Zharkova et al.: an update

Last year I wrote a couple of posts about a paper by Valentina Zharkova and colleagues, which suggested that global warming was partly due to the Earth moving closer to the Sun as the Sun moves around the Solar System barycentre. If you read my posts, the comments, and this pubpeer thread you’ll notice a number of people pointing out that this was wrong; the Earth-Sun distance doesn’t change as the Sun goes around the Solar System barycentre.

Recently, an editorial response and comments from two reviewers have been released, as has a response from the authors. They’ve also released an erratum. You can find some discussion of this on Geoff Sharp’s post.

Unsurprisingly, the editorial response and the recviewers agree with the criticisms of the paper. The editor says:

There is a legitimate, and based on my reading, completely correct concern about the Earth-Sun distance calculation.

Similarly, reviewer 1 says:

Earth is orbiting about the sun and the barycentric motion of the sun is irrelevant for computing/predicting Earth’s temperature. In a simpler term, the Sun is very close to being the prime focus of the Earth orbit.

In their erratum, the authors acknowledge that they were wrong to suggest that the Earth-Sun distance changed as the Sun goes around the Solar System barycentre. However, they then go on to claim that there is still quite a substantial change in the Earth-Sun distance on century timescales. However, as pointed out in Geoff Sharp’s post, what they’ve done is to consider only the perihelion distance. This is the distance between the Sun and the Earth when the Earth is closest to the Sun. It’s well known that there is a slow change in the eccentricity of the Earth’s orbit; it goes from being almost circular, to being slightly elliptical.

This does indeed then lead to a slow change in the perhelion distance. However, there is an equivalent, but opposite, change in the aphelion distance, so that the average distance remains unchanged. This is illustrated in the figure in this comment. So, despite acknowledging a pretty fundamental error when discussing the motion of the Sun around the Solar System barycentre, they’ve gone on to make another pretty fundamental error.

I don’t know what the outcome of this is yet. It’s the authors who have released the editorial response, reviewers comments, and their erratum. Their erratum is still not published and it may be that it goes through further review. Would certainly hope so.

This entry was posted in Global warming, Philosophy for Bloggers, Research, Sound Science (tm), The philosophy of science and tagged , , , , , . Bookmark the permalink.

77 Responses to Zharkova et al.: an update

  1. Joshua says:

    It’s surprising that she acknowledged any errors at all, given that the errors were pointed out by alarmists

    Following criticism of the paper, lead author Valentina Zharkova, of Northumbria University, described Rice as a “climate alarmist” in an online discussion.

  2. I was accused of being in denial during our email exchange.

  3. Everett F Sargent says:

    What goes up …

    Must come down …

  4. EFS,
    Where did you get that from. The reason I ask is that the figure in Zharkova’s seems to show the distance decreasing, while yours shows it increasing (which is what I was expecting). I’m not entirely sure quite what they’ve shown in their erratum.

  5. Everett F Sargent says:

    On the above plots. I mentioned in the old thread that you can obtain the full ~30.4 kyr JPL ephemeris data from here …
    http://iaaras.ru/en/dept/ephemeris/online/
    I did that for daily data for a total of 11,100,020 data points.

    In this particular exercise, I did a time domain analysis for minmax (the peaks and valleys of the annual cycle, also did zero crossings before that, to define the respective minmax ranges). I even fit a quadratic to each 3-day minmax (to get a form of interpolation for fractional date and distance, BTW makes no substantive difference).

    The full 30 kyr look like …


    (same units as previous plots)

    One final note. The above are a factor of ~six lower then as shown in their Figure 3 (the rate of change is 6X lower then their claim for the current ~1000 year era (1650-2650 CE).

    I actually don;t know what all they are doing. Each interation of their paper only gets worse (and they lost an author, four is now three in their erratum). :/

  6. EFS,
    Great, thanks. That’s what I was getting. I was assuming that they’re plotting perihelion (because the values are all less than 1) but I’m really not entirely sure.

  7. Everett F Sargent says:

    ATTP,

    I think I’ve done the above correctly, I used this subset (minmax and zero crossings) to define the eccentricity per the standard definition. These two curves define the envelope of Earth’s motion in AU.

