Multiple Positions Plotting

[bigmasterdrago]

I was wondering if there is a way to plot in an IA the positions of a comet or asteroid of the same name from different element sources (MPC, Bowell's ASTORB & JPL). Well, maybe not JPL as ST does not allow MP imports from there. Mostly to see how the positions differ in the real sky. The majority of software only makes available singles sources for an asteroid of comet element set. I noticed that SkyTools 4v has several different sources for the elements.

[theskyhound]

Hello,

There isn't normally a good reason to do such a thing. The basic idea is to ensure that you have the a good set of orbital elements and then use those to make your observations. Under typical circumstances, there wouldn't be any reason to plot positions based on different sets of elements. The only automated sources for elements are the MPC and ASTORB. SkyTools 4 does allow import of Horizons elements, they just aren't automated. Regardless, SkyTools doesn't care (nor really should it) what the source of the elements is. It keeps sets of elements for multiple epochs, so it can plot accurate positions over a wider range of dates. If you add elements from another source at the same epoch, it will simply overwrite them.

Lets say you logged a close pass of an asteroid. When you returned to this observation years later, if you used the orbital elements from that future date, as all other commercial software would, it would not provide an accurate re-creation of what you saw, because the elements on hand would be for a date in the future. SkyTools keeps the original set of elements for you so you can continue to calculate accurate positions for that date, and it even embeds the set of elements in the log entry for safe keeping. The source of those elements is of little concern.

There probably isn't any more complex part of SkyTools than the minor planet orbital elements database. This is because of all of the tricks used to allow so many objects to be calculated, searched for, and especially plotted, without long waits for the intense calculations for so many objects. I honestly don't know how I would change things in a way that would allow you to manually choose one set of elements over another based on the source alone, especially since it will always select the one with the closest epoch for the orbit, which means you would have trouble controlling which set it used any any given moment of time.

[theskyhound]

Presumably you want to compare elements from different sources. One thing you could do would be to plot positions with one set and store that way. Then replace the elements with those form another source, and plot again. But I have to say that I don't really see the point in doing this sort of thing. The elements are primarily defined by the observations used to generate them, and those are ALL FROM THE MPC. That is because every observer submits their data to them, not JPL or Lowell. The only other difference is that the same elements can be recalculated for a different date, using an n-body model of the solar system. ASTORB does this recalculation and so does JPL. For close approaching asteroids, the MPC makes elements available for the current date, which amounts to exactly the same thing.

The only other difference would be automated weighting of observations, or human involvement in selecting the data to be included in the orbit calculation, and in extreme cases, a human might provide reasonable constraints for an orbit with little data available. But these things only apply to objects with very few observations. New observations come in all the time, and the orbit will change. This is why SkyTools always overwrites old elements with new ones for the same epoch. But even if you found a difference for a specific asteroid that resulted in more accurate positions, it would be a unique case, and drawing broader conclusions would not be warranted. The results next time could be completely different.

[bigmasterdrago]

Thanks. I played with your suggestion. For the comets I tried, the difference in position for November 2nd was only a few arc seconds using epoch date elements from Sept 17, 2021 and current from JPL/Horizons. So yes, little need, but then I'm an experimenter by heart and always will be. The same experiment with small rocks proved a bit different. But hardly worth plotting as the difference was just 15 arc seconds.

Very time consuming but hey, I'm retired ;=)

[theskyhound]

I'm sorry that your experimenting isn't easy, but I've always felt it important to stay focused on doing one thing (supporting observations) well.

[PMSchu]

Hi Greg,

I suspect we may be talking past each other on the MP elements. I recognize that the elements derived from the most recent observations will yield the most accurate positions & MPC is tasked with collecting the observations & providing elements for others to use, including we SkyTools users. Unfortunately the MPC only provides elements for Today's epoch not future epochs. Is that correct? Sorry if I'm wrong on that. I don't think BMD wants to compare positions calculated from different data sources, I think he wants to calculate multiple positions from elements at different epochs all at a specific time to compare where each element set places the MP.

The only source for elements at a future epoch that I'm aware of is HORIZONS which allows you to obtain elements for any time (within reason) that you chose. This feature seems very handy for the close approach scenario where the purturbations during the close approach change the elements constantly, thus a single element set becomes less accurate as the pass proceeds. I think what BMD would like to be able to do is to provide SkyTools with a group of element sets, for example 5 sets from HORIZONS with epochs spaced around the time of the close approach. Again, as an example, spaced at 3 hour intervals, with the middle one at the time of minimum Re. Then draw a chart at time, t=x and use each of the 5 element sets to calculate the MP's position on the chart. This would allow him to visualize the effects of the changes to the elements. He could also step the chart time forward through the pass to see how the predicted positions change. At each time the 5 element sets would calculate the 5 positions.

Granted this is a highly niche functionality, but SkyTools' excellent charting capability would really shine at this. It need not involve the MP DB at all, just the small group of elements for one object at several epochs. Calculations would be very fast.

I hope this makes sense, I'd really like to see how changing elements alter the predicted positions too. It would really highlight the importance of accurate elements. I think ST4 already has 99.9% of the code needed to do this (at least I hope so - you have enough on your plate already).

Phil S.

[theskyhound]

Phil, I think he was quite clear when he wrote about "sources." He even enumerated them. There is specialty software available to do what you want to do already, so I see no reason to expand SkyTools beyond observing.

[PMSchu]

Hi Greg,

I stand corrected about BMD wanting to compare elements from multiple sources, thanks. I interpreted him to mean that the different sources were also providing elements for different epoch dates, which seems to be the case often.

What software provides the capability to display positions of MPs based on several sets of elements? Do you have a link, or is Google my friend?

Will ST4 accept MP elements with designations of the form '2021 UU1a, 2021 UU1b, ... 2021 UU1e'? That seems like it could be a workaround for what I was interested in. Just input ellements with similar designations at different epochs & ST4 will process them the same as all the others - problem solved.

TIA,

Phil S.

[PMSchu]

Hi BMD,

I was able to input multiple sets of osculating elements into the ST4v MP DB by appending a letter to the MP's designation string as shown: [attachment=2104]

I did this for 6 element sets for 2019 XS using elements obtained from HORIZONS at 3 hr intervals. The CNEOS website has close approach occurring on Nov 9 @3:48. I forgot that the CNEOS times are UT so my times are off by -5 hrs, but it still seems to work. After the elements were added to the MP DB, I opened the IA using 2019 XS as the target object. The IA displayed the position of the MP using the elements obtained from MPC for Nov 3 as well as the 6 extra elements sets as shown: [attachment=2106]

Here's a close up view: [attachment=2105]

The extra MP designations are searchable & you can time step the IA to see how the different elements change the predicted positions of the MP.

This seems to do exactly what I was asking Greg to add to ST4v, so if this works, problem solved.

Before anyone else tries to do this, let me do some more testing to make sure that I haven't corrupted the MP DB in some way.

Phil S.

[bigmasterdrago]

Yes, different epochs. Tonight, 2021 UO7 passes close at 18:08CDT for me at ~17th magnitude 43° below my ENE horizon. The Lowell and MPC position were only 0.03" apart. The JPL position was ~14" NW. Far to small a difference to be concerned about. JPL epoch June 30, Lowell epoch Oct 12, & MPC epoch July 4. Remember that JPL is concerned about impact risks, MPC passes elements to Lowell where the epoch date is standardized, so which is the most believable? I have no idea.

An interesting experiment can be demonstrated about the need to have gravity calculated in plotting. Simply run this simulation: http://www.astrograv.co.uk/blog.html#16_04_22
in the free software.