Return to Home Page
Dear all, I have written a lot of times about the way I observe stars. My goal has always been to get a V-like final magnitude when reporting and collecting visual data. I am happy with the results and I always try to improve them. I know that most of you disagree with my method, but let me tell you a story that has influenced my results and may serve as a moral. As you know, my method makes the way you look at a certain star dependent on that star's color. To put it simple: Naked eye and bright stars (low power, whiter background): RED Vs BLUE = out of focus vision. RED Vs YELLOW= direct vision, quick glances to the red star. BLUE Vs BLUE= whatever, but always the same way. RED Vs RED= direct vision. Quick glances for all Intermediate brightness stars (telescope and binoculars): RED Vs BLUE = direct vision. RED Vs YELLOW= direct vision. BLUE Vs BLUE= direct vision. RED Vs RED= direct vision. Quick glances for all DEEP RED Vs. whatever= direct vision to the very red star until it's seen very clearly. Faint stars/ Limitting magnitude (telescope specially, high power, dark background): RED Vs BLUE = direct vision (if not, red star undetected, blue star overestimated). RED Vs YELLOW= direct vision, specially to the red star. BLUE Vs BLUE= averted vision, specially if they are very faint. (not accurate but allows detection) RED Vs RED= direct vision until you see them well. DEEP RED Vs. whatever= direct vision to the very red star. Stare until the point source is apparent (Let Purkinje effect act until you see the star with no difficulty) YELLOW vs YELLOW= no problems. Direct vision in all cases. The big problem when it comes to get close to V have always been the use of different colored comparison stars, which sometimes is unavoidable and needs to be taken in account. A result can't be dependent on the comparison star. IMO, we need to calibrate our instrument as CCD or PEP photometrists calibrate theirs. When I developed my method based on what I observed and PEP published data (Mermilliod's GCPD has been a great tool all these years!!) there were a little bunch of stars that were not represented in any sample I could use to calibrate my eyes. They have been also a huge problem to photometrists when it comes to convert data to the standard system: deep red stars and the lack of very **very** red standard stars... Although you are able to get some published value for a certain carbon star, that star is for sure variable, so you can't make any comparison to your data for a given date. To get that calibration I relied on V data from an e-mail back in February 2001, when I was developing these technics. The message (from Taichi Kato) made a comparison between visual and CCDV magnitudes of R Lep: -------------------------------------- ----- Original Message ----- From: "Taichi Kato" <tkato@kusastro.kyoto-u.ac.jp> To: <vsnet-chat@kusastro.kyoto-u.ac.jp> Sent: Wednesday, February 07, 2001 1:19 PM Subject: [vsnet-chat 4067] Re: Visual estimates accuracy Re: [vsnet-chat 4043] Visual estimates accuracy > If we are not aware of these different responses of the eye and randomly > observe with one vision or another, without taking care, results may be > dissapointing, specially for stars like R Lep, that are very red and may > completely disappear if we used averted vision for observing them. (I've > noticed 1.4 mag. differences for that star, which is saying we are observing Incidentally, Kiyota-san obtained BVRc photometry of R Lep around the early 1999 maximum, which was well observed by a number of observers. The star had an average color index of B-V = +4.12. The two observers who reported magnitudes closest to Johnson V were S. Otero (v-V = +0.43) and M. Sato (v-V = +0.41). (v denotes visual observations) The largest deviation from Johnson V was seen in R. J. Bouma (v-V = +1.7) and R. J. Johanns (v-V = +1.6), but most of other observations clustered around v-V = +1.0. ------------------------------- I was disappointed by that 0.4 mag. difference and no way I could take it as acceptable. I used that relation to calibrate my observing techniques to obtain a good match for very red stars, that were "obviously" much fainter than V the way I was observing them. So I let Purkinje effect act a little, specially for fainter stars. Those days, there were no ASAS-3 like stuff!!!!! Some months ago, an interesting paper was published on red stars (Platais et al., A&A, 397, 997, 2003). There you can get carbon (SAAO) and red star (Siding Spring) observations with the JD included. So I made a comparison. For the whiter stars as the RCB type ones, the match is good. Even for red stars as R Cen (mean B-V around 1.9-2.0) the accuracy is at the 0.00-0.05 mag. level using direct vision. See the lightcurve at: http://ar.geocities.com/varsao/Curva_R_Cen.htm But for V Hya I got a result 0.37 mag. brighter than the Siding Spring value. For V Aql and W CMa the values were also brighter although not that different (directly related to their color indexes) I didn't know what to think then. I had observed those stars consistently in the way those V observations required. My mistake was to be inside a paradigm that went: "if it's measured electronically, then its' true" In the following months, I learned not to trust V values blindly. In the V838 Mon campaign I could see the differences between different observers. Those R Lep measurements were not meant to be chosen as a "calibrator".... Now we have ASAS-3 data. And I compared my results for CGCS 2792, BH Cru, UY Cen, V Hya and others with ASAS-3. Carbon stars (only those and specially the bright ones) are typically 0.2-0.4 mag. brighter. I was doing it okay and changed because I thought I could trust V observations from the beginning. So the staring thing that you all know it was banned, it's banned even if we want to get V magnitudes for very red stars. Fortunately I can correct my observations and calibrate my techniques again, but my point is: be careful with any data you use, no matter if it has a "V" after the magnitude... Regards, Sebastian.
Return to the Powerful Daisaku
vsnet-adm@kusastro.kyoto-u.ac.jp
