Underwater Welding Deaths: Causes, Statistics, and Safety

A 19-year-old diver descended into a power plant pump basin to check whether the stoplog gates were in place—and never came back up; his umbilical was sucked into a running pump in an adjacent bay that nobody had locked out. In North Carolina, a 47-year-old diver became unresponsive seconds after his first-ever descent on a bridge job and died despite his crew’s CPR. These are not freak acts of God; they are documented, preventable OSHA investigations into the deadliest welding job on Earth.

The Job Where Welders Die: Setting the Scene on Underwater Welding Deaths

In July 2021, a 19-year-old commercial diver went down into an electrical generating station to check whether the stoplog gates were in position. His umbilical was pulled into a running pump in an adjacent bay that nobody had locked out. He never came back up. OSHA’s accident report Nr. 137498.015 records the investigation in full: an ordinary task, an ordinary shift, and a single unverified energy isolation that turned a routine descent into a fatality.

Months later, in December 2024, OSHA issued a news release about a 47-year-old diver who became unresponsive seconds after descending roughly 19 feet. It was his first day on the job with Coastal Gunite Construction Co., working on the Manns Harbor Bridge in North Carolina. His crew performed CPR. He did not survive.

These two cases share something important: neither was an act of God, and neither happened while the diver was doing anything unusual. Both are documented, preventable failures found by federal investigators. That pattern is the reason this article exists.

So How Dangerous Is Underwater Welding, Really?

The honest answer: extremely dangerous, and more dangerous than most people outside the trade assume. Commercial diving sits near the top of nearly every list of the world’s most hazardous occupations, and NIOSH treats commercial diving as a high-risk maritime profession with its own dedicated guidance. Between 1989 and 1997, federal surveillance counted 116 occupational diving fatalities in the United States — an average of about 13 deaths a year in a workforce that, at the time, numbered only a few thousand full-time divers.

To understand why, it helps to see that underwater welding stacks two already-hazardous jobs on top of each other. The diver must survive the water; the welder must survive the arc, the heat, the fumes, and the metal. Remove either discipline’s safety net and the consequences are immediate.

The main hazard categories we will work through are:

• Drowning and asphyxiation — severed umbilicals, free-flowing regulators, entanglement, lost breathing gas, zero visibility.
• Differential pressure (delta P) — pump intakes, drain lines, and pipe flow that trap divers with startling force.
• Decompression illness, embolism, and gas toxicity — the silent, delayed killers of depth work.
• Electrocution — stray current, ungrounded machines, and alternating current in conductive water.
• Explosion — hydrogen and oxygen generated by the arc itself inside flooded spaces.

What This Article Covers — and What It Doesn’t

This is an informational breakdown of underwater welding deaths: the real statistics, the causes investigators actually document, the OSHA rules written in response, and the practices that keep divers alive. We will cite federal investigation records and peer-reviewed reviews rather than anecdotes. What we will not do is sensationalize. You will not find scare quotes, horror-movie framing, or unverified “insider” claims here.

We also stay focused on the fatality question itself. Career-path details, pay expectations, and topside training routes belong elsewhere — our aluminium welding course guide covers classroom and shop-based paths, and our aluminum boat welding guide handles hull work done above the waterline. Wet underwater welding has its own rules, and those rules are the subject of this piece.

One more thing worth stating plainly: neither of the opening cases involved a catastrophic equipment failure or a freak current. Both involved a checklist item that was skipped — lockout/tagout in one instance, fitness-for-duty and emergency readiness in the other. Keep that in mind as you read. The statistics that follow are large, but the failures behind them are usually small, specific, and correctable, which is exactly why regulators, training agencies, and OSHA’s commercial diving program focus so heavily on procedure rather than on the arc itself.

Let’s start with what the numbers actually say.

What Underwater Welding Actually Is: Wet, Dry, and Robotic Welding

Before we count anyone’s deaths, it helps to be precise about the job itself. Underwater welding is not one trade but a fusion of two demanding ones: commercial diving and arc welding. The welder is a qualified diver first, in full life-support gear, who then strikes an electric arc below the surface to join, cut, or repair metal. Every hazard later in this article flows from that combination, so the three branches of the trade are the fastest way to see why.

Wet Welding: The Arc and the Water in the Same Frame

Wet welding, also called in-water welding, is the most common and most visible form. The diver descends with a shielded metal arc welding (SMAW) stinger and waterproof electrodes — stick rods with a thick, moisture-resistant flux — and welds directly in open water, exposed to whatever visibility, current, and temperature the site offers. Because it differs so much from topside work, OSHA addresses it specifically: the commercial diving standard lays out welding and burning requirements for wet welding in 29 CFR 1910.422(g), distinct from cutting and from work in a habitat.

The environment changes the arc’s physics and chemistry in ways topside welders take for granted:

– **The arc atmosphere is contaminated by water.** At arc temperatures, surrounding water dissociates, feeding hydrogen and oxygen into the arc zone. The flux coating must generate its own shielding bubble to protect the puddle, which is why underwater stick electrodes look chunkier than ordinary rods.
– **Cooling is brutally fast.** Water acts as a huge heat sink, quenching the puddle and the heat-affected zone. Fast cooling promotes hard, brittle microstructures, and the hydrogen present raises the risk of cold cracking — a reason procedure qualification matters far more underwater than on a bench. Our guide to the advantages of stick welding explains why SMAW’s self-shielding simplicity wins in awkward positions; underwater, that simplicity is the whole argument.
– **Bubbles scatter light and hide the puddle.** Hydrogen and oxygen stream off the arc, so visibility swings between blown-out glare and near-zero.
– **Electrode choice is constrained.** Position, current type, and material all narrow the field, much as in alloy steel welding electrode selection, but with no chance to reposition mid-run.

