2018 FE4

[bigmasterdrago]

So my question becomes why does it take so... much longer to download and process the NEAs at Today's Epoch than the Near Earth Asteroids (MPC NEA)? 5m30s vs 10s. Especially if it's just a file generated by the MPC which should only take a few seconds to generate. I'm showing 29,813 including rock 2022 SZ4 discovered 6 days ago, 145656 (4788 P-L, 2003 SV84) & 6344 P-L which had a 15LD pass in Oct 1960.

[PMSchu]

BMD,

If the epoch for the regular NEA file is the standard 2022 Aug 9 00:00 UT, then the file doesn't change very much. There are just a few new additions to the MP DB. For the NEAs at Today's Epoch file, all of the MP's elements need to be updated to the new epoch. Have you ever watched the ST4v dialog while the update is being processed to see how many objects are Added, Updated & Skipped? Every time I download the NEAs at Today's Epoch file there are 0-30 objects Added, ~29.7k Updated & ~2 Skipped. I think that the updating is what's taking the extra processing time. The Technoking could confirm this.

Other than the time savings & the addition of the newly discovered NEOs, I don't see much point to downloading the MPC NEA file. It hardly changes. There might be a daily update available. I prefer to download the NEAs at Today's Epoch file. That provides the new discoveries & the most recent set of osculating elements for the ephemeris calculations.

That's how I believe that the MP processing works. Greg could provide more info, if I've gotten something wrong.

I noticed that CNEOS still doesn't list a close approach for this MP. It's almost within reach of one of the iTelescopes. Maybe I'll try to image it.

Hope this helps,

Phil S.

[bigmasterdrago]

I ran an experiment using just my PC with no other programs running. I downloaded just 29,598 of the NEA MP. I left out those with older epochs (~215) and the two P-L rocks. So just the rocks from the MPC NEA file with epoch dates 2459800.5JD. I let the PC process all of them to an epoch date of 2459846.90428378JD (current PC time). That process took about 35 seconds on my old W7Pro64 desktop. So I actually think that ST is doing the processing after downloading the NEA rock file. It may take ST longer if the epoch date of the file is farther from the current "today's epoch".

---

BMD,


I noticed that CNEOS still doesn't list a close approach for this MP. It's almost within reach of one of the iTelescopes. Maybe I'll try to image it.

Hope this helps,

Phil S.

I noticed that Spaceweather also ignores this rock. I suspect it is due to the fact that it is lost with such a short arc (10 days). Odd, as its period is ~770 days (2.11 years).
If you run their positions to the discovery date, they're only an arc second apart or less (Lowell, JPL, & MPC). Clearly, it was nowhere near Earth the two most recent passes (4/28/2020 & 6/7/2022) when ~1AU from the sun. Although much closer on Sept 26, if we believe the elements given.

[PMSchu]

I ran an experiment using just my PC with no other programs running. I downloaded just 29,598 of the NEA MP. I left out those with older epochs (~215) and the two P-L rocks. So just the rocks from the MPC NEA file with epoch dates 2459800.5JD. I let the PC process all of them to an epoch date of 2459846.90428378JD (current PC time). That process took about 35 seconds on my old W7Pro64 desktop. So I actually think that ST is doing the processing after downloading the NEA rock file. It may take ST longer if the epoch date of the file is farther from the current "today's epoch".

---

BMD,


I noticed that CNEOS still doesn't list a close approach for this MP. It's almost within reach of one of the iTelescopes. Maybe I'll try to image it.

Hope this helps,

Phil S.

I noticed that Spaceweather also ignores this rock. I suspect it is due to the fact that it is lost with such a short arc (10 days). Odd, as its period is ~770 days (2.11 years).
If you run their positions to the discovery date, they're only an arc second apart or less (Lowell, JPL, & MPC). Clearly, it was nowhere near Earth the two most recent passes (4/28/2020 & 6/7/2022) when ~1AU from the sun. Although much closer on Sept 26, if we believe the elements given.

Sorry, BMD, I don't do dates in JD. What are they in regular date notation? Are you saying that ST4v changed the epoch dates of the elements? I didn't know that it could do that. Greg has mentioned that SkyTools does a lot of preprocessing to speed up the plotting of MPs on the charts. I think that's what's happening after the file is downloaded.

