Culvert and Outfall Inspection
By Ray Whitfield, ADCI-certified commercial dive supervisor. Reviewed by Dana Whitcombe, technical reviewer.
Underwater Inspection | Commercial Diving Contractors Alabama The day was grey and about 45 degrees. I was suited up on the upstream face of a 40-year-old hydro plant, and the intake was running 2,000 CFS. My job was to assess the condition of the 10-foot concrete culvert that carried the tailrace under the powerhouse. I had to do it blind if the silt hadn't settled. Culvert and outfall inspection is a mix of old-school feel and new-school tools. It's not a glamorous gig, but it's the kind of work that keeps bridges, dams, and discharge lines from becoming news stories. So if you're planning one, here's how I run a job.
Pre-dive planning and site assessment
Before we splash, I'll call it a half-day of homework. I pull the as-builts from the owner, check the age of the concrete, and the original mix design if they have it. I also look at the last few inspection reports from the plant engineer. Then we talk about flow: can we reduce it? More than once I've had to talk a plant operator into dropping the discharge to 1,000 CFS so I could get within working distance. That's a negotiation, not a demand. And the safety case is clear: a 60-foot, 30-inch pipe with a 10-foot head has enough energy to trap a diver. I also walk the site from the outfall end. If the receiving stream is turbid, we may need to shut down or schedule for low flow. Those are decisions the plant operator makes, not me. But I'll state my minimum conditions in writing. That leads to a small digression: I once got called to a culvert under a railroad embankment where the owner hadn't shut anything off. When we jumped in, the current was ripping. We managed to tie off a surface marker buoy and work upstream in the dead zones, but the whole time I'm thinking, "This is how people die." That's not the kind of hero story it sounds. It's the opposite.
Dive gear and survey tools
For the vast majority of culvert work, I run a Kirby Morgan KMS-48 helmet with a zip suit for the dive side, and a full redundancy on the umbilical. Air is fine for shallow work, but if I'm going deeper than 30 feet or working in silty conditions, I'll switch to a nitrox mix to keep bottom time reasonable. The real tool kit is the survey equipment. We usually deploy a Teledyne BlueView M900 scanning sonar from the surface or on a pole, but for close-up structural work, I'll carry a hand-held sonar unit with a stabilized head. We also use a Garmin GPS to map the intake and outfall locations when the culvert runs offshore or across a lake. Camera systems? We use a clean-water video head, a 4K camera in a waterproof housing from Light & Motion, and a set of 6-inch underwater lights. And we never go in without a tape measure and a sounding line. The sonar gives us a map, but the tape gives us dimensions. We also measure water quality parameters: dissolved oxygen, pH, temperature, and turbidity using a Hach probe. That's not just environmental ticking-the-box — it tells us if the metal in the culvert is being actively corroded. If the pH is below 6.5, I'm looking at a different level of galvanic attack.
The survey itself
On the dive, we work a grid pattern. We set a baseline along the centerline of the culvert and take measurements at 5-foot intervals. We record joint spacing, offset, and any signs of debris or sediment buildup. We note the concrete condition: honeycombing, spalling, exposed rebar, cracks — their width, orientation, and activity. A static hairline crack is one thing; a crack that's actively moving is a call-out. Then we move to the outfall, where erosion often does the worst damage. I've seen culverts with the bottom half completely scoured away, and the pipe hanging like a birdhouse over a hole. That's why we survey the invert as carefully as the crown. And we look at the receiving channel: is there a scour pool deeper than the pipe's bottom? Is the slope starting to armor downstream? The culvert ends at the outfall, but the engineering problem doesn't. We always capture video and stills, with a real-time commentary, so the owner knows exactly what we saw. And I'll tell you right now — some divers don't narrate. They think it's an afterthought. No. That narration is the product. The report is the product. The video is the product. You have to bring that back.
Common defects and what they mean
Around culverts, I see a few types of failure over and over. First is differential settlement — the pipe moves more at one end than the other, and you get step-type offsets at the joints. Second is interior abrasion: sand and aggregate and debris, moving at high velocity for years, can wear through steel or concrete. I remember a 36-inch corrugated metal pipe under a county road where the invert was worn through to soil in two places. The road was still open, but it wouldn't be for long. Third is infiltration or exfiltration — water pulling soil into the pipe. That leads to voiding under the structure, and eventually, a sinkhole or collapse. We look for fine sand piles in the invert, or discharge that's grit-laden. And we measure, if we can, inflow from outside using a flow meter — but I'm honest: that measurement is rough. Call it a field estimate, not a certified flow measurement. The point is to flag the condition. And then there's the obvious: debris jams, beaver dams, tires, tree limbs. I always carry a grip on schedule but I don't force them. Some debris protects the invert; you have to judge what's harmful and what's better left alone.
Reporting and follow-up
The report I write after the dive is what the owner gets on their desk. It includes the inspection date, diver's name and ADCI certificate number, conditions (visibility, flow, water quality), then a section on each defect, a photo reference, and a priority rating. I use a simple A-B-C system: A means immediate action needed, B means repair in the next cycle, C is monitor. That's not a formal ASTM rating, but it's clear enough for a plant engineer to put in a work order. I also recommend a follow-up frequency. For a well-built reinforced concrete culvert, every 5 years is usually fine. For an older metal pipe or an outfall in aggressive soils, every 2 years. Depends on the plant, of course. And I keep one copy for my files, because in 20 years someone will call me and ask what we saw that day.
Frequently Asked Questions
What pricing model should I expect for an underwater culvert inspection? It's usually per day, per tank, or per job, depending on the crew and the equipment. Depth, current, access, and whether you need sonar or an ROV all move the number. I'd ask for a quote that lists mobilization, diver time, and reporting separately. Can you inspect a culvert without a dive team using a camera? If the culvert is small, dry, or free-flowing, a remotely operated vehicle like a VideoRay or a pan-and-tilt camera on a pole can get decent visuals. But you can't feel around joints or measure abrasion contact. For a critical asset, I want a diver in there. How do you handle zero visibility? Zero-vis is just another Monday. We use sonar to map the overall shape, then do a tactile inspection by hand, working a systematic grid. We also have a way to move the diver's helmet so the water clears just enough to see the pipe wall for a few inches. It's slow, but it works. Do you need to take the structure out of service? Not always. If flow can be reduced to under 500 CFS, a diver can work inside a large culvert. But there's always a risk assessment and a standby line. For small culverts, we often need to blind off one end or put in a dam. That's a plant decision, and you'll want to involve the dive supervisor early.