Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.stsci.edu/ftp/documents/archive-manual/06_searchstrat.ps Äàòà èçìåíåíèÿ: Wed Mar 24 18:10:13 1993 Äàòà èíäåêñèðîâàíèÿ: Sat Dec 22 18:50:14 2007 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: reflection nebula

Planned Versus Executed Information 61
CHAPTER 6 :
Archive Search
Strategies
In This Chapter...
Planned Versus Executed Information / 61
Choosing the Right STARCAT Screen / 62
Search Strategies / 65
Assessing Data Quality / 77
The AEC and the EC / 82
This chapter describes the types of information provided by
STARCAT and approaches that can be used to quickly find information and
datasets in the HST archives that are of interest to you.
Planned Versus Executed Information
The science tables of the DMF catalog contain information about both
the parameters of each HST observation as they were executed (the so
called exposure level keywords) and about the observational parameters as
they were planned (the so calledproposal level keywords). As described
below, in the section on choosing a STARCAT screen, the STARCAT
screens are organized within the menu structure so that screens that
principally report either exposure level or proposal level keywords are
easily distinguished. In addition, there is one screen
(PLANNED­VS­DONE under the POST­OBS menu) that allows a
comparison between the planned and executed values of several
observational parameters. If you need to determine whether a field refers to
an exposure level or proposal level keyword, refer to the online
61

62 Chapter 6: Archive Search Strategies
documentation describing each screen, and to Chapters9 and 11 of this
manual. Most fields displayed on STARCAT screens correspond to
keywords in the headers of HST data files.
Choosing the Right STARCAT Screen
STARCAT includes many pre­defined screens with which to query the
HST catalog.
STARCAT screens are organized in a menu structure that groups
together screens that provide a particular type of information. Below, a
brief outline of the menu structure and a brief description of the contents of
each menu are presented. A description of the GENERAL screen is also
provided here, since the GENERAL screen will typically be the starting
point for most archival searches. Refer also to Chapter 9, where more
detailed documentation on the purpose and contents of each screen is
provided.
GENERAL Screen
The GENERAL screen is the all­purpose search screen for archival
research. It is usually the best screen to begin with when searching for
archival data. This screen provides a window on a mix of both exposure
level and proposal level information. It can be used to search for
observations at a given position in the sky (using Target RA and DEC), for
a specific target (using Target Name), and for observations of a given type
of source (using Target Description) 1 . The GENERAL screen can also be
used to search for observations taken with a given instrument, with a given
instrumental configuration, or with a given set of filters or gratings. The
screen provides exposure level information, such as the exposure start time
and duration, the fgslock employed, the exposure flag, and the PODPS
Data Quality value and comments. This information can be used to judge
the usefulness of the exposure for a given line of research (see also
``Assessing Data Quality'' on page77). The date at which the data set will
become (or became) public (i.e., available for archival research) is
displayed on the general screen (as the Release­Date). Note that the
GENERAL screen can only be used to search for science data with an
1. See also ``Search Strategies'' on page65.

Choosing the Right STARCAT Screen 63
archival class of CAL (i.e., astrometry data can not be retrieved with this
screen).
PRE­OBS MENU
The PRE­OBS menu lists screens that display proposal level
information---information that was provided by the observer during phases
1 and 2 of the proposal submission process. Most of the screens within this
menu provide information about both completed HST observations and
about HST observations that have been approved for execution, but that
have not yet been taken. However, the GTO, OBS, OPPARM and
PROP­OBS screens provide proposal­level information only for executed
HST observations. Obviously, you cannot retrieve data for an observation
that has not yet been executed.
POST­OBS MENU
The POST­OBS menu lists screens that display exposure level
information about executed HST observations. Some of these screens also
display proposal level information about each exposure. In addition to the
screens within the INSTRUMENT menu, these screens provide the main
mechanism for learning about the actual executed parameters of each
exposure.
TARGETS
The TARGETS menu lists screens that display observer­entered
information describing the target, such as alternate names, target
description, redshift, magnitude, or position. Most of these screens display
one record for each target within a given proposal ID. These screens are
useful when searching for targets having certain shared properties (e.g.,
optical magnitudes greater than some value, or redshifts greater than some
value) or when searching on source name or source type (see also ``Search
Strategies'' on page 65). In particular, the TARGSEARCH screen was
designed to make it easy to conduct searches for a particular source based
on source name or for a class of sources based on the target description, or
proposal title (see ``Search Strategies'' on page65). However, you should
keep in mind that the target­related information displayed in these fields
was entered by the observer and for some of the fields displayed on these

