Common Questions: Scientific Calculations
Questions about Scientific Calculations
(Latest revision: 9 December 2010)
Q:How do you choose the coordinates for NED objects?
- How do you calculate the Galactic extinction values?
- Galactic extinction is presented as total absorption Aλ in
magnitudes, as calculated by three different methods.
- The first estimate of Galactic extinction uses the
Schlafly & Finkbeiner 2011 (ApJ 737, 103, 2011) recalibration of the
Schlegel, Finkbeiner & Davis 1998 (ApJ 500, 525, 1998; SFD98) extinction map
based on dust emission measured by COBE/DIRBE and IRAS/ISSA. The recalibration assumes a
Fitzpatrick (1999) reddening law with Rv = 3.1 and
a different source spectrum than SFD98.
The original SFD98 extinction values are also returned for comparison purposes.
The individual values of the total absorption at each waveband are calculated
from the list of A/E(B-V) in Table 6 of Schlegel
et al (ApJ 500, 525, 1998).
We have adopted the standard Landolt UBVRI and SDSS ugriz filters for
the optical total absorptions, and the UKIRT JHKL' filters for the
near-infrared total absorptions. Please note that Schlegel et al.
calculated the values of A/E(B-V) for these specific bandpasses using a spectral energy
distribution for an elliptical galaxy.
Therefore, the numbers displayed by NED for a
specific object may not be appropriate for other closely related bandpasses
or other galaxy types. The total absorptions are nominally consistent with
an average R = A/E(B-V) = 3.1, but do not agree numerically with this
See Appendix B of Schlegel et al
(ApJ 500, 525, 1998)
and references therein for additional details.
Note, too, the list of caveats in
SFD98 Appendix C. In particular, they call attention to the areas within the
Holmberg radii of LMC, SMC, and M31 -- total reddenings through these
large galaxies are replaced by Galactic reddenings toward them. They
also note that no contaminating sources at Galactic latitudes |b| < 5 degrees
have been removed from their dust maps, so calculated reddenings at these low
latitudes are especially uncertain and untrustworthy.
They state that the formal uncertainty in normalizing the dust column density
to the reddenings is 10%; this should probably be taken as a lower limit on
the formal error of the calculated reddenings at |b| > 5 degrees.
A few other galaxies may have unreliable Galactic extinction values as well.
An example is M82 (see Johnson et al
ApJ 697, 1138, 2009, Section 6.3)
where emission by dust within the galaxy has apparently affected the Schlegel
et al estimate of the foreground Galactic extinction. We thank L. C. Johnson
for alerting us to the M82 problem.
The Burstein-Heiles total B-band absorption is given as
AB = 4 E(B-V) + 0.005, consistent with AB = 4 E(B-V) in
Burstein and Heiles (ApJS 54, 33, 1984 and references therein)
but with a small adjustment to the zero point (Burstein, 1988,
private communication). Burstein and Heiles used HI column densities
combined with the Lick galaxy counts to determine the extinction for
any object with |b| > 10 degrees (aside from a few small patches of sky
where they had no HI data). South of -23 degrees, where the Lick counts
stop, the Burstein/Heiles Galactic extinction estimates depend only on
the HI column densities.
We note that we have replaced the Burstein-Heiles extinction
toward M31 with values that are not affected by the HI emission of M31
itself (see ApJS 54, 33, 1984 for a discussion).
This affects not only M31,
but several thousand objects within that galaxy. Specifically, we have
replaced "data word numbers" 402-405 in "physical record number" 20 in the
Burstein-Heiles reddening file "redsouth.dat" with the E(B-V) values
0.068, 0.071, 0.074, and 0.077 (interpolated from surrounding areas),
respectively. The original values were 0.043, 0.034, 0.024, and 0.045.
(We thank Tod Lauer for alerting us to this problem.)
One additional note:
The zero points of these two reddening laws differ by 0.02 magnitudes in
E(B-V), with Schlegel et al. adopting a higher zero point than
Burstein and Heiles. (We thank David Burstein for this note.)
Willick (ApJ 522, 647, 1999)
adds the following notes and caution concerning the
Galactic extinction calculations:
"Two all-sky Galactic extinction maps are presently available:
the older Burstein-Heiles (Burstein and Heiles,ApJ 225, 40, 1978,
hereafter BH, and
AJ 87, 1165, 1982)
maps, which are based on 21-cm column density and faint
galaxy counts, and the more recently completed Schlegel, Finkbeiner, and
Davis (ApJ 500, 525, 1998,
hereafter SFD) maps, based on IRAS/DIRBE measurements
of diffuse IR emission. The SFD extinctions have been favored in
several recent analyses and, indeed, were used in Paper I
(Willick,ApJ 516, 47, 1999). Unlike
BH, the SFD extinctions are based directly on dust emission and
have comparatively high spatial resolution. However, it has not
been established beyond doubt that they are free of systematic
errors, such as could arise from the presence of cold dust invisible
to IRAS. The BH extinctions are also vulnerable to possible systematic
effects, such as a variable dust-to-gas ratio and galaxy count
fluctuations. Thus, it seems prudent to use both methods, or linear
combinations of them, and see what effect this has on the results."
We try to adopt the best published position for each
object. In many cases, this will be the position in
the catalog in which the object originally appeared:
IRAS sources originally had positions from the IRAS PSC or FSC,
NVSS sources originally had positions from NVSS, and so forth.
