If every Indian who can jump did so at the same instant, about 6.2 × 10¹⁰ kg of people would rise a quarter of a metre and land with roughly 1.5 × 10¹¹ joules, the energy of about 36 tonnes of TNT. By conservation of momentum the Earth would move about 2.6 femtometres, around three proton widths, and only a tiny fraction of the landing energy would become seismic waves, equal to a magnitude −0.4 to 2.3 quake if it all happened in one spot. Because it is spread over 33 lakh square kilometres, no one would feel it, though a nearby seismometer could log it.
- 6.2 × 10¹⁰ kg of people leave the ground, about 85% of India at 50 kg each.
- 1.5 × 10¹¹ J comes back at landing, the energy of about 36 tonnes of TNT.
- The Earth recoils by 2.6 femtometres, about three proton widths, then returns.
- Only 10⁻⁷ to 10⁻³ of the landing energy becomes ground waves: magnitude −0.4 to 2.3 at most, if concentrated.
- Spread at 445 people per km², nobody feels it; a packed stadium is what makes a signal.
This is a classic of back-of-envelope physics, usually asked about the whole world. India is the better test case: it is the most populous country, with 1,463.9 million people in 2025 by the UN's estimate, and runs on one clock, so a synchronised jump is at least imaginable.
Below I set up the problem, solve it step by step with real numbers, run the sensitivities, and check the answer against what seismologists actually recorded when stadium crowds jumped.
What are the assumptions?
The answer depends on four numbers:
- How many people jump (the population and the share able to);
- How heavy they are, averaged over all ages;
- How high they get, measured at the centre of mass;
- How much landing energy turns into ground waves, the seismic efficiency.
Has a jumping crowd ever shaken the ground?
On 8 January 2011, the Seattle Seahawks played the New Orleans Saints in an NFC wild-card playoff at what was then Qwest Field, in front of 66,336 people. With 3:38 left and Seattle four points ahead, running back Marshawn Lynch took the ball on his own 33-yard line and ran 67 yards for a touchdown, breaking nine tackles on the way. The stadium erupted. The Pacific Northwest Seismic Network recorded a small tremor, put at about magnitude 2.0, located at the stadium. Seattle won 41–36, and the run has been called the "Beast Quake" ever since.
Wikipedia: Beast Quake (citing the Pacific Northwest Seismic Network)That was 66,000 people, packed into one bowl, celebrating together. India has about 22,000 times as many people, but spread across a subcontinent. Here are the variables, the value I use and where each comes from.
| Symbol | Meaning | Value used | Range tested | Basis |
|---|---|---|---|---|
| N | Population of India | 1.4639 × 10⁹ | fixed | UN World Population Prospects 2024, mid-2025 |
| f | Share able to jump | 0.85 | 0.70 to 0.95 | My estimate: excludes infants, the very old, the ill |
| m̄ | Average body mass, all ages | 50 kg | 45 to 57.7 kg | Asian adult average 57.7 kg (Walpole 2012); children pull it down |
| h | Rise of the centre of mass | 0.25 m | 0.10 to 0.40 m | My estimate; fit police recruits averaged 0.53 m on a jump-and-reach test (Lockie 2024) |
| η | Seismic efficiency | 10⁻⁵ | 10⁻⁷ to 10⁻³ | Impact modelling: 10⁻⁷ to 10⁻⁶ in loose regolith, 10⁻⁴ to 10⁻³ in bedrock (Rajšić 2021) |
| M⊕ | Mass of the Earth | 5.972 × 10²⁴ kg | fixed | NASA Earth fact sheet |
The authors combined WHO data on body-mass index and height with UN population figures for 2005 to estimate the mass of every adult on Earth: about 287 million tonnes, with a global average body mass of 62 kg. Asia, with 2.815 billion adults, averaged 57.7 kg, against 80.7 kg in North America. I use 57.7 kg as the upper end for India and 50 kg for the whole population including children. The limit: the data are from 2005 and averaged over all of Asia, not India alone.
