The Earth Just Started Spinning Faster than Ever Before and Scientists Don’t Know Why

Which led me to recall, that the Moon and the Earth move around a barycenter, so would Kepler's law apply to this center, more than to the geographic center of the Earth? Leaving the question aside, here is what they write about the barycenter:
In astronomy, the barycenter (or barycentre; from Ancient Greek βαρύς (barús) 'heavy', and κέντρον (kéntron) 'center')[1] is the center of mass of two or more bodies that orbit one another and is the point about which the bodies orbit. A barycenter is a dynamical point, not a physical object. It is an important concept in fields such as astronomy and astrophysics. The distance from a body's center of mass to the barycenter can be calculated as a two-body problem.

If one of the two orbiting bodies is much more massive than the other and the bodies are relatively close to one another, the barycenter will typically be located within the more massive object. In this case, rather than the two bodies appearing to orbit a point between them, the less massive body will appear to orbit about the more massive body, while the more massive, body might be observed to wobble slightly. This is the case for the Earth–Moon system, whose barycenter is located on average 4,671 km (2,902 mi) from Earth's center, which is 75% of Earth's radius of 6,378 km (3,963 mi).
For an animation of how the Earth-Moon movement can be conceived with the barycenter of the system as the important location, there is this Tweet, otherwise the Wiki has an illustration:

To add to the complexity, the rotation of the Earth is not around the center of the Sun, but around the barycenter of the solar system.

This NASA site writes:
Our entire solar system also has a barycenter. The sun, Earth, and all of the planets in the solar system orbit around this barycenter. It is the center of mass of every object in the solar system combined.

Our solar system’s barycenter constantly changes position. Its position depends on where the planets are in their orbits. The solar system's barycenter can range from being near the center of the sun to being outside the surface of the sun. As the sun orbits this moving barycenter, it wobbles around.
Astronomers have decided to define a Barycentric Dynamical Time:
Barycentric Dynamical Time (TDB, from the French Temps Dynamique Barycentrique) is a relativistic coordinate time scale, intended for astronomical use as a time standard to take account of time dilation[1] when calculating orbits and astronomical ephemerides of planets, asteroids, comets and interplanetary spacecraft in the Solar System. TDB is now (since 2006) defined as a linear scaling of Barycentric Coordinate Time (TCB). A feature that distinguishes TDB from TCB is that TDB, when observed from the Earth's surface, has a difference from Terrestrial Time (TT) that is about as small as can be practically arranged with consistent definition: the differences are mainly periodic,[2] and overall will remain at less than 2 milliseconds for several millennia.[3]
From the 17th century to the late 19th century, planetary ephemerides were calculated using time scales based on the Earth's rotation: usually the mean solar time of one of the principal observatories, such as Paris or Greenwich. After 1884, mean solar time at Greenwich became a standard, later named Universal Time (UT). But in the later 19th and early 20th centuries, with the increasing precision of astronomical measurements, it began to be suspected, and was eventually established, that the rotation of the Earth (i.e. the length of the day) showed irregularities on short time scales, and was slowing down on longer time scales. Ephemeris time was consequently developed as a standard that was free from the irregularities of Earth rotation, by defining the time "as the independent variable of the equations of celestial mechanics", and it was at first measured astronomically, relying on the existing gravitational theories of the motions of the Earth about the Sun and of the Moon about the Earth.
The differences are small, but it may give an idea of how complicated it is to measure the length of a day,

The diagrams in the last post were from TimeAndDate.com, and here are some more data:
On this page: How Long Is a Day on Earth?, they will tell us how much off the current day is.
Exact Day Length* — Fri, 2 Dec 2022
Today's prediction: 24 hours, 0 minutes, 0.0004430 seconds (0.4430 milliseconds)

Yesterday's prediction: 24 hours, 0 minutes, 0.0004407 seconds (0.4407 milliseconds)

At the start of today, UT1 was 0.0200320 seconds behind UTC.
While diagrams showing the variations are helpful for an overview, the numbers reveal the magnitude of the differences more clearly and show the date of the shortest and longest day for a period of about 50 years:

Average Solar Day Length*
YearAverage dayTotal yearly differenceShortest dayLongest dayLeap second added
2023-0.27 ms-100.30 ms12 Aug -1.94 ms21 Mar +1.08 ms-
2022-0.25 ms-90.16 ms29 Jun -1.59 ms5 Nov +1.01 ms-
2021-0.18 ms-65.15 ms9 Jul -1.46 ms26 Apr +1.00 ms-
2020-0.00 ms-1.30 ms19 Jul -1.47 ms8 Apr +1.62 ms-
2019+0.39 ms+141.25 ms16 Jul -0.95 ms22 Mar +1.68 ms-
2018+0.69 ms+252.47 ms30 Jun -0.64 ms4 Feb +1.69 ms-
2017+1.03 ms+375.01 ms4 Aug +0.06 ms25 Apr +2.20 ms-
2016+1.34 ms+490.76 ms18 Jul -0.03 ms10 Mar +2.49 ms31 Dec
2015+1.25 ms+458.03 ms17 Jun +0.19 ms26 Oct +2.31 ms30 Jun
2014+0.99 ms+362.96 ms24 Jul +0.02 ms26 Apr +2.02 ms-
2013+1.02 ms+373.99 ms6 Jul -0.35 ms28 Mar +1.97 ms-
2012+0.83 ms+304.11 ms16 Jul -0.35 ms5 Apr +1.87 ms30 Jun
2011+0.76 ms+277.94 ms27 Jul -0.34 ms14 May +1.85 ms-
2010+0.70 ms+254.74 ms23 Jul -0.76 ms1 Mar +2.09 ms-
2009+0.80 ms+293.37 ms6 Jul -0.43 ms22 Apr +1.81 ms-
2008+0.87 ms+319.49 ms16 Jul -0.41 ms5 Apr +1.91 ms31 Dec
2007+0.85 ms+310.81 ms27 Jul -0.63 ms16 Apr +2.31 ms-
2006+0.82 ms+300.88 ms12 Jun -0.40 ms7 Oct +2.26 ms-
2005+0.43 ms+157.76 ms5 Jul -1.05 ms27 Feb +1.73 ms31 Dec
2004+0.31 ms+114.01 ms15 Jul -1.05 ms5 Apr +1.56 ms-
2003+0.27 ms+100.16 ms13 Jul -0.96 ms19 Mar +1.55 ms-
2002+0.48 ms+173.79 ms6 Aug -0.74 ms2 Mar +1.66 ms-
2001+0.57 ms+208.94 ms2 Aug -0.71 ms11 Mar +1.64 ms-
2000+0.72 ms+262.42 ms11 Aug -0.25 ms26 Oct +1.58 ms-
1999+0.99 ms+361.19 ms30 Jun -0.13 ms15 Apr +1.93 ms-
1998+1.37 ms+501.72 ms9 Jul +0.01 ms1 Mar +2.66 ms31 Dec
1997+1.84 ms+671.08 ms4 Jul +0.52 ms6 Apr +2.98 ms30 Jun
1996+1.82 ms+666.37 ms10 Aug +0.67 ms12 May +2.68 ms-
1995+2.31 ms+843.66 ms25 Jul +0.81 ms17 Mar +3.29 ms31 Dec
1994+2.19 ms+800.86 ms6 Jul +0.86 ms27 Feb +3.36 ms30 Jun
1993+2.36 ms+862.66 ms17 Jul +1.25 ms2 May +3.49 ms30 Jun
1992+2.22 ms+812.25 ms12 Jul +0.84 ms18 Mar +3.59 ms30 Jun
1991+2.04 ms+743.88 ms27 Jun +0.79 ms1 Mar +3.00 ms-
1990+1.95 ms+710.04 ms20 Jul +0.63 ms26 Mar +3.28 ms31 Dec
1989+1.52 ms+555.00 ms2 Jul +0.25 ms10 Nov +2.82 ms31 Dec
1988+1.31 ms+480.30 ms12 Jul -0.09 ms20 Feb +2.76 ms-
1987+1.36 ms+497.35 ms23 Jul -0.06 ms1 Mar +2.67 ms31 Dec
1986+1.24 ms+451.06 ms2 Aug -0.04 ms23 Apr +2.30 ms-
1985+1.45 ms+528.83 ms16 Jul +0.11 ms9 Mar +2.64 ms30 Jun
1984+1.51 ms+554.42 ms12 Jul +0.16 ms18 Mar +2.77 ms-
1983+2.28 ms+832.08 ms23 Jul +1.01 ms1 Feb +3.57 ms30 Jun
1982+2.16 ms+789.64 ms2 Aug +0.84 ms23 Apr +3.14 ms30 Jun
1981+2.15 ms+786.03 ms16 Jul +0.82 ms8 Mar +3.42 ms30 Jun
1980+2.30 ms+842.04 ms8 Aug +1.34 ms23 Oct +3.24 ms-
1979+2.61 ms+953.02 ms23 Jul +1.46 ms27 Mar +3.65 ms31 Dec
1978+2.88 ms+1051.83 ms31 Jul +1.49 ms9 Mar +3.83 ms31 Dec
1977+2.77 ms+1012.60 ms14 Jul +1.46 ms4 Apr +3.72 ms31 Dec
1976+2.91 ms+1064.67 ms26 Jun +1.87 ms21 Oct +3.90 ms31 Dec
1975+2.69 ms+980.87 ms20 Jul +1.54 ms1 Nov +3.72 ms31 Dec
1974+2.72 ms+991.99 ms30 Jul +1.57 ms5 Apr +3.79 ms31 Dec
1973+3.04 ms+1106.21 ms2 Jan +0.00 ms2 Apr +4.03 ms31 Dec

If one goes through the list, then the shortest day has varied between January 2 (1973) and August 12 (2023), though it is true that most have fallen in July. The longest day has fallen anywhere between February 1 (1983) and November 10 (1989).

