name: Copernicus Marine API FinOps
description: >-
  Copernicus Marine Service is an EU-funded public programme operated by
  Mercator Ocean International on behalf of the European Commission. API
  consumers incur zero direct cost for data access, toolbox usage, or tile
  visualization. All operational costs are borne by the EU Copernicus Programme
  budget. Consumer-side FinOps concerns are limited to indirect costs such as
  compute, storage, and data processing infrastructure used to work with the
  downloaded marine data.
specificationVersion: '1.0'
framework: FinOps Framework FOCUS-aligned
currency: EUR
costModel:
  type: Free EU Programme Service
  directCost: 0.00
  description: >-
    No subscription fees, per-request charges, data volume fees, or bandwidth
    charges are levied on API consumers. Copernicus Marine data is freely and
    openly licensed for use by any user worldwide, including for commercial
    purposes, under the Copernicus Data Policy.
spend:
  directAPISpend:
    amount: 0.00
    currency: EUR
    description: >-
      No cost to API consumers for any Copernicus Marine API service including
      the Toolbox, WMTS tile service, or CSW catalogue.
  indirectCosts:
    - category: Compute
      description: >-
        Processing infrastructure required to analyze, transform, or visualize
        the large-volume NetCDF, Zarr, and CSV files downloaded from the
        Copernicus Marine Data Store. Ocean datasets can be multi-gigabyte to
        terabyte scale depending on spatial and temporal extent.
      estimationMethod: >-
        Tag cloud compute costs associated with Copernicus Marine data
        processing workloads; track CPU/GPU hours for xarray and dask jobs
    - category: Data Storage
      description: >-
        Storage costs for caching downloaded ocean data products locally to
        avoid repeated downloads across workflows. NetCDF and Zarr files can
        be large depending on variable selection and geographic extent.
      estimationMethod: >-
        Track GB-months of storage allocated to Copernicus Marine datasets
        in chosen cloud or on-premise environment
    - category: Data Egress
      description: >-
        Cloud provider data transfer charges when Copernicus Marine data is
        downloaded into cloud workloads from public internet endpoints. Large
        subsets of global ocean products can involve significant transfer volume.
      estimationMethod: >-
        Monitor data-in transfer costs for downloads from data.marine.copernicus.eu
        and related endpoints in cloud provider billing
    - category: Engineering
      description: >-
        Development and maintenance effort for integrations using the
        copernicusmarine Python library or CLI, including handling dataset
        version updates, format changes, and product deprecations.
      estimationMethod: >-
        Track engineering hours attributed to Copernicus Marine integration
        and data pipeline maintenance
optimization:
  recommendations:
    - name: Use Subset API to Minimize Download Volume
      description: >-
        Use the copernicusmarine subset command with tight geographic bounding
        boxes, limited time ranges, and specific variable lists rather than
        downloading full global products. This reduces storage costs and
        processing time significantly.
      savingsType: Reduced storage and compute costs
    - name: Use open_dataset for Exploratory Analysis
      description: >-
        Use the open_dataset function for exploratory data analysis to lazy-load
        data remotely without downloading full files. Only materialize and store
        data subsets that are actually needed for production workloads.
      savingsType: Reduced storage costs and egress fees
    - name: Cache Zarr Datasets Locally for Repeated Access
      description: >-
        When running iterative analyses over the same ocean region and time
        period, download once in Zarr format to a cloud object store or local
        cache. Zarr's chunked format supports efficient re-reads of subsets
        without re-downloading.
      savingsType: Reduced repeat download costs and API load
    - name: Select ARCO Service Type Based on Access Pattern
      description: >-
        Use arco-geo-series service for broad spatial queries over short time
        windows, and arco-time-series service for long time series over small
        geographic areas. Choosing the right service reduces server-side
        processing time and improves download throughput.
      savingsType: Reduced compute and network transfer time
tags:
  - EU Programme
  - Free Tier
  - Open Data
  - Zero Direct Cost
  - Copernicus
  - Marine