Dry Hyperbaric Welding: Welding Inside a Pressurized Habitat

Dry hyperbaric welding solves the water problem by removing it. A habitat — a chamber placed over the weld zone — is sealed to the structure, pumped dry, and pressurized with a helium-rich gas mixture carrying controlled oxygen. The diver-welder works inside that atmosphere, or the habitat is tended from the surface, and the arc burns far closer to normal shop conditions.

The payoff is quality: no water quench, no bubble interference, no arc contamination from dissociated water, so weld properties and inspection results approach topside levels. That is why critical offshore structural repairs often specify dry welding. The cost is complexity — habitat fabrication, gas handling, hot work controls, chamber tending, and decompression planning — which is why it is used selectively rather than routinely.

ROV-Assisted, Hydraulic, and Robotic Systems

The third branch removes the welder from the water where possible. Remotely operated vehicles carry cutting heads, welding manipulators, and hydraulic tooling to depths or inside spaces no diver should enter, and hyperbaric habitats are increasingly deployed by ROV rather than by hand. Hydraulic shears, abrasion tools, and coral-cutting gear handle much of what previously demanded a diver with a torch, cutting exposure to the moments when a human hand is genuinely needed. Every task an ROV completes is one less task a diver performs blind, in a current, next to a suction source.

Where the Work Happens — and Who Does It

All three methods show up across the same infrastructure map: ship hulls and tanks, offshore platform jackets and risers, bridge pilings and fender systems, dam spillways and penstocks, submerged pipelines, and the water tunnels of nuclear and fossil-fuel plants. Anywhere steel sits in or under water and can’t come into a shop, underwater welding eventually gets called. Even familiar territory has its own rule set; hull repair on small craft follows different practice than classification-grade offshore work, as our walkthrough of aluminum boat welding shows topside.

The workforce is small. CDC and NIOSH cite an estimated 3,380 commercial divers in the United States as of 2017, concentrated in coastal states and around the Great Lakes. That number frames what follows: this is a niche trade, which is why fatality statistics swing wildly year to year and a single preventable accident can move the national rate.

With the trade defined — a diver striking a wet arc in open water, a welder working dry inside a pressurized habitat, or a machine doing both at a distance — we can look honestly at what happens when any of the three goes wrong.

Underwater Welding Death Statistics: What the Numbers Really Say

Ask ten people how many divers die underwater welding each year and you’ll get ten different answers, most of them wrong. The “15% death rate” claim that circulates on social media and law firm blogs has no methodology behind it—it implies that one in seven underwater welders dies on the job, a figure no government agency, insurer, or academic study has ever produced. The honest answer is messier but far more useful: good data exists, it just needs to be read carefully.

What NIOSH Found in the Landmark MMWR Analysis

The most cited peer-reviewed benchmark comes from the National Institute for Occupational Safety and Health (NIOSH), published in the Morbidity and Mortality Weekly Report in 1998. Researchers analyzed OSHA drowning and asphyxiation investigation records from 1989 through 1997 and identified 116 deaths in occupational diving across all industries. Of those, 49 occurred among an estimated 3,000 full-time commercial divers—the workforce that includes underwater welders.

Forty-nine deaths in roughly 3,000 workers over eight years works out to about 180 deaths per 100,000 divers per year, which the authors noted was approximately 40 times the national average for all U.S. workers. You can read the original analysis on the CDC MMWR archive, and NIOSH’s current commercial diving hazard overview is maintained on the CDC/NIOSH maritime program page.

For context, that rate is comparable to some of the most dangerous military and extraction occupations—orders of magnitude above the roughly 3–4 per 100,000 rate for all private-industry workers.

Source	Period	Key Figure
NIOSH MMWR	1989–1997	49 deaths among ~3,000 full-time divers; ~180 per 100,000 per year
NIOSH NTOF database	1989–1994	56 occupational diving deaths (cause breakdown below)
U.S. Bureau of Labor Statistics	2011–2017	39 fatal injuries; 460 nonfatal injuries with days away from work
Cambridge/EMI Alliance review	2022	Estimated 3–10 fatalities per 10,000 divers per year (U.S. and U.K.)

What Actually Kills Divers: The Cause-of-Death Breakdown

NIOSH’s National Traumatic Occupational Fatalities (NTOF) database recorded 56 occupational diving deaths from 1989 to 1994, with a revealing cause distribution:

• Drowning: 73% of cases
• Asphyxia: 14% (usually breathing gas loss or contaminated supply)
• Embolism: 7% (gas entering the bloodstream from lung overexpansion)

Notice what’s missing: electrocution, explosions, and decompression sickness are comparatively rare as recorded causes of traumatic death. Drowning and breathing-gas failures dominate, which is why later sections on marine welding safety standards put so much emphasis on umbilicals, standby divers, and gas supply redundancy. The arc itself rarely kills; the environment around it does.

