Fri, 10/13/2023 - 17:21
Hello Everyone,
The AAVSO is partnering with the SuperNova Early Warning System (SNEWS) team to create a new campaign centered around supernova progenitors. The goal is to keep track of bright stars that have the potential to go supernova reasonably soon so that we have a record of the star leading up to such a monumental event. This project is open to all observing types and has a large target list so everyone should be able to find a way to contribute. More information on this project can be found here. Please use this forum to ask any specific questions you may have.
Clear Skies,
Bert Pablo
Staff Astronomer, AAVSO
Alert Notice 836 announces the observing campaign partnering the AAVSO and SNEWS in monitoring bright potential supernova progenitors. The notice points to SNEWS project and to information on the campaign with the AAVSO.
There is a thread for this campaign under the following AAVSO forum:
- Campaigns and Observing Reports: https://www.aavso.org/snews-campaign-forum
Please subscribe to this thread if you are participating in the campaign so you can be updated, ask questions, and join in the discussion!
Many thanks, and Good observing,
Elizabeth O. Waagen, AAVSO HQ
It looks to me, specifically from the SNEWS source pages, that things are not yet well concentrated for this project. The lead group seem to be neutrino people, not visible-light/near-IR astronomers in the traditional sense. Their publications seem to be centered on a "wait for the neutrinos" approach, which is unsatisfactory given that there's likely to be only hours notice at most for an actual event. Then why bother with a list of 200 stars? Perhaps someone more familiar with supernova observing in the visible and near-IR could weigh-in and prescribe more closely what could be done now (and in the future) by AAVSO and other capable observing groups.
Just from a naive standpoint I think the candidate list is too long, and surely contains stars not likely to become core-collapse supernovae soon, or ever. So the list probably could be winnowed down considerably. Expert advice needed here. Some assessment needs to be made (and given out) regarding the status of observations. Since the stars are bright, one wonders about visible-light photometry that is being done by wide-field surveys at high cadence, such that irregular and poorly calibrated data are not that useful. Can anything about the interior structure of the stars (which is what matters for the SN prediction) be obtained from small-amplitude, high-frequency pulsation modes from TESS? What about the high-cadence data from Evryscope? How many people are running simple security cameras with wide-field lenses staring at the sky all night every night? The campaign alert suggests only V-filter observations, but probably those need to be leveraged with data at other much different wavelengths. I would imagine (again naively) the most useful filters might include Johnson U (or Sloan u), V or I, and the Wing 1.04-micron band, perhaps the J band at 1.2 microns. The 1.04-micron filter especially gives us a handle on actual luminosity changes in these stars, whereas V is responding mainly to temperature changes. One might want to do a general survey of H-alpha spectroscopy of _all_ the stars maybe once per year --- preceded by a literature survey to see what's already been done. Is that even a particularly useful diagnostic? Following the eruption of 2023ixf in M101, I would imagine that folks are looking for additional greatly obscured dusty SNe progenitors in the Galaxy, possibly with JWST in the mid-IR. The broad idea from all this would be to attempt to pick out the 20 'best' stars (say), a list that might evolve as we learn more about which are the most likely candidates.
\Brian
Hey Brian,
Thanks for the feedback. You are right the star list is large right now and that's a bit by design. Bright stars like this are typically not covered by surveys as they are too bright and will saturate the detectors. However, these bright stars are evolutionarily also the most likely to go supernova. That evolutionary state is one of the main precursors to supernova at least from current theory. It's also worth noting that this list is taken from a larger list that was taken from an even larger sample, so there has been a decent amount of trimming to get to this number. The truth is that we don't know exactly what stars will go supernova and that's part of the problem. Close supernova are extremely rare and we typically have no data on them before the supernova happens and that's what we hope to correct as time goes along. The idea right now is to get regular observations on these stars to see what is happening with them and if there is anything unusual. Hopefully, we can narrow down the list over time to better candidates, but at present just having a record at fairly low but regular cadence is useful. Yes more types of observations will of course be useful. Multi-color in particular is a great diagnostic, but because of the fairly long target list at present, we are looking for the simplest diagnostic. All of the other things you mentioned likely can and should be done, but getting all of that data is difficult and time consuming and you still need a place to start from to determine the best candidates before you do. This is currently what we are attempting to do, by getting regular data on these stars, examining the results and deciding which are good candidates for further observations.