The time between the shortest and the longest day varies. 1973 appears as a span of three months, and 1980 had less than three months, from the shortest day, August 8, to the longest day, October 23.

The variations in the length of day are difficult to explain. One will need a lot of mathematics and physics to explain the variations, more so since there are not only the shortest and the longest day. Within one year, there are variations with minor ups and downs. It is far from a straight line.

If one goes through the list, then the shortest day has varied between January 2 (1973) and August 12 (2023), though it is true that most have fallen in July. The longest day has fallen anywhere between February 1 (1983) and November 10 (1989).
January 2nd as the shortest day of 1973 is an error (not the only one in DateAndTime's list), shortest day of that year was August 9th (check in the plots below), while year 2023 is still in the open so to say, which brings the question "How do they on DateAndTime predict their values in advance"?

Leaving those guys aside, IERS data can be found and checked on Index of /eoppc/eop/eopc04, where also the list containing LOD values from 1962 til mid November 2022 is located (which is used for that wiki plot). From those numbers we get a similar plot (yellowish line are the yearly means (average day duration in a year), centered at the middle of the respective year):

It's similar to plot in SAO's post, from where the plot below of Sun's activity (solar sunspots cycles) was borrowed:

It seems there's a correlation between Sun's (electro)magnetic activity and the duration of the day (LOD) on Earth, at least it's apparent that the days are getting shorter while Sun's going asleep in the roughly same period.

That brings us back to discussing shortest and longest days and when they occurred within a year.
Here's monthly distribution of their occurrences:

Apparently, shortest days in a year seem to follow Kepler's law as they happened dominantly around the time when Earth's close to aphelion in its orbit, but what's with longest days in a year? If Kepler's the reason behind variations in day duration, why don't we see longest days in a year in those months around perihelion? Even more shocking, apart from that one entry of December 7th for 1962 (first year of the data), there's no longest days in a year in December and absolutely no entries in January when Earth is supposed to really reach perihelion in its orbit!

To have a better look at the issue, the plot above is split into 11 years overlapping intervals:

Not only that days at the end and beginning of the calendar years were not among the longest days in a year, but except for few years, in that period of the year we notice drop in day lengths! It seems there's something else at play there, which effects on day duration are apparently more pronounced than those that come from Kepler's law (and gravitation).

Potential explanation, which on the other hand opens a whole new can of other questions, might be found in an old article from early 1920s: https://articles.adsabs.harvard.edu/pdf/1921PA.....29..325H, where we see that Earth on its way around the Sun passes the plane of Sun's rotation in June (ascending node) and December (descending node), while in March and September is furthest on the South and North from the Sun's equator.

The fact that the axis of Sun's rotation is inclined to ecliptic at approx. 7.25 deg usually goes 'unnoticed' in wider audience, while in fact that (perpendicular/equatorial) plane contains almost all angular momentum of the Solar system (not counting hypothetical Sun's twin), because the mass of Jupiter, as the largest planet in the system, is only about 0.001 of solar mass. Also, the orbit of Mercury, the only planet basically without precession and closest to the Sun is inclined about 7 deg to ecliptic (if it lies in or close to this Sun's plane of rotation, I don't know ATM, need to check). Well, that and other 'coincidences' are for another time and topic, what concerns us here is the potential effects on length of days on Earth.

As mentioned in another post:
I checked with good ol' Ben Davidson, and his take on the acceleration is that it is caused by the severe drop in the Earth's geomagnetic field. With shields down, this allows more energy into the system. "Higher supply, faster rotation - not unlike an electric motor".
and by crossing the plane of Sun's rotation, the Earth would be getting this "higher supply", if nothing else then via increased flux of cosmic rays and particles coming from Sun, like Solar wind, whose outflow from the Sun would be highest in this equatorial plane of Sun's rotation. That crossing of the equatorial plane of Sun's rotation would be enough to shorten the length of day, as seen from regular annual drop in LOD values in time intervals around these 'node points', i.e. in month(s) shortly after them.

