Historical Climate and Scientific Data Rescue
Pre-industrial climate is written by hand, in the tabular columns of ship logbooks and station registers. Leo transcribes this material while preserving its grids and figures, helping data-rescue teams turn fragile observations into structured, quality-controllable datasets.
6 articles
Leo Team · 17 September 2026
How to Read an Old Ship's Log: Columns, Conventions, and the Sea Day
How to read an old ship's log column by column, including sea-day dates, observed versus dead-reckoned positions, and non-Greenwich meridians, without treating those conventions as plain facts.
Read the article
Leo Team · 17 September 2026
Where to Find Old Ship Logbooks: Archives, Digitised Images, and the Condition They Arrive In
Where old ship logbooks are held by fleet and institution, how digitised images vary in condition, and why catalogues, images, climate datasets and transcriptions are not the same thing.
Leo Team · 18 August 2026
Historical Weather Data Transcription: Getting the Numbers Off the Page Without Losing the Grid
Converting handwritten weather observations from station registers, ship logbooks and observatory ledgers into structured data without losing the grid, and how volunteer and machine methods compare on that task.
Leo Team · 18 August 2026
Ship Logbook Transcription: Turning Handwritten Navigation and Weather Records into Structured Climate Data
Ship logbook transcription for climate data: how naval and merchant logs are structured, which date and unit conventions corrupt datasets, and what volunteer, OCR, and HTR routes deliver.
Leo Team · 6 August 2026
How to Write HTR Into a Data Rescue Grant Proposal: Deliverables, Validation, and Cost Lines That Survive Review
How to draft HTR into data rescue grant proposals through observation-counted deliverables, fail-able validation plans, and costed pipeline lines that survive review.
Leo Team · 22 July 2026
Climate Data Rescue: Recovering the World's Weather from Handwritten Records
Explains the seven-stage climate data rescue pipeline for recovering pre-industrial weather records, focusing on why handwritten transcription is the bottleneck and comparing crowdsourcing, OCR/LLMs, and specialist HTR as solutions.