Community Notes on X/Twitter

last updated 22 March 2024

1. Scientific Basis
2. Peer-reviewed Research
3. Fact-checks
4. Aliens/Angels
Note author Celeste Labedz

Since the major earthquakes in Turkey on 6 February 2023, a seismic potential that was emphasized by our CEO Frank Hoogerbeets in a tweet three days earlier, our accounts on X/Twitter have been a continuous target of community notes claiming that "there is no scientific basis for these earthquake forecasts." These notes are rated as citing "high quality sources." On X/Twitter we have repeatedly stated that these notes are actually of low quality citing sources exclusively from the United States, a country that is known for being very conservative in the field of seismology. In Asian countries there has been much more progress, especially in the field of ionosphere-lithosphere coupling (atmosphere-crust connection), a relationship largely rejected by United States seismologists. Thus, the community notes should be seen in light of the conservative attitude primarily coming from the United States.

Article continues below the image.

First of all, let's address the purpose of X/Twitter's community notes. They are supposed to target potentially misleading posts. In the case of the SSGEOS, none of our posts are misleading. Our research, models and statistics are documented and open for research to anyone. It is telling that none of the cited sources in the community note actually disprove or even address the models and statistics that we use. So let's break down the four sections of this repeatedly copy-pasted community note.

In this section the first source links to a text on the USGS website. The first line of that text states that: "No. Neither the USGS nor any other scientists have ever predicted a major earthquake." This is incorrect. In October 1989 USGS geologist Jim Berkland predicted an earthquake in California based on specific signs in the region. Four days later the Loma Prieta earthquake occurred. After his correct prediction Berkland was suspended because he was not allowed to make earthquake predictions in his capacity as a USGS scientist.

In the text the USGS goes on to explain that in order for a prediction to be valid, three key elements must be included, which are time, location and magnitude. It is important to realize that it is the USGS that sets these requirements. Not even weather forecasts are required to have this level of precision. As with the weather forecast, we believe that it is acceptable to forecast a greater potential, say 60% or 70%, for a major earthquake in the next 2-3 days, if the forecast is based on established models, as we have developed at the SSGEOS in recent years.

So the first source in the community note should be regarded as opinion, as it neither provides fact nor evidence. The second source links to a text from Caltech explaining more about earthquake probabilities and forecasts from their point of view.

It is important to understand that neither of the first two sources addresses models that include positions of the Sun, the planets and the Moon at the time of larger earthquakes. This is crucial, because in recent decades no comprehensive study has been done in this field and there is almost no literature about this subject. In fact, the whole issue of planetary geometry is suspiciously absent in the scientific community. A quick look at our about page that explains why and how the research started, will tell anyone with common sense the significance of such modeling and how erroneous scientists are in automatically rejecting it, simply because they regard the research approach itself as unscientific.

This section of the community note supposedly provides evidence against our methods. One source, a paper by Susan E. Hough addresses the question if large earthquakes occur on "preferred days of the calendar year or lunar cycle." First of all, lunar cycles have nothing to do with any human invented calendar. We have already demonstrated that not every Full/New Moon is the same and that a thorough analysis that includes the positions of all planets is essential. Limited, superficial research has little meaning. Apart from the fact that lunar cycles are unrelated to preferred days of the calendar year, we emphasize that analysis of specific days on calendars is not part of our research or forecast models. This issue is therefore irrelevant and there are better papers addressing effects of lunar cycles, including on seismic activity.

The other source, a paper by Pierre Romanet addresses the question if planet/sun conjunctions can be used to predict large (moment magnitude ≥ 7) earthquakes. Again we emphasize that limited, superficial research has little meaning. This also applies to this paper. Here we list five key points to consider regarding this paper:

1. The author is a seismologist and not an expert in planetary geometry. Unsurprisingly, he does not distinguish between critical and non-critical geometry; he incorrectly assumes that any conjunction should be a major earthquake trigger. This assumption is false. Given the fact that this paper has been peer-reviewed, those who reviewed it are obviously no experts in this field either.
2. Planetary geometry is matched against magnitude 7+ earthquakes only. Magnitude 6 earthquakes (clusters in particular) are ignored.
3. A huge conjunction tolerance of +/- 3°(!) is used. Apparently, the author did not make any effort to refine this tolerance (our SSGI models use a tolerance of 0.17°). To illustrate how bad this rough approximation of 3° is, we apply it to a Mars-Sun-Saturn conjunction (which is by no means critical, but it is included in the paper results), which would add a whopping 18 conjunction days to the total that serves as proof in the paper. If this conjunction were critical (which it is not) we would mark at most about 4 days following the conjunction as critical. But it gets much worse when outer planets reach a conjunction with the Sun and Mercury moves within 3° of them, such as in early 2024, when Mercury stays within 3° of Jupiter and Uranus from 27 January to 22 April. Using a 3° tolerance the peer-reviewed method adds 86(!) days to the conjunction day-count, whereas in reality Mercury is in actual conjunction with Jupiter and Uranus three times (9 Feb, 13 Mar, 9 Apr), accounting for no more than 14 conjunction days by SSGEOS standards.
4. Convergence of specific planetary conjunctions, as well as clustering of stronger earthquakes, both of which are very significant, are not taken into account. It makes the paper result all the more useless because "conjunction days" are added for each conjunction, even if they overlap in time.
5. Our SSGI models are not addressed, even though they consistently show a relationship between (clustering of) stronger seismic activity and critical planetary geometry.

