If every vehicle in India honked at once, roughly 27 crore horns would pour out about 840 megawatts of sound while drawing around 16 gigawatts of electricity, and an average spot in India would sit near 89 dB for as long as the horns held. Sri Lanka would hear nothing. At a horn's pitch of about 335 to 435 Hz, still air absorbs about 2 dB per kilometre on top of normal spreading, so after the 48 km gap to Talaimannar the combined sound falls below the threshold of hearing. Retune every horn to a ship's 70 Hz whistle and the answer flips.
- About 27 crore horns on the road, each making about 3.1 W of sound from 60 W of electricity.
- 840 MW of sound in total, a sound power level of about 209 dB.
- 89 dB at an average spot in India, if the horns were spread evenly.
- 2.17 dB per km of air absorption at 400 Hz kills the sound before Sri Lanka: about −45 dB on the far shore.
- At 63 Hz absorption falls to 0.09 dB/km and the same horns would arrive at about 72 dB.
Indians honk more than most drivers on Earth, and a CSIR-National Physical Laboratory survey of 100 Delhi sites found road traffic and honking were the dominant noise sources everywhere it measured.
So here is the absurd version of a real problem, solved like an engineering exercise: count the horns, find the sound power of one, add them up, then push the sound across the Palk Strait using the standard model of how air absorbs sound.
Has any sound ever crossed an ocean?
Yes, once in recorded history on a grand scale, and it tells you what kind of sound travels. The 1883 eruption of Krakatoa was heard thousands of kilometres away, because most of its energy was at very low pitch.
At 10:02 am local time on 27 August 1883, Krakatoa, the volcanic island between Java and Sumatra, blew apart. The British ship Norham Castle was about 40 miles away; her captain wrote in his log that the blasts had burst the eardrums of more than half his crew. About 100 miles off, a pressure gauge at the Batavia gasworks jumped by more than 2.5 inches of mercury, which works out to over 172 decibels. On Rodrigues, an island near Mauritius about 3,000 miles away, people heard it as distant gunfire. Barometers around the world logged the pressure wave passing again and again, roughly every 34 hours for up to five days.
Nautilus (Aatish Bhatia): The sound so loud that it circled the Earth four times (2014)Hold that against a car horn. Krakatoa was heard about 4,800 km away. The question here is whether 27 crore horns together can manage 48 km.
How many horns does India have?
India had 36.59 crore vehicles registered on the Parivahan portal by March 2024, according to the Ministry of Road Transport and Highways data cited by Infomerics. That is a cumulative count of registrations, so it includes vehicles long scrapped.
I assume 75% are still on the road, about 27.4 crore, and test 60% to 90%. Every one of them must carry a horn by law.
| Symbol | Meaning | Value used | Range tested | Basis |
|---|---|---|---|---|
| N | Horns on the road | 2.74 × 10⁸ | 2.2 to 3.3 × 10⁸ | 36.59 crore registrations (Parivahan, Mar 2024) × 75% on road (my estimate) |
| L₇ | Horn level at 7 m | 100 dB(A) | 83 to 112 dB(A) | AIS-014 limits: 83 to 112 dB(A) for small two- and three-wheelers, 93 to 112 dB(A) for others |
| f | Horn pitch | 400 Hz | 335 to 435 Hz | Disc horn data sheets, e.g. Ring REH510 low note at 335 Hz |
| Pe | Electrical draw per horn | 60 W | fixed | 12 V × 5 A (Ring REH510 spec) |
| α | Air absorption | 2.17 dB/km | 1.7 to 2.8 dB/km | ISO 9613-1 at 20 °C, 70% relative humidity, computed |
| D | India to Sri Lanka | 48 km | fixed | Length of the Ram Setu shoal chain |
The Indian standard for horns specifies how sound is measured: on an open site, at 7 m in front of the vehicle, with the microphone 0.5 to 1.5 m above ground and background noise at least 10 dB(A) below the horn. Measured that way, horns on two- and three-wheelers of up to 7 kW must produce 83 to 112 dB(A), and all other vehicles 93 to 112 dB(A). That 7 m reference is what lets us convert a horn's rated level into watts of sound. The limit: these are type-approval ranges, not what any particular horn on the road actually emits.
Read the standard (free)How much sound power does one horn make?
About 3 watts of sound from about 60 watts of electricity. The theory is the inverse square law: a source of acoustic power \( W \) spreads its energy over a growing surface, so intensity falls with the square of distance.
