The deep stop debate: what the science actually says

The deep stop debate has been one of the most contested topics in decompression science for the past two decades. If you've done a technical diving course, you've almost certainly been told to add a "deep stop" — a pause at around 50% of your maximum depth on the way up. It sounds logical. It feels conservative. And for a long time, most divers accepted it as gospel.

But the science has caught up. So let's see where we stand today.

Where deep stops came from

The concept of deep stops was popularized by Richard Pyle, a fish taxonomist and experienced technical diver who noticed that he felt better after long dives when he paused at depth to collect fish specimens. He developed a rule of thumb: add a stop at 50% of your deepest depth, and stay there long enough to double your bottom time before ascending.

Pyle wasn't wrong that he felt better. But feeling better doesn't necessarily mean doing better decompression. It could have meant other things entirely — reduced ascent rate at a critical phase, time for his body to settle, or simply the reassurance of slowing down.

His observations sparked real interest in the decompression community. Algorithm designers started building deep stops into their models. VPM-B (Varying Permeability Model) and RGBM (Reduced Gradient Bubble Model) both incorporate deep stops as a core mechanism, based on the idea that stopping deep "crushes" bubble nuclei before they can grow. Bühlmann-based algorithms, by contrast, focus the bulk of off-gassing time in the shallower stops where nitrogen gradient is actually high enough to drive gas out of your tissues.

The NEDU study changed everything

In 2011, the US Navy Experimental Diving Unit (NEDU) published a landmark study that directly compared deep stop and shallow stop decompression profiles. The results were not what the deep stop community had hoped for.

Divers using deep stop profiles actually had higher rates of venous gas emboli — measurable bubbles in the blood — than those using shallow stop profiles. More bubbles. Not fewer.

The explanation, in retrospect, makes physiological sense. When you stop deep and spend time there, your fast tissues (which saturate quickly and off-gas quickly) are still absorbing gas. Meanwhile, you're delaying the time you spend in the shallow zone, where the gradient is steep enough to actually drive nitrogen out of your slow tissues. You end up arriving at your shallow stops already behind on off-gassing.

Deep stops were designed to address bubble mechanics. But they were doing so at the cost of dissolved gas loading — and the dissolved gas was winning.

What this means for instructors and divers

This doesn't mean your past dives were dangerous, or that you should panic if you've been using a VPM-B-based computer or algorithm. Most recreational and technical divers who follow any structured decompression plan are operating with comfortable safety margins. The real-world DCS risk difference between these approaches for typical divers is small.

But it does mean that the blanket instruction to "add a deep stop" as a conservative measure is no longer well-supported. If you're teaching decompression diving, teaching technical courses, or guiding divers who use dive computers, this is worth understanding and communicating clearly.

The honest answer today: Bühlmann with GF (gradient factors) — particularly Bühlmann GF Low/High — has strong evidence behind it. A 30/85 GF gives you a conservative, shallow-stop-weighted profile that keeps gradient manageable throughout. VPM-B remains in use, particularly in the cave and trimix community, but the case for it rests more on tradition and theoretical models than on outcome data.

Turtles doing no safety stop whatsoever

The practical takeaway

Stop guessing and start understanding your algorithm. Know whether your dive computer or dive planner uses deep or shallow stops, and understand what that means for your off-gassing profile. When in doubt, spend more time shallow — the nitrogen gradient is working hardest for you there, and the science backs it up.

Decompression is not religion. The moment we got outcome data, the conversation changed. That's not a reason to distrust the science — it's exactly what the science is supposed to do.