    Again, I have absolutely no idea what all they are doing anymore.

    The both claim, at the exact same time or era, a so-called GSM (within the next few decades) and state that Earth will warm by 0.5C/Century over the next ~six centuries.

  8. EFS,
    Looks very similar to what I got with my code (the x-axis should be relative to about now).

  9. Everett F Sargent says:

    Also, h/t to Geoff Sharp, as that made me look into their Figure 3. It would appear that they all are still hung up on the four ice giants and their own misunderstanding of what the JPL documentation explicitly states,

    So that plot might be something to do with including or excluding the 3/4 ice giants, who knows.

    They go all conspiratorial here in the erratum …

    “Note. We wish to point out that after we submitted the previous reply to the Editorial Board in September 2019, this option to include all four large planets for the Sun-Earth distance in the JPL ephemeris has been removed from this software from October 2019.”

    They use some 3rd party software instead of the JPL or INPOP or EPM. I don’t know why. :/

  10. Everett F Sargent says:

    ice giants = Gas giants = Jupiter, Saturn, Uranus and Neptune = those four large planets outside of the Solar System Asteroid Belt, or some such.

  11. Everett F Sargent says:

    VV et. al. are using this software in their ephemeris calcs …
    http://www.alcyone.de/
    Like really up to date stuff if your clock stopped a decade or so ago …

    “The ephemeris calculation is based upon Steve Moshier’s analytical ephemeris using trigonometric expansions for the earth and planets and the lunar ephemeris ELP2000-85 of Chapront-Touzé and Chapront for the moon, both adjusted to Jet Propulsion Laboratory’s DE404 (see http://www.moshier.net). There are further adjustments in Alcyone Ephemeris, some optional, to JPL’s more recent DE406, the most accurate long-term ephemeris.”

    The trial version of this software dates to 2013-04-03, coming up on seven years old. The http://www.moshier.net no longer works.

    This so-called paper has drowned repeatedly and often, no amount of resuscitation will ever revive it, it should have been declared DOA from the get go.

    The GSM might actually happen, but this paper isn’t proof, it is more like Roulette, enough people bet and the table is 100% covered, someone will always win, and that is what this is all about, someone will get lucky and unfortunately famous for guessing correctly. :/

  12. Everett F Sargent says:

    My bad, The http://www.moshier.net link works. Don’t know how recent their version of the sourced software is. I installed the trial version it dates to 2011-12-03 version v4.3.0.680 in the About pane. :/

  13. Everett F Sargent says:

    I just wanted to post the main link to the review and response comments at …
    http://mpee.northumbria.ac.uk/staff/slmv5/kinetics/news.php

    There are a total of three relatively recent documents (not counting the erratum) …

    “2a. Reviews N2 of our published paper in December 2019 by two referees and the Deputy Editor of Scientific Reports after the paper published” (2019-12-06)

    http://mpee.northumbria.ac.uk/staff/slmv5/kinetics/Summary%20feedback%20for%20the%20editors.pdf

    ”2b. Our replies N2 to the reviews above by two referees and the Deputy Editor of Scientific Reports after the paper was published” (2019-12-17)

    http://mpee.northumbria.ac.uk/staff/slmv5/kinetics/reply%20to%20editorial%20comments2_subm.pdf

    “5. Reply to the reviewer comments N1 done for the Editorial Board in September 2019 after the paper was published on 24 June 2019” (2019-10-24)

    http://mpee.northumbria.ac.uk/staff/slmv5/kinetics/Editor_Reviewer%20comments_v3b.pdf

    The 2nd of these is what I have called the crazy train to crazy town document (promotes a bogus conspiracy theory) …

    “Furthermore, after we submitted the answer to the Editorial Board in September 2019 and the plots of S-E distance derived from the current JPL ephemeris considering all four large planets (Jupiter, Neptune, Saturn and Uranus) taken from the Alcyone ephemeris software website, this option has been removed in October 2019. This looks like a cover up of someone’s wrong results, which they do not want to be known to other researchers and the general public.