A horn mounted near the road radiates into a half-space (the ground reflects the rest), so the surface is a hemisphere, \( 2\pi r^2 \).
- Sound power level from a measured level. For hemispherical spreading,
\[ L_W = L_p + 20\log_{10} r + 8 \]With \( L_p = 100 \) dB at \( r = 7 \) m:\[ L_W = 100 + 20\log_{10}7 + 8 = 100 + 16.9 + 8 = 124.9\ \text{dB re } 10^{-12}\ \text{W} \]
- Convert to watts.
\[ W = 10^{(L_W-120)/10}\ \text{W} = 10^{0.49} \approx 3.1\ \text{W} \]
- Efficiency. A disc horn draws 5 A at 12 V, 60 W, so it turns about \( 3.1/60 \approx 5\% \) of its electricity into sound. The rest heats the coil and diaphragm.

How loud is 27 crore horns together?
About 840 MW of sound power, a sound power level of 209 dB. Horns are not phase-locked to each other, so their sound adds incoherently: powers add, not pressures. That means N identical sources raise the level by \( 10\log_{10}N \).
- Total acoustic power.
\[ W_{tot} = N W = 2.74\times10^{8} \times 3.1\ \text{W} \approx 8.5\times10^{8}\ \text{W} = 850\ \text{MW} \]
- Total sound power level.
\[ L_{W,tot} = 124.9 + 10\log_{10}(2.74\times10^{8}) = 124.9 + 84.4 = 209.3\ \text{dB} \]
- Electrical load. \( 2.74\times10^{8}\times 60\ \text{W} \approx 16.5\ \text{GW} \), drawn from vehicle batteries, not the grid.
What you would actually hear at home
Nobody hears the national total. You hear the horns near you. To find the level at a typical spot, spread the horns evenly: 2.74 × 10⁸ over India's 3.287 million km² is about 83 per km².
Each one contributes intensity \( W e^{-ar}/(2\pi r^2) \) at distance \( r \), where \( a \) is the absorption rate. Summing rings of sources of width \( dr \):
The integral grows only logarithmically, so distant horns add little and air absorption cuts off the tail. With \( \sigma = 83\ \text{km}^{-2} \), nearest horn at \( r_0 \approx 55 \) m and absorption at 400 Hz, I integrated this numerically and got about 89 dB.
That is an idealised flat country with evenly spread vehicles and no buildings; real cities would be louder near the road and villages far quieter. For comparison, the NPL Delhi survey measured hourly averages of 66.5 to 84.7 dB(A) at its 100 sites on an ordinary day.
The team measured hourly equivalent noise levels at 100 Delhi sites (19 commercial, 9 industrial, 16 residential, 43 silence zones, 3 highways and 10 intersections) between 9 am and 8 pm, built a noise map and ran a public perception survey. Hourly levels ranged from 66.5 to 84.7 dB(A), with 90% of sites between 70 and 80 dB(A), well above India's ambient limits, even in silence zones. Road traffic noise with honking was the dominant source; the paper cites earlier work that honking adds 2 to 5 dB(A) to road traffic noise. The limit: daytime spot measurements, not continuous monitoring.
Read the paper (free)Why does air swallow a horn's sound?
Because oxygen and nitrogen molecules soak up sound energy as they vibrate, and they do it far more at high pitch than low. This is called molecular relaxation, and the international standard ISO 9613-1 gives the formulas for it as a function of frequency, temperature and humidity.
I coded the standard's equations and checked them against its published table: 2.79 dB/km at 500 Hz and 4.98 dB/km at 1 kHz at 20 °C and 70% humidity, matching the table's 2.80 and 4.98.
See the data as a table
| Atmospheric absorption (dB/km) | |
|---|---|
| 63 Hz | 0.089 |
| 125 Hz | 0.335 |
| 250 Hz | 1.124 |
| 335 Hz | 1.729 |
| 400 Hz | 2.173 |
| 500 Hz | 2.791 |
| 1 kHz | 4.978 |
| 2 kHz | 9.039 |
Source: Computed from ISO 9613-1:1993 equations (author's calculation), checked against the standard's table · Disc horns sit at about 335 to 435 Hz, where air removes roughly 2 dB every kilometre.
