<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<p>Dear Mireille and Semantics,<br>
<br>
Thank you also for the reply to my "wavelength-dependent UCD
question". I am happy that you conclude that my selection process
for this UCD word is appropriate. And I note with interest your
work on providing a fuller description of the filter profiles.<br>
<br>
Kind regards, Bo<br>
</p>
<div class="moz-cite-prefix">On 3/8/22 9:44 PM, Mireille LOUYS
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:4104f494-b021-3953-9b4d-b16533fcf134@unistra.fr">
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
Dear Bo, Dear Semantics,<br>
thanks for this interesting feedback.<br>
<br>
<br>
The SVO Filter profile service contains a lot of Filter
descriptions and you can query what is stored with their search
box : <br>
<a moz-do-not-send="true"
href="http://svo2.cab.inta-csic.es/theory/fps/index.php?id=Subaru/Suprime.IB505&&mode=search&search_facility=Subaru&search_instrument=Suprime&search_text=suprime#filter">http://svo2.cab.inta-csic.es/theory/fps/index.php?id=Subaru/Suprime.IB505&&mode=search&search_facility=Subaru&search_instrument=Suprime&search_text=suprime#filter</a><br>
<br>
You see that many properties are recorded for a filter, much more
than the UCD label you can assign on the photometric column. <br>
<br>
In the IVOA we can describe filters properties using the
Photometry data model v1.0: <br>
<a href="https://www.ivoa.net/documents/PHOTDM/"
class="moz-txt-link-freetext" moz-do-not-send="true">https://www.ivoa.net/documents/PHOTDM/</a><br>
The way to provide more information to describe filters and
calibration zero points at the top of a VOTable catalog was
proposed in the <br>
PhotDM1.0 Appendix C.2 Photometric Data in Cone Search in the
doc pointed above.<br>
<br>
The current update of this spec : PHOTDM1.1 is in progress and
discussed in the community .<br>
You can see details on the revision process , people involved ,
etc, here : <br>
<p class="MsoNormal"><a
href="https://wiki.ivoa.net/twiki/bin/view/IVOA/PhotDM11RFC"
class="moz-txt-link-freetext" moz-do-not-send="true">https://wiki.ivoa.net/twiki/bin/view/IVOA/PhotDM11RFC</a></p>
<p class="MsoNormal">The goal is to set-up a mechanism to allow
any filter and calibration information to be described in an
annotation block at the beginning of a VOtable which in turn can
be interpreted by a parser application based on a library. </p>
<p class="MsoNormal">This work is in progress and will be shown at
the next interoperability meeting (April) probably .</p>
<p class="MsoNormal">Your example is up-to-the point and in sync
with our current development.</p>
<p class="MsoNormal"> I hope we can show you examples of this
representation for your filters in this syntax very soon. <br>
</p>
Best regards , Mireille.<br>
<br>
PS: So your selection process for ucd , based on the reference
wavelength/frequency/energy, is appropriate. It is just that not
all information can be wrapped in just one label.<br>
<br>
<div class="moz-cite-prefix">Le 02/03/2022 à 18:00, Bo
Milvang-Jensen a écrit :<br>
</div>
<blockquote type="cite"
cite="mid:3c0df75c-bac9-b672-9c4a-9d055d74fc7d@astro.ku.dk">Dear
Mireille and IVOA semantics list, <br>
<br>
Thank you for inviting me to send you my unsolved questions
about UCDs, I have two: one in this e-mail and one in the next <br>
<br>
Background: These questions have arisen while writing UCDs for
the around 1200 table columns in the COSMOS2020 catalogue, paper
recently published in Weaver+2022 (<a
class="moz-txt-link-freetext"
href="https://ui.adsabs.harvard.edu/abs/2022ApJS..258...11W/abstract"
moz-do-not-send="true">https://ui.adsabs.harvard.edu/abs/2022ApJS..258...11W/abstract</a>),
before publication via ESO's Phase 3. The catalogue contains
photometry in around 40 photometric bands/filters, and photo-zs
and physical parameters (e.g. stellar mass) computed by two
codes. <br>
<br>
My first question (i.e. this e-mail) concerns how to select the
wavelength-dependent (em) UCD word for a given photometric
filter. I have consulted the UCD1+ document, but the only
guidance I found was the short description for each UCD word,
e.g. "Infrared between 1.0 and 1.5 micron". First, I am not sure
I understand the motivation for having overlapping words, as
shown by this example: <br>
<br>
em Electromagnetic spectrum <br>
em.IR Infrared part of the spectrum <br>
em.IR.NIR Near-Infrared, 1-5 microns <br>
em.IR.J Infrared between 1.0 and 1.5 micron <br>
<br>
My intuition is to select the most "specific" word (here
em.IR.J), but that could be wrong. <br>
<br>
Second, my main problem is that I do not know to select the most