    The removal in October 2019 of the option to consider the gravitational effects of four large planets for calculation of S-E distance in the Alcyone Ephemeris software speaks for itself revealing that someone does not want the scientists and general public to learn about this missing SIM effect, which was not considered in the well-advertised models predicting the terrestrial temperature variations.

    We hope that the Editorial Boar of Scientific Reports will not be tricked to take part in the cover up process of someone’s scientific omissions and will not retract our paper.”

    Or maybe their 30-day trial licence expired on this eight year old piece of software and they are too cheap and lazy to buy a $50 licence!!! 😉

  14. David B. Benson says:

    Just consider conservation of energy.

  15. izen says:

    My shallow grasp of orbital mechanics suggests that not only does the average distance remain unchanged, but that the orbital speed is greater at perihelion as the eccentricity increases; and the planet slows and spends longer at the maximum distance.
    I lack the maths to determine whether this changes the total energy received.

    There has been a strand in soviet science that holds that the Sun is the primary control knob of the Earth’s climate. When TSI failed to comply by showing an increasing minimum as warming continued, attention turned to any other factor in the Earth – Sun relationship to support this a priori conviction.
    It is a measure of the motivated reasoning that any attempt to show that the correlations found between the barycentre, perihelion, axial tilt is too small and too slow to be an explanation of the observed warming will be dismissed as a conspiracy to den… suppress the ‘self-evident Truth’ that the Sun controls the climate.

    That the Russian economy is primarily dependent on the export of fossil fuels may have no influence on this.

  16. Everett F Sargent says:

    The motif of these two graphs (the top one is the same as Figure 3 in the erratum) …

    Is the same as the motif of the Alcyone Ephemeris software …

    Many more examples of Alcyone Ephemeris plots here …
    http://www.alcyone-ephemeris.info/ae/documentation/Index.html

    So, I conclude that Figure 3 is some form of artifact (or counterfactual) time series plot using the Alcyone Ephemeris software with an unknown number of Solar System bodies used.

    And boohoo, their trial licence expired or they have since forgotten the software settings that produced that plot. In other words, it would appear that they can no longer reproduce their own results from just a few months ago. Ergo their conspiracy theory. 😦

  17. Holger says:

    I followed that on Geoff Sharp’s blog. Quite interesting, especially using (stealing?) a graph from Sharp and claiming it as her own work. As Sharp wrote: “Recently I questioned Zharkova re stealing my plot and she stated it was just a coincidence and an artefact of her plot building software (the colors matched etc) which is clearly impossible”.

  18. David B. Benson says:

    Back to the main point, it looks very bad:
    https://m.phys.org/news/2020-01-climate-paris-goals.html

    Climate sensitivity is very high.

  19. Holger,
    Yes, using Geoff Sharp’s figure without credit and suggesting that they produced it themselves is very questionable. It is obviously his figure.

  20. The changes in aphelion and perihelion seems enough to start an iceage. The interesting thing is that on average there is no change at all in solar-energy globaly and over the year. Just very small redistributions over the seasons and latitudes. But this seems to be enough to start it some how.

  21. Dave_Geologist says:

    As a clue Raymond, consider two things: the unequal distribution of land and water between the Northern and Southern hemispheres; and whether or not there is a continent at either pole. Combine that with which way the Earth’s rotation axis is pointing at aphelion and perihelion. Other factors include whether a polar continent is surrounded by ocean or joined to temperate continent(s), and whether it is low-lying or is the continent with the highest mean elevation; and whether a polar ocean is land-locked.

    Of course that is all well understood, and has nothing to do with our current warming. Indeed IIRC it currently contributes to a cooling trend, and is part of the explanation for our own contribution to the present warming being more than 100%.

  22. Everett F Sargent says:

    1st, let me see if this shows up as an inline image …

  23. Everett F Sargent says:

    ATTP,

    Great! 🙂

    I have figured out how Figure 3 of the ERRATUM was generated in the Alcyone Ephemeris software. I can pretty much generate concave up or concave down or straight line with different amplitudes of AU up to the the eccentricity of whatever era you are in.

