Skip to content

Home

Jump to bottom
luna-garcia edited this page Jan 29, 2026 · 1 revision

System overview and documentation objectives

What is this infrastructure?

This documentation describes the operation of the infrastructure used in the laboratory for metadata management and data ingestion using the SensorThings API. The system enables:

  • The structured description of what is measured, where, when, and with which instruments.
  • The ingestion of observational data, including time series, images, and files.
  • The preservation of the scientific and operational context of the data, such as campaigns, dives, and resources.

The infrastructure is designed to work with oceanographic and environmental data, but the underlying model is extensible to other contexts.

Documentation structure

Why does this documentation exist?

The main objective of this wiki is to ensure the continuity of the system and the data, regardless of the people working in the laboratory at any given time. Specifically, this documentation aims to:

  • Facilitate that any member of the lab can:
    • correctly register metadata,
    • upload new data,
    • understand the structure of the system.
  • Ensure that data can continue to be ingested even when personnel changes.
  • Reduce errors during metadata and data ingestion processes.

This wiki should be understood as an operational manual, not as a theoretical document.

General system structure

The system is divided into two main blocks, which are related but fulfill different roles:

Metadata (MMAPI)

Metadata defines what exists in the system before any data is ingested. Among other elements, the following are registered:

  • stations
  • sensors
  • variables and units
  • resources
  • operations, activities, and dives

MMAPI is the single source of truth for metadata.

Data (SensorThings)

This block stores the actual observations:

  • time series (CTD, environmental sensors, etc.)
  • images and other files

Data can only be ingested once the corresponding metadata already exist.

General workflow

The correct and recommended workflow is always the following:

  1. Register metadata in MMAPI
  2. Export metadata to SensorThings
  3. Ingest data (observations or files)
  4. Verify that the data are accessible

Data should never be ingested before the corresponding metadata have been registered.

Clone this wiki locally