--- title: "Safe Ascents" description: "Why the slow, shallow end of the ascent does the most decompression work, how fast to come up, and where to spend your stop time." canonical: "https://donarundas.com/depth-notes/decompression/safe-ascents" author: "Donarun Das — TDI Trimix Instructor, KISS Sidewinder CCR Instructor" category: "decompression" published: "2025-05-22" tags: ["decompression", "ascent-rate", "off-gassing"] disclaimer: "Dive theory published by a working instructor for education. It is not medical advice and not a substitute for training. Decompression, gas, and depth decisions belong with your own training, your dive computer, and a dive medical officer." license: "© Donarun Das. Quote with attribution and a link to the canonical URL." --- # Safe Ascents ## A sponge and a soda bottle This is about the most overlooked skill in diving: how to come up. The ascent is the part that can hurt you, because that's when dissolved gas can leave solution too fast and form bubbles, which is decompression sickness. Picture your tissues as a sponge sitting in pressurised gas. On the way down you soak it up; on the way up you wring it out. The wringing only happens when there's a difference to drive it: your tissue has to hold more gas than the water around you is pressing back in. Decompression people call that difference a gradient, and without it nothing leaves. So you'd think the answer is simple. Drop the pressure as fast as you can, make the gradient huge, wring the sponge dry. That's where the second image comes in. Open a bottle of soda slowly and the gas eases out clear. It stays in solution and leaves through the surface. Crack the cap off fast and the whole thing foams over the moment the pressure drops. The gas was the same; only the speed changed. Your ascent is the cap. You need a gradient to off-gas at all (to drive the dissolved gas back out), but push it too far, too fast, and the dissolved gas stops leaving quietly and starts forming bubbles inside you. A good ascent balances those two: enough gradient to clear gas, never so much that it foams. (The soda image is a metaphor. Your tissues don't literally froth, but the same physics holds: gas kept in solution by pressure comes out as bubbles when the pressure drops too fast.) ## Why the shallow part is the dangerous part The risk isn't spread evenly up the water column. It piles up near the surface. What matters to the foaming isn't how many metres you move. It's how much the absolute pressure *changes in proportion* to where you already are. And that proportion gets brutal as you get shallow. Go from 40 m to 30 m and you drop from 5 bar to 4 bar absolute, about a 20% change. Go from 10 m to the surface and you drop from 2 bar to 1 bar. You've *halved* the pressure your tissues are pressing against. Same ten metres, wildly different shake of the bottle. For an ordinary recreational dive, a good pattern is about a minute at 9 m, two minutes at 6 m, and three minutes at 3 m. Call it six minutes total. It barely touches your gas, and each rung does a different job. The 9 m hold keeps your slow compartments under a deeper ceiling. The 6 m lets the medium tissues catch up. The 3 m is the classic stop where the fast compartments shed their last load before the surface. Does it actually help, or does it just feel thorough? Doppler studies, which count the silent bubbles drifting back through the veins after a dive, generally find that a controlled, graduated shallow ascent leaves a lower bubble load than bolting straight up. A direct ascent tends to leave the most; easing up the shallow zone in stages leaves less. *[Figure: bubble-reduction-stops — Schematic: a more graduated shallow ascent tends to leave fewer venous bubbles than a direct one. Illustrative of the general pattern seen across recreational Doppler studies, not a single dataset. — rendered in the HTML version of this page.]* One honest caveat. That fixed 9/6/3 ladder is the *recreational* pattern. On a decompression dive your stops and times come from the computer's gradient-factor schedule, not a memorised rule. The principle carries over exactly: more graduated stops mean a lower bubble load, and the algorithm just does the arithmetic for you. ## Flying the stop on a tech computer On a recreational dive you hold the stop by feel and a timer. On a technical computer you don't have to guess, because two live numbers tell you exactly how the dive is tracking against the plan. Both fall out of the [gradient factors](https://donarundas.com/depth-notes/decompression/gradient-factors) you set before you splashed. If your computer shows gradient-factor numbers, two of them tell you exactly how the dive is tracking. **[GF99](https://donarundas.com/depth-notes/decompression/gf99-and-surface-gf)** is how supersaturated your leading (most-loaded) tissue is right now, written as a percentage of its limit. **[Surface GF](https://donarundas.com/depth-notes/decompression/gf99-and-surface-gf)** is what that number would become if you surfaced this instant. As you hang and off-gas, both fall. When Surface GF has dropped to your planned **GF Hi** (the surfacing limit you chose before the dive), the model says you've earned the surface. > **Hold the depth, read the numbers** > > - **Hold each stop to about ±0.5 m, tighter as you get shallow.** A metre of drift at 6 m is over 6% of your absolute pressure, three times the swing of the same metre at 40 m. Drift up and your surfacing gradient spikes; drift down and you stop off-gassing efficiently. The model counted on you being *there*. > - **GF99 should be falling, never climbing.