64 Chapter 6: Archive Search Strategies
screens, the observer is not required to have filled in a value during the
proposal process. Thus, the information displayed will be both incomplete
and heterogeneous.
INSTRUMENTS
The INSTRUMENTS menu lists screens that display instrument­
specific information about HST exposures. There is at least one screen for
each HST instrument. More detailed information on the instrumental setup
and observing parameters used for an observation are shown on these
screens than on the more general screens in the POST­OBS menu. These
screens are useful for searching for all observations with a given
instrument in a specific instrumental configuration and for determining
more precisely the instrumental parameters employed in a given
observation.
ENGINEERING
The ENGINEERING screens provide a mechanism for finding
non­science data files, such as processor dump data, science instrument
engineering data, engineering subset data, guide star position data, science
mission schedule data, etc. By and large, each screen searches for one type
of file (i.e., a file with a given extension). The documentation for each
screen explains the naming convention for the files it retrieves.
ARCHIVE
The ARCHIVE menu contains screens that display accounting
information about the data sets and files contained on the optical disks that
comprise the HST archive. In general, archival researchers should not need
to access these screens, unless they are interested in knowing if a particular
type of file has been created for a given data set, or if they need to
determine the size of a given file.
ADMIN
The ADMIN menu contains administrative screens used by the Data
System Operations Branch of the Space Telescope Science Institute to
monitor STARCAT accounts.

Search Strategies 65
Search Strategies
Overview
There are some standard types of searches that an archival researcher
might be interested in doing. Among these are:
. Searching for HST observations of a specific source (e.g, 30­Doradus
or the jet of M87)
. Searching for observations of a given class of source (e.g, all observa­
tions of SNR or O stars)
. Searching for observations of targets which possess certain shared prop­
erties (e.g., a B magnitude greater than16 or a redshift greater
than 0.01).
Suggested strategies for carrying out such searches are described
below.
Searching for Specific Sources
There are two ways to search for observations of a specific source:
. Search on positional coincidence
. Search on the basis of source name
Since the source or target name is a generally an observer­entered
designation, for fixed targets, searches based on positional coincidence will
prove much more reliable than searches based on source name.
Positional Searches
There are two ways to search for positional coincidences with
observations in the HST catalog using STARCAT. You can either qualify
on target RA and target DEC, specifying a range of permissible values (this
defines a search rectangle---see the example on page 28), or you can use
the Center command to specify a RA and DEC and a radius around that
position within which to search (this defines a search circle); this is
described in ``Searching an Area Around a Centered Position'' on page89.
The Center command has the added advantage that it allows you to
specify the coordinate system and epoch in which the search will be done

66 Chapter 6: Archive Search Strategies
(e.g., it supports epoch 1950 or 2000 and allows searches based on Galactic
coordinates).
Searching on Source Name
Most of the STARCAT data screens have a field for the target name
(often called targname). In addition, the TARGSEARCH screen within
the TARGETS menu, contains both the target name and alternate names or
aliases. The archival researcher can qualify and search on the basis of
source name using these fields. Note that when using the TARGSEARCH
screen to search for a given source name, three separate searches must be
done:
1. First, qualify on target name
2. Then clear the qualifications for target name and qualify only on
alias1
3. Lastly, qualify only onalias2
When qualifying, the source name should always be embedded in wild
card delimiters (e.g., *MARS*, or *NGC1068*), because the name will
frequently be embedded in additional text (see below).
Since an astronomical source typically has many names, to understand
how best to qualify on the target name and alias fields, it is useful to know
how target names and aliases are assigned. Target names are used to
provide unique designations for targets within a given proposal. They are
entered by the observer during Phase II of the proposal process. There are
three different classes of targets (fixed targets, solar­system targets, and
generic targets). Observers are encouraged to use a specific convention
when naming fixed and solar system targets 2 . Below is a brief summary of
the conventions for each class of target, which should help you design
search strategies based on source name.
. Fixed Targets. When providing a target name for a fixed target, the
observer must specify one catalog name which will become the target
name (e.g., HD124897). He or she is encouraged to provide one or two
common names which will become the target aliases (e.g.,
ALPHA­BOO, ARCTURUS). The preferred order for catalogs to be used
to designate the target name is summarized in Table 6.1 below. A sum­
mary of the convention for naming uncataloged targets is also given.
Observers may also append a qualifier to the catalog name to indicate
2. Refer to thePhase II Proposal Instructions for a detailed listing of these con­
ventions.