Read the paper (free)How much energy is in the jump?
About 150 billion joules. The theory is gravitational potential energy: lifting a mass \( m \) through a height \( h \) in gravity \( g \) stores energy \( mgh \), and it all comes back as kinetic energy at landing. Air resistance at these speeds is negligible.
- Mass in the air. Multiply the jumpers by their average mass.
\[ M_j = f\,N\,\bar{m} = 0.85 \times 1.4639\times10^{9} \times 50\ \text{kg} = 6.22\times10^{10}\ \text{kg} \]
- Energy at the top. Gravitational potential energy of that mass at height h.
\[ E = M_j\,g\,h = 6.22\times10^{10}\ \text{kg} \times 9.81\ \text{m s}^{-2} \times 0.25\ \text{m} = 1.53\times10^{11}\ \text{J} \]
- In TNT. One tonne of TNT is defined as 4.184 × 10⁹ J.
\[ \frac{1.53\times10^{11}}{4.184\times10^{9}} \approx 36\ \text{tonnes of TNT} \]
- Per person. \( 50 \times 9.81 \times 0.25 \approx 123\ \text{J} \), about the energy of a 1 kg bag of rice dropped from 12.5 m.
Thirty-six tonnes of TNT sounds alarming until you remember it is shared among 124 crore people over a whole subcontinent. The interesting physics is in what that energy does to the Earth.
Does the Earth move when we all jump?
Yes, by about 2.6 femtometres, which is three times the radius of a proton. The governing law is conservation of momentum, which follows from Newton's third law.
When you push off the ground, the ground (the whole Earth) is pushed the other way with an equal and opposite impulse. The centre of mass of the Earth-plus-people system stays put.
- Take-off speed. From energy conservation, \( \tfrac12 v^2 = g h \).
\[ v = \sqrt{2gh} = \sqrt{2 \times 9.81 \times 0.25} = 2.21\ \text{m s}^{-1} \]
- Total momentum of the jumpers.
\[ p = M_j v = 6.22\times10^{10} \times 2.21 = 1.38\times10^{11}\ \text{kg m s}^{-1} \]
- Earth's recoil speed. The same momentum, carried by the planet's mass.
\[ v_\oplus = \frac{p}{M_\oplus} = \frac{1.38\times10^{11}}{5.972\times10^{24}} = 2.3\times10^{-14}\ \text{m s}^{-1} \]
- Earth's displacement. Because the centre of mass stays fixed, \( M_\oplus\,\Delta x_\oplus = M_j\,h \).
\[ \Delta x_\oplus = \frac{M_j h}{M_\oplus} = \frac{6.22\times10^{10}\times0.25}{5.972\times10^{24}} = 2.6\times10^{-15}\ \text{m} \]
- Time in the air. \( t = 2v/g = 0.45\ \text{s} \). Then everyone lands, the Earth returns, and the net motion is zero.
A proton's charge radius is about 0.84 × 10⁻¹⁵ m, so the Earth shifts by roughly three of them for less than half a second.
One simplification to flag: India spans about 30 degrees of the globe, so the "up" directions are not perfectly parallel. The vertical pushes from Gujarat and Arunachal differ by around 30°, and since cos 15° ≈ 0.97, the net push loses only about 3%. That does not change the answer.
Would it register as an earthquake?
Only barely, and only if you counted it all as one event. When 124 crore people land, almost all the energy goes into flexing knees, compressing soil and making noise. The fraction that leaves as elastic waves through the ground is called the seismic efficiency, η.
For small meteorite impacts on Mars, Rajšić and colleagues found η of 10⁻⁷ to 10⁻⁶ for loose, porous ground and 10⁻⁴ to 10⁻³ for solid rock. Human feet on soil, roads and floors are softer still than a meteorite, so the low end is more plausible, but I keep the whole range.
