can be extracted. SLITLESS extracts the appropriate region of 
the grating image and determines the wavelength range of the extracted 
spectrum. The properties of some STIS gratings (also NICMOS grisms) are
pre-defined in an ASCII table read by SLITLESS. Program parameters 
allow the user to fine tune the extraction process, such as the extent 
of the object and the number and size of the background spectrum 
apertures, or these can be altered interactively. The data quality for 
the spectrum image is also used to exclude bad points from the extracted 
spectra. When used in its non-interactive mode, SLITLESS automatically 
extracts a given number of rows from the grating image and determines 
the level of the spectral background using regions of the grating image 
that are near the spectrum. For interactive extraction, SLITLESS assists 
the user by marking the appropriate part of the grating image and 
displaying the extraction profile, the background sky(s) and finally the 
extracted spectrum. SLITLESS displays the extracted, background subtracted 
spectrum, produces a text file listing the extracted spectrum (and 
background) and can optionally produce a FITS image of the background
subtracted spectrum.

The program can be downloaded from 


and will soon be available in the STSDAS contrib package.

STIS Pipeline Calibration News

The CALSTIS pipeline calibration software continues to evolve at a
rapid rate as more STIS data becomes available and is carefully
evaluated.  Observers should keep a careful eye on the new STIS WWW
pages, in particular the "Pipeline Software History", to stay abreast
of the latest changes between issues of the STAN.  Additionally, the
newly released Data Handbook includes a high-level description of
CALSTIS, including flow charts and more detailed descriptions of the
various calibration steps.  Finally, the latest news about calibration
reference files can also be found on the new WWW page "Reference File
History."  We summarize here the latest developments with CALSTIS,
reference file delivery, and data recalibration and analysis tools.


A number of new data recalibration and analysis tools have recently been
installed in STSDAS and are described in the Data Handbook. We call your
attention to these new tools, and refer you to the Data Handbook for further
details about their use: 
- chcalpar, for updating header keywords in preparation for recalibration
- inttag, integrates a timetag event list (*_tag.fits data file) to form
    a raw image
- echplot, particularly useful for browsing echelle spectra
- the calstis "standalone" calibration/recalibration tasks:
  basic2d, occreject, wavecal, x1d and x2d.


If you are attempting to use Starview/Bestref to retrieve the best reference
files to recalibrate your spectroscopic STIS data, you will find that the
system does not currently automatically retrieve the LAMPTAB reference table
for you. You will need this file if you wish to calibrate your data using the
contemporaneous wavecal to update the wavelength zeropoint, the default
operation of calstis. Until we have straightened out this problem (hopefully we
will have it squared away early in the new year), you will also need to
retrieve the latest lamptab table in order to recalibrate your data. To do so,
select the STIS Calibration Screen and qualify on Reference File Type: LMP and
Cal. Lamp = the value of the SCLAMP keyword in your rootname_wav.fits file.
Retreive the latest lamp file and use this to recalibrate your data. 


We list below several of the recent problems found with pipeline processing.
Most of these problems will be corrected in an upcoming release of the CALSTIS
software (Version 1.6), expected early in January. Check out the details and
look for the release announcements on the STIS WWW "Pipeline Software History"

   The Time-Tag Time Stream:
The timetag stream for all STIS timetag data processed through the pipeline
before October 28, 1997 with versions of OPUS earlier than OPUS 7.1b are
subject to error. The accum image made from the timetag data was generally

   PHOTFLAM Keyword Incorrectly Populated for Imaging Data:
The PHOTFLAM keyword is currently incorrectly populated for CCD and MAMA
imaging data that use filters. We have posted the correct values of the
PHOTFLAM keyword, still based on the ground component throughputs, on the STIS
Reference File History page under Sensitivities.  We expect to update these
values of PHOTFLAM based on on-orbit measured performance in early spring of

   Cosmic Ray Rejection in the Case of Rapidly Varying Sky:
Cosmic ray rejection in the pipeline currently does not handle the case of a
rapidly changing sky background between individual CRSPLIT images well.  For
more detailed information, see also the the STIS WWW "Data Foibles" page. 