Thanks,
Bert Pablo
Staff Astronomer, AAVSO
I want to amplify Brian's comment about temperature versus luminosity. Look at the light curve for alf Ori in V and MA bands. MA tracks the star's luminosity, unaffected by the molecular absorption that affects V band. You will see that while V varies by about one-half magnitude, MA changes by only 0.14.
Tom
I've been going through the literature collecting published UBVRI photometry for the LMC type II supernova 1987A. None of this is in the AAVSO database. In the early going, on IAU Circular 4318 from 1987 Feb 25, Brian Marsden quoted comments from Craig Wheeler and Bob Kirshner regarding what sorts of observations were desired especially by folks doing modelling: "C. Wheeler, University of Texas; and R. Kirshner, Center for Astrophysics, have expressed the need to inform the astronomical community of the types of observations that should be made of the object. If the supernova is of type I, as suggested on IAUC 4316, it can be expected to brighten, over the course of the next two or three weeks, to mag -1. If, as seems to be coming more probable, it is of type II, it should brighten to mag +1. Kirshner notes: "Suggested observations include accurate photometry and spectrophotometry over a wide wavelength interval at low dispersion, careful monitoring of line shapes at moderate dispersion, and observation of interstellar and possible circumstellar features at high dispersion. Frequent sampling is highly desirable. This supernova offers unique opportunities for polarimetry across scattering lines to determine atmospheric properties. High angular resolution methods of speckles and interferometry are especially important, as this source is bright enough and will grow large enough for present methods to be effective." This text is from here:
http://www.cbat.eps.harvard.edu/iauc/04300/04318.html
So it has been 37+ years since these suggestions were made, and we've learned one heck of a lot about supernovae eruptions since then. Thus it seems that the SNEWS project might do well to have some prescriptive advice from experts about what to do when the next Galactic SN actually happens. Looks as though the recommendations will be: spectra, spectra, more spectra, but some additional details would be helpful. The UBVRI color plots for SN 1987A were very interesting to me, but I suspect they are well understood now, so perhaps not so valuable(?) --- I'm sure they'll get taken anyways. Some examples of multicolor lightcurve plots:
https://ui.adsabs.harvard.edu/abs/1989MNRAS.237P..55C/abstract
Figure 3 in this paper, and:
https://ui.adsabs.harvard.edu/abs/1990AJ.....99.1146H/abstract
Figure 2 in this one. Notice especially the rapid drop in the U band as the fireball dimmed, and that the star became very red around 100 days after the eruption.
At least in the case of SN 1987A, there was essentially no warning less than 24 hours before the eruption. Should folks expect any greater lead time for other sorts of events?
\Brian
Thanks for the…
Brian,
Thanks for the message. If I understand correctly, you are asking if the SNEWS project/team is properly prepared for a galactic supernova event.
The SNEWS project is being lead by several universities including Purdue, MIT, and others. Additionally, AAVSO has a spreadsheet of 193 progenitor objects that are being continually monitored be AAVSO observers with AAVSOnet (I have 20 objects myself). I believe the spreadsheet was compiled with input from the PI (from Purdue I think). If any of the observers notices anything interesting, the plan is to notify the world so others can check if something is really happening. Also, since the neutrino detectors will likely go off in such an event, we should get quick alerts from Atel, GCN, etc. Possibly also LIGO/VIRGO if a GW event accompanies the outburst.
