Transients in the Palomar Observatory Sky Survey

[John J.]

Their whole story is based on the aliens moving objects into an orbit where they can see and be seen but immediately moving them away where they cannot be seen.

I think Villaroel's hypothesis is that the ETI satellites are in regular orbits, possibly for protracted periods (Villaroel et al.'s "Glint in the Eye" paper) but sometimes (perhaps due to rotation or whatever) a highly-reflective facet of their surface is at just the right angle to reflect sunlight in the relevant direction for a relatively brief time.

 

[NorCal Dave]

I think Villaroel's hypothesis is that the ETI satellites are in regular orbits, but sometimes (perhaps due to rotation or whatever) a highly-reflective facet of their surface is at just the right angle to reflect sunlight in the relevant direction for a relatively brief time.

But also, they reflect that sunlight more often within 3 days of a nuclear test?

 

[John J.]

But also, they reflect that sunlight more often within 3 days of a nuclear test?

Good point.
Maybe it's because the lens cover has been opened? ()

 

[JMartJr]

Hi, and thanks for the response. If you don't mind, I've got a couple of follow-up questions -- not for purposes of of playing "gotcha!" or anything, but to clarify my own understanding of stuff...

Orbits don't need to be maintained, because the objects in orbit are in freefall.

But don't various other forces (shifting gravitational tugging from the moon, or possibly more distant objects, atmospheric drag from whatever remnants of atmosphere are out that far, solar storms?) tend to mess with satellite orbits over time? We've apparently got millions of years worth of even tiny factors to mount up!

Geostationary orbits don't stay stationary, they drift a little and become geosynchronous if not maintained.

But for an object to leave orbit, it must either acquire energy (e.g. solar radiation) or lose energy (e.g. tides).

And would not both of those factors operate on alien satellite fragments over millions of years?

 

[Mauro]

Villaroel has kind-of opined that the causes of the transients are still there (in GEO), so presumably they have been in continuous GEO for many years (including prior to their being photographed in the 1950s, as per Villaroel et al.'s "Glint in the Eye" paper referenced by @flarkey, here).

But this (objects in orbit for a long time) means the transients at POSS-1 should not be correlated to nuclear tests. If they were always there, they should have been always spotted (*). So, they cannot both say 'they have been in continuos GEO for many years' and 'the reflections of the Sun they gave off are correlated with nuclear tests': one excludes the other (barring some ad-hoc excuse to explain why satellites always orbiting there only catch the Sun around the dates of nuclear tests). Holding both opinions as true as the same time is illogical thinking.

The irony is that, looking for a correlation with nuclear tests, they undermined their own hypthesis: they use the correlation as evidence in favour of alien satellites, but the probability of aliens actually goes down... because the alien's behaviour becomes weird, that is to say, unprobable. They had better stop at the Earth shadow analysis, but you know, agendas...


(*) and should have been seen in the all the full-sky surveys which followed, and in those which are ongoing now.

 

[jdog]

But this (objects in orbit for a long time) means the transients at POSS-1 should not be correlated to nuclear tests. If they were always there, they should have been always spotted (*). So, they cannot both say 'they have been in continuos GEO for many years' and 'the reflections of the Sun they gave off are correlated with nuclear tests': one excludes the other (barring some ad-hoc excuse to explain why satellites always orbiting there only catch the Sun around the dates of nuclear tests). Holding both opinions as true as the same time is illogical thinking.

The irony is that, looking for a correlation with nuclear tests, they undermined their own hypthesis: they use the correlation as evidence in favour of alien satellites, but the probability of aliens actually goes down... because the alien's behaviour becomes weird, that is to say, unprobable. They had better stop at the Earth shadow analysis, but you know, agendas...


(*) and should have been seen in the all the full-sky surveys which followed, and in those which are ongoing now.

Also, the UFO lore (which the paper cites as the rationale for testing the hypothesis) has always been that the aliens showed up around nuclear facilities, where the weapons were made and stored, and the facilities existed continuously, so there's no lore reason for observing mechanisms not to also be present continuously.

(I'd note that much of the mainstream coverage gets the gist of the paper wrong, like this USA Today headline: "Did aliens spy on our nuclear tests? Study finds signs of UFOs near US sites in 1950s" with the subheadline "UFO documentaries have long given credence to the theory that otherworldly visitors are interested in humanity's nuclear weapons. Now, researchers claim to have evidence of UFOs near test sites." Which has no relation to the content of the story, let alone the claims of the paper.)

 

[Mendel]

And would not both of those factors operate on alien satellite fragments over millions of years?

Sure. They operate on the moon as well.
The fact that the orbit undergoes slight changes matches the slight forces at play: you need energy to get to a higher or lower orbit, and that's hard to come by that far out.

 

[overlord]

Then after about April of 1956, despite ongoing nuclear tests, the aliens and thus the transients, buggered off for unknown reasons. Nevermind that's also when the red emulsion was changed.

This has been stated a few times now, can someone quote the source for the bold part?

 

[beku-mant]

Villaroel says she intends to make the data and code available. This is science, and given what a monumental discovery it would be, if they made an error, it will ultimately be figured out. I very much doubt she is trying to pull one over.

They could have made an error, but that would not imply they intentionally falsified their data. Scientists make mistakes all of the time, and it can be very stressful for them when it happens. When we accuse them of fraud without evidence, we create an environment of fear that causes people to avoid taking risks. Scientists need to be able to afford to make mistakes if they are to tackle difficult hypotheses. In my opinion, Villaroel et al. are courageous, even heroic, for taking these risks to their reputation by testing a risky hypothesis that subjects them to stigma, regardless of whether their results pan out or not.

