Non-destructive inspection of submerged concrete combines visual and tactile inspection with sounding, ultrasound, cover measurement and corrosion potential. Underwater, almost every method loses reliability compared with its dry-land version, so the value of the work lies in the prior cleaning, the positional referencing and the honesty of the report.
There's a lot of concrete underwater, and almost all of it is ageing. Caisson quays, jetty piles, dikes, dam faces, water intakes and outfalls, gravity-based foundations, coastal defence works and offshore wind turbine foundations. All of it was built with an intended service life, and much of it has spent decades working in the worst possible environment for the material.
The owner wants to know what condition it's in without taking it out of the water, without stopping operations and without breaking anything. That's where the typical submerged non-destructive inspection job comes from. And that's also where the first thing that needs saying clearly comes from: underwater, most non-destructive tests give less information, and with more uncertainty, than they do on dry land. The job is knowing exactly how much less, and saying so in the report.
What happens to concrete underwater
Understanding the deterioration mechanisms is what lets you decide what to look for and where. In a marine environment the main ones are few and well known.
The most important is chloride attack. Seawater carries chlorides that penetrate the concrete by diffusion, reach the reinforcement and destroy the passive layer that protects it. From there the steel corrodes, the rust takes up more volume than the original metal, and the concrete cracks and spalls. That spalling that leaves the bars exposed is the final state of a process that's been under way for years without being seen.
Carbonation works through a different path — a loss of alkalinity caused by carbon dioxide — and is characteristic of the zone exposed to air, not the permanently submerged one. That's why it matters where you measure: on a marine structure, the tidal and splash zone is usually in worse condition than the part that never leaves the water, because it combines available oxygen, moisture and wetting-and-drying cycles. The permanently submerged zone has plenty of moisture but little oxygen, which slows down active corrosion.
On top of that come abrasion from suspended sand and wave action, impact from vessels and objects, chemical attack from sulphates, leaching in freshwater hydraulic works, and scour at the foot of the structure, which doesn't damage the material but changes the structural model and can be the most serious thing the inspection finds.
Visual and tactile inspection is still the foundation
No instrumental test replaces a diver working over the structure with judgement. Visual and tactile inspection picks up cracking and its pattern, spalling, exposed reinforcement and how much section it's lost, open joints, voids and honeycombing, out-of-plumb sections, loss of material from abrasion, and the condition of the ground at the foot of the structure.
The problem is that underwater, "visual" is an optimistic word. In many ports visibility is measured in centimetres, and inspection is tactile for a high proportion of the time. On top of that, the surface is colonised: algae, mussels, barnacles, calcareous tubeworms. A few centimetres of fouling completely hides a crack, spalling or an exposed bar.
Hence a rule that separates serious work from the rest: no cleaning, no inspection. The surface has to be cleaned down to the concrete in the areas being inspected or measured, and how it was cleaned has to be recorded, because the cleaning method affects what you see and can even alter the surface. Aggressive cleaning on already-degraded concrete strips away material and skews the assessment.
Methods usable underwater, and their real limits
Sounding
Tapping the surface with a hammer and listening to the response is the cheapest method and one of the most useful. A hollow sound gives away delamination: a zone where the cover has separated from the core, typically caused by corrosion of the reinforcement underneath. Underwater, sound travels differently, and a diver, wearing a helmet in a highly conductive medium, hears it differently from someone on the surface; the difference between a solid and a hollow sound is still detectable with practice, but the method is qualitative and depends on the operator. It's used to map suspect zones, not to quantify anything.
Rebound hammer
The Schmidt hammer is a classic tool on dry land and a source of problems underwater. It measures surface hardness, and its correlation with strength is indirect and sensitive to conditions. Water saturation alters the response, the submerged surface is rarely in the state the correlation assumes, and a diver's stability in holding the instrument perpendicular and firmly supported isn't that of a technician on land. It can be used to compare zones against each other within the same structure on the same day, but turning those readings into a characteristic strength is a mistake an honest report doesn't make.
Ultrasonic pulse velocity
Measuring the propagation velocity of an ultrasonic pulse through concrete does give information about homogeneity and about internal defects, voids or deep cracking. It needs access to two opposite faces for direct transmission, which is the reliable configuration, and that limits its use to elements accessible from both sides. In indirect transmission, on a single face, interpretation gets harder. Coupling the transducer underwater has the advantage that the medium is already liquid, but it needs a clean, regular surface. The velocity measured also depends on moisture content, which in a submerged element is at saturation, and that has to be taken into account when comparing against readings taken dry.
Corrosion potential
Half-cell potential measurement estimates the likelihood that the reinforcement is actively corroding, by comparing the potential of the steel against a reference electrode. On marine structures it's one of the most useful measurements because it detects activity before visible damage appears. Underwater it has one important quirk: seawater is an excellent, highly conductive electrolyte, which makes the measurement easier but also flattens the potential differences between zones, and the interpretation thresholds developed for structures in air don't carry across directly. It also needs an electrical connection to the reinforcement, which means exposing a bar somewhere, and it's skewed if the structure has cathodic protection installed. The correct reading is in terms of gradient and relative map, not an absolute value against a table.
Cover measurement and reinforcement location
The covermeter locates bars and estimates cover thickness by electromagnetic induction. It works reasonably well submerged if the equipment is built for it, and its value is twofold: it feeds the diagnosis — shallow cover explains early corrosion — and it stops a bar being cut later when a core needs extracting or a repair needs doing.