With Sun's (magnetic) activity being on the decrease as seen from the plot of solar sunspots cycles above, meaning less particles coming from Sun being confined by solar magnetic field, Earth-Moon system in general receives higher amount of influx energy (particles) from the Sun, which could be the reason why we see significant drop in day duration on Earth since 2016.

It's not perfectly correlated but there's definitely a general correlation. Maybe there's a lag - maybe the rotation speed changes and the sun takes a few decades to "catch up"? If so, we could hypothesize that the 1995-2003 dip in day length is correlated with solar cycle 24, the first one that was very weak. So roughly a 10-20 year lag. If so, then the dramatic dip that started in 2016 should reflect in the sun's behavior around 2026-2036. And that timeframe lines up with your link's prediction where the sun kinda bottoms out between 2028 and 2042.
Maybe it's the other way around with the "lag" - maybe it takes material Earth (and Earth-Moon system) some time to 'catch up' whatever changes have happened with and around the Sun (and on spiritual plane, so to say)?

Just out published today- study was done prior to the recent acceleration last June.

Is THIS why the length of Earth's days has been increasing? Our planet's inner core may have recently paused and could be REVERSING, study reveals​

the study

Two studies discussing the gravitational constant (big G) and a possible correlation with LOD. The second study listed is a refutation of the first paper. Both of these were posted in 2015. The inclinations of the researchers tends toward an explanation of experimental error or inadequate measurements.

If the rotation of the Earth speeds up, then centrifugal forces increase. Centrifugal force is greatest at the equator and zero at the poles.

Interesting article here, originally posted in 2017 and updated last year, discusses the possible effects of increasing the speed of rotation of the Earth. Mentions Earth and gravity changes.

What would happen if the Earth started to spin faster?​

Even a mile-per-hour speed boost would make things pretty weird.

There are enough things in this life to worry about. Like nuclear war, climate change, and whether or not you’re brushing your teeth correctly. The Earth spinning too fast should not be high up on your list, simply because it’s not very likely to happen anytime soon—and if it does, you’ll probably be too dead to worry about it. Nevertheless, we talked to some experts to see how it would all go down.

Let’s start with the basics, like: How fast does the Earth spin now? That depends on where you are, because the planet moves fastest around its waistline. As Earth twirls around its axis, its circumference is widest at the equator. So a spot on the equator has to travel a lot farther in 24 hours to loop around to its starting position than, say, Chicago, which sits on a narrower cross-section of Earth. To make up for the extra distance, the equator spins at 1,037 mph, whereas Chicago takes a more leisurely 750 mph pace. (This calculator will tell you the exact speed based on your latitude.)

The Earth does change pace every now and then, but only incrementally. This summer, for instance, it skimmed 1.59 milliseconds off its typical rotation time, making June 29 the shortest day on record. One hypothesis is that changes in pressure actually shift the planet’s axis of rotation, though not to the extent where regular human beings could feel the difference.

Little bumps aside, if the Earth were to suddenly spin much faster, there would be some drastic changes in store. Speeding up its rotation by one mile per hour, let’s say, would cause water to migrate from the poles and raise levels around the equator to by a few inches. “It might take a few years to notice it,” says Witold Fraczek, an analyst at ESRI, a company that makes geographic information system (GIS) software.

What might be much more noticeable is that some of our satellites would be off-track. Satellites set to geosynchronous orbit fly around our planet at a speed that matches the Earth’s rotation, so that they can stay positioned over the same spot all the time. If the planet speeds up by 1 mph, then the satellites will no longer in their proper positions, meaning satellite communications, television broadcasting, and military and intelligence operations could be interrupted, at least temporarily. Some satellites carry fuel and may be able to adjust their positions and speeds accordingly, but others might have to be replaced, and that’s expensive. “These could disturb the life and comfort of some people,” says Fraczek, “but should not be catastrophic to anybody.”

But things would get more catastrophic the faster we spin.

You would lose weight, but not mass​

Centrifugal force from the Earth’s spin is constantly trying to fling you off the planet, sort of like a kid on the edge of a fast merry-go-round. For now, gravity is stronger and it keeps you grounded. But if Earth were to spin faster, the centrifugal force would get a boost, says NASA astronomer Sten Odenwald.

Currently, if you weigh about 150 pounds in the Arctic Circle, you might weigh 149 pounds at the equator. That’s because of the extra centrifugal force that’s generated as the equator spins faster combats gravity. Press fast-forward on that, and your weight would drop even further.

Odenwald calculates that eventually, if the equator revved up to 17,641 mph, the centrifugal force would be great enough that you would be essentially weightless. (That is, if you’re still alive. More on that later.)

Everyone would be constantly jet-lagged​

The faster the Earth spins, the shorter our days would become. With a 1 mph speed increase, the day would only get about a minute and a half shorter and our internal body clocks, which stick to a pretty strict 24-hour schedule, probably wouldn’t notice.