It is very easy to accuse people of malfeasance. And people often do this on a hunch, without sufficient evidence. Vladimir Kramnik's accusations against Daniel Naroditsky come to mind. The stress of being accused of cheating definitely caused him a great amount of suffering, and may have played a major part in his tragic death. On the flip-side, Kramnik's allegations were baseless, but that doesn't make him a murderer.

We live in a world where people weaponize ridicule, accusation, and the discovery and excessively negatively framed publication of people's mistakes. That's part of a much larger, more general problem in today's world, especially in the information age where social media and internet forums are so influential, and where tribes wage war against each other.

Villaroel has been clear in her interviews that she is only human, and it is possible she has been wrong, and that she wants open data and open source so others can help to find out. With so much pressure on her now, it must be very difficult and stressful just to anticipate this process and how it unfolds under such a highly charged environment. It doesn't have to be this way. We can give people the benefit of the doubt, let them try to test a risky hypothesis, and allow them to make mistakes, all without dehumanizing them.

And if a mistake is found and the huge statistical significance goes away, it doesn't mean that the study of these transients loses all its value. It may have been a long shot that we find evidence of a very large amounts of glints in the first place. Some of the more credible UFO reporting dates back to this period, and the reduced noise from human technology compared to now makes it a compelling potential source of discovery. If those old UFO reports were accurate, then it is feasible POSS-I could have captured evidence, even if it is very, very difficult to find and validate that evidence among the noise. Maybe one day the physical plates can be inspected with advanced methods, for example, and we can gain much more precise information about potential causes of the transient candidates. At that point, having this catalogue of candidates could be very valuable.

 

[Mendel]

When we accuse them of fraud without evidence, we create an environment of fear that causes people to avoid taking risks.

You are the only person in this thread who uses the word fraud.

It is very easy to accuse people of malfeasance.

You are the only person in this thread who uses the word malfeasance.
I see what you are doing, and I don't like it.

Villaroel has been clear in her interviews that she is only human, and it is possible she has been wrong, and that she wants open data and open source so others can help to find out.

Villarroel wrote that the data on her 83 alignments would be in the appendix, but there is no appendix.
Villaroel referenced Solano 2022 and two data sets, 5399 and 298165, without clarifying that she used a different data set for the alignments. None of the 24 transients she does give are in the final Solano 2022 set, which means they all have alternate explanations; and Villarroel knows this because she co-authored Solano 2022 and cited it as a reference for her methods. I don't see her providing a well-founded significance level for the alignments, and neither is the shadow deficit analysis robust enough to establish an orbit as a cause.

Now I have two options as to what to think:
* either she did a terrible job, or
* she chose her data set by throwing Solano 2022 criteria overboard until she got the result she wanted. (Which is not fraud, but it's not science, either.)
If there's a third option, please explain why she chose this data set, and did not explain her choices in this paper.

(You've still not revealed what deficit you saw when you ran the shadow analysis on the 5399 data set, btw.)

 

[beku-mant]

If they did make a mistake, it's very probably both accidental and not one of the mistakes or abuses that has been floated in this thread or the other thread. I'll say that much. I'm not interested in speculating about the mistakes I would guess are mostly likely if they did make a mistake in a public forum at this point. Especially as someone who doesn't have all of the expertise and information to understand the nuance, because this can easily lead to poorly validated or unsubstantiated rumors. That is just my feeling, your are welcome to your own.

If there is a mistake, I think one should independently and diligently determine that, so that a claim that the mistake was made is credible and backed by evidence and rigorous methodology, before floating it publicly. And if a mistake is discovered, that is fine, that is how science works, taking risks and making mistakes is part of how we discover and learn, and they should still get credit for trying and being brave enough to put their necks out there.

 

[Ann K]

But don't various other forces (shifting gravitational tugging from the moon, or possibly more distant objects, atmospheric drag from whatever remnants of atmosphere are out that far, solar storms?) tend to mess with satellite orbits over time? We've apparently got millions of years worth of even tiny factors to mount up!

A collision of a satellite with another object would cause all the parts to be propelled into different orbits (perhaps nearly the same, perhaps not, depending on the force of the collision). Such an event would probably have a much greater influence than cumulative tiny factors.

 

[Kyle Ferriter]

I think Villaroel's hypothesis is that the ETI satellites are in regular orbits, possibly for protracted periods (Villaroel et al.'s "Glint in the Eye" paper) but sometimes (perhaps due to rotation or whatever) a highly-reflective facet of their surface is at just the right angle to reflect sunlight in the relevant direction for a relatively brief time.

A very brief period of time per flash event (and each of those being very bright), but each object also glinting multiple times at on a long enough repeating period that it shows up as distinct dots with no streaking, spread out in an approximately linear arrangement in the long exposure plate, right? Someone would need to do the astronomical math to determine what time period between glints on a line would make sense for a geostationary object. This linear arrangement criteria is a key part of the filtering criteria they applied after other more obvious defects had been filtered out based on PSF. Unfortunately I think they also filtered out streaks. I'm not sure whether they only filtered streaks which were known sky catalog objects or whether other unclassified streaks were also filtered out because it happened to not be what they were looking for.

This VizieR service has a dat file with metadata for all the observations:
https://cdsarc.cds.unistra.fr/viz-bin/cat/VI/25
https://cdsarc.cds.unistra.fr/ftp/cats/VI/25/ReadMe

Loading this into a sql database and doing some queries and python, you can get some useful high level information. Like exposures per night in the series of 937 red and blue exposures. Grouping the exposures by 24 hour periods centered on noon (not midnight, so it's grouped by "night" not by "day", e.g. thursday at 3am is categorized as the same evening as wednesday evening), the modal number of exposures per night for the span of the survey is 0. There are some periods where exposures were taken on multiple consecutive days, or many days clustered together, while there are are other periods where none were taken for weeks or months.