    It is as I originally thought, given the density of data as shown in their Figure 3 which happens to be ~annual. They may not have done this on purpose though (e. g. Hanlon’s razor).

    They are picking a start time and time step, where the time step is close to the annual period (365.23 days in the above example) and the start time is in the right AU location (in this case 1600-02-28 6:00 UT) of the annual cycle. It should also be possible to do this with a time step, close to and above, the actual annual period (e. g. 385.27 date but at a different start point in the annual AU cycle).

    This is a classic though, for anyone who does time series analyses, and is well known as data aliasing. I won’t spend anymore time fine tuning the above to get to their graph either, as the above shows the necessary methodology needed to produce almost any aliased frequency and amplitude (within the bounds of the annual eccentricity time series).

  24. Everett F Sargent says:

    “e. g. 385.27 date”

    should be …

    “e. g. 365.27 days”

    in my previous post.

  25. EFS,
    Thanks. So, their figure starts in 1600, which is why it is showing the perihelion getting smaller?

  26. Everett F Sargent says:

    “Thanks. So, their figure starts in 1600, which is why it is showing the perihelion getting smaller?”

    Think of it like this: I am looking at that part of the cycle AFTER the actual perihelion of any given year.

    (I) Pick a start time where the Earth-Sun distance is less than one in AU In this case, we are aliasing the mean annual cycle time to complete exactly one orbit (or year) by taking the sampling interval with a period less than ~365.25636 days or whatever the annual mean cycle time is. One almost annual data point in the bag as it were. We are at that starting point in the orbit past perihelion but before the point where Earth is more then one AU of that cycle.

    (II) My next sample will just be short of one year. So that data point will be just ever so slightly closer to the actual perihelion of the NEXT full cycle..

    (III) Repeat the above just slightly shorter than a year sampling interval and you are walking very slightly towards the actual perihelion of a far future year, You appear to get closer to the sun because you are taking ever more time away from the actual mean cycle time,

    The aliasing period of the annual cycle time calculation follows …

    Mean cycle – 365,25636 days.
    Sampling interval = 365,23 days
    (vertical bars below mean absolute value(

    | 1/365.25636 – 1/365.23 | = 1.97597e-7 (frequency difference in 1/days)

    period = 1/frequency = 5,0608e6 days = 13855 years

    That long of a period also means that the long term change in perihelion (increasing) must also be taken into account (but the rate of aliasing (-delta1 AU per year) is greater in magnitude than the rate of perihelion growth (+delta2 AU per year)).

    If it still isn’t clear see …
    https://en.wikipedia.org/wiki/Aliasing
    https://en.wikipedia.org/wiki/Wagon-wheel_effect

  27. EFS,
    Oh, I see. It’s even worse than I had realised. They’re not plotting the perihelion, they’re plotting the distance at time intervals of slightly less than 1 year, so it appears to be getting closer, but it’s just an artefact.

  28. Everett F Sargent says:

    ATTP,

    The start year and sampling interval are my own selections using the 30-day trial version of the same software VV was using. Call it a proof of concept (All AU’s < one, concave up and decreasing AU versus time so that the resulting graph makes it look as if the Earth-Sun distance is decreasing, but it is not decreasing, because the time series is sampled at a slightly lower annual period and specific start point).

  29. Everett F Sargent says:

    “they’re plotting the distance at time intervals of slightly less than 1 year, so it appears to be getting closer, but it’s just an artefact.”

    Exactly! 🙂

  30. Everett F Sargent says:

    It also works for sampling periods slightly greater then one year. In that case, you pick a start time below one AU but where the cycle location is heading towards perihelion for that cycle (thus the absolute value of the frequency difference calculation); You get the same end effect, the aliased time series gives the impression of decreasing Earth-Sun distance.

