File talk:Positions of Near Earth Asteroids at time of discovery.png

Failed verification, possibly original research

I am copying over the relevant post from the article talk page at 99942 Apophis on the English Wikipedia, originally raised by me in November 2021:

The (very useful) image [1] gives a source that does in no way support the data presented in it. Looking at its file page, things only get worse, as the page makes claims that the source can never support. The Close Approach List given as its single source does not contain any positional data that could be used to create a plot like this, let alone physical explanations, like the stated (on the file page) connection to the opposition effect. Maybe someone can look into that, too, and determine to what extent that image is WP:OR and to what it just needs better sources. Renerpho (talk) 06:27, 27 November 2021 (UTC)

Hi, the issue seems to be with the lack of citing of full sources? The page referenced links to the individual pages for each asteroid. The data you are looking for is contained in those individual pages for each of the asteroids plotted. I don't think that citing the individual references to every single asteroid would be helpful to the reader, but I agree it may be helpful if it is mentioned below that the cited source contains the links. Rafflesgluft (talk) 14:00, 4 June 2022 (UTC)

I have removed the image from the Apophis article, for now, until the issues can be resolved. Renerpho (talk) 02:39, 3 May 2022 (UTC) Removed from the other articles as well (again, just for now). Renerpho (talk) 09:59, 13 May 2022 (UTC)

Thank you for clearly stating your concern that this is OR, which I'll be happy to address. Regarding removing the image the discussion is supposed to happen first I think, although I will admit that I've been a little slow spotting your message on my talk page - apologies. In the meantime, if you could restore the image to the articles, it would be very much appreciated. Of course if after discussing the matter we conclude (either by ourselves, or by involving arbitration) that it is original research, fair enough we will need to replace it with something else that's agreed on instead. Rafflesgluft (talk) 14:28, 4 June 2022 (UTC)

Discussing it was my intention, hence why I originally raised the issue on the talk page here AND on the Apophis article talk page, in November 2021. I gave it 6 months (MUCH longer than I had to, I believe), but after zero response, I removed the image until the issue is resolved. Also compare my edit summary.[2]. Renerpho (talk) 03:25, 1 July 2022 (UTC)

[...] the vast majority (87%) within a 45° cone. This is called the opposition effect. - No, it is not. The opposition effect is only relevant within a few degrees from true opposition (and only really within 1-2°, where it causes noticeable brightening). The effect described in the image is a combination of geometry (objects tend to approach Earth near opposition, at least if they are at some distance and outside Earth's orbit), phase angle (asteroids tend to be much brighter at low phase angles), and observation bias (observatories tend to image the region around opposition, not least because they can only observe at night, and generally avoid regions far from the ecliptic). Renerpho (talk) 17:12, 13 May 2022 (UTC)

When you say "phase angle (asteroids tend to be much brighter at low phase angles)", I am in hearty agreement. The phrase "Opposition effect" was chosen on the basis that there is no hard limit, the brightness decreases gradually as the phase changes from full. However it's perhaps not the right term. The wording that goes with the image just needs to capture the fact that there is a known effect here - do you have a alternative phrase which we could use? Rafflesgluft (talk) 14:44, 4 June 2022 (UTC)

That effect is called the "day and night cycle". Objects are found near opposition because that's where they are when they are visible at night. All other effects (including the actual opposition effect) are secondary. I don't think I'd call it an "effect", but I agree it is worth mentioning. Apophis was unusual because it was discovered at low elongation, during twilight. Renerpho (talk) 02:49, 1 July 2022 (UTC)

Verifiability / Original Research?

To recap WP:OR says "original research" (OR) is used on Wikipedia to refer to material—such as facts, allegations, and ideas—for which no reliable, published sources exist.

