Measurements · v2026.09.10 · one name per quantity across datasets

89 measurements. 94 series. 5 datasets. 320 M values, 1949 → 2026.

Everything CalCOFI has measured in the water and the air above it — temperature and salinity from the bottle casts of 1949 to this year's CTD profiles, nutrients, carbon, pigments and light, the underway weather, the picoplankton counted by flow cytometry — keyed to one name, so the same measurement means the same thing in every dataset. Start from the overview below, search, filter by category, or start from a dataset; every measurement has its own page.

A measurement is one quantity under one name; a series is one dataset's column for it. Two datasets' series count as one measurement only where they carry the same NERC P01 concept for the same kind of sample — 5 do today: Temperature, Potential density (σθ), Salinity, Dissolved oxygen (ml/L), Dissolved oxygen (µmol/kg) — so 94 series make 89 measurements. 50 of the 89 carry a NERC P01 concept; 39 carry none and keep their own names, because an empty concept means no concept says exactly this, never that nobody looked.

The machine-readable form is one JSON per measurement (/measurements/{key}.json) and sitemap.xml.

89
measurements: 94
series: 5
datasets: 29 M
values · release grain: 320 M
with full resolution: 50
with a NERC concept

Which measurements, in which datasets

measurements · category × dataset

The datasets

series · years · values

Category is the registry's (metadata/category.csv, the same the Explorer folds its picker by); Picoplankton & Bacteria is a biology category whose cell counts are measurements — rows of obs_env, not organisms with a taxon — so they are listed here and nowhere in the species catalog. Dataset colour is never the only cue: the name is always beside the dot, and the matrix is one sequential ramp.

Which measurements, in which datasets, in which years

One row per measurement, one bar per dataset that carries it, 1949 on the left and the release year on the right; hover a bar for the dataset, its series, years, values, sampling events and depths. Two bars in a row are two datasets measuring the same thing — the bottle's temperature and the CTD's temperature_ave, comparable across datasets — never a dataset's own bottle samples beside its sensor (the CTD files' btl_* series stay their own rows, so the bottle dataset is not counted twice). The values column is the release grain (obs_env); the CTD and METS full-resolution tables carry more behind the same names.