  31. Game, set, match.

    Zharkova et al. used their models to support the temperature trend prediction of Akasofu. Dr. Ken Rice showed Akasofu’s temperature trend prediction was wrong, which Dr. Zharkova initially mocked:

    https://pubpeer.com/publications/3418816F1BA55AFB7A2E6A44847C24#37
    https://pubpeer.com/publications/3418816F1BA55AFB7A2E6A44847C24#38
    https://pubpeer.com/publications/3418816F1BA55AFB7A2E6A44847C24#39

    Zharkova et al. now admit Akasofu’s temperature trend prediction was wrong, though they don’t adequately address the implications of their model supporting this false prediction. They also state their hypothesis is not an objection to anthropogenic, greenhouse-gas-induced warming:

    “We wish to emphasize that we clearly stated in the paper that we do not challenge the anthropogenic effects on temperature increase, which have its own remits [merits?] as indicated by one of the replies N38 by Dr. K. Rice.
    Our prediction of temperature increase based on a straight line in the curve by Akasofu, 2010. Dr. Rice has shown in his reply to Pubpeer N38 (see Figure 3 below, blue line plotted by Ken Rice) that the terrestrial temperature increase above the baseline is higher than predicted by Akasofu, 2010”
    https://web.archive.org/web/20200128075803/http://mpee.northumbria.ac.uk/staff/slmv5/kinetics/Editor_Reviewer%20comments_v3b.pdf  

    So when people cite Zharkova’s work to me, that’s one of the first things I’ll tell them; human-made effects are outside its scope, and its predicted temperature trend effects are dwarfed by anthropogenic factors.

    A few sentences after that, they then say that “AGW people” (as they seem to call mainstream climate scientists) can include the temperature effect of TSI variations in their climate models… as if Zharkova et al. are unaware that climate scientists have done that for decades. 

  32. Everett F Sargent says:

    AS.

    You mean like this …
    “The PMIP4 contribution to CMIP6 – Part 3: The last millennium, scientific objective, and experimental design for the PMIP4 past1000 simulations”
    https://www.geosci-model-dev.net/10/4005/2017/

  33. Everett F Sargent says:

    I mentioned Hanlon’s razor in a previous comment.

    So I did a reverse engineering on this VV graph (the bottom plot) …

    And this is my result …

    The Earth-Sun distance is still aliased even though VV chose a Gregorian Year (365.2425 days) as the sampling interval.

    Start time: 2013-01-29 09:32
    Stop time: 2610-01-29 23:00
    Step size: 365.2425 days

  34. Ben McMillan says:

    Nice work.

    I wonder what the reputational consequences are of this kind of thing.

  35. Ben,
    There must be some reputational consequence. However, as far as I can tell, most will be largely unaware of this, and there will be little in the way of a substantive reputational impact.

  36. Ben McMillan says:

    Interestingly, I talked to a solar physicist who knew about this independently, so I’m thinking it might be pretty widely known.

  37. Okay, maybe it is more widely known. I guess one issue is that researchers, in the sciences at least, seem to be reluctant to publicly criticise others in a way that seems personal. It’s one thing to criticise someone’s work, but another to say something (publicly) that might be a criticism of them as a scientist.

  38. JCH says:

    So from medicine, the researcher who added animal blood in with human samples to somehow skew his result in his favor has some sort of professional invisibility?

    Oh, and the obligatory, that he did this in medical science casts a pall over all climate science.

  39. Everett F Sargent says:

    IMHO, the most disturbing thing about all this is that VV wants all of this to be known. No, not the objective facts. But the conspiracy that is trying to hide their broken SIM theory. Particularly if this paper is eventually retracted.

    There is a reason that the paper currently has an Altmetric of ~1038 (It has been in a continuous slow rise ever since the paper’s release) …
    https://www.altmetric.com/details/62615862

    It is social media (particularly the Twitless and the Utwob) that is maintaining such a (bad) metric. The GSM crowds, the anything-but-AGW crowds (yes, there are still many of those end-of-world types, it just aim’t due to AGW) and the conspiracy theory crowds are literally eating this stuff up.

    Of course, Nature Scientific Reports (NSR) is TOTALLY ignorant of these current social media trends. In fact, NSR is playing the fool, they just don’t know it, yet.

  40. Everett F Sargent says:

    Still not convinced of Earth-Sun distance data aliasing?