Search Strategies 67
the location in the source where the observation is centered, for exam­
ple, NGC4486­JET, LMC­R136A, or NGC224­004012+40548
(where the name of the parent body is followed by rounded J2000 coor­
dinates). Finally, a code designating the target purpose must be
appended to the target name when a target is observed specifically to be
used as an external calibrator (­CALIB), as an astrometric reference star
(­REF), or as an offset for target acquisition (­OFFSET), for example:
BD+284D4211­CALIB, 3C273­OFFSET
Table 6.1: Preferred Catalogs for Object Classification
Object Type Preferred Catalog Order
Stars Henry Draper Catalog (e.g., HD140283)
Durchmusterung (BD, CD, or CPD) (e.g.,
BD+30D3639)
General Catalog of Variable Stars (e.g., RR­LYR)
Star Clusters and Nebulae New General Catalog (e.g., NGC6396)
Index Catalog (e.g., IC418)
Perek­Kohoutek designation (e.g., PK208+33D1)
Sharpless catalog number (e.g.,S106)
Galaxies and Clusters of
Galaxies
New General Catalog (e.g., NGC4536)
Index Catalog (e.g., IC724)
Uppsala Catalog (e.g., UBC11810)
Abell Catalog (e.g., ABELL2029)
Quasars and Active
Galaxies
See Veron­Cetty and Veron, ESO Report No. 7, 1989
(e.g., 3C273)

68 Chapter 6: Archive Search Strategies
. Solar System Targets. The target name for solar system targets can
either be chosen from the standard list of solar system targets repro­
duced in Table 6.2, or a name can be defined by the observer. As with
fixed targets, qualifier's can also be appended to the name (e.g.,
MARS­FEATURE1).
Table 6.2: Solar System Target Names
Level 1 Objects Level 2 Objects
Sun Moon Adrastea (1979J1) Janus (1980S1)
Mercury Phobos Thebe (1979J2) Epimetheus (1980S3)
Venus Deimos Metis (1979J3) Helene (1980S6)
Earth Io Thethys Telesto (1980S13)
Mars Europa Dione Calypso (1980S25)
Jupiter Ganymede Rhea Pandora (1980S26)
Saturn Callisto Titan Prometheus (1980S27)
Uranus Amalthea Hyperion Atlas (1980S28)
Neptune Himalia Iapetus Cordelia (1986U7)
Pluto Elara Phoebe Ophelia (1986U8)
Pasiphae Ariel Bianca (1986U9)
Sinope Umbriel Cressidea (1986U3)
Lysithea Titania Desdemona (1986U6)
Carme Oberon Juliet (1986U2)
Ananke Miranda Portia (1986U1)
Leda Triton Rosalind (1986U4)
Mimas Nereid Belinda (1986U5)
Enceladus Charon Puck (1985U1)

Search Strategies 69
Searching for a Class of Object
Archival researchers may be interested in finding all (or most) HST
observations of a given class of objects (e.g., observations of elliptical
galaxies or Wolf Rayet stars). There are several ways to carry out such
searches:
. Search on the target description
. Search on proposal title, proposal keywords, or proposal abstract
. Cross correlate the contents of the HST catalog with a catalog or person­
ally created list of sources on the basis of position (see the section on the
Archived Exposures Catalog, below).
Searches on Target Description
A target description and category are entered for each target by the
observer during Phase II of the proposal process. These fields allow an
archival researcher to search for HST observations ofclasses of objects.
The target description and target categories appear on the GENERAL
screen, the PROP­OBS screen within the PRE­OBS menu, and the
TARGSEARCH screen within the TARGETS menu. The target description
spans two lines (or fields) which musteach be qualified separately in order
to completely search the target description for a given string. For example,
if you wanted to search for all observations of nova stars within our galaxy,
you could use the GENERAL screen, qualify oncategory as star and
qualify on Descrpt1 as *nova*. For completeness, the search should
then be redone, unqualifying onDescrpt1 and qualifying onDescrpt2
as *nova*, in order to find all observations withcategory equal to
STAR and the string nova in either the first or second target description
line. Note the use of the wildcard delimiters (
*), since the word ``nova'' will
likely be embedded in a longer text (see below).
The target description is assigned by the observer as part of the Phase II
proposal process, using the guidelines in the Cycle 2, Phase II proposal
instructions 3 . Each target must be assigned a single primary Category
followed by at least oneDescriptive Keyword. Observers are encouraged to
use as many as five descriptive keywords per category and are allowed to
3. Although Cycle 0 and 1 observations were not initially described using this set
of guidelines, target descriptions for all Cycle 0 and Cycle 1 GO and GTO observa­
tions were redefined in the DMF data base to comply with the Cycle 2 Phase II pro­
posal instructions, ensuring that a consistent set of terms are used to describe the
archival observations.

