Forum Replies Created
-
Posts
-
Just spotted this:
The reason I am doing this is that Astrometrica gets this wrong. In fact it reports two different SNRs in two different places
If by two different places you mean two different positions within the same image, it may not be wrong. As I noted in my earlier response the PSF can be position-dependent. The sky brightness will almost certainly be position-dependent (try measuring LBVs in the Andromeda galaxy!). Anyway, a fainter star (lower signal) will have a lower SNR than a brighter one even if the background is the same.
I am not quite understanding what you mean, in other words. I am but a bear of very little brain.
If you really want to get rid of stars, DAOPHOT does a damn good job in my experience.
The work flow is to identify all the objects, then a set of unsaturated and unblended stars. From the latter you build a model of the point spread function. Note that the PSF may be position-dependent. Fit one or more PSF models to each detected object, removing ones which are “good enough” in some sense as you go. Repeat until you no longer have any “good enough” fits.
Much the same could be said of the usefulness of publishing predicted magnitudes as faint as 15, let alone well below 20.
BAA members can now readily observe things which are significantly fainter than mag. 22.
My dome is controlled by LesveDome and very successfully too.
The major problem, from my POV, is that LesveDome is Windoze only and I want to move to a Linux-based TCS. The only thing stopping me is the lack of an INDI driver for the Velleman K8055 controller board. If anyone knows of one, please let me know.
Steer a (virtual) radio telescope around the sky. Listen to whistlers from Jupiter, the repetitive pock-pock-pock from a pulsar, the loud noise from the sun and the constant faint hiss from the CMB no matter where the receiver is pointed.
Many years ago I observed the sun with a small dish controlled from the visitor centre at Jodrell Bank.
This is going off at a bit of a tangent and might be more suitable for older people in astronomy but we should keep in mind that astronomy is much more than imaging.
Celestial mechanics was traditionally illustrated with orreries.
Mechanical models of solar system objects to scale (either by size or by relative separations) are relatively straightforward to make.
Cerenkov telescopes pick up flashes of light from individual incoming gamma rays. Modern neutrino telescopes pick up individual flashes of light too, and also have an angular resolution of a significant fraction of a radian. (Early ones were omnidirectional and were lucky to pick up one collision per day.) Throwing ping pong balls at an observer, or at a sheet held by the same, would illustrate this effect nicely. Alternatively, a number of “pings” from speakers scattered around a fixed source provides a sonic analogue.
Spatially resolved spectroscopy measurements permit the development of three-dimensional models of external galaxies and the way in which they rotate. C.f. solar system models.
Astrometry from Gaia allows three-dimensional models to be made of our local stellar environment and the motions of the constituent stars. C.f. orreries.
I’m sure that other examples can be given with a little thought.
A few years ago I was given a chocolate teapot as a Xmas present. It worked entirely according to spec.
It was designed to melt when boiling water was added, the liquid then to be used as a fondue-like coating for other edibles such as pieces of fruit, biccies, etc.
Thanks for prompting me to re-visit the S-G algorithm. As a result I found https://cran.r-project.org/web/packages/ADPF/ADPF.pdf
which describes an implementation which automatically optimizes the polynomial order. The window width still needs to be chosen by hand.
Two standard approaches, one lazy and the other more rigorous.
Lazy: guess the width of the signal peaks and set the smoothing window to be around 3 times that. Use degree 2 or 3 polynomials unless the window is very large (over 20 say) when you could choose degree 4 or 5.
Rigorous: Compute residuals for a range of window widths and polynomial degrees. Perform statistical tests on them and choose the filter which best passes your acceptance requirements of high noise to signal ratio (that way round because you are trying to remove all the signal from the noise). The Durbin-Watson statistic, which measures autocorrelatIon at lag 1, is a simple test which tends to work well, but a fuller autocorrelation at a variety of lags might be more appropriate.
Wikipedia has good articles on S-G, D-W and several other tests for signal in the presence of noise. Easy to use implementations are all over the place. I tend to use R as it is free and portable, unlike Matlab and Mathematica.
Fascinating!
Have you tried good old Savitzky-Golay smoothing? It was specifically designed for smoothing (and differentiating) spectra. If so, how does it compare with the wavelet approach?
Here it is: https://ui.adsabs.harvard.edu/abs/1997ApJ…482..420L/abstract
Hydrogen burning continues in the fully convective body of a low mass white dwarf until the temperature drops below 2000K.
White dwarfs in binary systems very often undergo hydrogen fusion. The BAA-VSS observes them all the time. I admit this is stretching the terminology for “energy output in white dwarfs” but still …
The temperature may be quite low but the density is so high that hydrogen fusion still occurs in white dwarfs. I have seen an estimate that about 5% of the luminosity of a 10,000K WD is due to this process. I will see if I can dig out the reference if anyone is interested.
“But for the amateur discovery and initial classification of SN 2018gwo, it might have been missed.”
Nice one Robin.
https://tvlapalma.com/not/18598/incendio-palma-continua-estabilizado-vecinos-evacuados-aun-no/ has a good report and an aerial view showing the burnt area and (apparent) lack of active fires. It explains that evacuees can not return until the authorities are certain the fire will not start up again.
If you don’t read Spanish, Google Translate will come to your aid.
The fire is mostly under control, though there are still some hot spots that need attention.
A major problem has been the weather. About 20% humidity, winds up to 50km/hour and temperatures up to 35C.
Last night I opened up the observatory and took some images. The dome temperature never fell below 20C and the wind-induced seeing was about 6 to 7 arcseconds. No good for pretty pictures but photometry was possible, albeit at the cost of markedly increased exposure times.
Wind is really howling now …
To be fair to Python (which is hard for me because I have a profound dislike of the language) the various point releases (e.g. 3.7 versus 3.8) appear to be very backward compatible. Something written for 3.7 will almost certainly run without any issues on 3.8 and, when it arrives, 3.9. Problems can arise in the other direction but old code tends to run well.
The real problems arose when Python 2.x reached end of life earlier this year. Vast amounts of code is now on emergency life support with kludged-together installations of Python 2. Much more is unavailable. To give just one example, my favourite GUI network scanner, zenmap, is no longer part of the Linux systems I run.
So do I. The calimas around here have been dreadful recently. Last night had noticeably better seeing but poor transparency. Saharan dust and a full moon are not conducive to good imaging. Some more data was taken, down to an altitude of 20 degrees this time, but my expectations are not high.
It is many years since I last had contact with Graeme. We used to see each other moderately often when we both worked at Oxford University.
Any chance that you could pass on my contact details to him please?
