This has been a long journey - started in earnest about 3 or 4 years ago. Now, having read my head off and got to grips with processes, software, analysis, severe headaches and night-sweats, I finally come up with my first light-curve for an asteroid.

I've starting off simple so this example is comparitive photometry with no filters. I haven't gone down the road of nightly zero points and extinction co-efficients - yet...

I chose an object with a well know light curve, bright, and in a good position right now. So (132) Aethra was it. The current period (on the MPC page) is given as 5.168 with a variation of 0.10 to 0.52hr.

So over two nights (seperated by a couple of weeks) I managed to get a period of 5.1 +/- 0.01hr. Not bad. The data at the far end of the period (see the chart below) is a bit ropey - so I'll try and get a third night and clean that up a bit. I would have gotten the dat the other night but I was distracted looking at some pesky supernova

Of interest, checking out the logs generated by the software (Canopus) I'm getting down to a level of 0.003 mags differentials. Now I'm not certain if that's just generated data or actual performance. But for sure, if you look at the scale on the chart (try the larger version), my system is working at the millimag level. That's very encouraging.

I'm still on a fairly steep learning curve (pun intended) but this is good progress and as a proof of concept, shows that I can probably mix it with other photometrists.

Anyhow, the full version of the chart is here: www.astroshack.net/images/00132Aethra_LQ.jpg

And here's a (very) condensed version for you to look at right now:

Dave that's excellent stuff. There's always somewhere further an amateur can take himself - there's so much that the amateur can do.

How many hours capture are you working with and do you have any strategies in mind for smoothing the light-curves?

Very best of luck
Mark C

Thanks Mark - for sure, Astronomy is one field where amateurs have a lot to contribute in terms of meaningful data. Certainly the pros recognis this and appreciate it very much. Its always been a goal for me that "science" would form the core of my astronomy work. For others, a different goal beckons. And that's the beauty of astronomy.

Regarding (132) Aethra, I chose it knowing the period was 5 hours or so and therefore, two nights of data would get most if not all of the light-curve. It was very much a proof-of-concept exercise for me. I managed to get a tad over 7 hours of usable data plus another two or so hours of unusable data (clouds, star too close to the target etc). So that's what the graph shows - hence the overlap of course. The blip at the end needs to be cleaned up - which I might do, or not...

The problem with asteroids of unknown light-curves, is that it could be a short period curve (easy peasy) or a long period curve running into many hours/weeks ro God forbid, months. If it is say 12 hours long, you'd need a good four or five or more nights of data - a challenge from Ireland certainly. An additional problem would be that comparing night-to-night data becomes difficult where they are spread over weeks or months. That's where filtered data needs to be taken and you put it onto a "standard" system - nightly zero points, 1st order and 2nd order co-efficeints etc. The advantage is that you can then directly compare nights and indeed, compare your data to any data from other observatories that are using a standard system - a requirement therefore for collaborations. But that's further down the road for me. I want to get a few more asteroids under the belt as they say and branch out into variable stars and exoplanets.

As regards "smoothing" the curve, there isn;t any need to if I was just reporting the period. Though I'd like to fix the "blip" at the end of the period alright. The error bars are as small as 3 millimags - that's pretty good and certainly enough to report an accurate period. Certainly, more data would make the curve smoother - but the period would remain the same, pretty much.

Of course, if I wanted to go down the road of report a Pole position, more data would be required. And even more data would allow shape-modelling to be carried out. But given the number of asteroids out there with no period reported, there is a huge amount of the basic work to be done.

For now, I'll just keep at the basic data gathering and analysis and gently complicate things as I go along!

Cheers
Dave

Dave,

Welcome to the world of differential photometry (if I read your message above correctly).

The curve is quite impressive and something which is new to me since my work has been exclusively around variable stars and exoplanets.

As for the weather, we are also stuck in the middle of an extended front and with no end in sight. To add insult to injury, all of this is occurring around new moon.

Anthony.

Hi Anthony - of course, I've done this before with TReS-3. But now I'm serious

We've been lucky the last few nights with extended clear spells. So, I've been able to update my data

I've re-calculated the period:

(132) Aethra
Period: 5.17 +/- 0.005h
Amplitude: 0.2

This concurs perfectly with the published data - the latest reference giving a period of 5.168h (no error given).
I checked my errors and I'm getting as low as 3 millimags. Average would be about 10 millimags or so.

As far as the "bump" at the far end of the curve is concerned, the new data really didn't clean it up. In fact, it seems to have confirmed that there is in fact a "real" bump. I'm not sure what the reason could be. I'm going to see if the MPB is interested in publishing the data, though published lightcurves (well, one...) do already exist.

Anyhow, proof of concept complete, now to do some more

Cheers
Dave


Full res light-curve: www.astroshack.net/images/00132Aethra-3.jpg


Condensed light curve:

That's very interesting Dave.
I must try and give that a go.

Michael.

Its a worthwhile use of scope time - especially as one can publish the data in a peer-reviewed journal which would be referenced by the professionals.

Its not easy but if anyone had an interest, I could my best to help out.

