Attached is a word document taken from the September 2011 BAA VSS Circular. This should help in the over-observing of select stars and variable star types. Kevin B. Paxson - PKV
Oops... Here is the attachment...
Thanks, PKV. This listing offers at least something for our observers to currently refer to in planning their night's observing run. It would be nice to see AAVSO, after some thought, officially post such a listing of its own both prominently and permanently.
I would note that in the case of the AAVSO in years gone by and stretching far into the past it was always cited by HQ that a ten day interval between observations of Mira variables and slowly varying semi-regulars was the most desirable, since our lightcurves were always based on 10-day means of the data. While this arrangement may sometimes prove inconvenient where the observer rarely has clear skies, the reasoning for the rule I feel was quite valid. It was meant to prevented any one observer's data from gaining an excessive weight in a 10-day mean by his submission of multiple estimates within the interval.
I suppose that if we were all "perfect" observers then no harm would really be done by over observing. Unfortunately, this is hardly ever the actual situation, as I quickly learned through my editorship of the Circular. One needs only a few low quality, or significantly zero point shifted, estimates within a given 10-day period to distort a lightcurve, or imply activity that never really occurred. That's something we really don't need, especially when it can be so easily avoided.
I totally agree. Recommended observation intervals may be buried in the Observing Manual somewhere, but I posted this for the benefit of the newer members and observers. Revising this document and prominently posting it somewhere on the AAVSO website is a great idea. Anything to help improve the quality of our data (visual and CCD) is worthwhile.
Though I am not a veteran observer by a long shot, my visual cadence is usually 5 to 7 days for Mira's and 3 to 5 days for RV Tauri stars. All other stars are 5 to 7 days, except for Z Ands, NR and CV's. My CCD cadence for Mira's is about 6 to 7 days. The rest of my CCD observing is for CV outbursts and my cadence there is every few days at a maximum.
One question that could be asked is "What is the desired cadence in estimates per cycle for a perfect sine wave shaped light curve?" It guess it depends on period and amplitude, but a sample every 0.3 to 0.5 magnitudes seems adequate. Thoughts anyone?
Kevin Paxson - PKV
To a degree the answer depends on just how the question posed was meant, Kevin. On a personal level, except in an instance where the individual observer wishes for some reason to generate a lightcurve based solely on his own data, the desired cadence of observations of any slowly varying star remains quite adequate as 10-days. The beauty of working collectively is that regardless of the period of the variable others are anticipated to be filling in the lightcurve with their combined data. This best creates the supposedly unbiased mean lightcurve of the object over a complete cycle.
Of course, if one is talking about a lightcurve generated by HQ, then we are speaking of an entirely different matter. In that instance of course, selecting the proper interval between generated "means" would be their educated choice for producing the best/smoothest (if it happens to really be so) curve. I suppose that any one of us can do the very same these days if so desired for fun through the use of the our wonderful AAVSO LCG, a creation beyond the wildest of dreams when I first joined the AAVSO 50 years ago.
The recommended observing frequency is listed in Table 6.2 of the Visual Observing Manual; for LPVs like Miras, it recommends once per week. I think once/10days may have been fine in the years when we had lots of active visual observers. Today, I think a higher frequency is needed, if nothing else than to ensure that the Legacy stars at least get observed by someone, even if the light curve gets defined by only a handful of observers. Once/week also seems like something easier to remember.
The traditional cadence in the digital world is 1/100 period for periodic variables so that you get a well-sampled light curve. I think it is important to consider using this cadence for visual observing as well. The Bright Star Monitor, for example, has found that many Miras have bumps and wiggles in their light curves that may be missed with longer cadence by visual observers (also noted in the LPV humps campaign), and LPVs come in many flavors, from about 100day periods to over 1000days. Using a 10-day cadence on a 100-day variable means you aren't sampling it very often. So a general rule may be good for a starting point, but I'd suggest adjusting as needed. I'd like to hear from the expert visual observers like John regarding their cadence on the short-period stars.
The other extreme is observing too often. Observing every 5 minutes on a Mira doesn't have any known astrophysical value - high frequency oscillations are not predicted for these big stars; planetary transits take many days since the orbit has to be outside the stellar surface; flares are not known on these big stars. Over-observing a star means you have to under-observe something else on the Legacy list; I'd rather those stars get well-covered. If you are a CV observer on the other hand, high-cadence observations have value since these small bodies can flicker with quite short periods. CVs can benefit from almost all cadences - the once/day measures can catch new outbursts, the 5-second digital estimates can measure the eclipses, etc.
Thanks for the clarification. Perhaps the recommended minimal observation intervals can also be listed for each variable star type on the page entitled "Types of Variables" under the "Variable Stars" tab. I believe that this is too important to only be listed in the Observing Manual only.
I recognize a good idea when I see it, so I have created a table of variable star types and observing cadences and linked it from the Observers Page under For New Observers.
How Often Should I Observe My Stars?
This is one of those questions that comes up often with new observers and is debated by experienced observers whenever serious discussion of variable star observing, particularly LPVs and semi-regulars, ensues. I agree that it needs to be easily located on our website, something that is often easier said than done. This is the most logical place I can think to put it. Thanks for the suggestion.
Thanks to our observing kindred of the BAA VSS for this resource.
Heck, why not have the chart plotter put the recommended interval right on the bottom area of the chart according to the var type? Of course, that might be making things too easy... :)
It might be prudent to also put the following boilerplate at the bottom of all AAVSO charts:
You can never be too careful these days.
Didn't AAVSO previously recommend a five-day interval for observations of RCB and symbiotic stars when they are not active? I am vaguely remembering some document from perhaps 20 years ago whose contents I memorized for the stars that I was observing.
I have long wished that "recommended frequency of observation" be printed directly onto the AAVSO charts along with the other basic information.
I would like to warn against a couple of pitfalls.
1. Beware of information overload. I would beware of putting so many features on the LCG or Chart Generators that they become difficult to use. These are accessed by everyone from oldtimers to the new observers. They need to be quick and simple to use.
2. New observers especially- and there are not enough of them to start with- need to be able to find and use things easily and quickly. Especially for these people- the KISS principle needs to be applied religiously. I find in my experience that most new amateurs think Variable Star astronomy is a very advanced activity that they don't have anywhere the experience to even attempt. At star parties i alweays get a huge WOW! response when explaining this type of astronomy- and I am currently only doing Visual estimates. Lets not exacerbate this.
The trouble is that every star is different. Theres' no escaping having to learn what to do, and need not follow a general instruction. I treat a Mira with a big amplitude and 90 day period, or one with a very fast rise, different to a 450 day one. And so on.
This is what mentors are for.
I just looked at the observing page.
This is where this type of thing belongs.
Nice and easy to find and use for new observers and keeps the site clean too.