It is important to realize that Romanet did not inquire as to our forecast models, in particular the geometry between specific planets, the Sun and the Moon that we emphasize. He simply made several assumptions on which he then based his paper, which is the result of just three weeks of research in March 2023. By contrast, the models that we use are based on nine years of research and software development. If you think about it, it is amazing how quickly and easily Romanet's paper was accepted and peer-reviewed.

Here we touch on a major issue in science regarding the influence of the planets, a debate that has been going on for decades if not centuries. Claims that planets have no influence are usually backed by assumptions and reasoning based on the theory of classical mechanics, rather than actual observation and research. Superficial, limited research as done by Romanet is already rare. In-depth, long-time research as we do at the SSGEOS, is virtually non-existent.

This section provides the so-called "fact-checks" of our "pseudoscience". The cited sources are carefully selected mainstream media articles, which are not fact-checks at all. At best they present expert opinion. No evidence is presented against our forecast methods (which are not pseudoscience, as they are based on careful observation, measurement and research). We could easily select mainstream media articles that actually confirm our forecasts. It is generally called cherry-picking. Mainstream media do not have authority, nor are they in any position to "fact-check" science.

To illustrate how manipulative mainstream media can be, we take the 18 May 2015 article from Slate, that states: "No, a Planetary Alignment on May 28 Won’t Cause an Earthquake." It addresses the second ever forecast by our CEO Frank Hoogerbeets, in which he expressed his concern about the final days of May 2015 due to a convergence on 27 May of critical planetary conjunctions, which he recognized from his initial 2014-2015 study. Media outlets were quick to dismiss the forecast as pseudoscience. But when a magnitude 6.8 earthquake occurred in Alaska on 29 May, followed by a magnitude 7.8 earthquake (initially estimated 8.5) on 30 May south of Japan, the media were silent and did not refer back to the forecast. Hoogerbeets' first ever forecast covered the magnitude 7.8 earthquake in Nepal on 25 April 2015. Anyone can read about the why and how of this research!

This is the final section of the community note. It claims that the user's (SSGEOS) last "prediction" site claimed to get "info from aliens/angels." We can easily debunk this claim. The SSGEOS was founded in 2022, while the note refers to an article from 2015 that does not address the SSGEOS. So this claim is verifiably false and it makes this community note all the more ridiculous.

On 22 October 2023 Celeste Labedz (Postdoc UCalgary PhD Caltech) revealed herself on Bluesky as X/Twitter community note author "Jazzy Sandalwood Penguin". In her posts she explained her motivation as "my hobby of annoying awful people" and "it is super fun". We released this information in a thread on X/Twitter where we explain how Dr. Labedz has been ignoring our statistics and data since March 2023. Since our publication she has locked her Twitter account and removed the posts from her Bluesky time line.

It is important to know that all sources selected by Dr. Labedz are from the United States and do not provide a balanced view of an international scientific community. Much more progress is made by scientists from other countries. Scientists from all over the world, including, seismologists, geologists and meteorologists, follow our research with growing interest. Some sources in the community note are irrelevant or unrelated, thus misleading and deliberately added in an attempt to discredit the SSGEOS. Other sources express opinion. Neither the peer-reviewed papers nor the mainstream media "fact-checks" disprove our forecast models. In fact, these models, in particular SSGI, which are based on years of observation and research, are not addressed at all! They are simply ignored and disregarded as "pseudoscience". A comprehensive study that conclusively disproves our models and statistics does not exist, because no similar research has been done in recent decades.

Obviously, Dr. Labedz' claim that "there is no scientific basis for these earthquake forecasts" is false. Unfortunately, X/Twitter does not actively check or control their community notes. Anyone with a specific (especially scientific) opinion that is supported by a sufficient number of people can target undesired posts with community notes, even if these notes contain irrelevant, unrelated, biased or incorrect information, as long as the content does not violate X/Twitter's terms of service. This policy opens the door to harassment and misuse of community notes as we have seen since the Turkey earthquakes.