Thermography and ground-penetrating radar
Infrared thermography, which on dry land detects delamination through differences in solar heating, doesn't work underwater: water evens out the temperature and absorbs infrared radiation. Ground-penetrating radar can be used in a submerged version, but the high conductivity of seawater heavily attenuates the electromagnetic signal and reduces penetration; it performs better in fresh water. It's a specialist method, with specific equipment, and its performance depends as much on salinity as on the moisture content of the concrete itself.
When the non-destructive test isn't enough
Almost every serious diagnosis ends up needing a targeted destructive check: core extraction. A core lets you test compressive strength on real material, measure the chloride penetration profile at different depths, check the depth of carbonation and examine the microstructure. It's the only way to turn a qualitative map into defensible quantitative data.
The right logic is to combine them: non-destructive methods locate and prioritise; cores, few and well chosen, calibrate. Extracting cores underwater has its own complications — fixing the drilling rig, torque reaction, recovering the sample without losing it, and sealing the hole afterwards — but it's known work and forms part of the standard kit of an underwater inspection company.
The underlying problem isn't the method: it's the medium
It's worth being explicit about why all of this is harder than on land.
Visibility limits visual inspection and complicates any kind of positioning. Current pushes the diver and the instrument around; a test that requires constant, perpendicular pressure degrades when the operator is an unstable platform. Access to the underside of a slab or the inner face of a caisson can be poor or non-existent. The material's moisture saturation affects several methods at once. And there's a time limit: the diver has bottom time, and every measurement costs more than it does on land.
The practical result is that the realistic density of measurements underwater is much lower than in an equivalent inspection on land, and that extrapolating from a few points to the whole structure is the weak point of every one of these reports. Acknowledging that doesn't take value away from the work: it's what makes it usable.
Documentation: where the job is won or lost
A test with no traceability is worthless. Everything measured underwater has to be locatable afterwards on a drawing, and that means setting up a reference system before starting: a grid physically marked on the structure, or references to identifiable elements — pile number, joint, level relative to a known datum — with one consistent approach across the whole job.
The record has to include identification of the equipment used and its current calibration, the inspection conditions (visibility, current, sea state, tide level), the cleaning method applied, the position of each measurement, the raw value obtained and not just the interpreted one, and associated photography or video. Whoever signs the report must be identified, and their qualifications — in inspection and, where relevant, in non-destructive testing — documented. All of this is structured much like any inspection report from a commercial diving company: what's specific is the testing, not the format.
The report also has to clearly separate three things that often get mixed up: what was observed, what was measured, and what was concluded. And it has to state what couldn't be inspected, and why. A drawing with zones marked as inaccessible is valuable information; a report that leaves them out is a report that misleads.
Who does this work
Inspecting submerged concrete sits on the boundary between two trades. On one side, commercial diving, with its training, its safety organisation and its ability to work efficiently underwater. On the other, structural engineering and non-destructive testing, with their own specific qualifications.
In practice it's resolved in one of two ways: a diver with additional training in inspection and testing, or a mixed team where the technician directs from the surface via video and comms while the diver carries out the work. The second option is common on large projects; the first is what opens up an interesting career path for someone who already has the basic commercial diving training and wants to specialise in something paid for by judgement, not just bottom hours. Internationally recognised underwater inspection qualifications are the usual route, and they're covered among the sector's certifications.
It's also work with staying power. Maritime infrastructure built in recent decades is entering the age where maintenance stops being optional, and an owner who needs to decide whether to repair, reinforce or replace needs a diagnosis first. That diagnosis always starts with someone in the water, cleaning a square metre of concrete to see what's underneath.
Frequently asked questions
Can concrete strength be known without extracting cores?
Not reliably. Indirect methods such as the rebound hammer or ultrasonic pulse velocity give indications of homogeneity and let you compare zones against each other, but turning those readings into a characteristic strength requires calibrating them against real cores from the same structure. Underwater, that uncertainty is even greater.
Why is cleaning before inspecting so important?
Because a few centimetres of algae, mussels or barnacles completely hide a crack, spalling or an exposed bar, and no instrument reads well over a colonised surface. Cleaning has to reach the concrete in the areas being measured, and how it was done has to be documented, because an aggressive method on already-degraded concrete can strip material away and skew the diagnosis.
Does corrosion potential measurement work in seawater?
It works, but it's interpreted differently. Seawater is highly conductive, which makes measurement easier but flattens the differences between zones, so the thresholds designed for structures in air don't transfer as-is. It's read as a relative map and gradient, not an absolute value, and it isn't valid if cathodic protection is active.
Which part of a marine structure is usually worst?
Usually the tidal and splash zone, because it combines moisture, available oxygen and wetting-and-drying cycles. The permanently submerged part has plenty of moisture but little oxygen, which slows down active corrosion. An inspection that only looks underwater can miss the most damaged zone.
Can any commercial diver do these tests?
The diving part, yes; the testing part needs specific training. On site it's resolved with a diver who has added underwater inspection and non-destructive testing qualifications, or with a mixed team where a technician directs from the surface by video and comms while the diver carries out the measurements.
What should the report always state?
What was observed, what was measured and what was concluded, kept separate; the equipment used and its calibration; the actual inspection conditions; the positional reference system; and, very specifically, which zones couldn't be inspected and why. A report that leaves out the inaccessible zones gives a false sense of coverage.
Published 10 September 2026 · By Arthur More, professional saturation diver
Photo: NOAA Photo Library · CC BY via Openverse