Now, do we need more detail on how to proceed in the event of a galactic supernova? You bet we do! I'm not aware of any further behind the scenes activity, so you might contact George Silvis (observer code SGEO I believe). He may be able to give you more detail. If you have any ideas, please post them to the forum so we can all take them into consideration.
I hope I've answered your question...if not let me know and I'll try to do better.
Dave
Perhaps a fair current answer to my general query (which was intended mainly to stimulate discussion) might be found by doing a literature survey of what happened with the event a year ago in M101. It was not overly bright, but simply seeing what resources were thrown at it ought to give an idea of what people thought was important/useful. Perhaps this could be published as a report in the JAAVSO. Someone like Michael Richmond could pontificate authoritatively from the perspective of the backyard amateur observer (or anybody...).
\Brian
Greetings,
What Brian said... and...
The supernova rate in the Milkyway is somewhere around 1 every 100 years. The next one to pop from the selected SNEWS list could be tomorrow or could be after we are all gone to dust. My bet is on the latter. If the BSM telescope(s) have lots of free observing time, looking for projects to do, why not just have them do a "survey" of the stars on the list and upload the data to the AAVSO db automatically every clear night. The data will be archived at the AAVSO and available for those still around to anaylze it. Also there are plenty of bright variables of all types, LPVs, Delta Sct, etc. brighter than the other "surveys" upper magnitude limit that might add useful information.
Jim (DEY)
A survey is an interesting idea. I don't know how much extra capacity is in the AAVSOnet, but they still keep taking observing proposals (I have a few myself). One of the AAVSOnet gurus would have to clarify that. The system seems to work well, so maybe a survey can be considered. Of the original 193 objects in the SNEWS program, I believe there are still a couple dozen that don't have an observer to monitor. Maybe this could serve as a "mini survey" to test the idea.
Another idea to consider is to come up with a checklist of what to do in the event that we (or our great grandchildren) detect an event. There is a lot of information scattered around about possible reporting procedures (various professional papers, e.g., arXivs, as well as the "Time Domain Astrophysics" link on the HEN website which discusses GCN, GW/LIGO/VIRGO, ATel, etc. techniques). But what we would need is a simple checklist that can be quickly used to report an event.
Anyway, food for thought.
Dave
In the olde days there was one place to send new discoveries: Those who are old enough, or located in/around Cambridge, MA will probably remember the Central Bureau of Astronomical Telegrams. That was the central place to report new discoveries of all types. Brian and Dan, et al. were very good at confirming claims before reporting to the community for follow up.
Today the Astronomers Telegram is probably the best place to report new or rapidly developing things in astronomy. However, you must know someone or be someone with an account to post there. There may be other places however. AAVSO SNEWS group lead needs to get with one of the original PIs and coordinate reporting of AAVSO data that may be worthy of reporting up the chain and to the entire astronomical community.
I see the SNEWS home page has an alert system and an app for that. Should have the AAVSO lead coordinate and have one interface between AAVSO and SNEWS.
It should be rather trivial to automate in software at HQ monitoring of the incoming photometry in the AAVSO db and send a report to AAVSO staff for follow up by others in AAVSO, confirmation, etc. of photometry exceeding normal magnitude limit windows before reporting to SNEWS and the astronomy community in general. The data anaylzer whose name is Artificial Intelligence (AI) could be applied to the problem!
Just some ideas...
I must admit I missed reading Bert's original posting on this topic. Everyone should read the link embedded in his sentence "More information on this project can be found here" ---> SNEWS Campaign | aavso
Jim (DEY)
I agree that the system for reporting needs to be smoother and clearer.
I always scan the galaxies in every image whilst waiting for downloads etc, but I'm not sure what to do if I see something.
I followed the AAVSO links once to see what came up and the websites are very clunky and hard to navigate.
I know this is a separate issue from our Milky Way SNs, however given that these are more likely there should be a clear and easy system f reporting.
