FDO2map

Resolve a FAIR Digital Object (FDO) and open it in an interactive map dashboard — automatically.

Given only an RO-Crate persistent identifier (PID), fdo2map discovers the dataset and visualization dashboard from the RO-Crate annotations and opens it in the browser. No prior knowledge of the data or dashboard is needed.

How it works

RO-Crate PID
    │
    ▼
┌─────────────────────────┐
│  1. Load from ROHub     │
│  2. Find ViewAction     │
│  3. Resolve dataset URL │
│  4. Apply URL template  │
│  5. Open browser        │
└─────────────────────────┘
    │
    ▼
Interactive map dashboard

The script uses schema.org annotations embedded in the RO-Crate:

• schema:potentialAction — links a dataset to a ViewAction
• schema:ViewAction — describes how to view the data
• schema:urlTemplate — the dashboard URL pattern (e.g., https://dashboard.example.com/#{dataset_url})
• schema:object — the dataset resource to visualize
• schema:instrument — the dashboard application

Web interface

The easiest way to use FDO2map is via the web interface — no installation required:

https://fair2adapt.github.io/FDO2map/

1. Paste an RO-Crate PID (or click one of the examples)
2. Enter an API key if the dataset is private (leave empty for public datasets)
3. Click Resolve and open dashboard

You can also link directly to a specific RO-Crate:

https://fair2adapt.github.io/FDO2map/#https://w3id.org/ro-id/24600867-23ca-4b97-be14-3aa63883c056

Command-line interface

For scripting or automation, use the Python CLI.

Prerequisites

• Python 3.10 or later
• A ROHub account (free, needed to query RO-Crate metadata)

Clone the repository

git clone https://github.com/FAIR2Adapt/FDO2map.git
cd FDO2map

Set up a Python environment

python -m venv venv
source venv/bin/activate   # On Windows: venv\Scripts\activate
pip install rohub

Configure ROHub credentials

Create ~/rohub_credentials.json with your ROHub account:

{
  "username": "your-email@example.com",
  "password": "your-password"
}

Run

python fdo2map.py <RO-Crate PID>

For private datasets, the script prompts for an API key. For public datasets, press Enter to skip.

Examples

RIOMAR ocean model (public, HEALPix DGGS)

python fdo2map.py https://w3id.org/ro-id/24600867-23ca-4b97-be14-3aa63883c056

Ocean temperature, salinity, and sea surface height from the GAMAR/GLORYS hindcast simulation, regridded to HEALPix level 13. Bay of Biscay, Oct–Dec 2023.

Sentinel-2 reflectance (public, HEALPix multiscale pyramid)

python fdo2map.py https://w3id.org/ro-id/fdc1c071-76d7-44df-a565-8217ebcc59fe

Sentinel-2 L1C Band 02 (Blue) reflectance on a HEALPix multiscale pyramid (levels 10–20, ~10km to ~10m resolution).

NorESM2 ARCTIC reanalysis (private, API key required)

python fdo2map.py https://w3id.org/ro-id/1f0b5044-ae4f-483d-b7a2-48a5a6ac3965

Monthly SST and 3D ocean temperature from NorESM2/BLOM (JRA-OC20 reanalysis), 2010–2018. Requires an API key — the script will prompt for it.

How the RO-Crate is structured

Each RO-Crate contains:

Resource type	Purpose
Dataset	URL of the Zarr data store
Web Service	URL of the dashboard application
ViewAction (annotation)	Links dataset to dashboard via a URL template
Product (annotation)	Scientific metadata (variables, spatial/temporal coverage, resolution)

The ViewAction makes the RO-Crate machine-actionable: any tool can discover how to visualize the data without hardcoded knowledge of the dashboard or data format.

Part of FAIR2Adapt

This tool is part of the FAIR2Adapt project, which transforms climate and ocean data into FAIR Digital Objects for climate change adaptation.

The visualization dashboard is gridlook / riomar-dashboard, a WebGL-based globe viewer supporting HEALPix, curvilinear, regular, and other grid types.

License

MIT License