GUARDIAS Emerging Trends Dashboard

Interactive Quarto dashboard for exploring emerging trends for marine species across the European Large Marine Ecosystem (LME) regions.

Features

• Species autocomplete — type any species name; suggestions appear as you type
• Interactive Leaflet map — LME regions containing the selected species are highlighted; grey regions hold no data for that species
• Click-to-view plots — clicking a highlighted region opens a popup with an interactive Plotly chart for that species × LME combination
• Interactive plots — zoom, pan, hover for exact values

Data sources

File	Source	Update frequency
data/lme_polygons.geojson	Derived from lme_eu.gpkg (guardias-eu/build-eu-cube), simplified at 0.1° tolerance	Manual (via scripts/prepare_data.py)
data/emerging_trends_ranking_list.csv	emerging_trends_ranking_list.csv (guardias-eu/emtrends)	Weekly (via GitHub Actions)
data/appearing_species.csv	appearing_species.csv (guardias-eu/emtrends)	Weekly (via GitHub Actions)
data/reappearing_species.csv	reappearing_species.csv (guardias-eu/emtrends)	Weekly (via GitHub Actions)
data/indicators_plots_rdata/	RData files extracted from ZIP archives in indicators_plots/ (guardias-eu/emtrends)	Weekly (via GitHub Actions)
data/indicators_plots_json/	Plotly JSON files converted from ggplot2 objects in RData files	Weekly (via GitHub Actions)

Data flow

The indicator plots follow this processing pipeline:

1. Source: ggplot2 objects stored in RData files within ZIP archives in the emtrends repository
2. Download: Python script downloads ZIP files and extracts .RData files (ZIP files are split into chunks per LME)
3. Convert: R script loads .RData files, converts ggplot2 objects to Plotly, and exports as JSON
4. Display: Observable JavaScript loads JSON files and renders interactive charts in Leaflet popups

Automated data updates

The indicator plot files and species CSV files (emerging_trends_ranking_list.csv, appearing_species.csv, reappearing_species.csv) are automatically updated weekly via a GitHub Actions workflow (.github/workflows/update-data.yml). The workflow runs every Wednesday at 00:00 UTC and can also be triggered manually via workflow dispatch.

To manually trigger an update:

1. Go to the Actions tab
2. Click "Run workflow"
3. Select the branch and click "Run workflow"

Re-generating the data files

To refresh data/ from the latest upstream sources, run:

# For LME polygons (requires geopandas)
pip install geopandas
python scripts/prepare_data.py

# For indicator plots and species CSV (automated weekly via GitHub Actions)
# Step 1: Download ZIP files and extract RData files
pip install requests
python scripts/download_emtrends_data.py

# Step 2: Convert ggplot2 objects to Plotly JSON (requires R)
# Install R packages first:
#   install.packages(c('ggplot2', 'plotly', 'jsonlite'))
Rscript scripts/convert_plots_to_plotly.R

Note: The indicator plots and species CSV are automatically updated weekly by GitHub Actions, so manual updates are typically not necessary.

Dependencies

Python

• Python 3.11+
• requests (for downloading data)
• geopandas (only for regenerating LME polygons)

R

• R 4.0+
• ggplot2 (for reading ggplot objects)
• plotly (for converting to interactive plots)
• jsonlite (for JSON export)

Quarto

• Quarto for rendering the dashboard

Running the dashboard locally

Install Quarto, then:

quarto preview

Quarto opens the dashboard in your browser (usually http://localhost:4848). The page live-reloads on every save.

Building the static site

quarto render

The rendered site is written to docs/ and can be served by any static host (e.g. GitHub Pages).

Funding

This package is being developed in the framework of the GuardIAS project. GuardIAS receives funding from the European Union's Horizon Europe Research and Innovation Programme (ID No 101181413).