I have images several but they are not in the VSX database. I cannot report them, so there isn't any point to imaging them.
The SNEWS candidate stars of interest can be listed from the TargetTool: just sort on 'SNEWS' in the Notes column.
Which of these stars are not in VSX?
George
Brad,
I need some more detail of the context of the problem to be able to help.
- Are you working within VPhot? If so, please share your image with SGEO and I'll take a look.
- Are you getting this red x when you try to submit data to webobs? Please send a copy of the submission file. ( gsilvis@aavso.org )
I'm here to help prevent software tech problems stopping members from being able to submit data.
George
Using the AAVSO bright star monitoring network (BSM) of remote telescopes could be a way to start with the weekly or biweekly photometry for the SNEWS Campaign and get some AAVSO observers involved. I propose the following:
1. AAVSO observer: Commit to performing 5 - 10 SNEWS target stars with a cadence of one or twice weekly.
2. For coordinating star selection, I made a copy of the target list spreadsheet provided for the SNEWS campaign and added a column "Observer". Once you select your stars, type your observer code to indicate your covering that star. The sheet is editable and set for global access, so you can update it with your target selections. For maximum target list coverage you would want to select stars that no one else is observing.
3. You may want to try the AAVSO target tool as described in the campaign website. I found that using the target tool and sorting on SNEWS in the notes section resulted in a list of 609 stars. It would be helpful if the campaign "Alert Notice 836" string could be added to the note section to allowing sorting only the 192 target stars.
4. As part of selection process, check the target FOV column and the chart in the spreadsheet to make sure you will have several comp stars in the BSM telescope FOV. The FOV for the BSM network for each telescope can be found in the AAVSONet information website. Typically the FOV is 90' X 60' for the BSM telescopes. BSM-TX has a FOV of 150' X 100'.
5. I suggest limiting your star selection to mag 6 - 9.9. Based on my experience with AAVSOnet telescopes this would result in exposure times of 10 sec - 30 sec max, with a good S/N and avoid saturation issues. I welcome others to chime in with guidance on mag vs exposure time using AAVSOnet.
6. Once you've selected your stars and added your observer code to the spreadsheet, fill out an AAVSONet proposal as usual and submit.
Here is a link to the spreadsheet:
https://docs.google.com/spreadsheets/d/13g_cnZwEbwFDS0Auxo5dDRqOHPgKXjTJRiIZhAMqVr4/edit?usp=sharing
We'll see how this goes. If we could get 5 observers using BSM, that would be great and provide long term coverage for 50 or more of the targets. BTW, I did post a query on the AAVSOnet forum titled "AAVSOnet support for Alert Notice 836 and SNEWS" hoping that folks would chime in. I got one response from an observer who was interested and he has selected his stars and updated the spreadsheet.
If you have any suggestions or comments on this approach , feel free...
Brian Ramey
RBCA
One additional tool might be to routinely look at ASAS-SN Sky Patrol photometry. These are recent observations from a few hours to a few weeks old. This is on the VSX dropdown menu for each star. The default time is 20 days, so any significant activity should be detectable. I took a look at several LC and SRC stars on the target list and the data for all objects was no more than about a week old. If significant activity is apparent for a given star, the observer responsible for that star can quickly post to this forum. Does this make sense?
Dave H. (HDHA)
BTW, I am potentially interested in monitoring several stars with BSM.
Hi Dave,
That is not a good approach for this project.
ASAS-SN starts saturating around mag. 10. All stars in this project are brighter than mag. 10 at maximum.
The data for the 4 or 5 faintest stars in the list might be okay, or it will be okay for the stars with large amplitudes that become much fainter, but as you go brighter, scatter will start increasing. Brighter than 9 it makes no sense to check.
Since most objects in the list have small amplitudes, scatter will hide real variations and the zero point will be incorrect.
This is one of the reasons this project is important: we are looking at the blind spot of surveys like this.