Can also look into stuff like day of week trends. For example when an exposure was taken after a period of at least two prior days of no exposure being taken, what day of the week was that evening?

Edit: also the inverse, nights with an exposure for which there was at least one exposure in the prior 2 nights. Again not a uniform distribution.

 

[John J.]

If there is a mistake, I think one should independently and diligently determine that, so that a claim that the mistake was made is credible and backed by evidence and rigorous methodology, before floating it publicly.

Peer and lay criticism of published work is perfectly legitimate (although by no means always correct).
The apparent ambiguity over data sets, and the current lack of data transparency, might indicate Villaroel's own evidence and methodology is flawed (it certainly impedes replication) but she "floated it publicly".
I don't think critics should be held to a higher standard than authors (though I agree they should also aim for accuracy and transparency).

Where claimed UFO sightings have been explained/ debunked, or when studies of UFO reports have concluded that there is no evidence of alien involvement, UFO enthusiasts are often vociferous in their criticism. That is their right.

And if a mistake is discovered, that is fine, that is how science works, taking risks and making mistakes is part of how we discover and learn, and they should still get credit for trying and being brave enough to put their necks out there.

Agree, to some extent. But again, extraordinary claims require extraordinary evidence, and that evidence should be accessible to aid replication.
Villaroel et al. don't have any proprietary rights over the NGS-POSS-1 plates, and it's hard to see how their data/ working might be commercially sensitive, lead to issues over intellectual property rights, contain confidential personal information etc.
It's not unreasonable that they list the dates of nuclear tests with which they claim transients correlate.
I accept that many published papers don't contain the data sets used, and authors might only share these with "legitimate" enquirers, but personally feel this is unfortunate. It might have been valid when data was largely hard-copy; pages and pages of data would put up the price of journals and researchers didn't always have the time and money to compile, and post, reams of paper to anyone who asked. But digitised information largely avoids those issues.

Not sure we should equate being wrong with bravery; that could end up with people who are consistently wrong or badly mistaken being lionized as risk-taking mavericks by some. I don't think Martin Fleischmann and Stanley Pons' announcement of the discovery of cold fusion (Wikipedia) is best characterised as "brave"; it was premature and deeply flawed.
Both men were respected in their fields, and I don't believe for one minute that they had any intention to deceive. But they were wrong, and their reluctance to accept this was not (IMHO) courageous as much as ill-judged.

Researchers sometimes make statistical errors and/or draw incorrect conclusions, and diligent findings published in good faith are often contradicted or superseded by later research. But the vast majority of those researchers are not making extraordinary claims or appearing on YouTube videos hosted by people known for promoting extraordinary claims on sometimes dubious evidence.
I can't really quantify it, but I feel there is a difference between someone publishing a relatively mundane finding (or something that is mainly of interest to specialists in a particular field) which is later demonstrated to be mistaken, and someone making an extraordinary claim which is later shown to be mistaken. Maybe it's because the latter tends to get much greater publicity and might negatively influence some people's understanding of science (there are people who believe Fleischmann-Pons cold fusion is a thing, EmDrives are a breakthrough technology, etc.), or maybe it's the poor track record of extraordinary claims (re. e.g. UFOs, psychic abilities).
Maybe the greater publicity associated with extraordinary claims means that those making them face more criticism and perhaps ridicule if their findings can't be replicated or are otherwise shown to be wrong, but to some extent that might be the result of the media that they choose to engage with.
Villaroel has chosen to discuss her views on NewsNation with Ross Coulthart. This doesn't affect the content of her papers, but it might raise questions about her beliefs; it's well established that the majority of errors found in academic papers favour the hypothesis held by the author (where that is known).

 

[MonkeeSage]

Ufologists never want to give the aliens credit for technologies so far beyond ours we would never even notice they were here.

Well obviously they aren't space aliens, they are cryptoterrestrial Anunnaki lizard people who co-evolved with humans and live deep underground or in bases on the bottom on the ocean (duh!). Their technology is only 50-60 years ahead of ours and when we started putting things in space they had to decommission the satellites they were observing the us with so we wouldn't detect them. But they figured out how to disguise themselves as Bill Gates, Barack Obama and Hilary Clinton so they could continue observing us unnoticed. (Sorry, couldn't help myself. As far as I know Dr. Villarroel has not made such claims, but the podcasts/shows she has been promoting her work on have.)

 

[Mendel]

Grouping the exposures by 24 hour periods centered on noon (not midnight, so it's grouped by "night" not by "day", e.g. thursday at 3am is categorized as the same evening as wednesday evening),

I recommend working in UTC

 

[Mendel]

mistakes or abuses

Again, you are the only person in this thread who uses these words (though "error" has come up).

We are talking about the paper (mostly), not the author. The problem is that, having analysed the paper, the author has made non-obvious choices and left them unexplained, which leads to speculation.
You don't simply grab the wrong data set "by mistake", this was a choice.
They only ran the shadow/altitude analysis for two altitudes, this was a choice. (I can't stress enough how easy this is to do for more altitudes as soon as you've automated it once.)
These choices are questionable, that's why I am questioning them.

Flat Earthers also bravely "put their necks out there". Would you say the same about them?

 

[HoaxEye]

This has been stated a few times now, can someone quote the source for the bold part?

Hambly & Blair (2024) mentions Kodak emulsions.

