WFPDB metadata
The Wide-Field Plate Database (WFPDB)[1] contains descriptive information for astronomical wide-field (>1°) photographic observations stored in numerous archives all over the world.
The WFPDB provides for each observation information for the corresponding archive, the parameters of the observational instrument, the observation parameters (position on sky, observation time, object name, method, exposure time, emulsion type, filter type, spectral band, plate size), as well as data on the plate quality, comments, and observers.
The entire observational information is stored in four metadata files as published on VizieR[2]: maindata, notes, quality and observer. (The other file archives mentioned there is strictly speaking intended for the Catalogue of WFPA[3].) These files are written in plain ASCII text und use fixed formats. Below is a description for each file structure with examples.
Maindata
The file maindata contains metadata for one plate per line. Each line is 104 characters long.
General notes:
- the column Bytes shows the character positions assigned to a metadata unit in the line
- the column Format shows the type and character length of a metadata unit
- the column Units shows the measurement units used for a metadata unit
- the column Label gives an arbitrary label of a metadata unit
- the column Explanation gives a short description of a metadata unit
- WFPDB observatory identifier is a three-letter obervatory code as given in the Catalogue of WFPA
- WFPDB plate number is not necessarily the same with the exact original plate number, sometimes differences in suffix letters do arise
Bytes Format Units Label Explanation ------------------------------------------------------------------------------------------------ 1- 3 A3 --- IDobs WFPDB observatory identifier 4- 6 I3 cm IDins Instrument aperture 7 A1 --- IDsuf1 [A-Z] Suffix to the instrument identifier 8- 13 I6 --- IDno WFPDB plate number 14 A1 --- IDsuf2 [A-Z] Suffix to the WFPDB plate number 15- 16 I2 h RAh Right ascension (hours) (J2000.0) 17- 18 I2 min RAm Right ascension (minutes) 19- 20 I2 s RAs Right ascension (seconds) 21 A1 --- DE- Declination sign (J2000.0) 22- 23 I2 deg DEd Declination (degrees) 24- 25 I2 arcmin DEm Declination (arcminutes) 26- 27 I2 arcsec DEs Declination (arcseconds) 28 A1 --- CCOD [EMU] Code for Error, Missing data, or Uncertainty of coordinates 29- 32 I4 yr DATEy Date of observation, year (UT) 33- 34 I2 month DATEm Date of observation, month 35- 36 I2 d DATEd Date of observation, day 37- 38 I2 h UTh Observation time (hour) (UT) 39- 40 I2 min UTm Observation time (min) 41- 42 I2 s UTs Observation time (sec) 43 A1 --- TCOD [EMU] Code for Error, Missing data, or Uncertainty of observation time 44- 63 A20 --- OBJNAM Object or field designation 64- 65 A2 --- OBJTYP Object type code 66- 67 I2 --- METHOD Method of observation code 68- 69 I2 --- MULTEX Multiplicity of exposure 70- 75 F6.1 min EXP Exposure time 76- 86 A11 --- EMULS Emulsion type 87- 93 A7 --- FILT Filter type 94- 95 A2 --- SPEC Spectral band 96- 97 I2 cm DIMx X dimension of plate 98- 99 I2 cm DIMy Y dimension of plate 100 I1 --- PQUAL [0,1] Pointer to file 'quality' 101 I1 --- PNOT [0,1] Pointer to file 'notes' 102 I1 --- POBS [0,1] Pointer to file 'observer' 103 I1 --- PAVA [0,1] Pointer to file 'availability' 104 I1 --- PDIG [0,1] Pointer to file 'digitization'
WFPDB's codes for object type are:
A1 - planet
A2 - moon
A3 - sun
A4 - asteroid
A5 - comet
A6 - meteor
A7 - artificial satellite
S1 - star
S2 - double or multiple star
S3 - variable star
S4 - stellar cluster
S5 - HII region
S6 - nebula
S7 - planetary nebula
S8 - supernova or SN remnants
S9 - fundamental star
SA - stellar association
SD - dark nebula
SH - Herbig-Haro object
SM - molecular cloud
SP - pulsar
SR - reference star around a radio source
G1 - galaxy
G2 - QSO
G3 - group of galaxies
G4 - cluster of galaxies
G5 - supercluster
G6 - void
G7 - radio galaxy
GR - gamma-ray source
F - field
XR - x-ray source
RS - radio source
IR - infrared source
U - object of unknown nature
For stars (codes S1-S3), it is advisable to follow SIMBAD's object type of each particular star and assign WFPDB's object type codes as follows:
1 - direct photograph
2 - direct photograph (multi-exposure)
3 - stellar tracks
4 - objective prism
5 - objective prism (multi-exposure)
6 - Metcalf's method
7 - proper motions
8 - no guiding
9 - out of focus
10 - test plate
11 - Hartmann test
12 - with mask
13 - focusing
14 - sub-beam (Pickering) prism
15 - raster scan/trail
24 - objective grating
25 - objective grating (multi-exposure)
Example:
POT017 004781 M193403052016 Y Dra S30101 60.0Superpan G 091211100
This metadata set describes the focus value during the observation and an observer note.
Quality
The file quality contains metadata for one plate per line. Each line is 80 characters long. The idea behind this file is to provide information about the quality of the plate itself, observing conditions and any other aspects which might have affected the observation.
Example:
POT017 004798 temperature: +6; cirri during the last 10 minutes
This metadata set provides air temperature and sky conditions during the observation.
Observer
The file observer contains metadata for one plate per line. Each line is 57 characters long. Its only purpose is to mention observers' names.
Example:
POT017 004781 R.Mueller
This metadata set just tells the name of the person, who performed the observation of the given plate.
Discussion
The WFPDB approach to handle metadata certainly has many advantages, the most important of which is the easier possibility to parse metadata by software. Nevertheless, there are still a lot of unclear, problematic or redundant points therein. Some of them are:
- the allocated length of text in each line in the files notes and quality is restricted to only 65 characters, which is often not enough; a similar shortage of space is present for the object and emulsion names in the file maindata
- there is no ready solution within the WFPDB metadata structure to handle plates with two different overexposed fields (e.g., the field of the observation target and a comparison field)
- if multiple close objects are simultaneously the observation target, then it's impossible to assign a single object type code, when these objects are of different types
- the exposure time in WFPDB has to be given in minutes with one decimal, which is not a standard way in astronomy (seconds)
- the files notes, quality and observer only have a continuation sign at the byte position 15, it is not included in the file maindata, which is a rather confusing fact
- the code M for missing data (coordinates and/or observation time) is strictly speaking not necessary, because the absence of data itself is already self-descriptive
- the explanations published on VizieR include original plate number, which is just incorrect, because it has to be a WFPDB plate number, which can be different from the original plate number
- the future status and structure of the files availability and digitization is still unclear
See also
Extended WFPDB metadata structure