From ship to screen — turning 75+ years of California Current ocean data into open apps, an API, and a versioned database.
CalCOFI — the California Cooperative Oceanic Fisheries Investigations — has sampled the California Current since 1949, making it one of the longest-running ocean observing programs anywhere. The data are extraordinary; getting to them used to be hard.
CalCOFI.io is the open-data software ecosystem that fixes that: from ship to screen. Reproducible
workflows ingest each dataset — ichthyoplankton, CTD casts, bottle chemistry, marine mammals — into a
single versioned integrated database. That database then powers interactive mapping apps, a
browser-based SQL playground, a REST API, and the calcofi4r R package.
As technical lead, I designed the pipeline and the stack behind it — DuckDB and Parquet for frozen releases, PostGIS and an H3 hexagonal tile server for the maps, FastAPI for the API, all wired together with Docker — serving scientists, resource managers, students, and the public alike.