Skip to content

Datasets

A Dataset describes one or more related data files produced by an experiment, and the variables in them.

What is a dataset?

A dataset typically corresponds to a single data file (e.g., one CSV of CTD profiles), but it may include multiple files when they share the same platform, instruments, and data submitter. The key organizing principle is that a dataset represents data collected on a single platform using a consistent set of instruments, managed by a single data submitter. If data come from different platforms or instrument configurations, they should be separate datasets.

Dataset Types

  • FieldDataset: data collected during in-situ experiments, such as CTD casts, bottle samples and sensor deployments. It records the platform that collected the data, using the NERC L06 platform types (see Platform).
  • ModelOutputDataset: output from a model experiment. How the output was produced is described here, through the simulation type, period, output frequency and hardware.

Rules

These apply on top of each field's own requirements:

  • Data access (both dataset types): open_access is for data you can reach today and requires data_access_link. Use scheduled_access for anything not openly available yet; it requires data_access_date, the date the data will open. data_access_link is optional for scheduled and conditional access.
  • Perturbation runs (model output): simulation_type can list several values. If it includes perturbation, describe the forcing in mcdr_forcing_description.

Linking Datasets to Experiments

Each dataset names the experiment it belongs to with experiment_id, so one experiment can have several datasets, such as CTD profiles, bottle samples and sensor time series.

Variables

Each dataset lists its variables, metadata for each measurement or column in its files. Field datasets use the field variable classes, with specific classes for pH, total alkalinity, DIC, CO₂ and other common measurements. Model output datasets use ModelOutputVariable. See Variables for how to choose a class.