Discover ocean data without hunting across portals

600,000+ datasets · ~100 sources · one searchable catalog. Meet the tool, then put it to work: find a real Hawaiʻi glider, load its track, and check the satellite against it — in Python or R.

Start the tutorial → Explore AQUAVIEW

or view the instructor storyboard →

STAC API live Python + R 60 min No sign-in for discovery

Tue, Sep 15 · 2:00–3:00 PM Central

aquaview.org/explore — Hawaiʻi

AQUAVIEW Explore map centred on the Hawaiian Islands, five ocean-data sources selected, datasets clustered into hex bins over the islands

5 sources · live hex bins over the fleet

How it works

One catalog, three moves

1

Discover

Search 600,000+ datasets from ~100 sources — satellite, floats, gliders, buoys, models, biodiversity, vessel traffic — in one catalog.

→ search by keyword, no IDs to memorise

2

Narrow

Pick a place, a time window, and the sources you care about. One call reaches across every source at once — no per-server hunting.

→ bbox + datetime + collections

3

Use & validate

Count before you fetch, find a real glider, load its track — then match the satellite to it and get bias, RMSE and correlation. From Python or R, the Explore map, or plain-English AI search.

→ glider track → satellite SST → bias · RMSE · r

Your tutorial example

The Hawaiʻi glider fleet

Satellite vs glider · passive acoustics + CTD

Is the satellite right, along the track?

r 0.94

satellite SST vs glider SST · 14 days · bias −0.04 °C · RMSE 0.11 °C — found among 4,228 datasets in the box

Satellite SSTWavesHF-radar currents AIS vessel trafficSpecies (OBIS)Seaglider sg626 track

See the validation →

Run it yourself

Same workflow, both languages

Py Python output

Python: glider vs satellite SST along the sg626 track — time series and scatter; bias −0.04, RMSE 0.11, r 0.94

The payoff — satellite SST vs the glider’s top-10 m along a two-week track: bias −0.04 · RMSE 0.11 · r 0.94.

View Python tutorial → ⬇ .ipynb

R R output

R: the identical glider-vs-satellite SST time series, produced with rstac, ncdf4 and ggplot2

The same validation via rstac + ncdf4 + ggplot2 — same glider, same numbers.

View R tutorial → ⬇ .ipynb

Python: pystac-client · pandas · xarray · netCDF4 · matplotlib · requests • R: rstac · readr · dplyr · ncdf4 · ggplot2

Running this in JupyterHub? ▾

The best experience is to open the downloaded .ipynb inside the Glider Rodeo Hub and run it there — every package is pre-installed and it runs top-to-bottom in seconds. A .ipynb is a JSON file, so a plain browser shows it as raw text until you open it in Jupyter — use Read the rendered tutorial above if you just want to look.