But if we were rotating 100 mph faster than usual, a day would be about 22 hours long. For our bodies, that would be like daylight savings time on boosters. Instead of setting the clocks back by an hour, you’d be setting them back by two hours every single day, without a chance for your body to adjust. And the changing day length would probably mess up plants and animals too.

But all this is only if Earth speeds up all of a sudden. “If it gradually speeds up over millions of years, we would adapt to deal with that,” says Odenwald.

Hurricanes would get stronger​

If Earth’s rotation picked up slowly, it would carry the atmosphere with it—and we wouldn’t necessarily notice a big difference in the day-to-day winds and weather patterns. “Temperature difference is still going to be the main driver of winds,” says Odenwald. However, extreme weather could become more destructive. “Hurricanes will spin faster,” he says, “and there will be more energy in them.”

The reason why goes back to that weird phenomenon we mentioned earlier: the Earth spins faster around the equator.

If the Earth wasn’t spinning at all, winds from the north pole would blow in a straight line to the equator, and vice versa. But because we are spinning, the pathway of the winds gets deflected eastward. This curvature of the winds is called the Coriolis effect, and it’s what gives a hurricane its spin. And if the Earth spun faster, the winds would be deflected further eastward. “That effectively makes the rotation more severe,” says Odenwald.

Water would cover the world​

Extra speed at the equator means the water in the oceans would start to amass there. At 1 mph faster, the water around the equator would get a few inches deeper within just a few days.

At 100 mph faster, the equator would start to drown. “I think the Amazon Basin, Northern Australia, and not to mention the islands in the equatorial region, they would all go under water,” says Fraczek. “How deep underwater, I’m not sure, but I’d estimate about 30 to 65 feet.”

If we double the speed at the equator, so that Earth spins 1,000 miles faster, “it would clearly be a disaster,” says Fraczek. The centrifugal force would pull hundreds of feet of water toward the Earth’s waistline. “Except for the highest mountains, such as Kilimanjaro or the highest summits of the Andes, I think everything in the equatorial region would be covered with water.” That extra water would be pulled out of the polar regions, where centrifugal force is lower, so the Arctic Ocean would be a lot shallower.

Meanwhile, the added centrifugal force from spinning 1,000 mph faster means water at the equator would have an easier time combating gravity. The air would be heavy with moisture in these regions, Fraczek predicts. Shrouded in a dense fog and heavy clouds, these regions might experience constant rain—as if they’d need any more water.

Finally, at about 17,000 miles per hour, the centrifugal force at the equator would match the force of gravity. After that, “we might experience reverse rain,” Fraczek speculates. “Droplets of water could start moving up in the atmosphere.” At that point, the Earth would be spinning more than 17 times faster that it is now, and there probably wouldn’t be many humans left in the equatorial region to marvel at the phenomenon.

If those few miserable humans would still be alive after most of Earth’s water had been transferred to the atmosphere and beyond, they would clearly want to run out of the equator area as soon as possible,” says Fraczek, “meaning that they should already be at the Polar regions, or at least middle latitudes.”

Seismic activity would rock the planet​

At very fast speeds—like, about 24,000 mph—and over thousands of years, eventually the Earth’s crust would shift too, flattening out at the poles and bulging around the equator.

“We would have enormous earthquakes,” says Fraczek. “The tectonic plates would move quickly and that would be disastrous to life on the globe.”

How fast would the Earth spin in the future?​

Believe it or not, Earth’s speed is constantly fluctuating, says Odenwald. Earthquakes, tsunamis, large air masses, and melting ice sheets can all change the spin rate at the millisecond level. If an earthquake swallows a bit of the ground, reducing the planet’s circumference ever so slightly, it effectively speeds up how quickly Earth completes its rotation. A large air mass can have the opposite effect, slowing our spins a smidgen like an ice skater who leaves her arms out instead of drawing them in.

The Earth’s rotation speed changes over time, too. About 4.4 billion years ago, the moon formed after something huge crashed into Earth. At that time, Odenwald calculates our planet was probably shaped like a flattened football and spinning so rapidly that each day might have been only about four hours long.

“This event dramatically distorted Earth’s shape and almost fragmented Earth completely,” says Odenwald. “Will this ever happen again? We had better hope not!”

Since the formation of the moon, Earth’s spin has been slowing down by about 3.8 mph every 10 million years, mostly due to the moon’s gravitational pull on our planet. So it’s a lot more likely that Earth’s spin will continue to slow down in the future, not speed up.

“There’s no conceivable way that the Earth could spin up so dramatically,” says Odenwald. “To spin faster it would have to be hit just right by the right object, and that would liquify the crust so we’d be dead anyway.”