70 Chapter 6: Archive Search Strategies
assign up to two categories (this is appropriate, for example, for absorption
studies of the interstellar medium using background quasars). The first
category is then entered in thecategory (or targcat) keyword and the
two lines of the target description are filled with the complete description in
the form:
category,keyword,keyword...;category,keyword,keyword...
The appearance of the second category is denoted by a semicolon and the
second category is entered in the secondcategory field (or targcat2).
The categories and corresponding keywords that can be used to describe
fixed targets by the observer are given in Tables 6.3 through 6.11. For solar
system targets, the descriptive keyword convention is give in Table 6.12.
Obviously, searches based on target description will be most successful if
the same terminology used to describe the targets by the observers is used
by the archival researcher. Keep in mind that assigning target descriptions
is an inherently subjective process. There are many ways to describe the
same target and different observers will inevitably describe the same target
in different ways. You may, therefore, want to apply broad criteria when
first qualifying on target description. If you are looking for a complete list
of all HST observations of a given type of source, you are encouraged to
also search based on the proposal title, abstract, and
abstract keywords, and to cross correlate positional lists of sources
with the Archived Exposures Catalog.
In Tables 6.3 through 6.11, information in parentheses next to the
keyword is explanatory only and is not part of the catalog. For example,
Table 6.4 has a keyword identified as ``LMXB (Low Mass X­Ray Binary)'',
LMXB is the keyword that appears in the target descriptor, the text is
explanatory only.

Search Strategies 71
Table 6.3: Keyword Categories
Category Keyword Contents
STAR Galactic stellar object
EXT­STAR Star in an external galaxy
STELLAR­CLUSTER Galactic star cluster, group, or association
EXT­CLUSTER Star cluster in an external galaxy
GALAXY Galaxy or AGN
CLUSTER OF
GALAXIES
Galaxy groupings, clusters, or large­scale structure
ISM Interstellar medium of the galaxy
EXT­MEDIUM Interstellar medium of an external galaxy
UNIDENTIFIED Unidentified object
CALIBRATION Calibration observations

72 Chapter 6: Archive Search Strategies
Table 6.4: Fixed Target Keywords for Describing Stars (Galactic Stellar Objects)
and Stars in External Galaxies
Descriptive Keywords for Stars (STAR or EXT­STAR)
Brown Dwarf Wolf Rayet (WC) Wolf Rayet (WN) Main Sequence O
Supergiant O OE SF SDO
WDO BO­B2 V­IV B3­B5 V­IV B6­B9.5 V­IV
B0­B2 III­I B3­B5 III­I B6­B9.5 III­I BE
BP SDB DB DA
DC DZ A0­A4 V­IV A4­A9 V­IV
A0­A3 III­I A45­A9 III­I AE AM
AP Horizontal Branch Star AGB Star Composite Spectral Type
F0­F2 F3­F9 FP Late­type Degenerate
G V­IV G III­I K V­IV K III­I
M V­IV M III­I R, N, or S Types Carbon Star
Long Period Variable Irregular Variable Regular Variable Luminous Blue Variable
Dwarf Nova Classical Nova Nova­Like Recurrent Nova
Polar (AM Her Star) Intermediate Polar (DQ
Her Star)
Symbiotic Star T Tauri Star
FU Orionis Star Shell Star Eta Carinae Star Pulsar
Interacting Binary X­Ray Novae X­Ray Burster X­Ray Transient
RS CVn Star LMXB (Low Mass X­Ray
Binary)
MXB (Massive X­Ray
Binary)
W UMa Star
Beta Lyrae Star Algol System Barium Star Blue Straggler
Binary Pulsar FK Comae Star Pulsating Variable PG1159 Star
ZZ Ceti Star Cepheid Supernova Supernova Type Ia
Supernova Type Ib Supernova Type II RR Lyrae Star Planetary Nebula Central
Star

