Skip to content
Tide Analysis

Tides, computed from the sea itself

Watch the tide at Indonesian ports taken apart into the waves it is made of, then predicted forward. Every number is computed here rather than copied from a table.

Open the record for Sabang, AcehHow this is computed
  • Observed the water level the gauge actually recorded
  • Predicted what the fitted constituents say it should be
  • Residual the difference — weather, wind and instrument trouble
Tide chart for Sabang, Aceh over 24 days: a black line for the recorded water level, a blue line over it computed from 8 constituents, and an ochre band beneath for the difference, on the same time axis. Residual RMS 0.105 m.7.508.008.509.0029/0404/0509/0514/0519/05-0.500.000.50Residual
ObservedPredictedResidual
Computed when this site was built, from the record at Sabang, Aceh · 24 days from the end of the fitted part · Residual RMS 0.105 m

Stations

8

Days of record

1686days

Hourly readings

38.818

No harmonic constant table ships as data. Every amplitude and phase on this site came out of a fit to the record displayed beside it.

What you can do

Five ways to interrogate one record

  • Record

    The record, the fit laid over it, and what the fit could not explain.

  • Constituents

    Amplitude and phase per constituent wave, with the nodal corrections applied.

  • Resolution

    How long you must watch to separate a pair of constituents — and what happens to a real record when the window shortens.

  • Comparison

    Least squares against the Admiralty method on the same record.

  • Prediction

    The fitted constants carried forward, with the high and low waters they imply.

How it works

How it works, in three steps

  1. 01

    The sea is recorded

    A gauge in a harbour writes down the water level every hour for seven months. That is the black line on the chart: as it came, weather and instrument trouble included.

  2. 02

    The record is taken apart

    A tide is a sum of steady waves driven by the Moon and the Sun. We work out how large each one is at that place, and how late it arrives.

    At Sabang, Aceh the largest of those waves turns out to be M2: 0.50 m tall, arriving 3.0 hours behind the Moon. Neither number existed anywhere until the record was fitted — the record page shows where both came from, row by row.

  3. 03

    What is left over is shown

    Whatever those waves cannot explain — weather, wind, surge — is not hidden. It is drawn as its own band beneath the chart.

Tidal character

Four ports, four characters, one physics

The Formzahl number F = (K1 + O1) / (M2 + S2) classifies tidal character, and the archipelago holds all four classes: Sabang has two nearly equal tides a day, Jakarta Port has one, and the rest sit between them. Every number below came out of a least-squares fit to that station's own record — none of it is quoted from a table.
Formzahl number and amplitudes per station
StationFTypeM2S2K1O1
Sabang, Aceh0.204Semidiurnal0.5090.2400.1060.047
Padang, Sumatera Barat0.394Mixed, mainly semidiurnal0.3580.1490.1230.077
Benoa, Bali0.413Mixed, mainly semidiurnal0.6430.3660.2530.164
Bitung, Sulawesi Utara0.578Mixed, mainly semidiurnal0.3410.2250.2080.120
Ambon, Maluku0.811Mixed, mainly semidiurnal0.5050.1530.3070.227
Surabaya, Jawa Timur1.242Mixed, mainly semidiurnal0.3770.2240.4750.272
Semarang, Jawa Tengah1.597Mixed, mainly diurnal0.0950.0760.2120.061
Kolinamil, Pelabuhan Jakarta3.486Diurnal0.0650.0490.2540.144
0.251.53
semidiurnalmixed → semimixed → diurnaldiurnal

Records

Stations

Every record carries its source, licence, period and datum. Nothing is assumed to be referenced to MSL.
  • Sabang, Aceh

    SemidiurnalF 0.20

    01 Jan 2026 — 01 Agu 2026 · 212 days · Datum ioc-sensor-zero

  • Padang, Sumatera Barat

    Mixed, mainly semidiurnalF 0.39

    01 Jan 2026 — 01 Agu 2026 · 212 days · Datum ioc-sensor-zero

  • Benoa, Bali

    Mixed, mainly semidiurnalF 0.41

    01 Jan 2026 — 01 Agu 2026 · 212 days · Datum ioc-sensor-zero

  • Bitung, Sulawesi Utara

    Mixed, mainly semidiurnalF 0.58

    01 Jan 2026 — 01 Agu 2026 · 212 days · Datum ioc-sensor-zero

  • Ambon, Maluku

    Mixed, mainly semidiurnalF 0.81

    01 Jan 2026 — 01 Agu 2026 · 212 days · Datum ioc-sensor-zero

  • Surabaya, Jawa Timur

    Mixed, mainly semidiurnalF 1.24

    01 Jan 2026 — 01 Agu 2026 · 212 days · Datum ioc-sensor-zero

  • Semarang, Jawa Tengah

    Mixed, mainly diurnalF 1.60

    01 Jan 2026 — 01 Agu 2026 · 212 days · Datum ioc-sensor-zero

  • Kolinamil, Pelabuhan Jakarta

    DiurnalF 3.49

    10 Jan 2026 — 31 Jul 2026 · 202 days · Datum ioc-sensor-zero

Licensing

Sources still behind the gate

The adapters for these are written and tested, but the licence gate keeps them shut until their terms are verified. None of their data is on this site.
  • unverifiedUniversity of Hawaii Sea Level CenterPortal UHSLC melampirkan syarat atribusi per negara yang ditetapkan operator pemilik alat; untuk stasiun Indonesia operator itu BIG, dan syaratnya belum diverifikasi. Adapter tersedia, gerbang lisensi menutupnya.
  • unverifiedBadan Informasi GeospasialSyarat redistribusi rekaman stasiun maupun produk Model Pasut belum diverifikasi. Tidak ada data BIG dalam repositori ini.