Also, you *may* need to consider purchasing software, though Canopus is relatively cheap. I'm currently looking at Mira Pro - there's a good deal on it at the mo (for the next couple of weeks anyhow) and I might pull the trigger...

You *can* simply use Excel but if I recall, its laborious.

Anthony - what do you use?

Dave

Dave that is excellent stuff. The data model presented looks excellent. One question, the second minimum presented is fainter than the first. Is that a real feature of the asteroid, or is it just that maybe the latter half of the data was taken with a higher air-mass?

Wonderful stuff and look forward to seeing more. A really good example of more useful work that amateurs can do.

Dave.

dmcdona wrote:

Anthony - what do you use?

Dave,

With respect to the differential photometry, all of my work is done via AIP4Win V2.2.

The graphics generation is done using the online facility at ostrava.astronomy.cz/plotter.php with some intervention at my end thereafter since the online facility is really for variable stars and where the only item of interest is the local minimum or maximum and in contrast to exoplanet transit work where we care about three points in time, namely ingress, midtransit and egress.

Anthony.

Dave,

Thats a good resolved light curve.

I've used MIRA PRO in aperture photometry.
It's a good program.

Eamonn A

www.kingslandobservatory.com

DaveGrennan wrote:

Is that a real feature of the asteroid, or is it just that maybe the latter half of the data was taken with a higher air-mass?

Yes - its a feature of the potato-ness of the asteroid

Since the comparison stars reamin in the FOV during the entire run, airmass differences drop out of the equation. That said, to minimise errors, you should always shoot at an airmass of less than 2. And of course, choose your comparisons carefully - no variable stars etc and preferably solar in colour. Nothing too blue or red.

The "bump" at the end of the period is odd and I've no epxlanation for it. Though I did see a paper suggesting that (132) Aethra is thought be particularly lumpy... The curve follows the shape of the published data very well - bar the lumpinees at the end of mine. I don't think its an artifact of my system either - two of the three runs that covered that aprt of the curve showed pretty much the same pattern. I guess more data would tease it out and characterise it better.

With amateur asteroid discoveries reaching the end of the road, photometry seems to be be a very worthwhile goal with many objects requiring study. But I'd like to have a go at any object that varies...

Anthony - I hadn't considered AIP4WIN. I'd be interested on your thoughts in terms of its perfromance for exoplanets - especially where you're well into the millimag zone.

Eamonn - how easy is Mira Pro? And do you know if it can handle zero points and first and second order extinctions? i.e. does it have the algorithms to get you on the standard (Landolt) system?

Cheers
Dave

dmcdona wrote:

Yes - its a feature of the potato-ness of the asteroid

Now don't start getting all technical on me McDonald.

So the obvious question. Have you classified the 'potato-ness' ot this asteroid? Rooster or Kerr Pink? When are you going to image a 'turnippy' asteroid? Lets see if you can sell that classification system to the minor planet centre.

dmcdona wrote:

Anthony - I hadn't considered AIP4WIN. I'd be interested on your thoughts in terms of its perfromance for exoplanets - especially where you're well into the millimag zone.

Dave,

I am quite happy with AIP4Win and as indicated by its continued use at my end for the past few years. One criticism which I have heard on occasion, and which I do not know if it is true or not, is the possibility that it slightly underestimates the standard errors.

With my exoplanet work and using AIP4Win, I have been able to detect exoplanet transits with a relatively shallow depth.

It is my understanding that Maxim/DL also does differential photometry but it is something which I have not looked into.

As an aside and further to the earlier comment above, I always try and match my variable star to comparison and check stars which have the same spectral type, virtually the same magnitude and the same air mass coefficient so as to have the latter factor cancel out.

Anthony.

Atually Dave, I sent Gareth an email a few weeks back on a proposed categorisation system for describing the non-spherical attributes of asteroids.

He hasn't gotten back to me - yet. He's probably asking for further clarification from Dan Green or NASA or something.

In increasing order of non-sphericalness...

1. Onion
2. Potato (new - Cypriot or Egyptian)
3. Potato (old - e.g. Maris Piper)
4. Tomato (plum)
5. Swede
6. Turnip
7. Cabbage
8. Garlic
9. Potato (sweet)
10.Banana

Maybe it was item 10 that threw him - its not a vegetable...

Maybe it was item 10 that threw him - its not a vegetable...

Neither is a tomato Dave....

As for the potato-scale....sounds a little half-baked....

Knowing Gareth well, I had already pre-empted the "tomato" question:

"The question of whether the tomato is a fruit or a vegetable found its way into the United States Supreme Court in 1893. The court ruled unanimously in Nix v. Hedden that a tomato is correctly identified as, and thus taxed as, a vegetable, for the purposes of the 1883 Tariff Act on imported produce. The court did acknowledge, however, that, botanically speaking, a tomato is a fruit."

In view of the stony silence from Haravrd, perhaps I should just just drop it - I'd hate to be accused of a breach of the peach.

michaeloconnell wrote:

As for the potato-scale....sounds a little half-baked....

Ok can i just 'Chip' in here?