Why Exact Underwater-Welder-Only Numbers Don’t Exist

Here’s a limitation readers should understand: the Bureau of Labor Statistics does not track “underwater welders” as a separate occupation. They fall under the broader commercial divers category, alongside helmet divers, tender-support roles, and scientific divers. So when BLS reported 39 fatal occupational injuries among commercial divers during 2011–2017, plus 460 nonfatal injuries involving days away from work, that figure includes welders but cannot isolate them.

The 2022 University of Cambridge Institute of Energy and Environmental Flows report, commissioned by the EMI Alliance, reviewed U.S. and U.K. fatality data and estimated commercial diving risk at 3 to 10 deaths per 10,000 divers per year—a range rather than a single number, reflecting how inconsistently different countries record these incidents. The full Cambridge paper is publicly available and worth reading before anyone quotes a statistic at you.

Reading the Trend Honestly

Raw counts also need care. Thirty-nine BLS fatal injuries across seven years sounds lower than NIOSH’s older rate, but BLS census-based counts depend on death certificate coding and often undercount self-employed and small-crew divers, while NIOSH used a smaller, denser time window. Both sources point the same direction: commercial diving fatality rates remain roughly an order of magnitude above the all-industry average, and drowning is the mechanism in the large majority of deaths.

Those statistics are what make OSHA’s commercial diving rules—covered next in this article’s regulatory section—so prescriptive about planning, communication, and standby procedures. The numbers aren’t abstract; they’re the reason the rules read the way they do.

Drowning and Delta P: The Leading Causes of Underwater Welding Deaths

If the arc gets the headlines, water gets the bodies. In NIOSH’s analysis of occupational diving deaths from 1989 to 1994, drowning accounted for 73% of fatalities and asphyxia for another 14%. Combined, roughly 87% of the commercial divers who died on the job never burned, never took a lethal shock, and never died in an explosion. They simply ran out of something to breathe.

That reframes the entire risk conversation. The arc is a genuine hazard, but the killer is usually the environment around it: a severed umbilical, a regulator venting into open water, a suction line nobody flagged, or silt so thick the diver cannot see his hands.

How a Diver Drowns on a “Breathing Air” System

Commercial diving is not scuba. The diver is tethered by an umbilical carrying breathing air, comms, and tooling, helmet sealed to his face. That design concentrates failure modes into one bundle, and most diving drownings trace back to one of these:

• Umbilical severing or snagging. Moving gates, sharp flange faces, rebar, and debris can cut or pin the air hose. In one power plant pump basin case, a diver’s umbilical was drawn into an unlocked pump.
• Free-flowing regulators. A stuck demand valve dumps air continuously, drowning out comms while the supply drains fast; cold water and ice make it worse.
• Entanglement. Wire rope, fishing line, formwork ties, and loose cable hold a diver until his gas runs out.
• Loss of breathing gas. Compressor failure, a closed valve topside, contamination, or a ruptured hose downstream of the filter.
• Zero visibility. Wet welders stir sediment and arc smoke at the faceplate. In a silt-out, a diver who loses comms can drown feet from an exit he cannot find.

Underwater wet welding tightens these constraints. The diver works one-handed, close to the structure, holding a live stick electrode in moving water while the arc boils off coating gases and kicks up debris. Many arrive with topside experience such as aluminum boat welding before ever adding a helmet and an umbilical.

Delta P: The Suction Hazard That Pins a Diver

Differential pressure, or delta P, is the difference in pressure between two connected points. Underwater it is a trap, because a diver pressing against an opening forms a seal. The numbers are deceptively small: a few inches of water column is enough to hold a diver against a suction. Once the seal forms there is no edge to grip; pulling harder only tightens it, and a standby diver hauling on his umbilical rarely wins. Worse, the trapped diver’s own breathing gas works against him, pushing air from his lungs toward the restriction.

Delta P is dangerous precisely because it is invisible: no spinning blade, no smoke, no spark, just water quietly moving from a higher-pressure point to a lower one.

Common delta P source	Where it hides
Pump intakes and suction bells	Circulating water bays, cooling water tunnels, sumps
Gravity drains	Drain lines with a partially open valve downstream
Flow through a line	A valve left cracked between isolated sections
Flooding compartments	A head difference between two connected flooded spaces
Pipe elbows and tees	Velocity rises inside the fitting; the pull feels mild

The last row is not theoretical. OSHA accident report Nr. 121668.015 documents a 38-year-old diver killed in November 2019 while installing an 18-inch blind flange inside a nuclear power plant service water tunnel. He was found trapped in the elbow of a 30-inch pipe with multiple blunt force trauma. Nothing sounded extraordinary topside: one blind flange, one tunnel, one dive.

Why OSHA Requires a Hazard Analysis Before Every Suction Dive

Because divers cannot see pressure, the paperwork has to. OSHA’s enforcement directive CPL 02-00-151 requires a hazard analysis to determine standby diver use near underwater differential pressure flows such as pump intakes and gravity drains. In practice, the dive plan must identify every suction, confirm how each is isolated, and decide before anyone gets wet whether a standby diver waits at the entry.

The requirement sits inside the broader 29 CFR Part 1910 Subpart T framework, which expects the safe practices manual and pre-dive planning to address these environmental hazards rather than the diver’s judgment at depth.

Ask any supervisor for the isolation plan, the standby diver assignment, and the comms check before you clip in. Delta P does not announce itself, and experience does not reduce it. In the NIOSH data, drowning and asphyxia together claimed roughly 87% of the divers who died on the job, a share that shrinks only where the hazard analysis was honest and the isolation verified twice.