    What follows is exactly the same as my last reverse engineering effort above. The only change is step size from a Gregorian Year (365.2425) to a Sidereal Year (365.256363004 days) …
    https://en.wikipedia.org/wiki/Year#Astronomical_years

    Ruh Roh!

    Start time: 2013-01-29 09:32
    Stop time: 2610-01-29 23:00
    Step size: 365.256363004 days

  41. Everett F Sargent says:

    ATTP said …

    “I guess one issue is that researchers, in the sciences at least, seem to be reluctant to publicly criticise others in a way that seems personal. It’s one thing to criticise someone’s work, but another to say something (publicly) that might be a criticism of them as a scientist.”

    You mean like signing this …
    https://clintel.org/world-climate-declaration/

    And posting three links to the petition on their ERRATUM page?
    https://web.archive.org/web/20200129015535/http://mpee.northumbria.ac.uk/staff/slmv5/kinetics/news.php
    https://archive.is/4QQnC

    Ant then scientists keeping silent because #But_Scientist … :/

    You all do understand what all is going on here? A scientist (her not you) is 247 engaging in ClimateBall Bingo! 😉

  42. Willard says:

  43. Everett F Sargent says:

    Oh crap. ClimateBall Bingo at its finest …
    https://www.thesun.co.uk/tech/10864950/mini-ice-age-earth-weather/
    https://www.thegwpf.com/scientists-divided-over-impact-of-solar-minimum-on-global-temperatures/
    CHILL IN THE AIR Earth about to enter 30-YEAR ‘Mini Ice Age’ with -50C temperatures in coldest regions, scientists warn

    They misspelled “scientists” it should be “so-called scientist” or some such.

  44. Everett F Sargent says:

    Hmm, very interesting …
    Oscillations of the baseline of solar magnetic field and solar irradiance on a millennial timescale
    https://arxiv.org/abs/2002.06550
    Document properties for creation/modification is 2/17/2020, 8:00:58 PM

    Is this some form of normal procedure (publishing a copy of a paper to arxiv after publication in an open access journal) or could this mean what I think it might possibly mean? The R-word, and no I don’t mean …

  45. Everett F Sargent says:

    ATTP,

    “This article has been updated”

    “Change history
    04 March 2020 This article has been retracted.”

    https://www.nature.com/articles/s41598-019-45584-3#change-history

  46. dikranmarsupial says:

    Hidden the retraction notice fairly well, reminded me of the bit from The Hitchikers Guide to the Galaxy about where Arthur Dent says “… yes I did. It was on display in the bottom of a locked filing cabinet stuck in a disused lavatory with a sign on the door saying ‘Beware of the Leopard”” ;o)

  47. Marco says:

    There is now an official retraction notice:
    https://www.nature.com/articles/s41598-020-61020-3
    Not quite unexpected:
    “S. I. Zharkov agrees with the retraction. V. V. Zharkova, E. Popova, and S. J. Shepherd disagree with the retraction.”

    I’m expected a long whine to come out of the GWPF relatively soon.

  48. Marco,
    Thanks. I saw that quoted on Retraction Watch. Some rather unfortunate circumstances there.

  49. dikranmarsupial says:

    “S.I. Zharkov is Sergei Zharkov, of the University of Hull, Zharkova’s son. ”

    ouch! 😦

  50. Everett F Sargent says:

    I happen to have a digital copy of Figure 1a of the retracted paper. The so called summary curve (dark blue line), direct from the lead author, received on 2019-10-07. Figure 1a covers the period 1000 BCE to 2200 CE. Compare the Figure 1a era summary curve (1-2200 CE) to the similar era summary curve as shown in their Figure 3a (cyan line). Same era for both figures but different summary curves!

    One of these curves will generate their ~2000 year low frequency so-called baseline magnetic field, while the other (my digital copy of Figure 1a) will not (there is no low frequency spectra in the digital copy sent to me).

    Having played around with the Shepherd (2014) curve fitting equations, I now know what can happen if those equations do not produce a piecewise linear time series (e. g. undersampling will produce so-called clipped time series containing low frequency artifacts).

