Every June, a government agency in New Delhi announces a date. Not a forecast of the season, not a range, but a single day on which the southwest monsoon has arrived over the Kerala coast. Newspapers print it as a headline. Commodity traders adjust positions on it. Farmers in the interior, hundreds of kilometres north, read it as the starting gun for a schedule they will follow for the next four months.

That announcement is the clearest illustration of how this pillar works. The monsoon is treated locally not as weather but as a calendar entry — an event with a date, a normal date, and a known habit of being a few days early or late. Once you hold it that way, a great deal of Indian economic and social life stops looking like custom and starts looking like scheduling.

A date, not a season

The onset is defined operationally, which is why it can be announced at all. The India Meteorological Department uses a fixed panel of fourteen rain gauges across Kerala. If at least sixty per cent of them record 2.5 mm or more of rain on two consecutive days, and the wind field at lower levels has turned to the west or southwest as expected, the monsoon is declared to have arrived. The normal date is 1 June. The declaration in any given year can be a week early or a week late without anyone considering it abnormal.

This is worth pausing on. A national-scale phenomenon — one that waters an entire subcontinent — is certified by fourteen instruments in one state. Kerala is used because it is where the current first makes landfall, not because Kerala matters more. The rest of the country then waits for the same wind system to advance north-west, and the progress of that advance is tracked and published in stages. So the country does not experience one monsoon; it experiences a queue, and each region has its own place in it.

Twelve months with rainfall, the two grain calendars, onion prices and the wedding season aligned to the same axis ONE AXIS, FOUR SCHEDULES JFM AMJ JAS OND rainfall about three-quarters of the year rice sow standing water harvest wheat sow cool, dry growth cut onion price peak after the new crop is still in store weddings main season tail second window the link Rain decides when the field is under water, so it decides the sowing date. The sowing date decides the harvest date, which decides when cash arrives. Cash and dry ground together decide when a family can host a wedding. The two grain calendars never overlap: rice needs standing water in the months the wheat field must stay dry. This is why a single date in June can be quoted as the start of the agricultural year across half the country. Rainfall bars are schematic; the onion curve is a typical seasonal shape, not a forecast.
The same twelve months, read four ways. The upper two tracks are hydrology, the lower two are the schedules that hydrology sets.

The crop calendar is a bet placed on that date

Rice is the crop that makes the wager explicit. It wants standing water. That single requirement means the field cannot be prepared, the nursery cannot be transplanted and the season cannot begin until the rain has actually arrived and the ground is saturated. In the deltaic districts of coastal Kerala and the Kaveri belt of Tamil Nadu, the sowing window opens within days of the declared onset. If the declaration comes late, the whole chain shifts.

Wheat makes the opposite bet on the same date. It is grown through the cool, dry months — sown after the monsoon has withdrawn in October and November, cut in March and April before the heat arrives. Too much rain at sowing time is a problem, not a blessing: it waterlogs the seedbed. So wheat farmers are not waiting for the monsoon to start. They are waiting for it to finish, and for the soil to hold the right amount of residual moisture without being wet.

This is why the two grain calendars in the diagram never overlap, and why the same four months — June to September — determine the income of two entirely different agricultural economies. One is waiting for water to arrive. The other is waiting for it to leave, and its success depends on how much water the first one left behind.

Prices move on the same axis

Onions are the standard demonstration because they store badly and everyone buys them. Harvested in the kharif and late-kharif seasons, held in ventilated sheds through the winter, they begin to spoil as the temperature climbs in April and May. So the price curve has a predictable shape: it falls when the new crop lands, sits low while storage is full, and climbs through the late spring until the next harvest breaks it.

What connects that curve to the monsoon is not the harvest itself but the timing of it. A late onset pushes the sowing date back, which pushes the harvest back, which shortens the interval before the store runs down — and the price spike arrives sooner and goes higher. This is not a metaphor about markets feeling the weather. It is a description of a physical stock being depleted on a fixed schedule that someone else's rainfall has just moved.

Weddings are scheduled around it too

The Indian wedding season is usually explained as astrology: auspicious dates, favourable planetary positions, an almanac consulted months in advance. That is real, and it matters. But the almanac is consulted within a window that is set by other things entirely — the ground has to be dry enough for a marquee and for guests to travel, the family has to have cash from the harvest, and the guests have to be free from field work.

Which produces the pattern in the bottom track: a heavy season from November to February, a smaller window in April and May, and almost nothing during the monsoon itself. Every one of those three conditions is a consequence of the date announced in June. Astrology chooses the day inside the window; hydrology draws the window.

What changes when you read it this way

Reading the monsoon as a calendar rather than a climate yields predictions that the climate framing does not. It predicts that a late onset will show up as a price movement before it shows up as a crop failure. It predicts that wedding venues are a seasonal business with a hard annual peak, not a year-round one. It predicts that irrigation is valuable precisely because it converts a date with a week's variance into a date with none.

And it explains a habit that otherwise looks like fatalism. When a farmer says the rain is late, the statement is not resignation about the weather. It is a report that a schedule has slipped — and that everything downstream of it, from the nursery to the repayment date on a loan, has slipped with it.

Sources and checks

  1. India Meteorological Department End-of-Season Report, Southwest Monsoon 2023, and the Criteria for Onset over Kerala (14-station 60% rule, 2.5 mm threshold, upper-air wind criteria)
  2. IMD Long Period Average rainfall 1961–2010: 868.6 mm for the country as a whole
  3. Ministry of Agriculture & Farmers Welfare Agricultural Statistics at a Glance, land use and net sown area series; rain-fed share of net sown area
  4. Gadgil, Sulochana "The Indian Monsoon and Its Variability", Annual Review of Earth and Planetary Sciences 31, 2003
  5. Census of India 2011 Primary Census Abstract, tables on cultivators and agricultural labourers by state

Last checked 30 October 2026 Yellowsea Huang Corrections