Search Strategies 73
Table 6.5: Descriptive Keywords for Stellar Clusters (Galactic Star Clusters,
Groups, or Associations) or Star Clusters in an External Galaxy
Table 6.6: Descriptive Keywords for Galaxies (or AGNs)
Table 6.7: Descriptive Keywords for Clusters of Galaxies (Galaxy Groupings,
Clusters, or Large Scale Structures)
Descriptive Keyword for Stellar Clusters
(STELLAR CLUSTER or EXT­CLUSTER)
Globular Cluster Open Cluster
OB Association T Association
Descriptive Keywords for Galaxies (GALAXY)
Spiral (Spiral Galaxy) Lenticular (Lenticular
Galaxy)
Elliptical (Elliptical Galaxy;
not a Dwarf Elliptical)
Dwarf Elliptical
Magellenic Irregular Amorphous Irregular Dwarf Compact (Dwarf
Compact/HII Galaxy)
Dwarf Spheroidal
BCM (Brightest
Cluster Member)
BGM (Brightest Group
Member)
LSB (Low Surface
Brightness/HI Rich Galaxy)
Seyfert
QSO (Radio Quiet) Quasar (Radio Loud) Radio Galaxy BL Lac (BL Lac or
BLAZAR)
Liner Starburst Ultraluminous IR Gal Interacting Galaxy
Lyman Alpha Cloud Protogalaxy Gravitational Lens Einstein Ring
High z Gal (Redshift
Greater than 0.5)
Descriptive Keywords for Clusters of Galaxies
Supercluster Void Group Rich Cluster
Poor Cluster High Redshift Cluster
(z>0.5)
BCM (Brightest Cluster
Member)
BGM (Brightest Group
Member)
Gravitational Lens Galaxy Pair Interacting Galaxy Einstein Ring
Blank Sky

74 Chapter 6: Archive Search Strategies
Table 6.8: Descriptive Keywords for Interstellar Mediums (of the Galaxy or of an
External Galaxy)
Table 6.9: Descriptive Keywords for Unidentified Objects
Table 6.10: Descriptive Keywords for Calibration Observations
Descriptive Keywords for Interstellar Medium (ISM or EXT­MEDIUM)
Herbig­Haro Object Planetary Nebula HII Region Reflection Nebula
Dark Cloud SNR (Supernova Remnant) Ring Nebula (Shock Ionized) HI Cloud
High Velocity Cloud Intermediate Velocity Cloud IRAS Cirrus Cometary Nebula
Molecular Cloud Bipolar Outflow Absorption Line System
(Extragalactic)
Damped Lyman Alpha
Cloud (Extragalactic)
IGM ICM Coronal Gas (10 5 ­10 6 K) Hot Gas (10 7 ­10 8 K)
Descriptive Keywords for Unidentified Objects
Radio Emitter Infrared Emitter Optical Emitter
Ultraviolet Emitter X­Ray Emitter Gamma Ray Emitter
Descriptive Keywords for Calibration Observations
Astrometric Photometric Wavelength
Point Spread Function Flat Field Polarimetry
Aperture Location

Search Strategies 75
Table 6.11: Discrete Features and Keyword Descriptors for All Categories
Discrete Features of Objects
Corona Ring Ansae Protoplanetary Disk
Wind Accretion Disk Jet Lobe
Hotspot Nucleus Halo Disk
Bulge Polar Ringh Dust Lane Spiral Arm
Shell Tidal Tail Bar Multiple Nuclei
Cooling Flow Emission Line
Nebula
BLR (Broad Line Region) NLR (Narrow Line Region)
Filament Ejecta Knot Star Forming Region
Shock Front Ionization Front Conduction Front