Also, for the few that might be okay to look, I would recommend requesting a much longer light curve. 20 days won't give you any feel of what is going on. These stars vary slowly and there aren't many points in the ASAS-SN light curve. You need to have more context to see if the star's behaviour is changing.
Cheers,
Sebastian
This is…
Thanks Sebastian,
This is good information. As you say, this makes the BSM observations that much more important. With that in mind, what metrics/parameters in the light curve would an observer use to determine that something important is happening and notify others to cross check the data? Rapid increase in amplitude, rapid fluctuations, slow long term increase or decrease in magnitude, etc?
Dave
I would say that the star deviating from its catalog range would be one of the things to keep in mind.
But we need to be sure this is happening, especially because we have very small amplitude objects in this list. Any zero point problem or miscalibration might end up giving results that might be outside of the catalog values.
Having more than a single observer is another way to add weight to any result indicating unusual values.
Cheers,
Sebastian
Thanks.
Hi Brian,
Certainly, AAVSOnet could be used for this campaign. However, remember that these are bright targets and finding good comp stars is often difficult. The BSMs have basically a 1x1.5 degree field of view with the existing cameras, so you might want to look at the target list and see which fields are suitable. I'm sure that there are plenty, but probably not all targets are suitable.
We're planning on having an informal AAVSOnet meeting at the Fall meeting during one of the lunch breaks. This is a good topic to bring up.
Arne
Very good point, thank you. The target spreadsheet the SNEWS team created includes a FOV column and a link to the chart. I've been replotting the chart with the FOV changed to 60 arcsecs and checking the number of comp stars in the field. For some reason the "Binoculars" option is set in the "Would you like a Special Chart" field as default. I been changing that to "None" to get all the comp stars available.
Also, I think BSM-TX has a somewhat wider FOV. Perhaps a small number of stars that require a larger FOV could be set up on that telescope as part of a specific proposal.
Brian
RBCA
Take a look at this paper: arXiv2307.08785v1 [astro-ph.SR] 17 July 2023. "Red Giant Candidates for Multimessenger Monitoring of the Next Galactic Supernova".
....."catalog of 598 highly probable and 79 likely red supergiants (RSGs) of the Milky Way".
In particular, look at Table 1 on page 20 (there are two Table 1's...this one is from MNRAS), Table 2 on page 4, and the github link which supposedly has all of the candidates. There is also a note that the complete list is on Vizier, but I couldn't find it.
If I understand this paper correctly, this could be very helpful in downselecting candidates for BSM monitoring.
Dave H.
Hello, Arne, and other people participating in this thread,
I wonder if there has been an update on this theme such as a revision of the target's list, or a selection of possible AAVSOnet BSM instruments that could be used in this campaign.
I have selected from the SNEWS list 78 targets visible from my vantage point, Dec -20 to -69, with magnitudes from 0.75 (alf Sco) to 10.9 (DQ Pup).
They are DCEP, LC, SDOR, SRC, SRD, and ZAND stars.
Has the subject been brought up in the recent AAVSO meeting?
Cheers,
Ari
George Silvis obscode: SGEO is managing the effort. See proposal #458 for details: https://app.aavso.org/aavsonet/proposal/458/
Best Regards,
Brian, RBCA
Thank you, Brian.
Your pointer to proposal #458 was all I needed to know.
Cheers,
Ari, SAMA
Now that George is organizing the effort, should individual observer participants submit their own proposals with the stars selected on the project worksheet or should we wait for further instructions? Thanks.
Dave (HDHA)
I am still sorting out the details of linking users with AAVSOnet output.
I will post documentation in the campaign page tomorrow.