Because NED's source hierarchy is based on physical
models for the sources, we try to use optical positions
for sources associated with galaxies, clusters, and so
on. However, if a galaxy, for example, is known to be
associated with a compact nuclear radio source, and if
a better position at a radio wavelength is available,
we will adopt the radio position. Similarly, if an IRAS source is
identified with a galaxy, and if an accurate optical
position is available for that galaxy, we will choose
the optical coordinates in preference to the IRAS position.
We are continually updating NED's positions as better
coordinates are measured and published. We also measure
positions (primarily from the Digitized Sky Survey) to help
sort out problem areas of the sky, or to resolve discrepant
Q:Is there an easy way to precess my coordinates?
A:NED offers an easy-to-use
coordinate calculator. This provides
coordinate transformation, precession, and position angle
calculations, as well as Schlegel et al Galactic extinction.
The coordinate calculator is flexible enough to convert accurately
between Besselian and Julian equinoxes, taking the epoch of observation
into account when needed. It assumes that Besselian dates refer to FK4
system, that Julian dates refer to the FK5 system, and makes the
Q:What precession routine has NED adopted?
A:NED adopted the FK5 system in 1992 with J2000.0 as
the default equinox. The precession routine is essentially that described in
the introduction to FK5, with minor changes to support the work then being
done at IPAC. J. Bennett, the author of the routine, has provided an
extensive description of NED's
precession calculations. Though NED does not yet explicitly support the
International Celestial Reference Frame (ICRF),
the FK5 optical system is consistent with ICRF to within the known errors of
the FK5 system (see e.g. Ma et al. AJ 116, 516, 1998).
Q:Is there an easy way to change coordinate systems?
A:NED's coordinate calculator
(see the previous question) will handle transformation from one coordinate system to another.
It currently handles transformations among the ecliptic, equatorial, Galactic,
and supergalactic coordinate systems.
Q:What kind of magnitude is listed in the Basic Data?
These are usually optical magnitudes taken from
the astronomical literature, and should be understood as being indicative
only. We are adding letters after the magnitudes indicating the band pass
to which the magnitude applies. For example, we use "U", "B", "V", "R", and
"I" for the standard Johnson and/or Cousins magnitudes in the optical; "p" for
photographic magnitudes from e.g. IIa-O or 103a-O plates, "g" for Gunn g-band
magnitudes, "j" for magnitudes from III-aJ plates, "J" for 2MASS near-IR
magnitudes, and so on.
We will eventually have all of NED's Basic Data magnitudes flagged with
the band passes. In the meantime, the magnitude may have already been
included in NED's table of of referenced
Q:What are the sources for the photometric data in NED?
Most of NED's detailed photometric data
are taken from the larger catalogs, though we have also loaded many
shorter lists from the journals as well. The Basic Data for each
object, displayed on the Object Search Results pages, are from
heterogeneous unreferenced sources, typically measured in the optical
or near-infrared part of the spectrum (see the previous question).
The data displayed on the
Photometric Data Search Results pages are fully referenced. These data
may apply to any part of the spectrum from gamma rays to radio.
>Q:Have the photometric data in NED been transformed or reduced in any way?
NED displays its collection of detailed photometry in two ways.
First, the "data as published" and enough information to understand and
use them are given. Then, NED applies the appropriate flux conversion factor,
if needed, to reduce the published data to a common set of MKS units:
W/m2/Hz (1 Jy = 10-26 W/m2/Hz).
NED has adopted effective wavelengths, band widths, and absolute
calibrations for the different photometric systems from the papers containing
the data, or from the papers originally defining the photometric systems.
If this information is not given in the original papers, we have assumed
the values in this table.
Q:Why are there negative fluxes and flux densities in NED?
NED attempts to represent the published literature as accurately as possible.
Sometimes this will lead us to include negative fluxes and flux densities,
and/or measurement errors larger than the detections themselves.
In these cases, we encourage you to read the original papers to check the
validity of the flux and flux density values.
Q:What kind of redshifts are used in NED?
We use heliocentric redshifts. When we display them as velocities, we use
the optical convention Vsun = czsun.
Note that we do not
make any relativistic correction to these velocities.
John Huchra at the Harvard-Smithsonian Center for Astrophysics
has kindly supplied more information on the
definition of extragalactic redshifts.
Q:Some of the reported redshifts in NED seem to show discrepancies
with my estimates. Why?
A:As with the celestial coordinates which we adopt
for the objects in NED, we attempt to use the best available redshifts.
However, occasional errors creep into NED, in spite of our best efforts.
If you find one of these cases, please let us know so that we can fix NED and
document the origin of the error.
Q:Is there any information which explains the jargon used in the photometry of objects in NED?
As with the redshifts, we encourage you to go back to the original paper
to fully understand the magnitudes adopted by NED.
Here are a few examples of magnitudes currently found in NED's detailed
- u, g, r, i, or z usually refers to the five-band SDSS photometric system
- B is a B-band magnitude on the Johnson system
- BT is a total magnitude in the B-band
- BT0 is a total magnitude in the B-band corrected
to "face-on" (i.e. inclination = 0 degrees)
- bj is approximately a B magnitude derived from photometry
on a IIIa-J plate
- R25 is an R magnitude at the 25th mag arcsec-2
isophote levelJ, H, or Ks usually refers to the three-band 2MASS photometic system