76 Chapter 6: Archive Search Strategies
Table 6.12: Solar System Keywords
Searching on Proposal Title and Abstract
Besides searching for individual HST observations of a given source
class based on the target description, you may want to search for proposals
that concentrate on, or include, observations of a certain class of source.
The PROP_ABSTRACTS screen within the PRE­OBS menu contains a
summary of information about each accepted proposal, including the
proposal title, and the keywords from the proposal abstract, as well as the
proposal abstract itself. To find, for example, proposals for observations of
Seyfert galaxies, qualify on proposal title as *seyfert*. To
generalize the search to proposals that pertain to Seyfert galaxies, repeat
the search, unqualifying on proposal title, and qualifying on the
abstract keywords as *seyfert*. Having identified specific
proposal numbers of interest, you can then return to the GENERAL screen,
qualify on proposal number, and bring up all data sets that were taken
for that proposal.
Keyword Comments
PLANET If observation is centered on planet, may be followed by planet
name, e.g., PLANET JUPITER
SATELLITE If observation is centered on object, may be followed by name of
object, e.g., SATELLITE GANYMEDE
COMET If object is centered on comet's nucleus, may be followed by
name or catalog designation, e.g.,COMET 1979X
ASTEROID If target is asteroid center, may be followed by common name or
catalog number, e.g., ASTEROID CERES
FEATURE Surface features are followed by the name of the parent body,
e.g., FEATURE IO
OFFSET May be followed by name of parent body, e.g.,
OFFSET COMET HALLEY
RING Will be followed by name of parent body, e.g., RING SATURN
TORUS Will be followed by name of parent body, e.g., TORUS SATURN
OTHER May be followed by description of observation type, e.g.,
ASTROMETRIC REFERENCE or ZODIACAL LIGHT

Assessing Data Quality 77
Searches Based on Target Properties
The screens under the TARGET menu can be used to search for
observations of targets that share certain properties, e.g., magnitudes less
than some value, or redshifts higher than some value. This can be useful to
the archival researcher who wants to see, for example, the signal to noise
produced by an FOS spectrum of a given duration of a star of a given
magnitude, or what kind (and quality) of HST images of high redshift
galaxies exist in the archive. Some of the screens in the T ARGET menu
display a target name and proposal identification number and not the
dataset name. You can either mark data sets for retrieval directly using
these screens, or you can use the screens to obtain proposal ids and target
names of interest, and then use the GENERAL or INSTRUMENT screens,
qualifying on both proposal id and target name, to obtain more
information about the relevant observations. Remember that the
target­related information displayed on the screens under the TARGET
menu were entered by the observer. For many of the fields displayed on
these screens, the observer is not required to have filled in a value during
the proposal process. Thus, the information displayed will be incomplete
and heterogeneous.
Assessing Data Quality
The quality of data in the HST archive varies greatly. While it is difficult
to assess the quality of the data in a given HST exposure from the keyword
information alone (i.e., without actually obtaining and analyzing the data),
there are several keywords that are written to the DMF catalog and
displayed by STARCAT screens that can help you to get a feel for the
quality of the data in a given data set. Among these are:
. The PODPS (Post Observation Data Processing System) Data Quality
keywords and comments
. The exposure flag (expflag)
. The fine guidance system lock used to track the target (fgslock)
Each of these is described below.
When interpreting the relevance of these parameters for a given
observation, bear in mind that the importance of these parameters to data
quality is relative; it depends both on the characteristics of the source being
observed and the purpose to which you will put the data.

78 Chapter 6: Archive Search Strategies
PODPS Data Quality Keyword and Comments
Each science data set is displayed and evaluated for data quality by an
operator after it has been calibrated by the Post Observation Data
Processing System (PODPS). The operator assigns a parameterized
PODPS Data Quality (PDQ) keyword to the data set and writes an optional
PDQ comment. The PDQ keyword and PDQ comments are written to the
PDQ table of the DMF catalog and can be displayed in STARCAT using
the GENERAL screen (which displays the PDQ keyword and the first 68
characters of the PDQ comment) or the PDQ screen (which displays the
PDQ keyword and the full PDQ comment) within the POST­OBS menu.
When you are interpreting the relevance of the PDQ keyword and
comments, bear in mind that:
. The keyword and comments are not assigned by instrument scientists
. Data quality parameterization is inherently subjective
. The relevance of the keyword and comments depends on the nature of
the target of the observations
Care must be taken to combine a knowledge of the characteristics of the
target with the PDQ keyword and comments when evaluating data quality.
For example, while an evaluation of no­source when the target is a
bright star and the integration time sufficient that a clear detection is
expected may well indicate poor data quality, the same evaluation when the
target is a distant, faint cluster of galaxies for which co­adding of many
exposures is required indicates little about the quality of the data.
As described above, the PODPS Data Quality evaluation is composed of
a parameterized keyword and a comment. The comment describing the
data is fully at the discretion of the operator. Table 4.1 is the list of possible
values that the operators can assign to the parameterized PDQ keyword,
and a brief explanation of their meaning. This list of permissible keyword
values was officially adopted as of February 1, 1992. Data processed prior
to that date were described with a similar set of values, however some of
the original values were found not to be useful and were dropped from the
current list, while other values were added. Most of the values on the list
have been used since shortly after launch. Note that while several of the
allowed values may apply to a given observation, the operator can assign
only one value to the keyword for each observation. It should also be noted
that when data is reprocessed, it is not re­evaluated for data quality.

