Skip to content

Research and safety notice

Read this before you treat anything on this site as a statement about a real dive, a real chamber run or a real person.

Last updated 31 August 2026.

OpenDeco is a research and information platform for decompression science. It publishes source material, model implementations, benchmark machinery and analytical software so that decompression methods can be studied and reproduced.

Operational boundary

OpenDeco does not provide operational dive plans, real-time decompression guidance, medical advice, diagnosis, treatment, fitness-to-dive assessment or safety clearance.

Do not use this site as the tool that plans or conducts a diving, chamber, aviation, altitude or suit-pressure exposure. Those decisions belong to the procedures, equipment, training and professional oversight that govern the activity.

What the model implementations here are

Every model implementation published on this site was written for OpenDeco. None of them reconstructs, transcribes or reproduces a published decompression model, and none of them claims equivalence to any named model family.

Their compartment half-times, tolerance limits and hazard parameters are configuration values chosen so that the arithmetic stays checkable by hand. Every parameter set shipped with them is synthetic, so no output computed on them describes a real exposure, a real person or a real outcome. A reported probability is a probability under a hazard specification written to exercise the pipeline, not a probability of decompression sickness or of any other real event.

The method pages state the same position, and where a record is a pipeline fixture it says so on its own page.

Reading an output

An output shows what one named implementation computes for one defined input under one recorded configuration. It does not establish that the input is safe for any person, exposure or operation.

A verification level describes how far the arithmetic has been checked against closed-form results, generated inputs and committed baselines. It never speaks to physiological validity. A comparison between two implementations is a comparison of arithmetic, not a safety ranking.

Historical material

Where a historical table or procedure is reproduced for study, its presence on this site is not a recommendation to use it. Age, provenance and the conditions a procedure was written for all bear on whether it means anything today.

Research software

Research software can contain defects. Published source, reproducible runs and recorded verification reduce uncertainty; they do not remove it. A baseline accepted by an automated build has been checked by a machine and not by a person unless the record says otherwise.

Third-party sources

Where OpenDeco links to third-party research, procedures, software or data, the original source remains authoritative for its own claims and for its own conditions of use.

Report a problem

Report a factual, unit, citation, dataset or model error through the correction route. Report a security issue through the vulnerability disclosure route instead, so it can be handled privately.