See: https://academic.oup.com/rasti/article/3/1/73/7601398?login=false

Furthermore, later work employed data from measures of second-generation POSS original plates employing fine-grained Kodak IIIa emulsions; whereas, the data analysed here are from copy plates employing older, coarser 103a emulsions (as were state-of-the-art at the time of POSSI). On the subject of modern (i.e. late 20th century) hyper-sensitized, fine grained Kodak emulsions, it is interesting to note that there is a large literature on the appearance of spurious microdots (see e.g. Good 1988, and references therein) but no such artefacts have been reported as occurring on the older coarse-grained varieties.

 

[overlord]

Hambly & Blair (2024) mentions Kodak emulsions.

See: https://academic.oup.com/rasti/article/3/1/73/7601398?login=false

Furthermore, later work employed data from measures of second-generation POSS original plates employing fine-grained Kodak IIIa emulsions; whereas, the data analysed here are from copy plates employing older, coarser 103a emulsions (as were state-of-the-art at the time of POSSI). On the subject of modern (i.e. late 20th century) hyper-sensitized, fine grained Kodak emulsions, it is interesting to note that there is a large literature on the appearance of spurious microdots (see e.g. Good 1988, and references therein) but no such artefacts have been reported as occurring on the older coarse-grained varieties.

"second-generation POSS" or POSS II started in 1985, therefore this is not relevant for the statement:

Then after about April of 1956 [...] Nevermind that's also when the red emulsion was changed.

 

[Shen]

But this (objects in orbit for a long time) means the transients at POSS-1 should not be correlated to nuclear tests. If they were always there, they should have been always spotted (*). So, they cannot both say 'they have been in continuos GEO for many years' and 'the reflections of the Sun they gave off are correlated with nuclear tests': one excludes the other (barring some ad-hoc excuse to explain why satellites always orbiting there only catch the Sun around the dates of nuclear tests). Holding both opinions as true as the same time is illogical thinking.

The irony is that, looking for a correlation with nuclear tests, they undermined their own hypthesis: they use the correlation as evidence in favour of alien satellites, but the probability of aliens actually goes down... because the alien's behaviour becomes weird, that is to say, unprobable. They had better stop at the Earth shadow analysis, but you know, agendas...


(*) and should have been seen in the all the full-sky surveys which followed, and in those which are ongoing now.

I think what they really imply but don't want to say is that it isn't alien satellites, but alien spacecraft dropping in to a suitable orbit to observe nuclear tests. Where they hang out the rest of the time is not specified.

 

[HoaxEye]

"second-generation POSS" or POSS II started in 1985, therefore this is not relevant for the statement:

Good catch.

I did some digging. Journal of the Royal Astronomical Society of Canada, Vol. 111, Issue 6, starting from p.252 contains A Short History of Astrophotography: Part 2 (by Klaus Brasch): https://adsabs.harvard.edu/full/2017JRASC.111..252B - sorry, could not find this text in text format.

In the 1950s and 1960s, Kodak produced several specialized spectroscopic emulsions for astronomical research, including the 103 series. Although designed for selected spectral sensitivity and reduced reciprocity failure, these were available in 35-mm format, but were also very coarse-grained. Because of this and although these films were also commercially available, pretty much all aspects of amateur astrophotography were a compromise. For instance, popular films like Kodak Tri-X Pan, though fast enough for short exposures to minimize the effects of atmospheric turbulence or seeing, were also very grainy, while more fine-grained films like Kodak Microfile or Plus X were rather slow and required relatively long exposures for the necessarily highly magnified planetary images. Much of that changed over time as better films were developed; culminating with the introduction of Kodak's ultra-fine-grained 2415 Technical Pan Film in 1981.

It's likely the source for the year 1956 is this Medium article that has been linked earlier in at least one related Metabunk thread:
Source: https://medium.com/@izabelamelamed/not-seeing-the-star-cloud-for-the-stars-a010af28b7d6
- "Not Seeing the Star Cloud for the Stars" by Izabela Melamed (July 2025).

By 1956, 103a-E was replaced with improved emulsions, and glass copies remained in use until digital photography took over in 2000. Curiously, the vanishing stars finally vanished around the time 103a-E was phased out.

I don't remember when we (Metabunk users) started to use year 1956 for Kodak emulsion change without fact-checking, but perhaps we did? POSS-I survey completed in 1958.

Despite the apparent error in the Medium article, it seems to be true that the "vanishing stars finally vanished" when Kodak 103-series emulsions were phased out. I've looked at POSS-II red plates and could not find any "glints" or "transients". My manual check was limited to handful of plates, though, but all POSS-I red plates are likely "peppered with such isolated detections" (see Hambly & Blair 2024, or the linked Medium post).

 

[Gary C]

Sorry if I missed this earlier but the alleged link to nuclear tests is still bugging me. The spherical angular distance between Semipalatinsk, USSR and Mt. Palomar, USA is approximately 101.4 degrees (per Google Gemini which appears to have used the correct law of cosines method in the calculation). That means from the POV of the instrument used in the survey, an object at GEO over Kazakhstan would be at or below the visual horizon as viewed from California. A quick eyeball check of the Western Pacific test sites suggests a similar improbable geometry.

We've already shown that the putative correlation with US tests is explained by weather, seasonality, and tests not taking place on weekends or national holidays. (See @jdog, et al - https://www.metabunk.org/threads/transients-in-the-palomar-observatory-sky-survey.14362/post-355561)

If we subtract out all these unobservable UK and Soviet test dates, what does that do to the author's claim?

(Edited to clarify locations)

 

[jarlrmai]

Well aliens can see the particles they need through the mass of the Earth, they just need to be close enough for the NukeTest (tm) Sensor 9000 to get enough signal.

 

[jdog]

Good catch.

I did some digging. Journal of the Royal Astronomical Society of Canada, Vol. 111, Issue 6, starting from p.252 contains A Short History of Astrophotography: Part 2 (by Klaus Brasch): https://adsabs.harvard.edu/full/2017JRASC.111..252B - sorry, could not find this text in text format.