Assessing Data Quality 79
Table 4.1: PDQ Keywords
The PDQ Files
The complete PODPS data quality report can be retrieved from DMF.
Every PDQ file is archived as ancillary data (classASA) with the same
rootname as the observation and an extension ofPDQ. In addition to the
data quality keyword and comments, the complete PDQ file contains
predicted as well as actual observation parameters extracted from the
standard header and science header files. To mark a PDQ file for retrieval,
use the FILES screen under the DATAFILES menu, qualify on
Keyword Meaning
BADCHAN Quality significantly compromised by dead CCD
column(s) or bad diode(s)
COSRAY Compromised by cosmic­ray contamination
DATA­DROP More than about 2% missing, or affecting probable area
of interest
EDGEPSF Got only an edge of the point­spread function
INCOMPLETE Got only some of the expected data groups
No­COUNTS Zero­level data
NO­EVAL Inadequate time or resources for data evaluation
NO­SOURCE No sources visible in image
NOISY High background, low S/N
OK No apparent problems
POOR Other problems affecting probably scientific utility (used
when no other keyword applies)
SATURATED Target saturated; counts rolled over, etc.
TRAILED Observation taken on gyros, trailed sources, but
otherwise OK
UNDEREXP Observation seems underexposed
UNKNOWN Uncertain of usefulness or quality of data

80 Chapter 6: Archive Search Strategies
extension, PDQ, and dataset_name using the full dataset name (e.g.
W0340A01T). Press for FindNext and for Mark.
OSS Observer Comment Files
Observer comment files may help an archive user determine the quality
of a HST observation. The SOGS Observation Support System (OSS)
monitors the health of the telescope and provides observers with a
real­time interface. Observer comment files contain updated mission
information obtained from the time the observation was executed. OSS
personnel also put keywords and comments into this ASCII text file.
However, these files are not created for every observation executed.
If an observer comment file exists for an observation, the file may be
retrieved from the ancillary data class (ASA) of DMF. The extension for all
observer comment files is OCX. The filename follows the standard
observation rootname naming convention for the first eight characters, but
the last character is either an ``x'' or a ``c''. To retrieve the OCX file, you
need to mark the file using the FILES screen under the DATAFILES menu.
Qualify on extension OCX and dataset_name using the first 8
characters of the dataset name followed by a* (e.g., to search for theOCX
file for dataset W0340A01T, qualify on dataset_name =
W0340A01*). For data taken prior to April 17, 1992, theOCX file was not
always archived separately, and in some cases was appended to the trailer
file (extension TRL under archive classCAL). If you cannot find aOCX file
for your dataset, and the dataset was taken prior to April 1992, then you
may want to retrieve theTRL file to see if it has anOCX file appended to it
(use the FILES screen and qualify on the fulldataset_name and ext
= TRL).
Exposure Flag
The exposure flag keyword (expflag) serves as a comment to the
exposure time keyword (exptime). It indicates whether the exposure
completed successfully, without interruption, and whether the actual
exposure time was different from the predicted exposure time. (Note here
the difference between predicted exposure time and proposed exposure
time. The proposed exposure time includes setup time, while the predicted
exposure time is only the predicted on­source integration time.) The
exposure flag can be useful to the archival researcher in that it allows
him/her to determine whether the integration time obtained for a given
F X