When the C's mentioned that there was less gravity in pre-Roman times (which allowed giants to survive), maybe this was due to a faster rotation of the Earth and more centrifugal force counteracting gravity as a result.

If I remember correctly, in very old calendars the year was several days shorter, 360 days I think. Which could be divided evenly into 12 months of 30 days.

When the C's mentioned that there was less gravity in pre-Roman times (which allowed giants to survive), maybe this was due to a faster rotation of the Earth and more centrifugal force counteracting gravity as a result.

If I remember correctly, in very old calendars the year was several days shorter, 360 days I think. Which could be divided evenly into 12 months of 30 days.
As for reducing the effect of gravity through centrifugal force, it is the rotation around one's own axis that matters, not the circling around of the sun.
There would have to be a shorter day.

If I remember correctly, in very old calendars the year was several days shorter, 360 days I think. Which could be divided evenly into 12 months of 30 days.
Those calendars would fall out of sync with the seasons so they would add additional days to the year as festival days in order to get back in tune with the seasons.

When the C's mentioned that there was less gravity in pre-Roman times (which allowed giants to survive), maybe this was due to a faster rotation of the Earth and more centrifugal force counteracting gravity as a result.

If I remember correctly, in very old calendars the year was several days shorter, 360 days I think. Which could be divided evenly into 12 months of 30 days.

Shorter days does not equate to fewer days.

Something to consider. We use time as a reference frame. Maybe the old aether guys were onto something. Einstein relativity came along and since then its been "don't sit under the apple tree with anybody else but me."

The aether drag hypothesis proposed that the luminiferous aether is dragged by or entrained within moving matter. According to one version of this hypothesis, no relative motion exists between Earth and aether. According to another version, the Earth does move relative to the aether, and the measured speed of light should depend on the speed of this motion ("aether wind"), which should be measurable by instruments at rest on Earth's surface. In 1818, Augustin-Jean Fresnel proposed that the aether is partially entrained by matter. In 1845, George Stokes proposed that the aether is completely entrained within or in the vicinity of matter.

Although Fresnel's almost-stationary theory was apparently confirmed by the Fizeau experiment (1851), Stokes' theory was apparently confirmed by the Michelson–Morley experiment (1881, 1887). Hendrik Lorentz resolved this contradictory situation in his own aether theory, which banished any form of aether dragging. Albert Einstein's special theory of relativity (1905) excludes aether as a mechanical medium.[1][2][3]

Frame-dragging is an effect on spacetime, predicted by Albert Einstein's general theory of relativity, that is due to non-static stationary distributions of mass–energy. A stationary field is one that is in a steady state, but the masses causing that field may be non-static ⁠— rotating, for instance. More generally, the subject that deals with the effects caused by mass–energy currents is known as gravitoelectromagnetism, which is analogous to the magnetism of classical electromagnetism.

The first frame-dragging effect was derived in 1918, in the framework of general relativity, by the Austrian physicists Josef Lense and Hans Thirring, and is also known as the Lense–Thirring effect.[1][2][3] They predicted that the rotation of a massive object would distort the spacetime metric, making the orbit of a nearby test particle precess. This does not happen in Newtonian mechanics for which the gravitational field of a body depends only on its mass, not on its rotation. The Lense–Thirring effect is very small – about one part in a few trillion. To detect it, it is necessary to examine a very massive object, or build an instrument that is very sensitive.

In 2015, new general-relativistic extensions of Newtonian rotation laws were formulated to describe geometric dragging of frames which incorporates a newly discovered antidragging effect.[4]

We formulate new general-relativistic extensions of Newtonian rotation laws for self-gravitating stationary fluids. They have been used to re-derive, in the first post-Newtonian approximation, the well known geometric dragging of frames. We derive two other general-relativistic weak-field effects within rotating tori: the recently discovered dynamic anti-dragging and a new effect that measures the deviation from the Keplerian motion and/or the contribution of the fluids selfgravity. One can use the rotation laws to study the uniqueness and the convergence of the post-Newtonian approximations, and the existence of the post-Newtonian limits.

This is likely just coincidence

Abstract –About a dozen measurements of Newton’s gravitational constant, G, since 1962 have yielded values that differ by far more than their reported random plus systematic errors. We find that these values for G are oscillatory in nature, with a period of P = 5.899 ± 0.062 yr,

From a year ago:

Something weird is happening onboard one of the most iconic NASA spacecraft of the 20th century. Voyager 1 is a certified space soarer, coming up on half a century of flight-time and still going strong. But it's beaming back an abnormal message to NASA
...
The problem seems to be coming from the attitude articulation and control system (AACS) computer onboard Voyager 1.