    In summary, there is not even one aspect of this paper that is worthy of publication.

    Any future so-called journal remotely considering publication of this nonsense, in whatever form, must demand all equations used with all constants defined explicitly as well as all digital data as presented in their paper.

  51. Geoff Sharp says:

    I am not sure sure if Zharkova’s fig. 3 showing the so called reduction in the Sun/Earth distance is an artifact of an aliasing problem.

    If you plot the Sun/Earth distance at perihelion you get the same result? She has just plotted what we would expect to see according to Milankovitch theory and the JPL Model.

  52. Geoff,
    Is that right? I thought we were currently in a phase where the eccentricity was getting smaller and, hence, the perihelion distance was slowly increasing.

  53. Everett F Sargent says:

    ATTP,

    I know that I am right …



    It is really hard to argue against the above facts. I am at p=1 on this one.

    Geoff, please show your own work here (in this thread) if you disagree. TIA

  54. EFS,
    Can you clarify the source of the above figures (are they yours)?

  55. Everett F Sargent says:

    Yes, all are from the 30-day trial version of the Alcyone Ephemeris software.

    You only need download said software and use my start, stop and time step to reproduce. Not to sound too paranoid, but this discussion needs to, no must be, done in an open forum/format.

    Do you have any other questions on any of my plots. I can also forward a communication I had with JPL.

    Flip I could have done the basic distance calculation in high school, if not before.

  56. EFS,
    Thanks. So, you’ve essentially reproduced the figures in Zharkova et als. update using the same software and using the procedure your describe. Pretty convincing 🙂

  57. Everett F Sargent says:

    Oops, my bad. The 1st figure originated from VVZ, The 2nd figure was my use of the same software that VVZ used to make the 1st figure. The 3rd figure is mine after the 2nd was completed using 365.2426 days and then using 365.25636005 days per year.

    Really sorry for any confusion on that 1st graph.

  58. EFS,
    I’m getting myself slightly confused now. What is the perihelion distance doing over the same period. My understanding is that it should be increasing. Why does your figure using 365.25636005 not show something similar (i.e., distance increasing, rather than decreasing)?

  59. Everett F Sargent says:

    “365.25636005” should be “365.256363004” days (sidereal year_
    365.2425 days is Gregorian year.

    Sorry for those typos.

  60. EFS,
    Just to clarify what I think is happening in those figures. You’re essentially sampling at different locations in the orbit, so the distance seems to be decreasing, even though the perihelion distance is actually increasing. Does that sound right?

  61. Everett F Sargent says:

    ATTP,

    In that case, or in general, for aliasing I can be anywhere in the annual cycle. If I am past the perihelion then my sampling period must be ever so slightly less then one sidereal year, conversely for aliasing before perihelion my sampling period is ever so slightly greater then one sidereal year.

    If I STARTED exactly at perihelion of a given year then in either case (period other then sidereal) I would be moving from perihelion to some other larger point in the annual cycle.

    If I STARTED exactly at perihelion of a given year then for exactly sidereal period then that would be the perihelion time series.

    There are two rates: perihelion growth rate and aliasing growth rate (the aliasing growth rate is larger (negative) than the perihelion growth rate (positive)).

    I do have a knack for Newtonian spatiotemporal stuff (which I’ve known since at least high school).

  62. EFS,
    Okay, thanks. I think I get it now.

  63. Everett F Sargent says:

    “You’re essentially at different locations in the orbit, so the distance seems to be decreasing, even though the perihelion distance is actually increasing. Does that sound right?”

    Yes.

  64. Pingback: Breaking good news - Ocasapiens - Blog - Repubblica.it

  65. Geoff Sharp says:

    Sorry Ken, I meant to say if you plot the aphelion distance you will get the same result as Zharkova’s plot. The Aphelion shortening as the earths orbit becomes more round.

  66. Geoff,
    Yes, but that should then be bigger than 1.

  67. Ben McMillan says:

    EFS: I’m wondering if a graph of a short period, with a few traces, would make the point more clearly (or at least in a more compact way):
    1) Daily sun-earth distance.
    2) Annual mean, sun-earth distance.
    3) Sun-earth distance sampled every 365 point whatever days.