Assessing Data Quality 81
observation is equal to the integration time the original observer expected
for that exposure. The details of how the value of this keyword is set are
complex and instrument­dependent. Table 4.2 gives a generic, description
of the possible values of this keyword and their meaning.
Table 4.2: Exposure Flag Keyword Values
Fine Guidance System Lock
The tracking mode employed during each HST observation is given in
the keyword fgslock which is displayed on the GENERAL and EXP
screens within the POST­OBS menu. There are currently three tracking
modes available during HST observations; coarse and fine lock which use
the Fine Guidance Sensors and gyro stabilization. When the HST is
stabilized with gyros, no guide star acquisition occurs and the absolute
error of positioning is 30". The anticipated guiding accuracies for the
three modes are 0.002"/sec drift rate for gyro hold, 0.015" RMS jitter for
Keyword Meaning
NORMAL EXPTIME was successfully calculated from telemetry
information, is equal to the predicted exposure time, and there is
no indication that the exposure was interrupted.EXPFLAG is
also set toNORMAL whenEXPTIME was successfully calculated
and the predicted exposure time was not available.
INTERRUPTED EXPTIME was successfully calculated from telemetry
information, is equal to the predicted exposure time, but there is
an indication that the exposure was interrupted.
INCOMPLETE EXPTIME was successfully calculated from telemetry
information and is not equal to the predicted exposure time.
EXTENDED EXPTIME was successfully calculated from telemetry
information and is greater than the predicted exposure time.
UNCERTAIN The Shutter Log Overflow flag was set, which indicates that not
all shutter open and close times are available.EXPTIME was
calculated from the shutter open and close times that are
available.
INDETERMINATE EXPTIME could not be successfully calculated from the
telemetry and the predicted exposure time was not available.
PREDICTED EXPTIME could not be successfully calculated from the
telemetry and EXPTIME was set to the predicted exposure time.

82 Chapter 6: Archive Search Strategies
Coarse track and 0.005" RMS jitter for Fine Lock. Obviously, the guiding
accuracy required by the archival researcher in a given observation
depends on the purposes for which the data are to be used.
The AEC and the EC
As described briefly in Chapter 2, there are two ASCII files that are
maintained in the exposure sub­directory of the documents directory
on the archive computer which may be useful to the archival researcher.
These are the Archived Exposures Catalog (AEC) and the Exposures
Catalog (EC). The AEC is a list of all the science observations (
CAL class
data sets) in the HST archive. It is updated monthly. The AEC contains
information about the observations including the target name, position,
instrument, mode, filters and gratings employed, and the date at which the
data become public (i.e., non­proprietary). The information provided in the
AEC is a selected subset of the keyword information in the DMF produced
through an automated search of the DMF catalog using STARCAT. Note
that observations taken with the Fine Guidance Sensors (FGS) arenot
included in the AEC. Positions in the AEC (and the EC) are in J2000
coordinates.
The EC is an ASCII listing of all guest observer (GO) and guaranteed
time observer (GTO) HST targets approved for observation. It contains a
listing of the proposal level parameters of all planned GO and GTO HST
exposures for which positional information has been entered in the
Proposal data base. Thus, it includes information for both executed and
planned HST observations. The EC listing is produced by the User Support
Branch of STScI. The principle purpose of the EC is to allow an
astronomer to determine in a straightforward way whether a given object or
position in the sky will be observed by the HST as part of an approved HST
program. This information is important to astronomers submitting
proposals to the HST, as it allows them to search for conflicts with
pre­existing programs. In addition to its role as a standalone catalog on the
archive host workstation, the EC has been made into a table within the
DMF catalog and can be queried by STARCAT using the EXP_CAT screen
under the PRE­OBS menu.
All of the information contained within the AEC and the EC is fully
available through the STARCAT interface, in the former case because the
AEC is produced by a STARCAT search of the DMF catalog and in the
latter case because the EC is a table within the DMF catalog that can be

The AEC and the EC 83
viewed with the EXP­CAT screen. However, the ASCII listings of these
catalogs provide the archival researcher with an added degree of flexibility.
First, because they are ASCII files, they can be examined easily, e.g., with
an editor, without the need to learn the STARCAT menu system. Second,
and perhaps more importantly, they can be used to cross­correlate existing
catalogs (e.g., the Abell Catalog, or a catalog of bright stars) with the
contents of the HST catalog. Such cross correlations are useful for looking
for all observations of a given source type (e.g., all positional coincidences
with Abell clusters) or for finding observations that might prove useful for
obtaining point spread functions. Cross correlating catalogs with the HST
catalog is not a capability currently supported by STARCAT and so can
only be done using these ASCII listings.

84 Chapter 6: Archive Search Strategies