In the 1950s and 1960s, Kodak produced several specialized spectroscopic emulsions for astronomical research, including the 103 series. Although designed for selected spectral sensitivity and reduced reciprocity failure, these were available in 35-mm format, but were also very coarse-grained. Because of this and although these films were also commercially available, pretty much all aspects of amateur astrophotography were a compromise. For instance, popular films like Kodak Tri-X Pan, though fast enough for short exposures to minimize the effects of atmospheric turbulence or seeing, were also very grainy, while more fine-grained films like Kodak Microfile or Plus X were rather slow and required relatively long exposures for the necessarily highly magnified planetary images. Much of that changed over time as better films were developed; culminating with the introduction of Kodak's ultra-fine-grained 2415 Technical Pan Film in 1981.

It's likely the source for the year 1956 is this Medium article that has been linked earlier in at least one related Metabunk thread:
Source: https://medium.com/@izabelamelamed/not-seeing-the-star-cloud-for-the-stars-a010af28b7d6
- "Not Seeing the Star Cloud for the Stars" by Izabela Melamed (July 2025).

By 1956, 103a-E was replaced with improved emulsions, and glass copies remained in use until digital photography took over in 2000. Curiously, the vanishing stars finally vanished around the time 103a-E was phased out.

I don't remember when we (Metabunk users) started to use year 1956 for Kodak emulsion change without fact-checking, but perhaps we did? POSS-I survey completed in 1958.

Despite the apparent error in the Medium article, it seems to be true that the "vanishing stars finally vanished" when Kodak 103-series emulsions were phased out. I've looked at POSS-II red plates and could not find any "glints" or "transients". My manual check was limited to handful of plates, though, but all POSS-I red plates are likely "peppered with such isolated detections" (see Hambly & Blair 2024, or the linked Medium post).

I saw a reference to "reciprocity failure," which jumped out to me as a photographer.

We have to remember that these plates were very nice emulsion on very nice glass, but you're exposing chemistry to light and the process differs significantly from digital photography.

Changes in aperture, shutter speed, and ISO (formerly ASA) are normally reciprocal; if you double your shutter speed to take action shots you have to halve your aperture to get the same exposure. ("Halve" as in reducing the aperture number, making it bigger, like going from f/2.8 to f/1.4.)

It's not much of an issue with modern digital cameras, but with film the chemical reactions aren't always linear. Your exposure times can be off and your colors can shift, so you get "reciprocity failure" that has to be accounted for.

So the crystals in photographic emulsion can have an uneven response to light levels that vary significantly from the light levels the emulsion was designed to capture.

 

[Kyle Ferriter]

I recommend working in UTC

I have the data stored in UTC, but for this analysis, the local timezone of the telescope (US pacific standard/daylight) determines what the boundaries of one contiguous night are.

 

[Mendel]

I have the data stored in UTC, but for this analysis, the local timezone of the telescope (US pacific standard/daylight) determines what the boundaries of one contiguous night are.

well, are there nights that start later than 4pm PST, which is when the UTC date change is? All Palomar nights are on the same UTC date.

 

[Kyle Ferriter]

well, are there nights that start later than 4pm PST, which is when the UTC date change is? All Palomar nights are on the same UTC date.

Hmmm good point. It would make the code too specific to PST/PDT though, I think. If there was another telescope dataset in a different timezone this logic may not work and localizing the times makes it easier. Like over the course of a few months, astronomical twilight in a location may move from is at 23:50PM UTC to 00:10 UTC the next day.

 

[NorCal Dave]

I don't remember when we (Metabunk users) started to use year 1956 for Kodak emulsion change without fact-checking, but perhaps we did? POSS-I survey completed in 1958.

I know when I originally used it, I did qualify it (bold by me):

But after March of 1956, and the possible changing of the problematic red emulsion, an additional 38 nuclear test resulted in 0 transients observed.

I found this 1956 paper from the American Astronomical Society:

https://adsabs.harvard.edu/full/1956AJ.....61..399G

They do seem to note that the 103a, red emulsion was problematic, particularly closer to the edges of the plates (screen shots):

Two plates of each kind were tested. The
means are plotted in Figure I which shows first
a s Click on the page for more article options
layer neaт
ПОеm em ,etaip enit in sgna am таcn re
tends to run out, the edge-effect. Thedistorted area
is about 2 to 3 cm deep from each edge, in agree-
ment with the results of other authors. Secondly
it shows marked differences in the behavior of
different types of plates. The displacements for
103a-0 and IHa-O plates are more than 3 times
as large as for the others.

They were not doing actual astro-photography with these plates in the this paper as far as I can tell. Rather, they were using them for stellar radical velocities studies, but were having problems with the red emulsion:

1. Introduction. During the determination of
stellar radial velocities, measurements of the Fe
comparison lines showed greater differences than
could be expected. As the Cassegrain spectro-
graph was always kept at the same temperature,
the dispersion of the spectra was the same and
the readings should differ only by the accidental
errors of plate measuring. Nevertheless, differ-
ences of 10 microns and more between the first
and last line over a range of 21 mm from A3922
to X4528 were quite frequent. After excluding
other possible sources of errors we concluded that
movements of the photographic layer must ac-
count for these differences, especially as we found
a relation to the type of emulsion.
The question of displacements of photographic
emulsions has been raised before,

They corrected most of the problems with a very specific development process:

As the results shown in Figure 4B were the
best of our various experiments, we recommend
the following method of processing: Kodak Trop-
ical Developer, 11 min; intermediate washing in
saturated sodium sulphate solution, ½ min;
Kodak Acid Fixing Powder with Hardener*, 10
min; sodium thiosulphate (4 oz. in 10 oz. of
H.O), 10 min; all baths at 20°C (68°F). Final
washing in running water, 60 min; natural dry-
ing at room temperature. The plates are kept
in a horizontal position throughout. The second
fixing bath is necessary only in cases like ours,
when plates have to be stored indefinitely as
records. Kodak Anti-Fog No. I (Benzotriazole)
may be added to the developer if required, and
Kodak Photo-Flo used ad libitum before drying.