Decompression Sickness, Embolism, and Nitrogen Narcosis: The Silent Killers

Not every underwater welding death happens in a flash. Some of the deadliest risks in commercial diving build slowly, silently, and invisibly, which is why the diving community calls them the silent killers. A diver can feel perfectly fine at depth, finish a clean pass on a bridge piling, and still be in serious trouble before he ever reaches the surface. Understanding how these physiological injuries happen is the difference between treating them as bad luck and treating them as predictable, preventable failures.

Decompression Sickness: The Bends

Below roughly 33 feet, the increased pressure forces nitrogen from the breathing gas into the diver’s blood and tissues. Ascend slowly, and that nitrogen comes out of solution safely, much like a soda bottle opened gently. Ascend too fast or stay down too long, and nitrogen comes out of solution inside the body, forming bubbles in joints, the spinal cord, and the bloodstream. The result is decompression sickness, commonly called the bends: joint pain, weakness, dizziness, paralysis, or loss of consciousness.

Bottom time and depth are the obvious variables, but welding adds others. An uncontrolled ascent triggered by a snagged umbilical, a free-flowing regulator, or a panicked entanglement skips decompression entirely. Repetitive dives over a long shift accumulate residual gas. And a diver who is cold, dehydrated, or exhausted tolerates a much narrower margin than a rested diver on a single morning dive.

Arterial Gas Embolism and Pulmonary Barotrauma

If a diver holds his breath during an uncontrolled ascent, expanding gas can rupture lung tissue. Once air enters the pulmonary veins, it travels to the brain, causing an arterial gas embolism: stroke-like symptoms, convulsions, and often death within minutes. NIOSH’s National Traumatic Occupational Fatality data attributes 7% of occupational diving deaths to embolism, and researchers note that hypothermia and air embolism are likely undercounted precisely because they are so difficult to diagnose after the fact (CDC/MMWR analysis of OSHA diving fatality data).

This is why the dive plan is not paperwork for its own sake. OSHA 29 CFR 1910.421(d) requires pre-dive planning to explicitly cover decompression and treatment procedures, including altitude corrections for travel after a dive, plus the residual inert gas status of every dive team member. Under 29 CFR Part 1910 Subpart T, divers must also be told where a decompression chamber is ready for use and warned about the hazards of flying after diving.

Nitrogen Narcosis and Oxygen Toxicity

At depth, nitrogen itself becomes an anesthetic. Nitrogen narcosis produces impairment similar to alcohol: slowed reaction time, poor judgment, overconfidence. For a welder making fine decisions about electrode angle, entrapment, and exit routes, impaired judgment is a hazard multiplier, not a nuisance. Deeper or longer exposure raises another risk, oxygen toxicity, which can trigger seizures underwater and drowning follows almost immediately.

Thermal stress compounds all of it. CDC/NIOSH identifies respiratory and circulatory complications and hypothermia among the primary hazards facing commercial divers, worsened by the number and duration of dives in a shift. Cold divers metabolize differently, fatigue faster, and lose the fine motor control that safe welding demands. A topside habit that translates poorly underwater is pushing through discomfort; the discipline that does work, whether you learned it on aluminum boat welding jobs or in a shop, is to stop before conditions degrade your thinking.

What the Rules Require

The regulatory answer is deliberately boring, because boring saves lives:

• Written decompression and treatment procedures with altitude corrections, per 1910.421(d).
• Dive team members advised of a ready decompression chamber and warned about flying after diving.
• Post-dive debriefing that documents how each diver actually feels, not how the logbook says they should feel.
• Medical fitness evaluations and rehabilitation after any decompression illness event.

The Diver Alert Network remains the standard reference for emergency consultation and evacuation planning, and its guidance echoes the regulations: recognize symptoms early, treat them as an emergency, and never let schedule pressure talk a symptomatic diver into one more dive.

The honest takeaway: no amount of experience stops nitrogen from behaving like nitrogen. Tables, chambers, standby time, and thermal protection do.

Explosions and Electrocution: The Hazards Unique to Underwater Welding

Drowning and decompression sickness can kill any diver. But explosions and electrocution are different: they exist only because the diver is holding a live welding arc. That separates underwater welding from inspection, rigging, or salvage. The tool that pays the wage can also kill in a fraction of a second, and these mechanisms sit behind many underwater welding deaths.

Hydrogen and Oxygen: The Bubbles You Can’t See Coming

When a welding arc burns underwater, its heat splits water molecules. The arc breaks H₂O into hydrogen and oxygen, so every bubble rising from the puddle carries an explosive mixture. Phil Newsum, Executive Director of the Association of Diving Contractors International, says underwater welding and burning can produce hydrogen and oxygen mixtures that are highly explosive. Hydrogen is one of the most detonable gases in existence, and it is generated at the point where an ignition source always exists.

Under normal open-water conditions, those bubbles disperse harmlessly. The danger appears in or near an enclosed space. Flooded compartments inside ship hulls, sealed structures, closed pipeline sections, and cofferdams can trap the gases. Hydrogen is lighter than water, so it collects at high points and in trapped air pockets. In a confined space, concentrations can reach detonable levels quickly.

That is why the industry rule is absolute: a flooded compartment is not automatically safe. OSHA’s commercial diving standards address this directly in 29 CFR 1910 Subpart T. Section 1910.422(g)(5) requires that closed compartments, structures, or pipes containing flammable vapors be vented, flooded, or purged with a non-combustible gas mixture before any welding or burning. OSHA’s enforcement directive, CPL 02-00-151, warns that a flooded compartment is not necessarily safe for cutting and welding because oxygen, hydrogen generated by electrolysis, and other gases can accumulate even when the space looks water-filled from outside.

