Exploring Global Tuna Atlas data via Shiny

Author

VLab Course Team

Objective: Learn how to interact with processed GTA datasets through dynamic Shiny applications, understand the benefits/limits of each app, and how CWP-formatted datasets fit in.
Prerequisite: Good understanding of GTA datasets (Levels, effort, CPUE).
Presentation: Hands-on demo + self‑exploration + short quiz.
Outcome: Learners can use a Shiny app to explore GTA datasets with or without writing code.

Quick start in VLab5

Shiny apps can be accessed in two ways:

  1. From the Shiny Apps tab in VLab → directly launch stable or dev versions of available apps.
  2. From the repository in ~/GitHubRepos → open the .Rproj, restore packages, and run shiny::runApp().

All Shiny apps have stable and dev versions, except darwin_core_viewer which is only available as a stable version in the Shiny Apps tab. Note: Depending on your level of write access, you may not be able to access the development version.
If you would like access, please contact us.
Please note that the list of deployed Shiny applications may evolve over time.

Outside Vlab 5

If you’re not (yet) a VLab5 member, you can still run the Shiny app locally using Docker.

Note: Docker images can be large and the app may require several GB of RAM.

We strongly recommend running this Shiny app on VLab5 RStudio Server because it can require a lot of RAM (often 32 GB, depending on dataset size and filters). VLab5 provides:

  • a consistent environment (R/RStudio, system libs),
  • higher memory availability than many laptops,
  • fewer installation issues (GDAL/PROJ, curl, etc. already present).

docker pull ghcr.io/<name-of-the-shiny-app-repo-eg.firms-gta/tunaatlas_pie_map_shiny> #(first time only)
docker run --rm --name tunaatlas \
  -p 3838:3838 \
  ghcr.io/<name-of-the-shiny-app-repo-eg.firms-gta/tunaatlas_pie_map_shiny>

and then open http://127.0.0.1:3838/

Shiny Apps in this lesson

Main apps (available in Shiny Apps tab)

Additional app

Example walkthroughs

A) Explore a CWP dataset with tunaatlas_pie_map_shiny

  1. From the Shiny Apps tab, open the stable version (recommended).
    (Alternatively, run locally with shiny::runApp() after restoring packages.)
  2. A formatted dataset is already loaded. You can as well choose a different dataset to explore in the ‘Choose dataset’ panel.
  3. Filter → by species, gear, year range, area.
  4. Visualize → interactive map (e.g., 5° grid) + plots; export as needed (see README).

Strengths: quick exploration, map‑centric, supports broad CWP datasets.
Limits: depends on schema conformity; heavy datasets may be slow.

B) Compare several GTA releases with shiny_compare_tunaatlas_datasets

  1. From the Shiny Apps tab, open dev or stable version.
  2. Datasets are already loaded from DOIs. If you want to explore other datasets, just update the DOI.csv file in the repository.
  3. Choose keys (year/area/gear/species).
  4. Inspect differences → tables/plots of differences, coverage, and changes.

Strengths: release comparisons, quick differences across versions.
Limits: assumes comparable schemas; interpret differencies carefully as the differences of processes for each datasets are complex.

Strengths vs. limitations of GUI tools

Strengths

  • Lower barrier to entry - no code needed.
  • Fast exploratory analysis & visual quality assurance.
  • Standardized filters/views reduce errors.

Limitations

  • Less flexible than writing code for bespoke analytics.
  • Performance can degrade on very large datasets.
  • Schema assumptions must be respected (CWP format).