These figures again show distinctly the
improvement by the new method of processing
particularly with the IIa-O and 103a-O emulsions.

I'm still looking for where the Medium article got the information about a change in the emulsion, but maybe it was a change in the development process?

 

[John J.]

It's likely the source for the year 1956 is this Medium article that has been linked earlier in at least one related Metabunk thread:

I hadn't really cottoned on to the paper referenced by @HoaxEye and discussed elsewhere, for others who might have missed it, it's worth a look.

"On the nature of apparent transient sources on the National Geographic Society–Palomar Observatory Sky Survey glass copy plates",
N.C. Hambly, A. Blair, RAS Techniques and Instruments 3 (1), 2024 https://academic.oup.com/rasti/article/3/1/73/7601398?login=false
Apologies if I'm revisiting material already posted. Emphases are mine.

In this paper, we present an independent analysis of the source photographic plate material and SSS scans used in passing by Villarroel et al. (2021). We argue against the premise that these are real above-atmosphere transients and suggest an alternative, mundane explanation as to their origin.

Earlier in the Introduction, the authors write

We examine critically recent claims for the presence of above-atmosphere optical transients in publicly available digitized scans of Schmidt telescope photographic plate material derived from the National Geographic Society–Palomar Observatory Sky Survey. We employ the publicly available SuperCOSMOS Sky Survey catalogues to examine statistically the morphology of the sources. We develop a simple, objective, and automated image classification scheme based on a random forest decision tree classifier. We find that the putative transients are likely to be spurious artefacts of the photographic emulsion. We suggest a possible cause of the appearance of these images as resulting from the copying procedure employed to disseminate glass copy survey atlas sets in the era before large-scale digitization programmes.

Recently a number of articles have appeared in the astronomical literature noting the presence of a significant population of apparently transient sources on POSSI 'E' (red) plate scans (Villarroel et al. 2021, and citations thereof). The data analysed came from both the Digitized Sky Survey (DSS, Lasker 1992) and the SuperCOSMOS Sky Survey (SSS, Hambly et al. 2001a, b, c). While the original discovery paper discusses various plausible non-astronomical origins for the nine transients identified, subsequent work (e.g. Solano, Villarroel & Rodrigo 2022; Solano et al. 2024) is firmly based on the premise that these apparent transients represent real, above-atmosphere astronomical phenomena... . ...Villarroel et al. (2022) propose a population of highly reflective, glinting objects in near-Earth orbit as the source of the transients. No such population has been noted in more recent, deeper surveys such as SDSS and PanSTARRS...

One of the authors, Hambly, worked on the SuperCosmos Sky Survey, producing data used by Villaroel. Their understanding and interpretation of the data is presumably relevant.

From the application of our trained classifier for field E0070 we note that all nine of the transients identified in Villarroel et al. (2021) are classed as bad with probabilities amongst the various classes as shown in Table 4.

Neither Villarroel et al. (2021) nor any of the follow-up works discuss the provenance of the photographic plate material scanned in creating the images on which their analyses are based yet the reproduction of glass copy plates from the survey originals is clearly an important consideration. Emulsion flaws will be present on original negatives, contact positives derived from them as the first step in creating copies, and any paper, film, or glass copy negatives derived from those positives. While the presence of the transients in independent scans (DSS and SSS) does indeed eliminate the respective digital scanning procedures, hardware and software as introducing the images under consideration here, it does not follow that those images must be present on the original negatives and are astronomical in origin.

The authors- and more significantly, it is strongly implied Villaroel- did not study the original POSS-1 plates;

Clearly it would be most instructive to analyse (or at the very least, visually inspect) the original POSSI negative plates and the intermediate copies.

Under "Conclusions",

We find that (i) the image profiles of the transients are significantly sharper than typical stellar images on the plates; (ii) that an ML decision-tree classifier badges the images as spurious with high probability; (iii) that similar examples of apparent transients are present on the copy plate of the adjacent field; and finally (iv) that there are many hundreds of similar images on both plates in the overlap region between the two fields. We suggest one likely mechanism for the origin of at least some of these apparent transients as being emulsion holes on the intermediate positive plates used during reproduction of the copy sets. We therefore caution that digitized all-sky survey catalogues derived from the POSSI glass copies are likely peppered with these isolated false detections and that great care must be exercised when interpreting the publicly available digitized images or when making samples of unpaired catalogue records derived from them.

The last line might be taken as Hambly and Blair suggesting that Villaroel et al. have been insufficiently careful in their interpretation of the (reproductions of the) images available to them.

I won't pretend to understand all (or much!) of what the authors write, but Hambly is clearly a relevant figure in photographic astronomy- Villaroel gets some of her data from his work- and the points made seem to be pertinent and follow logically from one another.
The language used is measured throughout (no ridicule, no stigmatising, no allegations of wrongdoing; perhaps a couple of lines that might be interpreted as implied criticism).

I hope I won't come across as overly cynical in predicting Ross Coulthart won't be interviewing Hambly and Blair on NewsNation.