( https://www.aavso.org/snews-campaign )
The sign up process for AAVSOnet images will be in the google sheet listed in the forum post, which will be called the Public sheet.
https://docs.google.com/spreadsheets/d/13g_cnZwEbwFDS0Auxo5dDRqOHPgKXjTJRiIZhAMqVr4/edit#gid=876428242
The Master sheet is that on the campaign page
https://docs.google.com/spreadsheets/d/1XuJIRjggs_nBi_3hT5jHuF7MUzmiL8BXRYGdTx1fH9s/edit#gid=876428242
For security reasons there needs to be a manual cut out between user entry and aavsonet mechanics, public and master. I'm working on how to be advised of changes in Public so I can do the manual updates.
I anticipate there will be a sorting out process as we figure out which targets are appropriate for the BSM scopes. (we will use NM and South ) . Some targets are best done by Pep and Visual.
So go ahead and make your target claims in the Public sheet.
I'm working on identifying fields that contain more than one campaign star. Most of these fields are untested on the BSM scopes, but are already in a plan so we hope to have them tested and tuned soon. Be aware that stars in these fields can't be claimed independently. Just another thing that might require some sorting out as we get things up and running.
Field Name,Additional target(s) in the field
BO Car,RT Car|V0730 Car
HD 232766,V0411 Per
HD 305632,CL Car
NSV 18809,V0910 Cen|V1092 Cen|HD 101007
NSV 4976,ASAS J104420-5803.9
NSV 929,V0648 Cas
PP Per,V0550 Per
RS Per,SU Per|AD Per|FZ Per|V0403 Per|V0439 Per
RW Cep,IRAS 22183+5547
T Per,S Per
V0361 Car,V0349 Car
V0441 Per,PR Per
V0528 Car,HD 95687
…
Thanks George & Daniel,
This is helpful. Looking forward to participating when things are ready.
Dave
Hi Brian and team,
I've started receiving data for SNEWS stars in VPHOT. I'm getting V and I images, and have a question on color transformation. Shall we do our photometry with two color transformation using (V-I)? If I focus on only the variable as listed, this should be possible in all instances so far. There are quite a few "other" variables in these images that are bright enough to also submit photometry, but the (I) channel can be too faint in some images to do two color photometry with V.
My question is, is two color transformation more important than submitting untransformed data? I believe the answer is Yes, but would like confirmation and some pointers on V-I to pick good comp and check stars.
Thanks,
Andrew
Andrew:
1. What criterion are you using to determine if a star is too faint in a specific filter/image? Why? How faint a star do you think you can report? Should you ask for a longer exposure?
2. If you are getting images in V and I filter, what other filter pair could you use for transformation? At this point, V and I were selected as good/sufficient indicators of potential change rather than including a B and R filter.
3. Transformed magnitudes are more accurate but, IF you do not have a choice, un-transformed magnitudes are useful. The SNEWS campaign is looking for changes that may be pre-cursors of a supernova event. Un-transformed mags would be better than none?
4. AAVSO Comps are going to be mid color (yellow/green) stars so they may not typically match the color of your target(s). Since the campaign targets are generally bright, finding comps that have a similar magnitude as the target (+- 1-2 mags) will be difficult. Stacking images will help improve the SNR of the comps. Using comps that can provide SNR~>100 would certainly be appropriate but may not always be possible? Do your best! If your check star has the same magnitude as the target you will obtain a representative precision similar to that of the target. Again, this may not be possible due to the brightness of the target.
5. IF you measure several other targets in your FOV, they will exhibit a range of magnitudes and colors, and bracketing will be useful and may be easier than for the brighter campaign target? Generally, you would use comps that have reasonable SNR range to improve error. This is a compromise for multi-target analysis, where the variables may range widely in magnitude and you don't want to use many different sequences!
6. Transformation will correct for color differences between comps and targets, so matching colors is not critical.
7. Selecting a comp that has the same color as the target makes transformation unnecessary BUT choosing very red comps to match a very red target is subject to potential error because many red stars exhibit some variability.
Comments/questions?
Ken
Thanks for the useful advice!