I do not believe this is original research as it is just a plot of data. It is a non-trivial plot, so had to be done in a spreadsheet, rather than by using wiki-charts, but nonetheless the data comes from the cited source (and pages it links to). As such someone else could re-construct the plot from the data available from NASA without any input from me. Ergo it is not original research. To the Verifiability point, as discussed above, it does not seem useful to link to every individual page which the data comes from, but the data is nonetheless obtainable. Please feel free to check it. Thanks Rafflesgluft (talk) 15:07, 4 June 2022 (UTC)

Your understanding of what is permitted under WP:OR is much broader than mine. Your source would have been sufficient if all the numbers used in the plot were explicitly stated on that page. What is said on pages that are linked to from there is irrelevant. We can plot data that is given, but we cannot expect users to know how to generate the data themselves. However, giving detailed instructions, as you do below, should be enough, as long as the procedure is absolutely clear. Renerpho (talk) 03:04, 1 July 2022 (UTC)

Also, the WP:OR issue was not just about the data itself, but also the image description. Nowhere in the source is the word "opposition effect" mentioned, or any of the other things claimed to be shown in the image. Plotting data is one thing; interpreting it is something else. Also see the section about synthesis of published material. If the interpretation is not given in the source then that is WP:OR. The image description needs to be rewritten to meet the standards of WP:OR and WP:VERIFY. Renerpho (talk) 03:43, 1 July 2022 (UTC)

For reference to get the details for 99942 from the cited reference:

• Follow the link to the NASA JPL NEO Earth Close Approaches page
• Select the 1LD filter as the maximum close approach distance you are interested in
• Select "All available data" (note, I advise to do this after you have selected the "<= 1LD" filter, or you will be downloading a very large amount of data)
• Select "Show 100 entries"
• In your web browser use Ctrl-F to search within the page for 99942
• Click on "99942 Apophis (2004 MN4)"
• See Discovery Circumstances section, and note discovery date: 2004 June 19
• Scroll back up to the title "99942 Apophis (2004 MN4) Classification: Aten [NEO, PHA]" and click " Related Links: Ephemeris"
• Edit Time Specification to include the discovery date, noted above
• Click "Generate Ephemeris"
• Read off the phase angle and elongation from the S-O-T and S-T-O columns (both approx. 55.7°)
• The above is enough information for a plot, but it's easier if you also use the delta (range of the target from the observer in AU) - the angle and range is what is shown in the plot.

Rafflesgluft (talk) 15:47, 4 June 2022 (UTC)

@Rafflesgluft: Thanks for your reply, and apologies that my own response comes so late. I'll add the image back into the articles. Regarding the reference, are you saying that you repeated that procedure for each and every asteroid in the plot? That's quite an effort. May I suggest instead to tell the reader that each data point can be generated using an API query of the form https://ssd.jpl.nasa.gov/api/horizons.api?format=text&COMMAND='2004 MN4'&START_TIME='2004 June 19'&OBJ_DATA='NO'&MAKE_EPHEM='YES'&EPHEM_TYPE='OBSERVER'&CENTER='500'&STOP_TIME='2200-01-01'&STEP_SIZE='300y'&QUANTITIES='23,20', where the first two inputs (the ones in bold), are changed to the designation and discovery date of each asteroid? The output of that query will be the two numbers that were plotted in the diagram (S-O-T and delta). Unless, of course, you have some method to automate that, so it doesn't need to be repeated for each of the ~1,000 asteroids in the table! Renerpho (talk) 02:49, 1 July 2022 (UTC)

The image has been added back to the Apophis article (see edit summary there).[3] The reference needs to be adjusted, with whatever results from the discussion here, before the {{failed verification}} template can be removed. But I am sure we can come up with something that works. If you don't mind, I would add it back to the other articles (NASA, etc.) once we have written a proper procedure to generate the data. Maybe adding a footnote where we describe the procedure would be a good idea? Renerpho (talk) 03:36, 1 July 2022 (UTC)

@Renerpho Thanks. Apologies for the delay again. Yes sounds good. The API query is very useful and yes I agree should be included and yes a footnote sounds like a good place to put that. 99942 is unusual in that the discovery date is included in the JPL data. For most NEOs, the reader will have to take an additional step of looking that up at the minor planet centre like so: http://www.minorplanetcenter.net/db_search/show_object?object_id=2021+YK. Unfortunately I do not have the time to do any updating right now, but please feel free to do so, if not, I'll get to it at some point - the amount of time I get to spend on Wikipedia varies quite a lot! Many thanks Rafflesgluft (talk) 09:09, 27 July 2022 (UTC)