In passing, I don't know if it's an example of gentle (but dry) humor, or simply due to the measured tone of the paper with the authors not introducing dramatic speculation about observations/ findings, but Hambly and Blair's main concern re. the transients, if they are physical objects, is not that they might be thousands of alien spacecraft in Earth orbit but that they might interfere with future sky surveys

No such population has been noted in more recent, deeper surveys such as SDSS and PanSTARRS, but if it truly exists there would be major implications for current and future deep, high time-cadence surveys... Such surveys are already facing significant contamination problems from constellations of artificial satellites currently being launched into low Earth orbit...

There's nothing like sticking to day-to-day practicalities during interesting times!

 

[joao_lucas]

View attachment 82851

Some Transients in the Palomar Observatory Sky Survey (POSS-I) May Be Associated with Above-Ground Nuclear Testing and Reports of Unidentified Anomalous Phenomena

https://www.researchsquare.com/article/rs-6347224/v1

Transient star-like objects of unknown origin have been identified in the first Palomar Observatory Sky Survey (POSS-I) conducted prior to the first artificial satellite. We tested speculative hypotheses that some transients are related to nuclear weapons testing or unidentified anomalous phenomena (UAP) reports. A dataset comprising daily data (November 19, 1949 -April 28,1957) regarding identified transients, nuclear testing, and UAP reports was created (n=2,718 days). Results revealed significant (p = .008) associations between nuclear testing and observed transients, with transients 45% more likely on dates within +/- 1 day of nuclear testing. Significant (p<.001) associations were also noted between total number of transients and total independent UAP reports per date, with the largest association observed for dates on which at least one transient was identified (Spearman's rho = 0.14, p = 0.015). For every additional UAP reported on a given date, there was an 8.5% increase in number of transients identified. Small but significant (p = .008) associations between nuclear testing and number of UAP reports were also noted. Findings suggest associations beyond chance between occurrence of transients and both nuclear testing and UAP reports. These findings may help elucidate the nature of POSS-I transients and strengthen empirical support for the UAP phenomenon.

On the Image Profiles of Transients in the Palomar Sky Survey

https://arxiv.org/abs/2507.15896

The VASCO project has discovered groups of short-lived transients on historical photographic plates that lack conventional explanation. Hambly & Blair (2024) examined nine such transients reported by Villarroel (2021) and found that they exhibit narrower, rounder profiles, attributing this to emulsion flaws. However, well-established optical principles and atmospheric physics imply that unresolved flashes lasting less than a second naturally appear sharper and more circular than stellar images, particularly on long-exposure plates where stars are significantly blurred by seeing and tracking errors. Such profiles are an expected consequence of sub-second optical flashes, making their findings consistent with the transient interpretation.

This is quite popular in UFO circles, as supposed evidence of pre-space era satellites or spacecraft that correlate with upticks in UFO sightings.

I'm skeptical.


See also this thread on more detailed analysis.
https://www.metabunk.org/threads/digitized-sky-survey-poss-1.14385/

I've been reading these papers, and one thing bothers me about the aligned multiple transients. In the paper, they identify a small number of such events, which they consider very improbable (around 4 sigma, I think).

However, that improbability applies only if we consider the local probability—that is, the chance of finding that alignment on a single plate. If I'm not mistaken, there are 635 unique plates in the analysis. The total probability of finding at least one highly unlikely alignment across all plates is not actually that unlikely. I think this part of the paper might suffer from the "look-elsewhere effect." Am I understanding this correctly?

I've attached an R script where you can run simulations to see that these alignments aren't as improbable as they seem. Any feedback is welcome! The one part I'm not entirely sure about is my choice of 10 boxes per plate—it could be fewer or more. What would be the most reasonable number? I do not know.

Best regards,

João

 

[Michael]

Very curious about this passage in the paper:

This analysis was restricted to dates on which at least one transient occurred (n = 310), an analysis that eliminates the substantial bias due to the large number of zero values in the transient data (there were no transients observed on 88.5% of the days in the dataset). This simple analysis revealed a very small but statistically significant association (i.e., beyond chance) between the total number of transients and total UAP reports on a given date (Spearman's rho = 0.138, p = 0.015). A scatterplot of this association is presented in Fig. 2.

What is the "bias" that being removed by excluding zero values? Why shouldn't they be considered in the analysis - particularly given they are 88.5% of the relevant dataset? (Figure 2 isn't exactly a compelling trend, but that's secondary compared to excluding so much data with an obvious explanation.)

 

[Mendel]

In the paper, they identify a small number of such events, which they consider very improbable (around 4 sigma, I think).

https://iopscience.iop.org/article/10.1088/1538-3873/ae0afe#paspae0afes6

We use Table 3 to adopt other values of pmax, setting it equal to FWHM of the smallest star in an alignment (e.g., for Candidate 1, FWHM = 2.″7). Doing this, we see that all 3-point alignments are non-interesting events with p > 0.05 (less significant than 2σ), with an exception of the borderline case of Candidate 2. This shows that for POSS-I data where the seeing in general is rather large, 3-point alignments of simultaneous transients do not provide significant proof against randomness.

 

[Mendel]

Very curious about this passage in the paper:

This analysis was restricted to dates on which at least one transient occurred (n = 310), an analysis that eliminates the substantial bias due to the large number of zero values in the transient data (there were no transients observed on 88.5% of the days in the dataset). This simple analysis revealed a very small but statistically significant association (i.e., beyond chance) between the total number of transients and total UAP reports on a given date (Spearman's rho = 0.138, p = 0.015). A scatterplot of this association is presented in Fig. 2.