This same chemistry explains why hot work on hulls and marine structures demands rigorous gas testing. Anyone studying aluminum hull repair should notice how much of the prep work in Aluminum Boat Welding: The Complete Guide for Strong Hull Joints concerns ventilation, compartment preparation, and atmosphere control rather than the weld itself. The arc is the easy part.

Electricity in Saltwater: A Different Kind of Danger

The second unique hazard is electrical. Underwater wet welding typically uses direct current, and the holder, cable, and work are all live while the arc is struck. Saltwater is an excellent conductor, making stray current a serious harborside problem. A faulty shore connection, a running cathodic protection system, or corroded bonding can leak current into the water. A diver swimming through that gradient can suffer uncontrolled muscle contraction, burns, or cardiac disturbance before touching a welding cable.

OSHA addresses the on-the-job side with hard requirements. Under 1910.422(g)(1), the welding current switch must be tended by a dive team member in voice communication with the diver, and it must stay open except while the diver is actively welding or burning. The welding machine frame must be grounded, as required by 1910 Subpart T. That switch exists for one reason: if a diver goes limp, drops the stinger, or gets into trouble, a topside person can kill the current instantly without fumbling in zero visibility.

ADCI consensus standards reinforce these practices across the commercial diving industry. Equipment grounding and circuit protection are shared with topside work, as covered in Air Arc Welding and Carbon Arc Gouging: Process, Safety, Applications, and Best Practices. The difference underwater is that escape is slow, visibility is often near zero, and a muscle-locking shock can prevent self-rescue.

Why These Two Killers Demand Extra Scrutiny

Both hazards are invisible until it is too late. You cannot smell hydrogen accumulating inside a flooded pipe, and you cannot see stray current in harbor water. That is why control is procedural rather than sensory: atmosphere testing, purging, venting, grounding checks, and a tended switch. DAN guidance on commercial diving safety emphasizes that these procedures are not bureaucratic padding. They are the only warning system against hazards the body cannot detect.

For anyone in adjacent trades, the lesson travels. Marine fabrication, shipyard repair, and harbor work all inherit versions of these risks. Even surface-side guides such as AJ Marine Welding: Processes, Standards, Safety and Best Practices stress compartment atmosphere testing before hot work, because the physics of trapped hydrogen and stray current does not stop at the waterline. Understand the hazard, verify the controls, and never let a deadline talk anyone out of a purge.

The Rules Built to Keep Divers Alive: OSHA Subpart T and Industry Standards

Every fatality reviewed in this article happened inside a regulatory framework that already existed. That is what makes commercial diving deaths so frustrating to investigate: the rules that would have prevented them were written decades ago, published as a federal standard, and in most cases ignored on the day of the dive. For a supervisor writing a job hazard analysis, or a newcomer weighing a career underwater, the regulations are not bureaucracy. They are a checklist of what a legitimate employer must be able to show you before you get in the water.

What 29 CFR Part 1910 Subpart T Actually Requires

OSHA has regulated commercial diving operations under 29 CFR Part 1910 Subpart T since 1977. It applies to diving conducted by employers in connection with their business and is organized around five requirement areas:

• Dive team qualifications. Every dive must be supervised by a qualified dive supervisor, supported by trained divers, tenders, and a standby diver or equivalent assistance ready to respond. Competence is documented, not inferred from years in the water.
• A safe practices manual. The employer must maintain a written manual covering its diving operations, and the dive team must follow it. An oral tradition of “how we’ve always done it” does not satisfy the standard.
• Pre-dive planning. Before the diver enters, the team evaluates the hazards of the specific job: depth, breathing gas, visibility, currents, entrapment, and energy sources. This is the step that fatal investigations most often show was skipped or faked.
• Procedures during the dive. Continuous communication, attentive tending of the umbilical, monitoring of depth and bottom time, and controls for the hazards identified in the plan.
• Procedures after the dive. Post-dive checkouts, recorded dive data, a debrief that includes fitness questions, and restrictions on flying after diving.

The full text is free through the eCFR, and OSHA’s commercial diving safety page gathers the related guidance in one place.

The SCUBA Restriction That Reshaped Ship Husbandry

One of Subpart T’s most consequential provisions limits SCUBA for work dives. Where the employer cannot demonstrate that surface-supplied equipment is impractical, SCUBA may not be used, and OSHA’s enforcement directive is explicit that a flooded ship compartment is not automatically safe for cutting and welding. The practical effect was an industry-wide shift: ship husbandry work that once ran on open-circuit scuba moved to surface-supplied diving, with an umbilical carrying air, communications, and monitoring. That single change eliminated an entire category of drowning causes, because a diver on a hose cannot silently run out of gas, swim away from a hazard, or drift out of contact in zero visibility. For a broader look at how these principles show up in marine work, see our guide to AJ Marine Welding: processes, standards, safety and best practices.

The Scientific Diving Exemption Does Not Cover Welders

Subpart T’s Appendix B exempts scientific diving performed under an American Academy of Underwater Sciences organizational program, but the exemption is narrow and conditional. It explicitly excludes construction, demolition, and cutting or welding, along with other heavy or hazardous work. An underwater welder cannot hide inside it. If an employer claims the AAUS program covers hot work, that claim is wrong on its face.

