GVI and CVI are the two basic levels of underwater visual inspection. The general one is done from a distance, without cleaning, to pick up obvious damage and decide where to look more closely. The close one is done at short range, on a cleaned surface, and looks for specific defects such as cracks. The difference between the two is one of method and preparation, not of eyesight.
A good share of the hours worked in commercial diving worldwide aren't spent building or repairing anything. They're spent looking. Structures in plataformas offshore, hulls, quay piles, gates, mooring lines: everything that's submerged and carries load has to be checked periodically, and someone has to go and look at it.
That work is called visual inspection, and it doesn't just mean looking. It means looking in a defined way, and keeping a record of what was looked at, how, under what conditions and with what result.
Two levels, two purposes
GVI: general visual inspection
The general inspection is done from some distance and without preparing the surface. The inspector works over the structure looking for whatever stands out: missing or displaced components, deformation, impact damage, loose elements, spent anodes, excessive marine growth or, conversely, unusually clean patches, which often give away rubbing.
Its job is to sweep. It's not trying to find fine defects: it's trying to cover a lot of surface and decide where it's worth looking closely. It's fast, relatively cheap, and often carried out by ROV at depths where diving isn't viable.
CVI: close visual inspection
The close, or detailed, inspection is done at short range and on a cleaned surface. This is where specific defects are looked for: cracks, gouging, loss of material, weld defects, localised corrosion.
The key word is cleaning. A CVI starts by stripping off marine growth, loose rust and anything else standing between the eye and the metal, until the area is fit to be examined. Without that preparation there's no CVI worth the name, and that has to be stated in the report if for whatever reason it couldn't be done.
How the surface is prepared
Cleaning is done with whatever means the plan calls for and the surface can take: hand or rotary brushes, scrapers, pressure water jetting. The choice isn't trivial. A method that's too aggressive can damage the protective coating, and a damaged coating that no one repairs becomes a corrosion hotspot created by the inspection itself.
The extent of the cleaning is also defined in advance: how many centimetres either side of the weld seam, to what finish, checked how. All of that is in the procedure, and the diver carries it out exactly as written even if it seems excessive.
Cleaning is also the physical labour of the inspection. It's where the bottom time goes and where the diver wears out. A solid point to brace against changes performance completely, which is why long campaigns rely on work platforms.
The record: what's actually delivered
The client isn't buying a dive: they're buying a report. And an inspection report is worth exactly what its traceability is worth.
Every observation has to be identified against the exact component it belongs to, its position and its orientation. Large structures have their own reference systems, with numbered nodes and elements, and using that naming isn't a formality: it's what lets this campaign be compared against the last one and shows whether something has changed.
Photographic and video recording always goes along with it. It's shot with the helmet camera, with a hand-held camera when the detail calls for it, and with scale markers in frame so that whoever views the image knows the size of what they're looking at. Lighting matters: a raking light reveals relief and cracks that a front-on light flattens out completely.
And it's narrated. The inspector describes out loud what they're seeing as they see it, and that narration is recorded along with the video. It's one of the reasons communication with the surface is treated as part of the inspection team, not an accessory.
Where visual inspection stops
Visual inspection has a clear limit: it only sees what breaks the surface with enough of an opening. A very tight crack, a subsurface defect or internal loss of thickness won't show up just from looking, however carefully you look.
That's what non-destructive testing is for, which uses other physical principles and requires specific qualifications. It's the next level, and it's often triggered precisely because a CVI found something that needs confirming.
Thickness measurement and checking the cathodic protection system usually go in the same work package, because they make sense at the same points and use the same cleaning.
Conditions call the shots
There's one factor no procedure can eliminate: the water. An inspection done in two metres of visibility and one done in twenty centimetres aren't the same inspection, even if the report uses the same words.
That's why conditions are recorded just like findings. Visibility, current, temperature, sea state and anything else that limited the work are part of the record, and not as an excuse but as a technical fact: whoever reads the report in two years' time needs to know under what conditions what they're reading was obtained.
In very poor visibility, the work falls back on tactile inspection: the inspector runs a gloved hand over the surface looking for steps, discontinuities and changes in texture, and describes over the radio what they feel. It's a legitimate method, widely used in ports and rivers, but it's different from a visual inspection and has to be stated as such. Calling a tactile inspection a CVI is one of the things that wrecks a contractor's credibility.
Who can do it
Not just any diver can inspect in the professional world. Underwater inspection schemes require a recognised inspector certification, on top of a diving qualification, which the client checks before accepting anyone onto the project. The best-known route in the offshore world is the certified underwater inspector, and the specific options depend on the market you're working in, as detailed on the page about certifications.
There's a good reason this specialism is so worth pursuing. Inspection is recurring work by nature: as long as the structure exists, you have to keep coming back. And it's less physically demanding than heavy construction work, which makes it one of the natural routes to extending the professional life of a diver.
How an inspection campaign is planned
An inspection campaign doesn't start in the water. It starts with the plan: which components need checking, at what level, in what order and against what acceptance criteria. That plan is drawn up by engineering from the structure's history and the approved inspection regime, and it reaches the dive team as a closed work package.
The diver or inspector studies it before boarding. Knowing in advance which node is number 14 and how to get to it saves minutes of bottom time that cost money, and avoids the classic mistake of surfacing with video of the wrong element. When the structure is already known, the prep work includes reviewing the previous campaign's findings: whatever was already flagged gets revisited first, because how a known defect has evolved matters as much as a new one showing up.
The most common mistake
The typical failing of an inexperienced inspector isn't missing a defect. It's giving an opinion.
The report describes: what was seen, where, under what visibility and cleaning conditions, with what means, and with what result. Whether that compromises the structure is for the responsible engineer to judge, with the design, the loads and the history in front of them.
An inspector who writes "not significant" on a finding has made a call that wasn't theirs to make, and has contaminated the data for good. The one who describes precisely and keeps the judgement to themselves is the one who ends up with work indefinitely.
Frequently asked questions
What do the acronyms GVI and CVI mean?
GVI stands for general visual inspection and CVI for close visual inspection, or detailed inspection. They're the two basic levels of visual inspection used on submerged structures.
What's the practical difference?
A GVI is done from a distance and without preparing the surface: it picks up obvious damage, missing parts, deformation or abnormal marine growth. A CVI is done at short range on a surface previously cleaned down to the metal, and looks for fine defects such as cracks.
Why does the surface need cleaning before a CVI?
Because a fine crack is invisible under marine growth, rust or lifted paint. Without proper cleaning, a CVI isn't a CVI: it's a GVI done up close, and that has to be stated in the report.
Does a CVI catch every crack?
No. Visual inspection only sees what breaks the surface with enough of an opening. For very tight surface defects or subsurface ones, non-destructive testing is used, which is a different level and requires specific qualification.
What qualification is needed to inspect?
Underwater inspection schemes require a client-recognised inspector certification, on top of a diving qualification. It's a common specialisation route for a commercial diver and one of the ones that best extends a career.
Who interprets the findings?
The inspector describes and records; structural assessment is for the responsible engineer. Mixing up those roles is a classic mistake: the report has to say what was seen and under what conditions, not whether the structure is fit for service.
Published 1 September 2026 · By Arthur More, professional saturation diver
Photo: U.S. Navy Seabee Museum · CC BY via Openverse