What is the "bias" that being removed by excluding zero values? Why shouldn't they be considered in the analysis - particularly given they are 88.5% of the relevant dataset? (Figure 2 isn't exactly a compelling trend, but that's secondary compared to excluding so much data with an obvious explanation.)

the big biases that the paper does not check:
(1) number of plates taken on that date
(2) number of astronomical objects photographed on that date

(1) scales with the number of plate defects, (2) may scale with the number of astronomical objects misidentified as transients (Solano 2022 found over 90,000 in the data set used in this study).

See https://www.metabunk.org/threads/transients-in-the-palomar-observatory-sky-survey.14362/post-355589 for the references.

 

[MonkeeSage]

The authors- and more significantly, it is strongly implied Villaroel- did not study the original POSS-1 plates;

Clearly it would be most instructive to analyse (or at the very least, visually inspect) the original POSSI negative plates and the intermediate copies.

In a recent Scientific American article, Hambly said Villrroel and team should physically inspect the original plates for defects, which also implies that has not yet been done.

The most prosaic explanation is that Villarroel's transients are simply artifacts in the photographic plates such as speckles of dust, blobs in the emulsion or even radioactive particles. Nigel Hambly, an astronomer at the University of Edinburgh, who has previously analyzed Villarroel's work and who has extensive experience with photographic plates from Palomar and other observatories, says one way to check would be to study the original plates themselves rather than using digital copies. "I've been caught out many times by apparently real things turning up in my data," he says, especially when working with plates that weren't stored in pristine conditions. "When you actually physically examine the plates under a microscope, you begin to get a feeling for what's real and what's spurious," he says. "There's no shame in being wrong."

Source: https://www.scientificamerican.com/...-photograph-ufos-orbiting-earth-in-the-1950s/

 

[Michael]

the big biases that the paper does not check:
(1) number of plates taken on that date
(2) number of astronomical objects photographed on that date

(1) scales with the number of plate defects, (2) may scale with the number of astronomical objects misidentified as transients (Solano 2022 found over 90,000 in the data set used in this study).

See https://www.metabunk.org/threads/transients-in-the-palomar-observatory-sky-survey.14362/post-355589 for the references.

The number of plates is a huge issue, given Palomar and Nevada aren't that far away from each other one can easily imagine correlations arising due to good weather (seasons!) being correlated at both locations.

 

[Eburacum]

If the distribution of anomalies is 'lumpy' anyway, then that may explain the relative absences of anomalies in the Earth's shadow. The Earth's shadow is only 1-2% of the entire sky; if the distribution of anomalies is sufficiently lumpy, there may be many segments of the sky with unusually low numbers of anomalies, and it is just due to chance that the Earth's shadow often falls into these voids.

The lumpiness of the distribution of anomalies may be traceable to differences in manufacture and processing, so it might be possible to explain this uneven distribution by looking at the dates these plates were made or developed.

 

[MyMatesBrainwashed]

Forgive me if I'm miles out, but rather than critiquing the paper (cos I doubt I can) I've been wondering how this could possibly move forward.

I was thinking around how it can't be that difficult to stop an orbiting satellite from reflecting sunlight back to telescopes on Earth. Just painting it matt black would surely go a fair way to achieving that.

Which leads to the thought that it must be pretty much impossible to stop an orbiting satellite from occulting stars.

Google's AI tells me that satellites do occult stars and it's a nuisance for astronomers. But that they also use these occultation events to classify satellite shapes and determine their size and orientation.

Maybe this is impossible with these plates and long exposure times. I dunno. Maybe there's other data from the period to look at that could be used instead. But if possible this could potentially help determine if there really was something there or not.

 

[Eburacum]

Another possibility is that the 'Earth's shadow anomaly' is entirely consistent with the distribution of stars in the Milky Way. To get a correlation with the location of the Earth's shadow, Villaroel et al had to consider the previously-discounted dataset of anomalies which were associated with known stars within 5" of angular separation. Even if all of these anomalies were in fact associated with real stars, there would still be an uneven distribution of anomalies in the data -because the region along the Milky Way contains many more dim stars than the regions towards the Galactic poles.

The Earth's shadow moves along a path which is inclined at (approximately) 60° to the angle of the Milky Way. This means it will encounter significantly fewer real stars in its path than if it were aligned with the Galactic Equator. If these plates were often or always exposed when the Earth's shadow fell in a part of the sky separated from the Milky Way, then the negative correlation with the anomaly database could be explained.

 

[FatPhil]

https://iopscience.iop.org/article/10.1088/1538-3873/ae0afe#paspae0afes6

We use Table 3 to adopt other values of pmax, setting it equal to FWHM of the smallest star in an alignment (e.g., for Candidate 1, FWHM = 2.″7). Doing this, we see that all 3-point alignments are non-interesting events with p > 0.05 (less significant than 2σ), with an exception of the borderline case of Candidate 2. This shows that for POSS-I data where the seeing in general is rather large, 3-point alignments of simultaneous transients do not provide significant proof against randomness.

That sounds like they found the green jelly bean that gives you cancer. This is a fishing expedition. Did they adjust p accordingly?

 

[Mendel]

If these plates were often or always exposed when the Earth's shadow fell in a part of the sky separated from the Milky Way, then the negative correlation with the anomaly database could be explained.

First, I entirely agree. The "transients" could be unidentified stars, and be distributed like stars in general, and we should check if they "avoid" the shadow, too.

Secondly, it may be more complicated like that. The transients included in the set were identified by Solano 2022 after 3 big star catalogues had already been matched. Hypothetically, if these star catalogs cover the equatorial regions of the sky better than the polar regions (because there's more atmosphere down low than when looking up), then newer, space-based catalogs may have covered the polar regions better than the older data, but be about equal in the more equatorial region.
This would bias the filtering done by applying these star catalogues towards data points in places that the Earth shadow never covers.

But we can only find this data set if we replicate Solano 2022 exactly.