One of three on the spacecraft, this computer is responsible for Voyager 1’s orientation, like controlling its thrusters and keeping the high-gain antenna pointed towards Earth so that data about the interstellar medium keeps trickling back.

The team can steer the spacecraft, and that, paired with the signal’s strength and a lack of fault protection activation, tells them that Voyager 1 is doing well. But the telemetry signal itself “does not make any sense,” producing either all zeros or the number 377, Dodd says.

If the spacecraft was in dire straits, “we would be seeing a degradation in our signal from our spacecraft,” she says, which they aren’t seeing.

“Somewhere in the interface with the flight data system … there’s something that’s causing the telemetry data to be mixed up, I guess, or nonsensical,” Dodd says. “And we don’t understand that yet.”

The natural logarithm of number 377 is 5.93

This is likely just coincidence

From a year ago:

Since you bring Newton's "G constant", here's an abstract from a post found in Ark's blog, (bold is his) :

"P.S.1. 01-04-23 A friend, physicist, sent me this morning a link to this video by Rupert Sheldrake:

Rupert Sheldrake - The Science Delusion BANNED TED TALK

And there, in particular:

.../... But I said "well, then what about big G?" The gravitational constant, known in the trade as "big G", it was written with a capital G. Newton's universal gravitational constant.
"That's varied by more than 1.3% in recent years. And it seems to vary from place to place and from time to time." And he said "oh well, those are just errors. And unfortunately there are quite big errors with big G."

So I said "well, what if it's really changing? I mean, perhaps it is really changing." And then I looked at how they do it, what happens is they measure it in different labs, they get different values on different days, and then they average them. And then other labs around the world do the same, they come out usually with a rather different average. And then the international committee of metrology meets every ten years or so and average the ones from labs all around the world to come up with the value of big G. But what if G were actually fluctuating? What if it changed? There's already evidence actually that it changes throughout the day and throughout the year. What if the earth, as it moves through the galactic environment went through patches of dark matter or other environmental factors that could alter it? Maybe they all change together. What if these errors are going up together and down together? For more than ten years I've been trying to persuade metrologists to look at the raw data. In fact I'm now trying to persuade them to put it up online, on the internet. With the dates, and the actual measurements, and see if they're correlated. To see if they're all up at one time, all down at another. If so, they might be fluctuating together. And what would tell us something very, very interesting. But no-one has done this, they haven't done it because G is a constant. There's no point looking for changes. "

If the rotation of the Earth speeds up, then centrifugal forces increase. Centrifugal force is greatest at the equator and zero at the poles.

Interesting article here, originally posted in 2017 and updated last year, discusses the possible effects of increasing the speed of rotation of the Earth. Mentions Earth and gravity changes.
Considering that Earth's rotation today reduces gravity at the equator for only about 0.3% (wiki), it seems highly unlikely that changes in rotation alone played a significant role in hypothetical 4.6% gravity increase during real fall of Roman Empire. To reach that amount of change in gravity only due to change(s) in rotational speed and pattern, even if then Rome stood on the equator, days on Earth would need to be less than 14h long. Effects of that kind of change on the planet would almost certainly be quite dramatic if not devastating for living world and hardly would go unnoticed/unmentioned by peoples around the globe of that time, like Chinese for instance.

Shorter days does not equate to fewer days.
It usually even means the opposite.

To reach that amount of change in gravity only due to change(s) in rotational speed and pattern, even if then Rome stood on the equator, days on Earth would need to be less than 14h long.
Bolded part is a bit off, because Trot=14h is the result obtained for centrifugal acceleration (2*pi/Trot)2 * R (where R=6731km is average radius of the Earth nowadays) amounting only 1% of the average gravity (g=9.81m/s2) today on the surface of the Earth. To reduce today's surface gravity for 4%, which would be in the ballpark of 4.6% increase in the past, Earth would need to make one rotation in roughly 7h.
That even more strongly disqualified changes in rotational speed and/or pattern as main culprits for hypothetical 4.6% change in Earth's gravity which, assuming that the giants weren't confined to the geographical region of old Roman Empire alone, would probably have been a global phenomenon/event. In other words, the change of being more strongly glued to the ground was probably felt by the living world all around the globe in those times.

It usually even means the opposite.
In the sense that when year remains unchanged, in the case of shorter days there would be more of them, not less, within that year.
For example, if days in old Roman times were just 7h long, as mentioned above in case of rotation being the reason for 4.6% change of gravity, standard year as we know it today (365.25 * 24h) would be more than 1250 such 7h days long.

Apologies for doubling posts.

Here are some potential data bits about Earth's rotation and possibly mechanism(s) behind/driving it.