One more question: if we load APASS stars in order to have a more useful sequence, where can we find the color and magnitudes (in the different filters) of the stars we have loaded?
-leo (FLEA)
Leo:
Have you loaded APASS comps into one of your images in VPhot yet?
If so, just open one/any or the comps in the upper left sequence table. Click edit sequence and you will see all the comp's standard magnitudes (and errors) in BVRI and Sloan filters. Colors (e.g., V-I) just require manual arithmetic.
Ken
I see it now!
Thanks Ken!
-leo
Everything Ken said is valuable but to add a little context. The overall aim is to have data to answer the question "What was it doing before it went supernova" if a star in this galaxy goes supernova. Obviously if there is no data at all, there can't be even the weakest answer, but it would be nice to have some kind of guidance regarding how weak an answer can be and still be of some use. It would be ideal if there were competing models that predict different things and thus we could compute what kind of cadence, precision, etc. would be needed to distinguish between the models. Unfortunately there is nothing like that. I do get the impression that if there is anything to be seen, it won't be subtle. Likely to be 1+ magnitude changes. In any event, I'm in favor of taking a pragmatic approach:
In short, I'd like to have this project become a place for ongoing discussion of the details and choices involved with acquiring, reducing, analyzing, and interpreting data rather than just an assembly line for pushing numbers into a database.
DanZ
Hi Ken,
Thank you for the detailed questions and advice. You are one of the best mentor/instructors! You helped me a lot through the VPhot course and I'm glad you are still here to help us along.
I'm going to take these questions one at a time, and start with why I was running into stars I thought were too faint. The variable image I'm looking at is of WY Gem, and it is a very rich field of variable stars that includes some rather bright variables besides WY Gem; BU Gem, PU Gem, and WW Gem. When I first saw the variables in this image, I thought to kill 4 birds with one stone, and see if I can get a good measurements on all 4 variables.
In V these 4 variables are no problem, they are all fairly bright with good SNR (higher than 200). WW Gem is at 213 SNR, and PU Gem SNR is a little high at 1538. Magnitude estimates in V have a rather large error of .054 (and I'm not sure why).
In I there is a big difference in SNR and WW Gem drops to 83 SNR. I would consider that too low (below 100). When I View the Photometry report, my error improves to between .013 and .001 (seemingly much better).
Also - Now that I'm really looking closely to analyze this data, I'm showing PU Gem listed as a Variable star and also listed as a comp star (58). One on top of the other in VPhot. Have you seen this happen before? I'll eliminate one or the other to continue.
I'm going to take a couple more runs at this to see if I can improve the error, and also get rid of the variable/comp overlap. and retry. I'll come back to your other questions on the next round. After I work through this image again, things may be much more clear. I'll also run the V-I to compare my previous results, and let you know how it goes.
Also just an FYI - With WY Gem, I received 4 images in V and 4 images in I. These were then stacked and average combined to improve the SNR. Then I started the photometry.
My goal here is to get a good result that I can then continue to submit every week, or as often as I receive these images, and not have any doubt that the data will be useful.
Thanks,
Andrew
Andrew:
You can certainly 'choose' to use a conservative statistical criterion for the precision (repeatability) associated with the target magnitudes you decide to report. You cited an SNR of 100 as acceptable since it typically yields an uncertainty (1/SNR=1/100) of about 1% (0.01). That is, if you measure the magnitude of a target in replicate images (e.g., 3 images), your mean magnitude may vary by about +- 0.01 (1 STD) between the images. In fact, the most reliable method of measuring the precision of your target analyses is to measure their magnitude in repeated images and calculate and report the mean and standard deviation. The Time Series tool in VPhot does this for you easily. (Don't worry that this involves a very brief 'time series'. It is just a simple math tool that yields both statistical measures).
