How this calendar is computed
Panchangam values are calculated from astronomical positions. Before a release, selected reference dates and times are checked against Drik Panchang. The build fails when a tested name differs or a tested time drifts by more than three minutes. These checks catch regressions; they do not prove that every festival convention agrees with every printed almanac.
What computes what?
| Quantity | Source | Notes |
|---|---|---|
| Malayalam date | kollavarsham | Kollavarsham solar reckoning. Chingam 1 opens the year. |
| Thithi, nakshathram, yogam, karanam | mhah-panchang | Lahiri ayanamsa, roughly 24°13′ for this decade. |
| Sunrise and sunset | suncalc | Upper limb with refraction, the standard −0.833° altitude. |
| Rahu, Gulika, Yamagandam | Derived | Eighth parts of the span between sunrise and sunset, assigned by weekday. |
| UK bank holidays | gov.uk | Kept separate for England & Wales, Scotland and Northern Ireland, because they differ. |
Which conventions were chosen, and why?
- The day runs sunrise to sunrise, not midnight to midnight. The headline thithi and nakshathram are the ones in force at sunrise, which is the Kerala convention.
- Handovers are shown, not hidden. “Dwitiya until 08:04, then Tritiya” is what a printed calendar gives you, and the changeover time is usually the thing you actually need.
- Mahashivarathri is fixed by nishita — the middle of the coming night — not by the thithi at sunrise. In 2026 those two rules fall on different days.
- Kerala festivals are computed at Kochi and shown on the same date everywhere. Onam does not move because you are in Birmingham.
How is Navarathri selected?
Navarathri belongs to the regular lunar month of Ashwinam, not simply any Shukla Prathama in the Malayalam solar month Kanni. An intercalary lunar month made that shortcut wrong in 2020. The opening, Durgashtami, Mahanavami and Vidyarambham dates are now checked against the Kochi Malayalam calendar for each year from 2020 to 2035.
For Vidyarambham, the calculation selects the day with the greatest Dashami overlap in the first fifth of daylight at Kochi. This is an implemented convention validated against those published dates, not a claim that all temples use that rule. A thithi touching sunrise for only a few minutes does not automatically choose the date.
Pooja Vaippu is separate from Durgashtami and uses Ashtami at sunset. When Ashtami touches neither sunset, as in 2022, almanacs can disagree: the calendar explains the uncertainty instead of inventing one settled Pooja Vaippu date. Confirm the ceremony with your temple. Government holiday dates and religious observance dates can also differ.
References: Drik Panchang’s Kochi Malayalam calendar;Kerala Government calendar 2026;the 2022 Pooja Vaippu convention disagreement. Navarathri reference checks expanded in September 2026.
Where can this be wrong?
Being straight about this matters more than looking authoritative:
- Islamic dates are expected, not settled. They are generated from the Umm al-Qura tabular calendar. The observed date is decided by moon sighting and can move a day either way, and UK mosques and Kerala do not always agree. Anything on this site marked “subject to moon sighting” should be confirmed locally.
- Sunrise can differ by a minute or two from another published source, depending on the horizon altitude assumed and the exact coordinates used for a city.
- Elsewhere in the UK is not London. Glasgow's sunrise differs from London's by enough to matter. Use the “Use my location” button on any day page to recompute where you actually are.
- Printed almanacs sometimes disagree with each other on festival dates where a thithi straddles a sunrise. Where that happens, a printed Kerala calendar is the better authority.
Found something wrong?
Tell me and I will check it against Drik and a printed calendar, and either fix it or explain why it stands. Contact Girish Kaitholil.