Did you watch the dialog as the processing was occurring to see how many objects were Added, Updated & Skipped?

According to CNEOS's Orbit Viewer, 2018 FE4 comes within 0.061 AU on 2022 Sep 29, but the event isn't listed in their close approach table. That's strange.

Phil S.

Phil S.

[bigmasterdrago]

Sorry Phil. What I meant was I processed the complete 29K+ NEOs from the MPC epoch of 2459800.5JD (Aug 9, 2022) to 2459846.90428378JD (2022 Sept 24 09:42:10UT) using just my PC to see how long it would take. I did it externally, outside of SkyTools. As an experiment. ~35 seconds. When I fetch the NEA at today's epoch using ST, it took ~5m30s yesterday to do all the processing which I assume is internal to ST using Greg's algorithm.

I did watch the dialog but do not recall the numbers. I ran it for now (Sept 24, 2022) and it took only 2m12s - downloading, reading data, updating data, 2,386,789 MP, writing epoch files, rebuild cross index, updated 29,491, added 5, skip 2. It shows today's date (Sept 24) for all the NEAs.

---

Phil, try plotting the Lowell elements and the MPC elements for 2018 FE4. When I do that, the two rocks are showing 91° apart in the sky. MPC rock 0.0069AU and Lowell rock 0.045AU - Sept 25 19:17CDT.

[PMSchu]

BMD,

I don't know what's involved in precessing the elements from one epoch to another. The old Astronomical Almanac used to have precessing matrices to change between J2000 & some other date, but that was for star positions, not MP elements. The ST4v preprocessing is related to the plotting of MPs on the charts. It speeds the plotting up significantly.

That's a huge discrepancy between the Lowell & MPC positions. That makes it hard to find  .

Phil S.

[bigmasterdrago]

I'm hoping that this rock gets recovered. Such an enigma. Here are the info pages from SkyTools. Just one point of fact. The positions generated by Horizons (JPL) appear to only be ~3° off the Lowell position vs 93.5° off the MPC position for date 25 Sept 19:00CDT 2022. For some reason, I was under the impression that the MPC fed elements to Lowell who then precessed to the common epoch.

The precessing is pretty involved. Maybe why it takes a good bit of CPU time via ST interface.

Ignoring relativity and oblateness, (minor stuff), it's mostly algorithms, rather than math.  Newton's law of gravitation is involved, as well the solution to Kepler's equation. But stand alone on my old PC, the fact that 29k+ element epochs can be rewritten in just 35 seconds. Someone else spent several months writing the software to solve Kepler's equation, with several improvements over the years. I'm just a mouse clicker. LOL

   

[PMSchu]

Hi BMD,

Those predictions are sure different. Your MPC elements are for the 'Standard' epoch date of 2022 Aug 9 00:00 UT rather than their epoch at Today's date. I don't know what's invovlved in precessing elements for a Keplerian orbit to a different epoch, as I've said. Precessing star coordinates to a different epoch is straightforward. 

I scheduled some observing time on an iTelescope for tomorrow night. We'll see if there's anything where the MPC predicts 2018 FE4 should be.

Phil S.

[bigmasterdrago]

Hope the weather Gods are nice. Here is the poop on that rock using elements at today's epoch.

9-24-22    

---

At least those positions are only 12.4' apart now. But... the JPL are only 3' away from the Lowell positions. Who ya going to believe?

[PMSchu]

Hope the weather Gods are nice. Here is the poop on that rock using elements at today's epoch.

9-24-22

---

At least those positions are only 12.4' apart now. But... the JPL are only 3' away from the Lowell positions. Who ya going to believe?

But the Lowell elements that you used are for the epoch 2022 Aug 9 00:00 UT & your new MPC elements at Today's Epoch are for 2022 Sep 25 00:00 UT. That's ~7 weeks different. Still, you have 2 pairs of positions ~90° apart from each other. The MPC's elements don't usually change that much unless there's been a close approach to perturb the orbit, correct?

Maybe the iTelescope will catch 2018 FE4 tonight. The weather was bad in New Mexico last night.

Phil S.