Using the iSALE-2D shock physics code, the team simulated how a small meteorite making a 1.5 m crater on Mars converts kinetic energy into seismic waves, for unfractured bedrock, fractured bedrock at 25% porosity and regolith at 44% and 65% porosity. They found the seismic efficiency is very sensitive to the target's sound speed and elastic threshold: about 10⁻⁷ to 10⁻⁶ in regolith and 10⁻⁴ to 10⁻³ in bedrock. The limit: these are hypervelocity impacts on Mars, so I use them only as bounds for how badly a soft landing on soil couples into the ground.
Read the paper (DOI)To turn seismic energy into a magnitude, use the Gutenberg–Richter energy–magnitude relation. The US Geological Survey gives it in ergs as \( \log_{10}E = 11.8 + 1.5M \); converting ergs to joules (1 erg = 10⁻⁷ J) gives:
- Seismic energy at η = 10⁻⁵. \( E_s = \eta E = 10^{-5}\times1.53\times10^{11} = 1.5\times10^{6}\ \text{J} \).
- Magnitude.
\[ M = \frac{\log_{10}(1.53\times10^{6}) - 4.8}{1.5} = \frac{6.18 - 4.8}{1.5} \approx 0.9 \]
- Range. At η = 10⁻⁷, \( M \approx -0.4 \); at η = 10⁻³, \( M \approx 2.3 \). Each step of 10 in η adds 0.67 to the magnitude.
Even the top of that range, around magnitude 2, is a quake most people never feel, and it assumes every joule came from one point. In reality it would be smeared over the country.
See the data as a table
| Magnitude if released at one point (M) | |
|---|---|
| η = 10⁻⁷ | -0.41 |
| 10⁻⁶ | 0.26 |
| 10⁻⁵ | 0.92 |
| 10⁻⁴ | 1.59 |
| 10⁻³ | 2.26 |
Source: Author's calculation: E = 1.53 × 10¹¹ J, Gutenberg–Richter relation (USGS), efficiency bounds from Rajšić et al. 2021 · Negative magnitudes are real: the scale is logarithmic and continues below zero.
Why would nobody feel it?
Because the jumpers are spread thin. India's 1.46 billion people over about 3.287 million km² is an average of about 445 people per square kilometre.
Seismic waves spread out and weaken with distance, so the shaking you feel is dominated by the people within a few hundred metres of you. In an average square kilometre that is a few hundred jumpers, not tens of thousands.
What a stadium shows
Compare a stadium, the one place where crowd jumping is well measured. Seismologists led by Jacqueline Caplan-Auerbach analysed Taylor Swift's July 2023 concerts at Seattle's Lumen Field, the same ground as the Beast Quake, about 72,000 people a night, on an accelerometer 150 m away.
Synchronised jumping and dancing showed up as narrow-band energy at 1 to 8 Hz, matched to each song's tempo, and the peak shaking was about 2.5 times that of the 2011 "Beast Quake" football celebration. Packing 72,000 people into one bowl is what makes a signal; spreading 124 crore over a country does the opposite.
Using station UW.KDK, an accelerometer about 150 m from the stadium, plus around 80 fan videos to time songs, the authors found low-frequency (1 to 8 Hz) energy from crowd motion whose fundamental matched song tempo (for example 2.67 Hz for a 160 bpm song), high-frequency (30 to 80 Hz) energy from amplified music, and song-by-song waveforms that correlated above 0.75 between the two nights. Sound checks without a crowd produced no low-frequency energy. The limit: the authors say a conventional magnitude is not meaningful for stadium events, and cite per-song magnitudes of −0.5 to 0.85 from a separate Los Angeles study.
Read the paper (free)Why timing matters: N against √N
There is a second reason the national signal stays small: timing.
- Perfectly synchronised landings add coherently: the wave amplitudes from N jumpers add, so amplitude grows like N.
- Randomly scattered landings add incoherently: amplitude grows only like √N.