I'll get me coat:dry:

Hi Dave Mc,

Mira is a powerful tool!
As the saying goes its only easy when you know how.

As long as you follow the instructions carefully it works.
It took me a while to grasp the methodology. It is complex, but that is my opinion.

Mira does handle both zero points and second order extinctions.
and also uses Landolt standard stars.

There are transformation equations to derive the zero point for each zero point obseration and then you average the zero points in each color to derive the nightly zero points. Zero points can change during the observing run if sky conditions vary. Therefore, if you are going to transform your data to standard Landolt magnitudes, you need to check periodically to make sure zero points haven't changed significantly. Mira has all those key elements. Basically you don't have to do any scripting.

A fast way to do this is to slew your telescope in order to pick north polar sequence stars east of the meridian (say 2 hours). The airmass through which they are observed changes very slowly and by small amounts. Therefore significant changes in extinction of these stars over a few hours, means that sky conditions are changing and you need to take new zero point measurements using another new group of Landolt standard stars near the meridian.


I have used MIRA over some years and have used it for photometry of exoplanets, but mainly for source extraction. This is important for detection and measurement of all objects brighter than a threshold value. I have applied this in research on very faint objects.

You probably know I have carried out a survey for outer planets in the outer solar system. Picked up 44 suspects. Source Extraction was applied to all these suspects because of there faintness. A very useful tool!

I hope that helps.

Eamonn

www.kingslandobservatory.com

Eamonn - that's very encouraging. The online information on Mira Pro doesn't go into details around NZP's and first/second order extinction. And sure, I think that at this level, anything is going to be complicated...

I've been looking at some alternative methodologies in the literature. Landolt fields are "the standard" but there are shorter options (e.g. using Henden fields). I'm guessing that getting onto the standard (Landolt) system is the best (but not necessarily required) though other methods can get you close enough. I suppose it depends on what you're doing and the nature of the data as to how far you need to go.

By the way, I was flitting around the web and I came across to other sets of standard fields - "Stetson" and "JHK". Any idea where they came from and if they have any use?

Thanks for the information on Mira pro - that's very helpful indeed.

Oh, and that's an excellent scheme for checking sky conditions. I've not seen that before.

Dave

Hi Dave,

I presume you won't be monitoring in the infrared (IR) because if you do you can use the JHK photometric system. JHK was specifically developed for IR bands not the visible.

Regarding the Stetson photometric standard fields, you'll find that the stars are widely spaced as compared to Landolt. Also the stars are in BVRI bands. Although this is no problem, I use Landolt myself when I was carrying out surveys. There are a lot more stars to choose. So I have a personal preference for Landolt stars.

If your planning colour photometry, watch out that the Johnson & Morgan system is being phased out. Although the Bessel and Cousin filters are still around, they also will be phased out. Of course both core types have slight variations in the UBVRI.

The reason for the phasing out of the these filters is because
chips were red peaking at around 0.7 microns. Now there peaking
at around 0.5 microns. This distorts the true photometric response.

Although I use Bessel filters, I have recently used SDSS filters. In time they will be the norm. My CCD camera sensor peaks at 0.7 microns. It's an old camera by todays types.

The reason I'm using SDSS filters is because I have been experimenting with another camera that peaks at 0.5 microns.

The SDSS provides wider bands and therefore more light. It also depends on your chip sensor spectral response and QE. I think your chip could be blue orientated.

Eamonn

www.kingslandobservatory.com

Eamonn - my current filter set (BVRI) are Custom Scientific to Bessel and Cousins specifications. I use a KAF1001E chip. I'll check out its response and see where that lies.

I'll check out the SDSS filters. I guess that if the "norm" is going to change then its better to be prepared.

Thanks for all the info!

Dave Mc,

I wouldn't just rush out now to purchase SDSS filters. Its going to be some years before its a norm. There is a significant amount of astronomers still using and publishing with Bessel & Cousins UBVRI filters.

The reason I am using SDSS filters is for a collaborative project
that I am involved in that specifically requires these filters. SDSS are expensive.

Dave, your Bessel filters from Custom Scientific are the best you can obtain.

Eamonn

www.kingslandobservatory.com

Actually Eamonn - they aere *really* hard to get hold off - I didn't see any of the mainstream supplier offering them though I think I saw Astrodon with them in the catalogue...

Anyhow - good news that I don't need fork out yet more dosh - the Custom Scientific filter set I got was the price of a small APO... But they *are* the bees knees

Dave,

Re SDSS filters, Barr Associates and Omege have them at Rolls Royce prices. Don't go near them, they specially make filters for
all the big observatories. Hubble, Keck to name a few.

I bought SDSS filters from Astrodon. The prices were expensive for the 50mm. so I redesigned the camera nose arrangement to take 29mm. The price came down dramatically and affordable.

Eamonn

www.kingslandobservatory.com

The 50mm Astrodon's are very expensive - 320 and up - muliplied by five becomes expensive... THe other two manufacturer's don;t even "do" prices on their website - that's scary...

Anyhow, thanks for the info. Let us know how you get on with the SDSS filters - I'd be very interested.

Cheers
Dave