The standard gives an analytic formula for the pure-tone attenuation coefficient of air from 50 Hz to 10 kHz, for temperatures from −20 to +50 °C and relative humidity from 10% to 100%, based on the relaxation frequencies of oxygen and nitrogen. At 20 °C and 70% humidity it gives about 0.09 dB/km at 63 Hz, 2.80 at 500 Hz and 76.6 at 8 kHz. I used it to compute the 2.17 dB/km at 400 Hz in this post. The limit: it covers absorption in still, uniform air only; ground effects, wind and temperature gradients, which can matter more over water, are in Part 2.
Read a summary with the standard's table (free)Could Sri Lanka hear India honk?
No. The nearest horns would be 48 km away across the Palk Strait shoals, and over that distance spreading plus absorption takes the sound far below hearing. Here is the working, for a listener on the Sri Lankan shore facing an Indian coastline full of evenly spread horns beyond 48 km.
- One horn at 48 km. Start from 100 dB at 7 m, subtract spreading and absorption:
\[ L_p = 100 - 20\log_{10}\frac{48000}{7} - 2.17\times48 = 100 - 76.7 - 104.2 \approx -81\ \text{dB} \]
- All of India's coast and hinterland. Integrating the same \( e^{-ar}/r^2 \) contributions over a half-plane of horns starting 48 km away (83 per km²) gives about −45 dB. Absorption alone, \( e^{-a\cdot48\,\text{km}} \), cuts intensity by a factor of about 10¹⁰.
- Threshold of hearing. 0 dB is the faintest sound a young ear can detect in perfect silence. −45 dB is 30,000 times quieter than that in intensity.
A single horn, for reference, fades to 30 dB, about a quiet room, at roughly 5.6 km and to 0 dB at about 15 km in still air by the same formula, before buildings, trees and wind take their share. Riders who have pushed their horns or exhausts past the rules can read the legal side in motorcycle exhaust noise limits in India.

| Scenario (400 Hz unless noted) | Level at listener | Audible? |
|---|---|---|
| Average spot in India, all horns spread evenly | about 89 dB | Very loud |
| Sri Lankan shore, horns spread across India | about −45 dB | No |
| All 27 crore vehicles parked behind Dhanushkodi (2,744 km² at 10 m² each) | about −15 dB | No |
| All 27 crore horns at a single point 48 km away | about +3 dB | At the edge of hearing, in total silence |
| Same as row 2, but horns retuned to 63 Hz | about 72 dB | Yes, loudly |
What would it take to make Sri Lanka hear it?
Lower the pitch. Absorption falls steeply as pitch drops, so at 63 Hz air takes only 0.09 dB per km instead of 2.17. Rerunning the same half-plane integral at 63 Hz, with each horn still radiating 3.1 W, gives about 72 dB on the Sri Lankan shore.
Ships already solved this
The international collision regulations (COLREGs Annex III) require vessels of 200 m or more to sound whistles with a fundamental of 70 to 200 Hz, at 143 dB at 1 m, for an audibility range of about 2 nautical miles in still air.
The engineered version, and its price
So the engineered version of this what-if is: replace every disc horn in India with a low-frequency whistle near 70 Hz. The catch is the size of the radiator.
A 63 Hz sound wave is about 5.4 m long (343 m/s ÷ 63 Hz), and a small disc a few centimetres across radiates such long waves very inefficiently, which is why ship whistles are big.
My rough estimate, not a quoted price, is a few thousand rupees per vehicle for a compact low-pitch horn plus wiring, so ₹2,000 × 27 crore ≈ ₹55,000 crore, and the result would be a country that drowns itself in low rumble to be heard 48 km away.
Retune every horn in India to a ship's whistle and Sri Lanka hears it at 72 dB. Leave them at 400 Hz and the sound dies in the shoals.The working of this post
Nature has done it with far more power. Krakatoa, from the story above, was heard about 4,800 km away on Rodrigues island near Mauritius, and its pressure wave circled the globe more than three times, according to the Royal Society's 1888 report summarised on Wikipedia. Most of that energy was at very low frequencies, where air absorption barely bites.

As India's horns are built, impossible: at 335 to 435 Hz, ISO 9613-1 absorption of about 2.2 dB/km plus inverse-square spreading puts the nationwide honk at about −45 dB on the Sri Lankan shore, far below the 0 dB threshold. Retune every horn to a ship-whistle pitch near 63 to 70 Hz and the same 3 W per horn arrives at about 72 dB. The deciding number is the absorption coefficient: 2.17 dB/km at 400 Hz against 0.09 dB/km at 63 Hz.
How sensitive is the answer?
Not very, because absorption dominates. Every input except pitch changes the Sri Lanka figure by a few decibels at most:
- Horn count, 22 to 33 crore: ±1 dB (10 log of a factor of 1.2).
- Horn level, 90 to 110 dB(A) at 7 m: ±10 dB, still at least 35 dB below hearing.
- Hot, humid coastal air (30 °C, 70%): absorption at 400 Hz is 2.19 dB/km, essentially unchanged.
- Pitch: the only lever that matters. From 400 Hz to 63 Hz the absorption over 48 km drops from about 104 dB to about 4 dB.
Where the model breaks down
- Sea-surface layers: real sound over warm sea can be bent by temperature layers, which can carry sound further than still-air formulas predict.
- Ground and buildings: they would cut the "average spot" figure well below 89 dB in practice.
Neither rescues the 400 Hz horn across 48 km. If you want the cultural side of why we honk at all, see why Indians honk, and for what to do about a neighbour's horn, how to file a noise complaint.
Use this to check what you just read, with your own horn count, horn rating, pitch and distance.
National honk: power and how far it carries
What's new
- CSIR study builds a model that maps Delhi's noise from road network and real-time traffic ThePrint (PTI), Sep 2026
- India's railway noise problem and its poorly understood health impacts Mongabay India, Jul 2026
What if next?
- What if all 1.4 billion of us wanted chai at the same minute? Count the milk, the gas and the queue.
- What if every Indian wedding of the year happened on one day? The baraat bands join the horns.
- What if a head wobble could generate electricity? A nod has energy too; the question is how many it takes to charge a phone.
Questions people ask
How loud is a car horn in decibels?
Under India's AIS-014 standard, a horn measured 7 m in front of the vehicle must produce 93 to 112 dB(A) for cars and larger vehicles, and 83 to 112 dB(A) for two- and three-wheelers of up to 7 kW. Aftermarket disc horns are often rated 105 to 118 dB.
How far can a car horn be heard?
In still air with no obstacles, a 100 dB(A) horn at 7 m falls to about 30 dB, a quiet room, at around 5.6 km and to the threshold of hearing at about 15 km, using inverse-square spreading and ISO 9613-1 absorption at 400 Hz. Traffic noise, buildings and wind cut that to a few hundred metres in a city.
Why do low sounds travel further than high sounds?
Air absorbs sound by molecular relaxation in oxygen and nitrogen, which is much stronger at high frequencies. At 20 °C and 70% humidity, ISO 9613-1 gives about 0.09 dB/km at 63 Hz, 2.8 dB/km at 500 Hz and 9 dB/km at 2 kHz. That is why ships use low whistles and thunder rumbles from far away.
Is honking illegal in India?
Honking itself is not illegal, but horns must meet AIS-014 limits, and Rule 119 of the Central Motor Vehicles Rules bans multi-tone horns and devices making an unduly harsh, shrill, loud or alarming noise.
Sources
- ARAI: AIS-014 Horn Installation Requirement, with amendments. Source for the 7 m method and 83/93 to 112 dB(A) limits.
- Chauhan B. S. et al.: Evaluation and analysis of environmental noise levels in NCT of Delhi, MAPAN 38(2) (2023). Source for 66.5 to 84.7 dB(A) and the honking finding.
- Wikibooks: Outdoor sound propagation, reproducing the ISO 9613-1 absorption table. Source for the check values 2.80 and 4.98 dB/km.
- ISO 9613-1:1993, Calculation of the absorption of sound by the atmosphere. Equations used for all absorption figures.
- Ring Automotive: REH510 disc horn specification. Source for 12 V, 5 A, 335 Hz and 105 to 118 dB.
- Infomerics: Automobile industry report, citing the MoRTH Parivahan portal (2024). Source for 36.59 crore registrations.
- COLREGs Annex III: Technical details of sound signal appliances. Source for 70 to 200 Hz, 143 dB and 2 nautical miles.
- Wikipedia: Ram Setu. Source for the 48 km length.
- Nautilus (Aatish Bhatia): The sound so loud that it circled the Earth four times (2014). Source for the story: 10:02 am, the Norham Castle log, the Batavia gauge and 172 dB, and the 34-hour pressure waves.
- Wikipedia: 1883 eruption of Krakatoa, citing the Royal Society's 1888 report. Source for the 4,800 km audibility.
- Wikipedia: India. Source for the area of 3,287,263 km².