appropriate em word. For each of my filters I know the central
wavelength and the full transmission curve. Some of my filters,
e.g. y and Y, have a transmission curve that transmits both
below and above 1 micron, which is one of the limits used in the
descriptions of the UCD words, so if one were to consider the
full transmission curve when selecting the em word, these
filters could only be given the very general word "em". I have
made the choice of only using the central wavelength when
assigning the em word, not the full transmission curve. <br>
<br>
Here is the full list of filters in COSMOS2020, with their
central wavelengths and widths (both in Angstrom), and with the
UCD I currently have assigned based on my two principles (most
specific word, and central wavelength) [see Weaver+2022 Fig. 2
for the filter curves]: <br>
<br>
Label Description lam_cen width UCD <br>
--------------------------------------------------------------------------
<br>
GALEX_FUV GALEX FUV 1526 224
em.UV.100-200nm <br>
GALEX_NUV GALEX NUV 2307 791
em.UV.200-300nm <br>
CFHT_u CFHT/MegaCam u 3709 518
em.opt.U <br>
CFHT_ustar CFHT/MegaCam u* 3858 598
em.opt.U <br>
ACS_F814W HST/ACS F814W 8333 2511
em.opt.I <br>
HSC_g Subaru/HSC HSC-SSP PDR2 g 4847 1383
em.opt.B <br>
HSC_r Subaru/HSC HSC-SSP PDR2 r 6219 1547
em.opt.R <br>
HSC_i Subaru/HSC HSC-SSP PDR2 i 7699 1471
em.opt.I <br>
HSC_z Subaru/HSC HSC-SSP PDR2 z 8894 766
em.opt.I <br>
HSC_y Subaru/HSC HSC-SSP PDR2 y 9761 786
em.opt.I <br>
SC_B Subaru/SuprimeCam B 4488 892
em.opt.B <br>
SC_gp Subaru/SuprimeCam g+ 4804 1265
em.opt.B <br>
SC_V Subaru/SuprimeCam V 5487 954
em.opt.V <br>
SC_rp Subaru/SuprimeCam r+ 6305 1376
em.opt.R <br>
SC_ip Subaru/SuprimeCam i+ 7693 1497
em.opt.I <br>
SC_zp Subaru/SuprimeCam z+ 8978 847
em.opt.I <br>
SC_zpp Subaru/SuprimeCam z++ 9063 1335
em.opt.I <br>
SC_IB427 Subaru/SuprimeCam IB427 4266 207
em.opt.B <br>
SC_IB464 Subaru/SuprimeCam IB464 4635 218
em.opt.B <br>
SC_IA484 Subaru/SuprimeCam IA484 4851 229
em.opt.B <br>
SC_IB505 Subaru/SuprimeCam IB505 5064 231
em.opt.V <br>
SC_IA527 Subaru/SuprimeCam IA527 5261 243
em.opt.V <br>
SC_IB574 Subaru/SuprimeCam IB574 5766 273
em.opt.V <br>
SC_IA624 Subaru/SuprimeCam IA624 6232 300
em.opt.R <br>
SC_IA679 Subaru/SuprimeCam IA679 6780 336
em.opt.R <br>
SC_IB709 Subaru/SuprimeCam IB709 7073 316
em.opt.R <br>
SC_IA738 Subaru/SuprimeCam IA738 7361 324
em.opt.R <br>
SC_IA767 Subaru/SuprimeCam IA767 7694 365
em.opt.I <br>
SC_IB827 Subaru/SuprimeCam IB827 8243 343
em.opt.I <br>
SC_NB711 Subaru/SuprimeCam NB711 7121 72 em.opt.R
<br>
SC_NB816 Subaru/SuprimeCam NB816 8150 120
em.opt.I <br>
UVISTA_Y VISTA/VIRCAM UltraVISTA DR4 Y 10216 923 em.IR.J
<br>
UVISTA_J VISTA/VIRCAM UltraVISTA DR4 J 12525 1718
em.IR.J <br>
UVISTA_H VISTA/VIRCAM UltraVISTA DR4 H 16466 2905
em.IR.H <br>
UVISTA_Ks VISTA/VIRCAM UltraVISTA DR4 Ks 21557 3074
em.IR.K <br>
UVISTA_NB118 VISTA/VIRCAM UltraVISTA DR4 NB118 11909 112 em.IR.J
<br>
IRAC_CH1 Spitzer/IRAC ch1 35686 7443
em.IR.3-4um <br>
IRAC_CH2 Spitzer/IRAC ch2 45067 10119
em.IR.4-8um <br>
IRAC_CH3 Spitzer/IRAC ch3 57788 14082
em.IR.4-8um <br>
IRAC_CH4 Spitzer/IRAC ch4 79958 28796
em.IR.4-8um <br>
<br>
However, my ESO Phase 3 scientist Laura (in cc) thought it could
be misleading to assign em.IR.J to the Y and NB118 filters, but
also said that ESO needed to look into standardizing this. If my
UCDs for these two filters could be misleading, so could those
for all the intermediate bands (IA, IB), the other narrow bands,
and the broad bands that don't have a UCD with its name encoded
(e.g. g, z). <br>
<br>
I would be interested in your view on this. <br>
<br>
Kind regards, Bo <br>
<br>
On 3/1/22 5:41 PM, Mireille LOUYS wrote: <br>
<blockquote type="cite">hi again , <br>
please do not hesitate to send your questions to me , or to
the ivoa semantics list in copy . <br>
this helps the community and maintainers like me to be aware
of the trouble you may encounter. <br>
<br>
many thanks , Mireille <br>
<br>
Le 25/02/2022 à 16:59, Bo Milvang-Jensen a écrit : <br>
<blockquote type="cite"> <br>
PS: I have a couple of other questions/problems with the UCD
standard that I would like to ask/report. What is the best
channel? <br>
<br>
Kind regards, Bo <br>
</blockquote>
<br>
</blockquote>
</blockquote>
<br>
<pre class="moz-signature" cols="72">--
--
Mireille Louys, MCF (Associate Professor)
Centre de données CDS IPSEO, Images, Laboratoire Icube
Observatoire de Strasbourg Telecom Physique Strasbourg
11 rue de l'Université 300, Bd Sebastien Brandt CS 10413
F- 67000-STRASBOURG F-67412 ILLKIRCH Cedex
Tel: +33 3 68 85 24 34</pre>
</blockquote>
</body>
</html>