ADCI Standards, State Plans, and Reporting Duties

Beyond the federal rule, the Association of Diving Contractors International publishes Consensus Standards for Commercial Diving and Underwater Operations. Many contracts adopt them by reference, and OSHA compliance officers consult them when judging whether an employer met recognized practice. State plans often mirror and sometimes reinforce that framework. Washington State’s DOSH Directive WRD 22.85 follows federal guidance and points directly to the ADCI Consensus Standards, giving state inspectors a documented benchmark.

Reporting is the piece employers most often get wrong. Under 29 CFR 1904.39, an employer must report any work-related fatality to OSHA within eight hours, with no exemption for small companies or low-hazard classifications. Late reporting is a violation in its own right, layered on top of whatever caused the death.

Medical fitness closes the loop. Subpart T requires each diver to be examined by a physician before starting work and evaluated periodically thereafter, with results reviewed by the dive supervisor. The Diver Alert Network’s commercial diving safety guidance documents how these evaluations catch cardiac and respiratory conditions before they become in-water emergencies.

What Compliance Actually Looks Like

When the rules work, they look unglamorous: a signed manual, a documented pre-dive briefing, a standby diver who stays dressed, a logged air purity test, a chamber identified in the emergency plan, and a supervisor willing to call the dive. Anyone training for the trade should learn that discipline early, and a structured aluminium welding course covers a standards mindset that transfers directly to hot work underwater. The regulations cannot stop a negligent employer, but they make negligence visible, and visibility is the first step toward prevention.

Preventable Failures: What Fatal Investigations Reveal About Common Mistakes

Read enough commercial diving fatality reports and a pattern jumps out: in almost none of these cases does the diver die at the arc. The technique side of the trade, whether that means choosing an all position welding rod or tuning amperage for a deep fillet, rarely shows up as the cause of death. What shows up instead is a decision made before anyone touched the stinger: a gate assumed instead of locked out, a fitness check never run, an air hose not pressure-tested in a year. Mining OSHA’s investigation records for underwater welding deaths turns up the same handful of human failures, repeated across states, job types, and decades.

The Coastal Gunite Case: Missing the Basics

In June 2024, a 47-year-old diver on the Manns Harbor Bridge replacement project in North Carolina became unresponsive seconds after his first-ever descent. His crew pulled him up and performed CPR, but he could not be revived. OSHA cited Coastal Gunite Construction Co. for four serious violations and proposed $40,329 in penalties, the maximum the agency could legally recommend for serious violations at that time (OSHA news release). The inspector’s findings are worth reading slowly, because none of them involve exotic hazards:

• No fitness-for-diving assessment. The company never evaluated whether the diver was medically fit for diving on his first day on the job.
• No rescue readiness. First aid supplies and a manual resuscitator were not on site.
• An improperly positioned air intake. The compressor intake could draw in contaminated or exhaust air.
• No regular air purity testing. Nobody checked that the breathing gas met the required purity.
• No annual hose pressure tests. The diver’s lifeline had not been proof-tested on schedule.

The Bulldog Diving Case: The Cost of Taking Someone’s Word

The other investigation cited constantly in this topic is Bulldog Diving’s power plant fatality, where a 19-year-old diver descended into a pump basin to confirm that stoplog gates were in place. Investigators found the crew relied on a coworker’s uncertain statement about those gates and never verified that the running pump in an adjacent bay had been locked out. Moments later, the diver’s umbilical was drawn into that pump and he was drowned before the surface team could respond.

Read those two cases side by side and the common structure is obvious. In both, the fatal condition was created upstairs: an unverified isolation, an unchecked diver, an air supply nobody tested. The arc never got the chance to be dangerous.

What Forty Years of Fatality Research Keeps Finding

This is not a modern pattern either. A 1984 review published in Aviation, Space, and Environmental Medicine examined fatalities among Gulf of Mexico and North Sea commercial divers and found the most significant contributing factors were diver experience and behavioral dysfunction, followed by equipment failure and errors by the supervisor or tender (study abstract). Behavioral dysfunction is a clinical way of saying decisions like skipping checks, suppressing doubts, and rushing a descent to keep a schedule on track. OSHA’s current findings keep listing the same categories in new language.

The Recurring Mistake Checklist

Before any dive, a supervisor should be able to walk this list and answer every item with verified evidence rather than someone’s recollection:

• Skipped pre-dive hazard analysis. OSHA Subpart T treats planning as a requirement, not a formality. “We’ll sort it out in the water” is how stoplogs go unchecked.
• Unverified energy isolation. Lockout/tagout and gate positions get physically confirmed, not accepted from a coworker’s uncertain answer.
• No standby diver or rehearsed rescue. Sites with suction, entanglement, or near-zero visibility need a recovery plan that has been walked through before the descent.
• Poor communication. Weak topside-to-diver links, an inattentive tender, or unclear hand and umbilical signals turn small problems into emergencies.
• Inadequate medical screening. Missing fitness-for-duty evaluations, as in the Coastal Gunite case, leave disqualifying conditions undetected.
• Cost-driven shortcuts. Skipped air purity checks, lapsed hose tests, and out-of-date regulators are where budgets quietly become fatalities.

