[vsnet-chat 3639] re overobserving

[crawl@zoom.co.uk]

Okay, lets not get the dogma stick out here....

ESA SP-402 "Hipparcos Venice 1997" contains at least one paper relating to
rapid variations in Miras.  Hipparcos short term photometry is
"problematic", but ground based detections of these events were known
previously, and also confirming work done.

The following [one line] link enables you to get to download the FULL
ARTICLE on this.

http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=1997ESASP.402..387L&;
db_key=AST&high=390983001608628

I don't know how follow up is going, but AAVSO were gonna do summat about
it observing wise.

In other words, short term nightly variations in Miras occur quite
regularly for some stars, but no bugger's gonna see 'em if they don't look.


As for the over observing / over reporting bit.  People should do what they
want on both counts.

I went on about this a while ago on the MSDG list.  Most Miras and SR stars
are looked at about two to three times a month, probably when there's
little moon about.  This means in a certain geographic region most folk are
probably observing the Mira in question at very much the same time [what
with weather etc], yet nobody ever bothers about the weighting this causes
in the data, and so on and so forth.

You can see massive 1 year and 6 months relative peaks in the spectral
windows of quite a few Miras, even near circumpolar ones like V Boo.
Whether that is one of the Carnevali effects or the swap from binocs to
scopes I dunno.  But I bet if I looked I could find fortnightly and monthly
timing regimes in the data too, and all due to how the Miras are observed
by the standard lore.

There's more weighting troubles nowadays from the fact that folk multi-post
to several organisations via t'web.  This can cause serious weighting when
merging data from several organisations.


Anyway, folk never seem to take into account how the data is actually
_analysed_, when they comment on these things.

When I used to observe I'd include all me less thans and me approximate
measures in me reports, yet when I analyse data I instantaneously pluck
these out and throw them away... ...they are just no use for LPVs.


So if folk want to observe some objects very regularly, they can.  If they
want to report 'em regular, they can.  Datasets include observer IDs,
analysts will just have to put in more effort, digit extraction wise.

For every argument agin it I can think up one, with examples, for it.

Okay, so it precludes the number of variables looked at by any one
observer, but that's the personal choice of the observer...

...and so on, and so forth, etc, ... ...I'll leave it there afore I go on a
bit too much and start getting ratty

John

VSNET Home Page

Return to Daisaku Nogami

---

vsnet-adm@kusastro.kyoto-u.ac.jp