I think you will admit that it is easy to detect a target with an SNR=100. It's quite bright! However, how faint a target can you visually detect? Try this in a few of your images. What stars can you just pick out on the image? What is its SNR? I suspect it will be somewhere around 3-5? If you were imaging a galaxy where a supernova was discovered and you wanted to identify the precursor and report the speed of the outburst, might you want to report the magnitude of the faintest star present at its position in your pre-discovery images? Would you limit your reported magnitudes to those with an SNR=100? I bet not! So, are there circumstances under which you want to ignore the SNR=100 (1%) criterion?
If you tell the users/researchers of your data that the precision of your measurements is less rigorous than 1%, perhaps 5% (1/20), 10% (1/10) or 20% (1/5), have you provided them with sufficient information to apply a realistic level of confidence to the reported target magnitude? If so, can you relax your reporting criterion?
Ken
Thank you Ken. You've opened my mind to these other possibilities.
Andrew
Hi Andrew,
I was late in realizing that the WY Gem field contained another SNEWS star: BU Gem. Please make adjustments as needed so you can report both these stars.
I shared some early images of this field with you on VPhot.
You can request adjustments to exposure length and center point to snews@gasilvis.net
And you can't go wrong following Ken's advice!
Cheers,
George
Hi George,
Thanks for the early images. The latest images look fine, and center point is great. The exposures are also fine. I'll let you know if anything needs to be chaged.
Thanks,
Andrew
Andrew:
Under some circumstances, a target may be found to be a variable star with a very small amplitude (e.g., PU Gem's amplitude is reported in VSX as 0.06 mag).
IF a comp star catalog happens to exhibit a precision of 0.02-0.04 mag, that same star may have been reported as a non-variable star.
So, your observation of a dual listing for PU Gem and 58 is possible but infrequent. It happens!
It is apparently a small amplitude variable and you should remove the designation as a a comp.
Ken
Andrew,
as Ken mentioned, some small amplitude variable stars may be labelled as comp stars but, if that is the case, you should look at the photometry table. You will find a remark there recommending that the star should only be used for visual observations.
When there are no other available options, they are kept in the sequences.
But always check the photometry tables, because there are some cases when larger amplitude variables are still labelled as comp stars. In those cases there will be no remark because we haven't still checked them. We are in the process of removing some remaining non-suitable stars from sequences.
Cheers,
Sebastian
Hi Sebastian,
Thanks for reminding me about the photometry tables. That's the right place to go check.
Thanks,
Andrew
Hi Ken,
Some more observations related to your questions.
2. - I've been submitting transformed magnitudes of T_CrB for almost a year (through AAVSOnet) and using B, V, R, I filters. It's nice to be able to use Transform Applier in VPhot and get transformed magnitudes for all filters in one shot. I've done very few two color transformations so V and I do work.
3 - Agree, untransformed are better than none. Using V and I filters, we should be able to get transformed magnitudes most of the time. I'm finding that every now and then, I'm getting the reported magnitudes as "zero" in VPhot, and not sure what I did wrong.
4 - Comps and check stars are certainly quite varied with these brighter variables. Thanks for the pointers. Doing my best is the best I can do.
5 - I'm not sure what you mean by "bracketing". Do you mean to take one variable at a time, instead of trying to do 3 or 4 at once?
6 - Confirmed.
7 - ok. I understand.
Thanks for all the detailed comments and questions Ken!
Andrew (CANG)
Hello, for information I change my target in the spread sheet. FERA
Hello,
Perhaps this is a silly question but: why are there two columns in the spreadsheet with the same "FoV/Lim mag" heading?
Ari
The second column is just there for stars with large amplitudes that have two different charts recommended, one for when they are bright and the other one for when they are faint ;)
Cheers,
Sebastian
Greetings,
Sorry to be cranky today, but my emal is full of multiple entries on SNEWS and it is overwhelming the other topics I'm interested in. I have observed one constant star for many decades in my AAVSO life, an NSVS star, and that would be enough for me! I would recommend you get this topic in its own forum entry.
Best regards, Jim (DEY)