None of these items require new technology or a bigger crew. They require someone on deck whose job is to say “show me” before the diver goes in. The craft knowledge that keeps welds sound on decks and hulls, including the marine and shipboard fundamentals in our AJ Marine Welding guide, still matters afterward. But when you line up what the investigations actually found, the lesson is consistent: underwater welding kills through unverified assumptions and skipped procedures far more often than through the arc itself.

How the Industry Is Getting Safer: Technology, Training, and Best Practices

Most of the fatality data in this article looks backward, and that is actually useful: the deadliest patterns were identified decades ago, and the modern commercial diving industry has spent the time since redesigning the job around them. Nearly every change now treated as normal practice maps onto one of the leading causes of death covered earlier in this article — drowning, gas supply failure, decompression illness, or hot work in a confined space.

Surface-Supplied Diving Replaced Scuba for Work Dives

The single biggest safety shift is the move away from scuba for working dives. The Divers Alert Network has noted a rash of fatalities tied to scuba use in ship husbandry work, and both the industry and OSHA’s commercial diving standard push that work toward surface-supplied systems instead.

The difference matters because of what the umbilical carries:

• Continuous breathing gas from a surface control panel, so a diver cannot quietly exhaust a limited supply.
• Hard-wired voice communication instead of hand signals in zero visibility.
• Hot-work cables and a pneumo hose run through the same umbilical, so depth calls and welding power stay supervised by a tender on the surface.

That single bundle removes several recurring causes of death at once: unexplained loss of breathing gas, communication failure, and the stranded-diver scenario when a diver has to swim back to a boat before the clock runs out. The tender also knows continuously whether the diver is moving, stuck, or in trouble.

Technology That Reduces Diver Exposure

The second trend is simply spending less time in the water. Remote-operated vehicles now scout and inspect structures before a diver ever suits up, so the first human in the water is not also the first person to discover an open valve or a live pump. Hydraulic and pneumatic tooling handles cutting, brushing, and grinding tasks that used to require a diver hovering next to the workpiece for hours.

Dry hyperbaric habitats changed the weld itself. Inside a pressurized chamber, the arc behaves far more like a topside arc, weld quality improves, and the diver is shielded from current, cold, and visibility problems. Real-time gas monitoring in habitats and bells continuously tracks oxygen, carbon monoxide, and combustible gases instead of relying on scheduled spot checks. And for deep offshore work, saturation diving lets welders live in pressurized bays for weeks at a time, eliminating repeated decompression cycles and sharply reducing decompression sickness risk.

Case in Point: The MV Golden Ray Removal

Progress is not theoretical. During the wreck removal of the MV Golden Ray in St. Simons Sound, Georgia, commercial divers executed more than 1,000 dives safely using underwater thermal cutting and welding procedures. A project of that scale — wreck sections, tight spaces, heavy hot work — is exactly the environment where older practices failed. It ran because procedures were written first, verified constantly, and treated as non-negotiable by every diver on the team. The same discipline underpins the guidance in DAN’s commercial diving safety resources and the oversight described by NIOSH’s commercial diving program.

What Aspiring Divers Should Do Before Signing On

If you are considering the trade, you can screen employers the way they screen you:

1. Train at an ADCI-recognized program. Look for a course that covers surface-supplied systems, not just scuba, along with topside welding fundamentals such as those in an Aluminium Welding Course: What You Learn, Costs & Careers.
2. Get a proper diving medical. Fitness screening before hire and periodically thereafter is a cornerstone of ADCI Consensus Standards, and it protects you as much as the employer.
3. Ask how work dives are run. If the answer is scuba for anything beyond shallow, short, task-limited work, keep interviewing.
4. Ask to see the job hazard analysis, the lockout/tagout procedure, and the chamber plan. Vague answers here predict vague practices in the water.
5. Check the company’s record. Open OSHA citations and turnover tell you more than a recruiting brochure, and a culture that encourages stopping a job is worth more than a pay premium.

For the topside half of the career — shipyards, fabrication, hull and repair work — the standards mindset continues in material like AJ Marine Welding: Processes, Standards, Safety and Best Practices and Aluminum Boat Welding: The Complete Guide for Strong Hull Joints.

The honest summary: underwater welding is still statistically one of the most hazardous jobs on Earth, but the tools, procedures, and training that exist today remove the failures that produced most historical deaths. The remaining risk is concentrated wherever planning, verification, or equipment standards slip.

Conclusion: Is Underwater Welding Worth the Risk, and What Should You Do Next?

After everything in this article, the honest answer is a qualified yes—and the qualification matters. Underwater welding still sits near the top of the list of the world’s most dangerous occupations. The NIOSH analysis of OSHA records, the CDC’s ongoing surveillance, and the 2022 Cambridge and EMI Alliance review all point the same direction: commercial divers die at a rate many times the all-industry average, and drowning leads the list. What the same record also shows, going back to the 1984 Bradley study of Gulf of Mexico and North Sea diver populations, is that mortality has trended downward for decades as training, equipment, and operating discipline improved. The job is not a death sentence. It is a job with an unusually thin margin for error—and that margin is almost entirely a human product.

That is the most useful thing this topic has to offer. Review the fatality investigations one after another and the same failures recur: a fitness-for-dive evaluation that never happened, an energy isolation that was assumed rather than verified, a standby diver who wasn’t there, an emergency kit that wasn’t on the barge, a pump still running in the adjacent bay. The arc itself almost never kills anyone. People are killed by the water around the arc, by the pressure differential beside it, by the electricity feeding it, and by decisions made—or skipped—before the diver ever hit the waterline. CDC/NIOSH continues to track these deaths, and the pattern remains stubbornly consistent: drowning is still the dominant preventable cause.