First plot below shows last year and a half of IERS data (link), with red line representing LOD values in ms obtained with new algorithm (new data) and blue line by algorithm from last year (old data). There are no significant differences between the lines which indicates consistency of two algorithms/datasets. There is also a dotted gray line on the plot showing Moon phases for the time period in question (going from 0, standing for New Moon, to 1 which marks Full Moon), and a bit darker vertical line marking Tonga volcano eruption on January 15th 2022 (Forum thread).

The oscillatory/cyclic pattern seems to be appearing with rather regular time period of roughly two weeks; check the horizontal (time) distance between two neighboring local maxima (peak points) and/or minima (dip points) across the plot. The approximate biweekly time period is also present around January 15th 2022 Tonga eruption event, which created ripples/waves in Earth's atmosphere too, suggesting that neither an energetic event on Earth's surface like a huge volcano eruption nor notable changes in pressure and/or density in the atmosphere all around the Earth play a significant role in changing Earth's rotation.

Absence of any visible change in the behavior at the beginning of February this year - slowing down for a week or so and then speeding up with the rotation for roughly the next week - around the time of Syria-Turkey earthquake on February 6th 2023 (wiki), puts also planetary seismic activity in the category of non-leading order effects in relation to Earth's rotation.

Comparison of LOD data to lunar cycles (Moon phases) shows no clear synchronicity between the two, indicating that also mechanical (Newtonian) Earth-Moon motion (and around their system's barycenter) does not play a leading role in this Earth's rotational cyclic (biweekly) dance. For example, two local minima (shortest days) before and after June 29th 2022, the day when the Earth spun "faster than ever before", together with that global minimum itself, roughly coincide with consecutive New and Full Moon phases in that time period. On the other side, similar 'coincidence pattern' can be observed around the start of April and October when local maxima (longest days) coincide with consecutive New and Full Moon phases. This 'interchangeable coincidence' of New and Full Moons with both, local minima and maxima, but at different times in the year, indicates that the exact lunar cycle (which exact Moon phase it is) apparently does not discriminate between two opposite trends, acceleration and deceleration of Earth's rotation.

LOD data show that Earth's rotation was speeding up (accelerating) from Last Quarter of (the previous lunar cycle) til New Moon on June 29th 2022, at which point Earth started to slow down (decelerate) with its rotation and continued to do so until roughly First Quarter (of that lunar cycle). Such behavior suggests that the direct influx of particles from the Sun also does not affect Earth's rotation in a significant notable manner, because there are no visible changes either on acceleration side of the slope/line when the Moon 'entered the space' between the Sun and the Earth (2-3 days before the New Moon), or on deceleration side when 2-3 days after the New Moon that region between the star and the planet was left by Earth's natural satellite.

Apart from that 'fastest Earth's rotation day ever' unexpectedly coincided with New Moon when lunar shadowing of the planet was (expected to be) the greatest during the lunar cycle, absence of slowing down when the Moon started to cast its shadow on the planet and absence of speeding up in Earth's rotation when it stopped, suggest that 'biweekly' oscillatory rotational pattern is not due to the 'currents' flowing in from the Sun, at least not in a significant manner that showed up on the plot. In a similar direction of indicating that direct high energy solar radiation effects on Earth's rotation are of a non-leading order, points also the absence of any notable traces of solar flares (wiki) reaching Earth events in LOD data.

Next plot, showing last 7 years of LOD data, sort of suggests similar things as the one above.

Although annual minima fall midyear around the aphelion, the absence of opposite behavior around perihelion time suggests that Keplerian motion is not behind the patterns with yearly periodicity in Earth's rotation. On the other side, not only there is absence of notable speeding up of Earth's rotation in March/April and September/October when Earth in its orbit is supposed to pass through solar equatorial plane, but data show that roughly during those months Earth's rotation exhibits its slowest spinning over the year. That behavior is sort of the opposite of the expected, if the flux of particles and radiation coming directly from the Sun plays a driving hand/force behind Earth's rotational pattern, because the largest 'outpour' from the Sun is expected to happen in and around solar equatorial plane (not only because of differential rotation of the Sun where equatorial regions on the surface rotate faster the polar ones).

Another 'puzzling point' comes from the comparison between the whole LOD dataset from 1962 and solar activity for that time period (link), as shown in next two plots below.

Since there is no (straight) coincidence between solar activity as represented by smoothed sunspots number (red line in the upper part of 2nd plot) and average yearly behavior of Earth's rotation over 61 years of LOD data (yellow line in 1st plot), neither in positions of minima/maxima on the plots nor in their time periodicity, it appears there's also no (clear) correlation between them.

I've also provisionally looked into possible (electro)magnetic origins for observed features in LOD values on different time scales, like simple interactions/couplings between solar and geomagnetic fields for example, but so far haven't spotted any (clear) signs or signatures even for them.