Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Unexpected year bar calculation
#1
Hi Greg,

Something I've just noticed; I'm not sure what the cause could be. I was looking at NGC 7681 and there are two aspects of its year bar that I cannot understand. In both cases I set the minimum Moon separation to 0, to remove a variable.

First, the year bar for a somewhat dark sky (44°03'00" N, 78°44'29" W) near Toronto. As seen, the target is only shown for August and September and a bit of October, but nothing for July when it's fairly up, as if it suddenly pops in the sky in August Smile

   

In the second screenshot, I merely changed the location to my home in Toronto (Bortle 8), less than 100km away from the above site. I expected something worse, of course, but the result was still surprising: August has no bar but then there's a continuous green between mid-September to the end of October.

   

As a final note, if the Moon separation is zero, shouldn't the gaps be gone between the time is up (unless the Moon happens to eclipse the object?) Of course, I wouldn't observe a galaxy with the Moon nearby, just asking about the calculation.

Thanks.
Reply
#2
Hello,

Please check two things:

1. That you are running the latest version (4.0e)

2. That you have entered the longitude and latitude correctly for your locations. I suspect you may have them reversed or something similar.
Clear skies,
Greg

SkyTools Developer
Reply
#3
Hi Greg,

I am running 4.0e. The coordinates are also correct, I double checked the values, that they are N and W and that they are in the right order. I'm attaching the screenshot with the first, the other is just a bit different but the numbers are in the right places.

Long Sault: 78°44'29"W 44°03'00"N
    Toronto: 79°25'23"W 43°51'44"N (I chose an observatory in Toronto instead of my home for privacy but shouldn't matter in thise case, it's close).

Are you able to reproduce it?

Thanks.

   
Reply
#4
Do you have an obstructed horizon set up for either of these locations?
Clear skies,
Greg

SkyTools Developer
Reply
#5
No, I haven't got to do that. In case you need any config files, let me know and I can zip them up, host them and PM you the link.
Reply
#6
Check your weather settings for each location.
Clear skies,
Greg

SkyTools Developer
Reply
#7
Keep in mind that the Yearbar doesn't use a simple geometric calculation, as other software would. It is a complex calculation that involves the location, weather & seeing, and imaging system. Given that you are seeing different results for the same imaging system, I suggest checking every detail of the locations, including the monthly weather settings, seeing, and light pollution.
Clear skies,
Greg

SkyTools Developer
Reply
#8
- locations checked. No difference except the minor delta between the coordinates.
- weather: I haven't touched it. It's left to current conditions.
- seeing: set to average in both cases.
- LP: Bortle 8 (18.1 mag /arcsec^2) at home, Bortle 5 (20.7 mag /arcsec^2) at Long Sault. In both cases, it's set based on the LP Database.

Three things that concern me:
- the year bar doesn't show anything for July while in the Target Selection the same object has IQ=B for tonight. Everything else held the same. Is there a difference between the calculations in these two screens?
- that some of differences between the two year bars is in the opposite direction I'd expect. It's showing the object as somewhat visible around full Moon in a Bortle 8 location (with a dip in the green bar) whereas not showing it at all in a Bortle 5. This doesn't make sense. If the algorithm is correct, then it's only the data that can be wrong. Is there any way to see what it's using?

Or, more generally, is there a debug log you can enable? If not, could you add this functionality? It'd be very useful to be able to just send you the log with the details about what values are used and so on (I've been programming for a long time and the values of application logs have been proven many times).

Thanks!
Reply
#9
Hi,

You need to check the monthly weather, as this is used (naturally) for calculations that are not for the "current" date.

FYI: I know you are assuming some sort of bug, but I think we should be past that point now. More likely this is a data entry error or misinterpretation.

Also, I don't know if it has ever been tested with a Moon Separation of zero, so please try putting it back to a more normal number, such a 30 degrees.
Clear skies,
Greg

SkyTools Developer
Reply
#10
Hi Greg,

I put back 30 degrees as Moon separation (as it used to be), no change in the result.

I haven't set anything special for the monthly weather - but with the defaults, given that the locations are close (65km straight line), weather (temp/humidity/seeing) alone cannot explain the two issues I listed in the previous email. I tested with another location in between, it was similar to Long Sault (gaps around full moon and complete absence except Sep and Oct). Then I tested with "North York" (part of Toronto) and Ottawa, both from the default ST location database. The latter two showed the same pattern as Toronto locations.

Then I kept a location static (Ottawa) and only changed the Bortle scale. That seems to be directly related to the problem: setting Bortle 9 explicitly shows the pattern similar to my screenshot with Toronto: continuous green bars between mid-September to end of October. Changing the sky to Bortle 1 shows some visibility in August and gaps around full moon. So something is not right, increasing the light pollution should not allow more visibility.

I hope it's the data but without visibility into the data that's used, I can't know one way or the other. It's somewhat frustrating since I cannot trust the year bar any more, I don't know when it's correct even for other objects, particularly when it says there's 0 chance of observing an object. I can mentally ignore cases that show the object visible on full moon, but to have zero in the year bar when the object is in fact up requires a check of the position, thus negating the value of the year bar. I would appreciate if one day you try to reproduce it with the locations. I know you have a busy plate so I'm not asking for priority on it, just for another data point. Maybe I'm not the only one affected.

Thanks again,
Razvan
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)