    Might need a plot + zoomed plot.

  68. Everett F Sargent says:

    Ben, ATTP, Geoff …

    From the ready made Internet …


    A short periodicity becomes a longer periodicity.

  69. EFS,
    Thanks, that illustrates it nicely.

  70. Everett F Sargent says:

    The above are already demeaned, so add one AU to get back to the original eccentricity time series (this is an oversimplified example of an actual orbital cycle as there would be higher superharmonics for an actual elliptical orbit).

    Tamino has a beautiful discussion, a decade or so ago, about limited telescope time and how one goes about removing aliased periods for very limited uneven temporal sampling. Of course, this only works for very narrow banded spectra, so we still need oversampling (and Nyquist) for things like ocean water wave spectra since that type of spectra are not narrow banded.

    I’ve been doing this stuff for way to long. 🙂

  71. EFS,
    Yes, I think I remember Tamino’s discussion about this. We do the same kind of thing when exoplanet hunting using the radial velocity method. We don’t try to sample a single orbit, we sample in such a way that hopefully we reasonably sample the orbital phase (i.e., you’re not sampling at the same time in the orbit).

  72. Everett said:

    “This is a classic though, for anyone who does time series analyses, and is well known as data aliasing. “

    There are also time-series that naturally undergo a process similar to data aliasing. Consider calculating exactly when eclipse cycles occur given the significant difference between the daily, monthly, and yearly signals involved. Arguably, this is not an artificial sampling issue but just something that one has to understand to calculate eclipse times.

    “A short periodicity becomes a longer periodicity.”

    Moreover when these aliased intersections occur naturally and they have the chance to amplify through a nonlinear interaction, they can in fact create behaviors such as super-tides, etc. Instead of physical aliasing, I think this is more well known as calculating a combination period, but the graphical construction is much the same.

    I am pretty sure that this does not apply to the Zharkova analysis but it can apply elsewhere such as in this recent paper where the term “physical aliasing” is actually used: “On the aliasing of the solar cycle in the lower stratospheric tropical temperature”

  73. Everett F Sargent says:

    PP,

    OK. Whatever.

    But here is the real backstory …

    I downloaded and installed said Alcyone Ephemeris software. I played with it for a while.

    In the PubPeer discussion I mentioned over seven months ago (commented July 26th, 2019 4:59 PM and accepted July 26th, 2019 4:59 PM) about potential temporal aliasing, but at that time I had no idea that this, now nearly decade old, software was used …
    https://pubpeer.com/publications/3418816F1BA55AFB7A2E6A44847C24#261
    “The main problem would be the ALIASING of that annual time series, since a value very close to the annual cycle is taken.

    No one who knows anything about time series analysis, would not and should not, do something so fundamentally stupid.”

    Now, only an ignoramus, who knows absolutely nothing about data analysis and/or Earth’s motions, would use a Gregorian Year as a metronome beat frequency for the annual cycle. No one except you know who.

    So, I said to myself, if I know absolutely nothing about astronomical years, what dumba$$ period would I use? It could be 365 or 366 or 365.25 or viola 365.2425 years, you know something to do with human calendar years. So my 1st try was 365.2425 and I stopped right there. I actually backed into the exact periodicity that VVZ used, it was then a matter of picking a start time, but I was already close enough for government work, as it were.

    Also, go back to the long discussion I had with VVZ over the JPL documentation …
    https://www.nature.com/articles/s41598-019-45584-3#article-comments

    I was not dealing with a normal person who understands the English language fully (heck neither do I, I often say I don’t have a 1st language, but not in this case against VVZ).

    In summary, I don’t think I’ve ever seen such a sloppy paper written and published in a peer reviewed publication. This paper wins the EFS Lifetime Award for worst paper ever.

  74. dikranmarsupial says:

    EFS, I have seen far, far worse!

  75. Everett F Sargent says:

    DM,

    Yes. Let’s just say I have OCD and tunnel vision to boot.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.