   Acq/Peak Pseudo Images for Very Bright Sources:
The acq/peakup pseudo image (contained in the fourth image extension,
EXTNAME=PEAKUP of the acq/peakup exposure) created using the total counts from
each dwell point of an acq/peakup can contain erroneous flux values for very
bright targets.  For a workaround in interpretting your PEAKUP image until this
problem can be corrected, consult or your Contact Scientist. 

   Varying Bias Pattern:
As explained under the Technical News section, above, the algorithm for the
overscan (bias level) subtraction is being changed to correct for the
variability of the bias pattern slope with time. 

  The location of the pipeline extracted 2-D spectrum can be erroneous if 
subarrays are used in conjunction with Patterns or Postargs.


  CCD Darks and Biases:
In-flight CCD darks have now been delivered to the pipeline and are available
through the archive for the period through 30-Oct-1997. We expect to update all
of the darks and biases for data taken prior to December 1997 in early December
1997 to correct for the impact of the varying bias pattern (see above). 

  G230LB Flat:
Problems were recently found with one of the previously delivered CCD
spectroscopic flat field files for grating G230LB, wavelength 2375
(h2308493o_pfl.fits). The pipeline flat had bad rows around the location of the
slit fiducials.  A new file has been delivered. 

  Aperture Description File:
The aperture description file is used for calculating the dispersion solution
and the wavelength zeropoint by calstis.  A new aperture description (_apd)
reference file that updates the locations of several of the apertures was
delivered the week of 24 Nov 1997. 

  First Order L Mode Sensitivities and Aperture Throughput Updates Pending:
New sensitivity throughput (_pht) files and a new aperture throughput file
(apt) based on on orbit calibration measures which update the sensitivities for
first order gratings are expected to be delivered early in December. Check the
WWW pages for updates. 

We expect reference file updates to be occurring fairly frequently throughout
cycle 7.  Between STANS, please consult the STIS WWW "Reference File History"
page for updates. 

Weekly superdarks are delivered to the pipeline within ~three weeks of an
observation and can be retrieved via StarView and used to recalibrate your CCD
data.  (Go to "HST Instrument Searches," choose "Reference," and choose the
"Recommended" files for the data set you wish to calibrate).  Additionally,
daily darks are also taken. These are not routinely processed by the STScI and
delivered to the pipeline, but you can retrieve them from the HST archive and
use them to make a superdark tailored to your particular day of observation.  A
one day old superdark removes 95% of the hot pixels, while a 5 day old
superdark typically removes 88% of the hot pixels.  In the future we will make
an easy script available for you to make your own daily darks.  In the interim,
for information on how to make daily superdarks, send inquiries to To find STIS CCD darks stored in the archive, taken, for
example, between October 29 and November 1, you can use starview as follows: 

        1) Go to "General Search" screen
        2) Enter the following:
                Target Name: DARK
                Instrument: STIS
                Config: STIS/CCD
                Start: 10/29/97 .. 11/1/97  (for example)

After extracting from the archive, check that the gains are the same. If not,
either throw out the ones you don't want, or correct them all to the same gain.

There will probably be several hundred hot pixels left after correction by
calstis with the best dark reference file. The cosmetics of the final image can
be improved using the tasks noao.imred.ccdred.cosmicrays and proto.fixpix. 

See also the hotpixels button on the Foibles WWW page for the latest updates. 

ERO Data Recalibrated

STIS ERO data have been recalibrated with the latest calstis codes
(calstis 1.5 and 1.5a) and the most up to date reference files as of
October 1997. The recalibrated data have also been rearchived and are
publicly available. Calibration steps not available at the time the ERO
were calibrated by the pipeline have been added for these recalibrated
data (i.e., 1-D extraction for echelle spectra, and use of the
engineering wavecals for zero point correction of the wavelength

Programs included in the recalibration are

7121    Imaging Spectroscopy of NGC4151
7122    Imaging of SN1987A and its Nebula
7123    MAMA and CCD spectroscopy of SN1987A
7124    Kinematics of the Nuclear Gas Disk in the Galaxy M84
7125    CCD Coronographic Imaging of the Disk of Beta Pictoris
7126    Early Release observations of Shocked Interstellar Gas in the 
        Vela Supernova Remnant
7130    STIS/Spectroscopy of the Brown Dwarf GL229B 
7137    STIS Observations of UV High-velocity Interstellar Lines in the 
        Carina Nebula
7138    Imaging and Spectroscopy of the Bright Arcs around the most 
        luminous X-Ray cluster RXJ 1347.5-1145 
7139    UV Spectral Dating of Galaxies