** If it rises mid-ascent you came up faster than your tissues could unload. Slow down, or drop back to the last stop until it settles. This is also where you buy cheap insurance on a dive that was harder than the model assumed. After a cold or strenuous dive your real off-gassing runs slower than the algorithm pictured, so extend a stop until Surface GF drops another 10% or so below your plan. It costs a few minutes of gas you can spare and buys a meaningfully gentler surfacing gradient. One display tip for that final leg. Switch your ascent-rate indicator from the cartoon bar to a numeric m/min readout. The last climb wants to sit near 3 m/min, a metre every twenty seconds, and a bar graph simply can't show you that precisely. ## Three things people get wrong **"Slower is always safer."** It isn't. A too-slow ascent keeps you deep longer, and your slow tissues are still *on-gassing* while the fast ones are trying to clear. Dawdle at 15 m and you're loading bone and cartilage even as you off-load blood and brain. There's a sweet spot, not a one-way "slower wins" dial. **"The deep safety stop is the important one."** It feels important, because it's deep and it feels disciplined. But the deep stop isn't where most of your gas leaves. The **shallow stop does the heavy lifting**, because that's where the surfacing gradient gets set. The gas you fail to clear at 5 or 6 m is the gas that follows you to the surface. **"A 5 m stop always helps."** On a shallow, easy dive it can do almost nothing. Your tissues may be barely supersaturated there, with a gradient-factor reading in single digits, so there's little gradient to off-gas against. On some profiles a slightly deeper hang at **6 m** clears gas better, because it sits you where the gradient is still working for you instead of where it's already spent. Stop depth is a tool, not a ritual. ## What's settled and what's still argued Settled enough to dive by: - Slow the **final** ascent. That's where the bottle foams. - Prioritise the **shallow stop** over the deep one for off-gassing. - The rate numbers are **conventions**, not physiology. Treat them as guardrails. Still genuinely contested: - The exact optimal ascent rate. The research narrows it but doesn't pin it. - Whether a brief deep stop helps at all on no-stop dives, or just delays the off-gassing that matters. That second debate is really a question about bubbles versus dissolved gas, which is its own story in [bubble trouble](https://donarundas.com/depth-notes/decompression/bubble-trouble). ## Your next ascent 1. From the bottom to your first stop, come up at about 9 to 10 m/min. 2. In the shallow zone, climb like a ladder: on an ordinary recreational dive, roughly a minute at 9 m, two at 6 m, three at 3 m. 3. Crawl the shallow legs at about 3 m/min, a metre every twenty seconds. 4. On a technical computer, surface only once Surface GF has fallen to your GF Hi, and never let GF99 climb. This note explains the why and the habits behind a good ascent; it is not a substitute for training. No-stop and decompression dives follow different rules, and a real decompression schedule comes from your planner and your course, not a memorised ladder. ## References 1. Divers Alert Network. *Ascent Rates.* Alert Diver, 2012. 2. Doolette DJ, Gerth WA, Gault KA. *Redistribution of Decompression Stop Time from Shallow to Deep Stops.* NEDU TR 11-06, US Navy Experimental Diving Unit, 2011. 3. Marroni A, Bennett PB, Cronje FJ, et al. *A deep stop during decompression from 82 fsw (25 m) significantly reduces bubbles and fast tissue gas tensions.* Undersea Hyperb Med. 2004;31(2). (DAN Europe DSL programme; on deep-stop bubble reduction, the effect later shown not to lower DCS.) *I drill ascent discipline, and the shallow stop especially, on every decompression course I run, from a diver's first staged ascent to deep trimix. [Ask me about training.](https://donarundas.com/contact)* ## Common questions ### How fast should you ascend when scuba diving? No faster than about 9 to 10 m per minute, and slower still in the shallows. The last few metres are the most dangerous part of the ascent because the proportional pressure change is largest there. ### Why is the shallow safety stop so important? Because the shallow zone is where off-gassing is fastest and your surfacing gradient is set. The gas you fail to clear at 5 to 6 m is the gas that follows you to the surface, so the shallow stop does more work than a deep one. ### Is a slower ascent always safer? No. Lingering deep keeps your slow tissues on-gassing even as the fast ones unload, so there is a sweet spot rather than a one-way slower-is-better rule. Ease off as you get shallow rather than crawling from the bottom. --- **Source:** [https://donarundas.com/depth-notes/decompression/safe-ascents](https://donarundas.com/depth-notes/decompression/safe-ascents) **About the author:** Donarun Das is a TDI Trimix and KISS Sidewinder CCR instructor teaching technical and rebreather diving in the Maldives, Sri Lanka, and India. He holds the deepest verified civilian dive in India (103 m, open-circuit trimix, Lakshadweep, March 2026). Courses: https://donarundas.com/courses — enquiries: https://donarundas.com/contact > Dive theory published by a working instructor for education. It is not medical advice and not a substitute for training. Decompression, gas, and depth decisions belong with your own training, your dive computer, and a dive medical officer. ## Sitemap See the full [sitemap](https://donarundas.com/sitemap.md) for all pages available as Markdown.