Business Insider claims that China's Three-Gorge's Dam will slow down Earth's rotation:

Raising 39 trillion kilograms of water 175 meters above sea level will increase the Earth’s moment of inertia and thus slow its rotation. However, the effect would extremely small.  NASA scientists calculated that shift of such as mass would increase the length of day by only 0.06 microseconds and make the Earth only very slightly more round in the middle and flat on the top. It would shift the pole position by about two centimeters (0.8 inch).

Is this true?

  • 2
    The article isn't saying this is "special". The factors are reservoir volume, height above sea level, and latitude. The mass of the dam is small compare to the mass of the water (so don't compare to mega-polis). A higher-elevation dam like Glen Canyon would have a bigger effect than Three Gorges. There are reservoirs that are much bigger than Three Gorges also. – DavePhD Nov 9 '19 at 18:52
  • 2
    I'm a little surprised at the magnitude of the claimed effects, but the basic physics is sound (even if there is an error in the math). Do note that the significant factor is not the mass, per se, but rather the elevation of the mass above the "normal" height of land. "Conservation of momentum" says that elevating the mass will slow the earth's rotation. – Daniel R Hicks Nov 9 '19 at 18:58
  • 1
    @DanielRHicks As anyone spinning on an office chair while moving the legs in and out can attest to. – pipe Nov 9 '19 at 19:13
  • 4
    @DavePhD - Per wikipedia's list of reservoirs by volume, the Three Gorges Dam ranks as #27 . Moreover, several of the larger ones are much higher above sea level and are much closer to the equator. There is indeed nothing special about theThree Gorges Dam reservoir in this regard. – David Hammen Nov 10 '19 at 9:53
  • 1

Yes, based on modelling, the rotation of the Earth was slowed down by 0.06 microseconds per "day".

That is, a day would be made about one seventeen millionth of second longer than it otherwise would be if the dam had not been built.

Business Insider references NASA's Jet Propulsion Laboratory (JPL} - the original link is broken, but the JPL still make the claim in a 2005 press release:

Dr. Richard Gross of NASA's Jet Propulsion Laboratory, Pasadena, Calif., and Dr. Benjamin Fong Chao, of NASA's Goddard Space Flight Center, Greenbelt, Md., said all earthquakes have some affect on Earth's rotation. It's just they are usually barely noticeable.


To make a comparison about the mass that was shifted as a result of the earthquake, and how it affected the Earth, Chao compares it to the great Three-Gorge reservoir of China. If filled, the gorge would hold 40 cubic kilometers (10 trillion gallons) of water. That shift of mass would increase the length of day by only 0.06 microseconds and make the Earth only very slightly more round in the middle and flat on the top. It would shift the pole position by about two centimeters (0.8 inch).

In 1995, Dr Chao published a paper showing how to model the effects of major reservoirs on the moment of inertia (and the location of the poles):

That paper looked at 88 major reservoirs, not including Three-Gorges Dam - it wasn't opened yet.

In 2002, Dr Chao was a co-author of another paper:

In this paper, they use the same modelling from the 1995 paper to look at the effect of the Three-Gorges Dam.

when filled, the Three‐Gorges Reservoir would cause an overall change in Earth's dynamic oblateness J2 by +3.0 × 10−13, J3 by +2.4 × 10−12, and Earth's rotation by only +0.060 μs in length‐of‐day, but as much as 0.64 milliarcsecond in polar motion excitation toward the direction away from the longitude of the Three‐Gorges.

That a dam would affect the rotation of the Earth is not surprising. As the JPL article explains:

"Any worldly event that involves the movement of mass affects the Earth's rotation, from seasonal weather down to driving a car," Chao said.

| improve this answer | |
  • 1
    So in 50,000 years we will add another leap second because of this. – gnasher729 Nov 10 '19 at 16:21

The Business Insider article referenced in the opening post was published over nine years ago, and this was in turn based on a then five year old press release from NASA. The Business Insider article briefly indicates that this 0.06 microsecond change in length of day (LOD) was a calculated effect.

The NASA press release is clearer in this regard: It explicitly says that both the 2.68 microsecond change in LOD due to the 2005 Sumatra earthquake and the 0.06 microsecond change due to the filling of the Three Gorges Reservoir are calculated rather than observed effects. The hope in 2005 was that the impulsive change due to the earthquake would eventually be observable. This turned out not to be the case. (Evidence: Lack of a press release from NASA touting that the calculated effect has been observed.) The ability to observe the Earth's orientation and rotation rate in 2005 simply was not good enough to observe even a 2.68 microsecond variation in length of day, let alone a 0.06 microsecond change.

While measurement quality of the Earth orientation parameters (length of day and the direction in which the Earth's rotation axis points) have improved by multiple orders of magnitude since the 1970s, the one sigma uncertainty in length of day has stubbornly remained at about 2 to 5 microseconds for corrected observations since the time of that NASA press release, and has remained at about twice that for recent measurements. (The International Earth Rotation and Reference Systems Service (IERS) updates their observations about a month after the fact.) For example, the one sigma uncertainty for recent measurements is about 6 microseconds for recent measurements per the most recent IERS Bulletin B.

If effects such as earthquakes and dam construction are not observable, why bother? Neither the Business Insider article nor the NASA press release address this rhetorical question. The reason is to make it possible to move changes in the Earth orientation parameters from the "only observable after the fact" column to the "predictable before the fact and verifiable after the fact" column. Uncertainties in and the predictability of the Earth orientation parameters have an increasing effect on commerce (e.g., predicting the orbits of GPS and geosynchronous satellites), government (e.g., spy satellites, various weapons), and science (e.g., microarcsecond radio astronomy).

The two NASA scientists behind that NASA press release are members of one of several groups that contribute to the IERS to improve accuracy and predictability. The ability to predict the effects of earthquakes and reservoirs goes hand-in-hand with the ability to predict the effects of more important contributors to variations in the Earth orientation parameters such as tides, atmospheric pressure, and the amount of snow that falls on Siberia and Canada.

| improve this answer | |

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .