Appearance
2.5 — The Monsoon
Around the first of June, in an ordinary year, rain reaches the coast of Kerala. Over the following six weeks it works its way north and west across the subcontinent, reaching Mumbai in about the second week of June, central India by late June, Delhi at the very end of June or the start of July, and Rajasthan and Punjab in early July. Then it sits over the country until September, when it withdraws in the reverse order.
That rain is about 70 to 80 percent of India's annual precipitation, delivered in about four months. Around half of India's farmland has no irrigation and depends on it entirely, and even irrigated land depends on it indirectly, since the monsoon is what refills the reservoirs and recharges the groundwater.
A monsoon that is 10 percent below normal is officially a drought year. Historically, back-to-back failures produced famine. No other weather system on Earth is watched this closely by this many people, and the machinery behind it is more interesting than the version taught in schools.
The word, and the incomplete school explanation
"Monsoon" comes from the Arabic mausim, meaning a season, by way of Portuguese sailors who used it for the seasonal winds that governed voyages across the Arabian Sea. It refers to the wind reversal, not to the rain — the rain is a consequence.
The classical explanation goes: land heats faster than sea in summer, so a low-pressure area forms over the Indian landmass, and moist air is drawn in from the ocean, giving rain. In winter the land cools faster, the flow reverses, and dry air blows offshore. It is a giant sea breeze on a continental scale.
That is not wrong, but it is not sufficient, and the reasons matter.
If differential heating of land and sea were the whole story, every large continent at that latitude would have a monsoon of comparable strength. They do not. Africa and Australia have monsoons, but nothing on the Indian scale.
Simple heating also fails to explain the abruptness. The monsoon does not creep in as the land gradually warms; it arrives. Rainfall at a given station can go from almost nothing to torrential within days. That is the signature of a system that switches state, not one that drifts.
And it does not explain why the monsoon fails in some years and floods in others, when the land–sea temperature contrast is broadly similar.
What is actually going on
The modern understanding treats the monsoon as the seasonal migration of the whole tropical rain belt, powered by a heat source raised into the middle of the atmosphere, and locked in place by the mountains. Four components.
1. The tropical rain belt moves with the Sun
Chapter 2.2 introduced the equatorial belt where air rises and rains — the Intertropical Convergence Zone, where the trade winds of the two hemispheres meet. It sits where the Sun is most directly overhead, and since that point moves north and south through the year, the rain belt moves with it, lagging by about a month.
Over the ocean it moves only a few degrees. Over land it moves enormously, because land heats so much more responsively. By July it has migrated north to sit over the Gangetic plain, around 25 degrees north. The monsoon is, first and most simply, the tropical rain belt visiting India.
And this is why the winds reverse rather than merely strengthen. The southeast trade winds of the southern hemisphere, on crossing the equator, encounter a Coriolis effect that now deflects to the right instead of the left. A wind from the southeast, turned right, becomes a wind from the southwest. The southwest monsoon of India is the southern hemisphere's southeast trades, bent by the change of hemisphere.
2. The Tibetan Plateau is a heat source in mid-air
This is the piece the school version leaves out, and it is what makes the South Asian monsoon the strongest on Earth.
The Tibetan Plateau is an area roughly the size of Western Europe at an average altitude of about 4,500 metres (Chapter 1.8). In summer its dark rocky surface absorbs intense sunlight at an altitude where the air is thin and there is little atmosphere above to intercept the radiation.
So it heats the air directly at a height of 4 to 5 kilometres. No other large heat source on the planet is elevated like that.
Why elevation matters so much. Heating the atmosphere at height, rather than at sea level, warms an entire column of air aloft. Warm air is less dense, so the pressure surfaces above the plateau bulge upward, creating a large high-pressure region at altitude over Tibet and a strong pressure difference between that and the air over the Indian Ocean at the same height. That difference drives a powerful flow aloft from Asia toward the ocean, which must be balanced by a return flow at low level from the ocean toward the land. That low-level return flow is the monsoon current, and it is far stronger than land–sea contrast alone could produce.
The evidence is direct: computer models with the plateau removed produce a much weaker monsoon, and the geological record shows the monsoon strengthening as the plateau rose. The Himalaya made the monsoon, which means Indian agriculture, Indian population density and a great deal of Indian history are consequences of a continental collision fifty million years ago.
3. The mountains also act as a wall
The Himalaya blocks cold dry Central Asian air from reaching the subcontinent. Compare northern India with places at the same latitude elsewhere — the Sahara, Arabia, northern Mexico — all of which sit under the descending branch of the Hadley cell and are desert. Northern India is not, because the mountain wall keeps the dry descending air out and the plateau pulls the wet flow in.
The wall also wrings the rain out. Air rising over the Western Ghats gives the Konkan and Malabar coasts several thousand millimetres a year while the Deccan interior, only 100 kilometres inland in the rain shadow, gets a fraction of it. Air funnelled up against the Khasi hills gives Mawsynram and Cherrapunji their extraordinary totals.
4. Latent heat makes it self-sustaining
Once rain begins, condensation releases latent heat into the atmosphere over India (Chapter 2.3). That heating strengthens the very pressure pattern that is drawing the moist air in. The monsoon feeds itself, which is why it switches on sharply and why it holds through cloudy periods when the ground is receiving little sunlight.
The two branches, and the sequence of arrival
The current splits around the Indian peninsula.
The Arabian Sea branch is the stronger. It strikes the Western Ghats, dumps most of its water on the windward slopes, crosses the drier Deccan, and part of it continues up into Gujarat, Rajasthan and the Gangetic plain.
The Bay of Bengal branch moves up the bay, hits the Arakan hills of Myanmar and the hills of the northeast, and is then deflected westward by the Himalaya to run back up the Gangetic plain. Because it is being deflected rather than blocked, rainfall along the plain decreases as you move west — Kolkata far more than Patna, Patna more than Delhi, Delhi more than Amritsar. That gradient is the reason eastern India grows rice and the western plains historically grew wheat and millets.
The two branches merge over the plain, and the whole system is organised into travelling low-pressure disturbances that move west-northwest along the monsoon trough and deliver the actual heavy rain days.
Active and break spells. The monsoon does not rain continuously. It comes in active spells of heavy rain lasting days to a fortnight, separated by break spells when the trough shifts north against the Himalaya and the plains go dry and hot. Long breaks in July or August, at the wrong point in the crop cycle, can ruin a season even when the seasonal total looks adequate. This is why the seasonal rainfall figure is a poor measure of how good a monsoon was for farmers, and why the India Meteorological Department now issues distribution forecasts rather than just a percentage.
The northeast monsoon, which most people forget
When the system retreats from October, the flow reverses to northeasterly — dry continental air. But as it crosses the Bay of Bengal it picks up moisture, and it delivers that to the southeastern coast.
Tamil Nadu, coastal Andhra Pradesh and southern Karnataka get most of their annual rain in October to December, not in the summer monsoon. Chennai's wettest months are October and November. This is why Tamil Nadu can be in drought while the rest of the country has had a good monsoon, and why the two halves of the peninsula run on different agricultural calendars.
What makes a monsoon fail
El Niño is the strongest known influence. Chapter 2.6 covers the mechanism in full; the short version is that in an El Niño year, the warmest water in the Pacific shifts eastward, and the great atmospheric circulation that normally has rising air over the Indonesian region and sinking air over the eastern Pacific weakens or shifts. Sinking air over the Indian and western Pacific region suppresses monsoon convection.
The correlation is real but it is not a rule. Historically, a majority of India's severe drought years have been El Niño years, but a substantial number of El Niño years have produced normal monsoons — 1997 had one of the strongest El Niños on record and a normal Indian monsoon. Anyone stating that El Niño means drought in India is overstating the evidence, and this book will not.
The Indian Ocean Dipole is a second influence, and it can offset El Niño. It is the difference in sea surface temperature between the western and eastern tropical Indian Ocean. A positive dipole — warm in the west near Africa, cool near Sumatra — tends to strengthen the monsoon; a negative one weakens it. The strong positive dipole in 1997 is the accepted explanation for why that El Niño year was not a drought.
Other influences with real but smaller effects include Eurasian snow cover in winter and spring, which affects how much heating the plateau can deliver, and shorter-term oscillations that govern the active and break cycle.
Forecast skill, stated honestly. The India Meteorological Department has issued seasonal forecasts since 1886. Skill for the all-India seasonal total is modest but real and has improved with dynamical models; skill for regional distribution and for the timing of breaks — which is what a farmer actually needs — remains low. The monsoon is genuinely one of the hardest forecasting problems in the world, and pretending otherwise has been a recurring failure of Indian science communication.
The monsoon in Indian history
This is not an aside. It is one of the threads that runs through Part 6.
Monsoon failure and famine. The Bengal famine of 1770, the great famines of 1876–78 and 1896–1900, and the Bengal famine of 1943 all had drought or crop failure among their causes — though in every case, as Chapter 6.17 shows in detail, the death toll was determined at least as much by policy, grain trade, hoarding and administrative choices as by rainfall. A drought becomes a famine through human decisions, a point Amartya Sen's work made rigorously and which Part 9 develops.
Monsoon and the economy. Well into the 1990s the Indian growth rate correlated visibly with rainfall, and budget forecasts were adjusted to the monsoon. That dependence has fallen substantially — agriculture is now a much smaller share of output, irrigation covers more land, and there are grain buffer stocks — but it has not gone. A bad monsoon still moves rural demand, food inflation, and therefore interest rates.
Monsoon and trade. The reversing winds were the basis of Indian Ocean commerce for two thousand years. Sail west to east in one season, east to west in the other. The Greek and Roman trade with India ran on this, as did Arab and later Gujarati and Malabar shipping. The knowledge of the wind reversal is what made the Indian Ocean the world's busiest trading system long before the Atlantic mattered at all, and Chapter 8.4 explains what happened to it when European ships arrived.
Monsoon and culture. The arrival of the rains is embedded in Indian music, poetry and festival calendars — the raga Megh Malhar, the Kalidasa poem Meghaduta in which a cloud carries a message, the Teej and Onam festivals timed to the season. Chapter 15.2 takes that up.
Where this shows up in your life
Food prices. Onion, tomato and pulse prices in Indian cities track monsoon performance with a lag of a few months, and food is a large share of the consumer price index, so the monsoon still moves the headline inflation number.
Urban flooding is not a monsoon problem, it is a drainage problem. Mumbai in 2005, Chennai in 2015, Bengaluru repeatedly, Delhi in 2023 — in each case the rainfall was extreme but the flooding was made far worse by lakes and wetlands built over, storm drains blocked or undersized, and floodplains occupied. The water arrived where it always used to go and found buildings there.
Personal planning. The onset dates above are averages with a standard deviation of about a week, so any single year can vary. The IMD publishes onset and withdrawal in real time, and its "monsoon onset over Kerala" declaration has specific technical criteria involving rainfall at a set of stations, wind depth and outgoing longwave radiation — it is not a judgement call.
What the next page covers
The monsoon is influenced from the Pacific, and El Niño was named twice in this chapter without being explained. Chapter 2.6 covers the ocean in motion — the great surface current gyres and why they exist, the Gulf Stream and what it actually does for Europe, upwelling and why a few small patches of ocean feed the world, the deep circulation that turns over the whole ocean on a thousand-year cycle, and El Niño in full: what it is, why it happens, and how a change in the Pacific wind reaches a farm in Maharashtra.