RPS2 Updates

A new version of RPS2 (7.1.2) was released 21 October 1997.  The primary 
purpose of the release was to support a compromise focus position for NICMOS
in preparation for the upcoming NICMOS3 campaign.  However, also included in
the release was the new buffer management design affecting STIS as well.  The
new design will provide both gains in speed and efficiency.  You will also
note PED (Proposal Editor) now provides online links to the proposal 
instructions and that the FOV (field of view) postscript plots are now 

On December 1 1997, RPS2 7.2 was released. RPS2 7.2 contains the TARGET=CCDFLAT
optional parameter which allows GOs to insert their own contemporaneous
Tungsten lamp flats (the so-called Fringe Flats) in their Phase 2. Fringe Flats
are recommended for all observations in G750L or G750M at wavelengths longward
of 7000 Angstroms. GOs are directed to STIS ISR 97-15 which describes the whys
and hows of inserting fringe flats. 
HDF-South Test Observations

Plans are in the works to observe a southern ``Hubble Deep Field'' in October
1998. The observations will be done in Director's Discretionary time and will
be non-proprietary. The field selected includes a QSO at redshift z~2.2, which
will be observed with STIS while WFPC-2 and NICMOS observe nearby blank fields.
Further information on the project can be found on the web at

A two orbit observation of the field was carried out at the end of October
1997. The STIS observations during this test include low-resolution spectra of
the QSO with gratings G230L and G430L, and a 50CCD clear image of the field
surrounding the QSO. These data can be found in the HST archive under proposal
7633. Data were reprocessed using the latest test versions of the dark-current
and sensitivity reference files. Those results can be found on the HDF web page
listed above. 

One feature that can be readily seen in the HDF raw data is the sensitivity of
the STIS CCD to scattered earth light. Those planning broad-band imaging with
the STIS CCD may want to take a look at these images. The enhanced background
can be avoided by invoking the LOW SKY special requirement in your Phase 2


All new or updated documentation related to the HST spectroscopic instruments
(STIS, GHRS, FOS) which has become available since distribution of the last
STAN will be noted here as appropriate.  Detailed information may be found 
in the "Documentation" links off the specific SIs home page.  New STIS 
advisories or newly identified problems with data may be found on the STIS 
home page Advisories or Data Foibles links, respectively.



  97-15  GO Added Near IR Fringe Flats
  97-13  Extraction Slits for First-Order STIS Spectra 


  - Problems with pipeline processing of STIS data
  - Using Starview/Bestref to retrieve reference files to recalibrate data  

  Data Foibles:

  - Poor Cosmic Ray Rejection for Rapidly Changing Sky
  - Obscurations in Slits
  - STIS Point Spread Functions


  - STIS Data Handbook - Volume 1 of the HST Data Handbook (current 
  - GO Primer - Brief intro to analysing STIS data [located on STIS Web page 
                under "Data Analysis User Support"



  - Are GHRS spectra background subtracted?


                         FOS and GHRS NEWS

Volume II of the Archive Data Handbook which describes how to work with data
from the retired or `legacy' instruments of HST, including major updates for
FOS and GHRS, is not yet out. It is under final revision and should be out 
early in 1998.  Updates to the FOS sensitivity and flat field reference files,
to take into account the post-costar time variations, are also not yet
delivered by expected on roughly a month timescale.  Updates will be announced
in this STAN. 


         Of General Interest to UV Spectroscopists: The IUE Archive

Data from the IUE Final Archive is now available through STScI. STIS observers
may find the IUE data useful both in planning and interpreting their HST data.
Access to IUE data (both NEWSIPS and IUESIPS) is provided through the same  WWW
and StarView interfaces used for data in the Hubble Data Archive. In addition,
experienced IUE staff members are available to assist researchers.  In
Starview, users can access the IUE Merged Observing Log of observations by
selecting "Non-HST Data Searches" in the Welcome Page and then "IUE." The IUE
archive can be searched using a variety of parameters, ranging from target name
and position to exposure time, observing date, and object class.  Check out the
STScI Archive WWW page for more information. 

                 APPENDIX: HST Spectroscopy 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 stis_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