Over a whole country, with people reacting to a signal at slightly different moments and waves arriving from different distances, the national sum behaves much more like √N. Stadium seismology shows exactly this: synchronised motion produces the biggest spikes.
A crowd shakes the ground by being packed and in step. A nation is neither.The working of this post

How sensitive is the answer to the assumptions?
The energy moves by a factor of about seven across reasonable inputs, and the magnitude by about one unit. That is the comforting thing about logarithmic scales: big uncertainties in the inputs make small changes in the verdict.
| Case | f | m̄ (kg) | h (m) | Energy (J) | Earth moves (fm) |
|---|---|---|---|---|---|
| Low | 0.70 | 45 | 0.10 | 4.5 × 10¹⁰ | 0.8 |
| Central | 0.85 | 50 | 0.25 | 1.5 × 10¹¹ | 2.6 |
| High | 0.95 | 57.7 | 0.40 | 3.1 × 10¹¹ | 5.4 |
A cross-check from the other direction: if one stadium song is worth a magnitude of up to about 0.85, and India has about 20,000 stadiums' worth of people (1.46 × 10⁹ ÷ 72,000), adding the energies gives \( 0.85 + \tfrac{2}{3}\log_{10}(20{,}000) \approx 3.7 \).
That is an upper bound, because a song is minutes of repeated jumping on a concrete bowl, not one jump on soil, but it confirms the order of magnitude: a small earthquake's worth of energy, scattered too thin to be felt.
Use this to check what you just read: plug in your own numbers below.
Jump India: energy, Earth recoil and equivalent magnitude
Could India actually pull off a simultaneous jump?
The physics is easy; the coordination is the hard part. India already has the clock for it.
- The source clock. CSIR-NPL generates Indian Standard Time with an accuracy of 20 nanoseconds, and under the Legal Metrology (IST) Rules notified in 2026 that time is pushed to banks, railways and telecom networks.
- The phone. A phone synced to network time is good to a fraction of a second.
- The human. Reaction times add a few tenths of a second of scatter, comparable to the 0.45 s flight.
So a well-announced countdown would put most of the country in the air within about a second of each other: synchronised enough to be fun, not enough to add coherently across a continent.
The Edinburgh calibration
The best real-world calibration comes from Edinburgh. When Taylor Swift played Murrayfield in June 2024, the British Geological Survey recorded ground movement of up to 23.4 nanometres at stations nearly four miles (about 6.4 km) away, and estimated the crowd transmitted about 80 kilowatts of power, CBS News reported.
In August 2025 Oasis crowds of about 70,000 a night beat that, peaking at 261 kW during "Cigarettes And Alcohol". The BGS put it down to people jumping and dancing in time, not loudness. Those are concentrated crowds of 70,000 jumping repeatedly for minutes. A national single jump puts a few hundred people in each square kilometre, once.

Conservation of momentum limits the Earth's response to about 2.6 × 10⁻¹⁵ m, and seismic efficiency of 10⁻⁷ to 10⁻³ turns the 1.5 × 10¹¹ J landing into at most a magnitude 2 or so if it were concentrated at one point. Spread over 33 lakh km² at 445 people per km², the shaking is far below what anyone feels. A seismometer next to a packed ground or a Kumbh crowd could record it; the planet would not notice.
Where does the model break down?
Three places:
- Seismic efficiency. The efficiency of people landing on soil, concrete and floors has not been measured directly that I could find, so I borrowed bounds from meteorite modelling.
- Magnitude. The Gutenberg–Richter energy relation was built for earthquakes, which are slip on faults, not distributed surface impacts, so "equivalent magnitude" is a comparison, not a forecast.
- Buildings. A crowd jumping in step on a floor near its natural frequency can excite resonance, which is a structural safety issue (see the warning above), not a seismic one.