What Should You Do Next?

The right next step depends on why you are reading.

If you are considering the trade. Start topside. Wet welding is shielded metal arc welding performed in all positions, in zero visibility, under a helmet you can barely see out of. Welders who show up to a commercial diving school with weak fundamentals are the ones who struggle most at depth. Build your arc skills first with resources like the All Position Welding Rod Guide, then decide whether the marine side of the trade is for you by exploring something like the Aluminum Boat Welding guide. After that, choose an ADCI-recognized diving program, get your hyperbaric medical evaluation, and ask any prospective employer direct questions before signing on: Is work done surface-supplied or on scuba? Is a recompression chamber on standby or a phone call away? What happened on your last recordable incident? If those questions make the employer defensive, that is your answer.

If you are a supervisor writing a job hazard analysis. Treat every item as a verification, not an assumption. Confirm energy isolation at the source, not from a paperwork trail. Survey differential pressure before anyone enters the water. Test breathing gas. Confirm communications. Confirm the standby diver is suited, connected, and briefed. Confirm medical fitness is current and documented. Every one of these steps maps directly to a requirement in OSHA’s commercial diving standard, 29 CFR 1910 Subpart T, which addresses nearly every hazard covered in this article—from explosive gas mixtures and delta P to electrical shock and decompression—through specific, citable rules.

If you are a family researching after an incident. Know that OSHA’s fatality investigation records are public, searchable documents. They name the equipment, the procedures, and the specific failures. They are the most reliable counterweight to speculation, and they turn a bewildering loss into something legible: what went wrong, and who was responsible for preventing it.

The Bottom Line

Underwater welding pays extraordinarily well precisely because it asks people to weld in the least forgiving environment on Earth. That premium is not free money—it is hazard pay, and it is only fair if the risks are named honestly, which is what this article has tried to do. The technology trend is genuinely encouraging: surface-supplied air replacing scuba for work dives, ROV-assisted tooling reducing diver exposure, dry habitats improving both metallurgy and safety, real-time gas monitoring, and saturation techniques making deep offshore work routine rather than heroic. But none of that technology matters on a job where the pump was never locked out.

If you take one thing from this article, take this: the divers who survive long careers in this trade are not the luckiest or the most experienced. They are the ones who verified, who briefed, who stopped the job when something felt wrong, and who refused shortcuts. Every fatality in the record was preventable. The industry’s job—and every individual diver’s job—is to make sure the next one doesn’t happen.

Frequently Asked Questions

How many underwater welders die each year?

Exact underwater-welder-only counts don’t exist because U.S. labor statistics group welders under commercial divers. OSHA data cited by NIOSH showed about 13 occupational diving deaths per year in 1989–1997, and BLS recorded 39 fatal injuries among commercial divers from 2011–2017, roughly five to six per year across the whole occupation.

What is the underwater welder fatality rate?

The best historical estimate comes from NIOSH: 49 deaths among roughly 3,000 full-time commercial divers over nine years, equal to about 180 deaths per 100,000 divers per year—around 40 times the all-industry U.S. average. A 2022 Cambridge review estimated 3 to 10 deaths per 10,000 divers annually across U.S. and U.K. data.

What is the leading cause of underwater welding deaths?

Drowning, followed by asphyxiation. In NIOSH’s 1989–1994 data, drowning accounted for 73% of occupational diving deaths and asphyxia another 14%. Fatal incidents typically involve breathing-gas loss, umbilical entanglement, or entrapment in currents and suction.

Can underwater welding cause explosions?

Yes. Electrolysis at the welding arc generates hydrogen and oxygen, and that mixture is highly explosive, especially inside enclosed spaces. OSHA requires closed compartments, structures, and pipelines to be vented, flooded, or purged with non-combustible gas before any welding or burning.

Is underwater welding electrocuted divers?

Electric shock is a genuine hazard from welding current, stray current in harbors, and faulty insulation. OSHA requires insulated gloves for the diver, a grounded welding machine frame, and a current-interrupt switch that is tended by a team member in voice contact and kept open except during active welding.

What regulations govern underwater welding safety?

In the United States, OSHA’s 29 CFR Part 1910 Subpart T, Commercial Diving Operations, sets requirements for dive team qualifications, pre-dive planning, hot-work procedures, decompression, and emergency response. Many employers also follow ADCI Consensus Standards, and state plans such as Washington’s DOSH add their own directives.

Do most underwater welding deaths happen offshore or inshore?

Both. The Cambridge review found the largest share of fatalities in the offshore sector, but inshore deaths are common in harbor work, bridge inspection and repair, nuclear plant water systems, and municipal infrastructure—where pump suction, confined spaces, and poor visibility dominate the hazard profile.

Is underwater welding being replaced by robots?

Not replaced, but increasingly assisted. ROV-mounted tooling, hydraulic systems, and automated tracking systems are used for long subsea pipeline girth welds, reducing diver bottom time. Most repair, inspection, and intricate fabrication work still requires a human diver with a stinger or torch.

How can I reduce my risk if I want to become an underwater welder?

Train at an ADCI-recognized commercial diving program, not a recreational scuba course, and complete a proper occupational dive medical before training. Evaluate employers on their safety record, surface-supplied equipment, standby diver practices, and willingness to walk away from a job that skips the job hazard analysis.