Skip to content
gnsscalc

Campaign processing

How does a station move? Drop daily observation files — or name a station in the IGS/EUREF archives — and each day is solved with precise point positioning, then fitted into a time series with a velocity, seasonal motion, jumps and outliers. Everything runs in your browser.

Stations and days

From the IGS and EUREF archives, solved day by day in your browser.

WTZR00DEU
One day every

24 days × 1 station = 24 station-days, about 120 MB to download. For a velocity, span two years or more — shorter series let seasonal motion leak into it.

Processing
Ocean tide loading: automatic

Near tidal coasts the crust moves by centimetres under the ocean tides. It averages out of a full day, but in sessions of a few hours it biases the height by up to a few centimetres (Brest, 2 h: 48 mm → 16 mm scatter). By default it is applied everywhere from our own global grid: the GOT4.10c ocean tide model (NASA) loaded onto an elastic Earth, typically within a millimetre of Onsala's coefficients.

To use a specific model instead, get your station's coefficients in BLQ format from the free Onsala ocean loading service (model FES2014b, centre-of-mass correction on) and paste them here; the station name must match the file's marker (first 4 characters). A BLQ entry takes precedence over the grid.

Advanced settings: recommendedconstellations, satellites, cut-off, signals, products, engine
Measured on the EPN benchmark: 24 h static, East/North/Up RMS in mm against the EUREF solution.
Constellations
Excluded satellitesLeft out for the whole session — a satellite with a known problem on the day.
Elevation cut-offLow satellites help only with the VMF3 weather-model troposphere: Up 3.7 → 2.6 mm float at 3°; with GPT3 they make it worse.
Signal pair per constellation

RINEX input. The precise clocks refer to the recommended pairs; another pair carries the satellites’ inter-frequency biases into the ambiguities. Antenna offsets follow the pair’s own bands where the calibration has them.

Expert

Settings apply from the next Process; days already solved with other settings are solved again.

Or use your own files: drop daily RINEX (2/3/4, Hatanaka, gzip) here or click to choose — any stations, any days, grouped by marker name. Receiver logs: convert them to RINEX first.

Campaign processing FAQ

What does campaign processing do?
It turns many days of observations from one station into a coordinate time series: each day is solved with static Precise Point Positioning, then an offset, a velocity, seasonal terms and jumps are fitted and outlier days are left out. The result is how the station moves — its velocity in ITRF2020 and relative to its tectonic plate — and how stable it is.
How accurate is it?
Checked against EUREF's published multi-year solution (EPN C2415, in IGS20 and ETRF2020) for six EPN stations — HOFN, ONSA, WTZR, JOZE, MATE and TRO1 — with daily solutions every 10 days from November 2021 to October 2026: horizontal velocities agree within 0.25 mm/yr (one station 0.5 mm/yr North), vertical ones within 1–2 mm/yr — EUREF's velocity averages each station's whole history since 1996, these five years. Day-to-day scatter is about 2–3 mm horizontally and 5–6 mm vertically. That suits teaching, CORS operators, surveyors and deformation monitoring; official reference-frame solutions (EUREF, IGS) use network software and more modelling.
How much data do I need?
Full days (24 h) give the best daily positions. For a velocity, span at least two years — shorter series let seasonal motion leak into the trend. Monthly or weekly days are enough for a velocity; daily data shows short-term motion and events.