The same kind of national-scale arithmetic, done for a far more consequential case, is in what if the monsoon failed two years in a row. For the instrument that keeps everyone on the same second, see what if India had two time zones. And for a look at forces and motion in machines you can actually ride, the Yantra world starts at the Motoverse.
What's new
- British Geological Survey finds Oasis crowds at Murrayfield shook the ground more than in 2009, peaking at 261 kW Anglo-Celt (PA), Aug 2025
- Taylor Swift fans in Edinburgh triggered seismic monitors nearly four miles away CBS News, Jun 2024
- GSA Today paper links concert seismic signals to synchronised crowd motion GSA Today, May 2024
What if next?
- What if all of India honked at once? Same crowd, different wave: sound instead of seismic, and the air is far less forgiving.
- What if all 1.4 billion of us wanted chai at the same minute? The milk, the gas and the queue, counted.
- What if every Indian wedding of the year happened on one day? Guests, generators and gold, all at once.
Questions people ask
Can everyone jumping at once cause an earthquake?
No. The total landing energy for India is about 1.5 × 10¹¹ J, and only a tiny fraction, 10⁻⁷ to 10⁻³ by impact-modelling bounds, becomes seismic waves. Even concentrated at one point that is a magnitude of roughly −0.4 to 2.3, and spread across a country it is not felt.
Would the Earth move out of its orbit if everyone jumped?
No. Jumping is an internal push between people and the Earth, so the system's centre of mass, and therefore the orbit, does not change at all. The Earth moves about 2.6 femtometres relative to that centre of mass and comes back when everyone lands.
Has a jumping crowd ever been measured?
Yes, many times at concerts. The British Geological Survey recorded up to 23.4 nanometres of ground movement about 6.4 km from Murrayfield during Taylor Swift's June 2024 shows, and higher readings for Oasis in August 2025, driven by synchronised jumping rather than volume.
Why do concerts show up on seismometers but a national jump would not?
Concentration and rhythm. A stadium packs about 70,000 people into one bowl, jumping repeatedly in time to music, so the waves add up near the sensor. A national jump puts about 445 people in an average square kilometre, once.
Sources
- Walpole S. C. et al.: The weight of nations, BMC Public Health 12:439 (2012). Source for the 57.7 kg Asian adult average and 62 kg global average.
- Rajšić A. et al.: Seismic efficiency for simple crater formation in the Martian top crust analog, JGR Planets (2021). Source for the 10⁻⁷ to 10⁻³ seismic efficiency bounds.
- Caplan-Auerbach J. et al.: Beast Quake (Taylor's Version), GSA Today (2024). Source for crowd size, frequencies, the 2.5× comparison and the −0.5 to 0.85 per-song magnitudes.
- US Geological Survey: Magnitude Types. Source for the energy relation log E = 11.8 + 1.5 M (ergs).
- Lockie R. G.: Percentile rank data for the countermovement vertical jump in law enforcement recruits, Sportlogia (2024). Source for the 52.63 cm mean (833 recruits), used as an upper reference.
- DT Next: India's population touches 1.46 billion, UNFPA State of World Population 2025 (2025). Source for the 1,463.9 million estimate.
- NIST CODATA: proton rms charge radius. Source for 0.84 × 10⁻¹⁵ m.
- Wikipedia: India. Source for the area of 3,287,263 km².
- NASA: Earth fact sheet. Source for the Earth's mass, 5.972 × 10²⁴ kg.
- Sharma L. et al., CSIR-NPL, Current Science (2018). Source for IST's 20 ns accuracy.
- CBS News: Taylor Swift concerts in Edinburgh caused seismic activity (2024). Source for 23.4 nm, 6.4 km and 80 kW.
- Anglo-Celt (PA): Oasis gig at Murrayfield more ground-shaking than 2009 (2025). Source for 261 kW and 70,000 a night.
- Wikipedia: Beast Quake. Source for the story: 8 January 2011, 66,336 fans, Lynch's 67-yard run and the magnitude 2.0 PNSN tremor.
