Submarine Cable Installation and River Crossings

By Ed Brannigan, Marine contractor. Reviewed by Dana Whitcombe, technical reviewer.

Marine Construction | Boat Lift Installation August 1987, I was standing on a spud barge in the St. Lucie River, watching a crew muscle a 4-inch HDPE duct across the channel with a winch line. The current was running 3 knots, and the drag on that duct was every bit of 8,000 pounds. That's when I learned you don't just pull a cable — you pull it over rollers, through a casing, and into a trench that's been jetted into the bottom. It's not a wire job. It's a marine construction job. Thirty years later, I've done a dozen river crossings for subsea cables, from fiber optics to 115kV feeders. The tools have gotten better, but the basics haven't changed. You still need to know the river, the bottom, and the limits of your gear. This is how I think about it. If you're reading this because you've got a cable to get across a river, listen. The last thing you want is to be the guy who hangs a cable on a snag mid-channel or leaves it exposed to a 5-ton anchor. Here's what I'd tell a young super.

Survey the Bottom First

Before I burn a single gallon of diesel, I spend at least a week on the bank. You're looking for old piles, wrecks, hard clay, and rock. I run a side-scan sonar and a sub-bottom profiler — something like a Klein 3000 or an EdgeTech 3100 — plus I pull a few vibracores. The core samples tell you whether you can jet or if you'll need a rock saw. Honestly, half of the river crossings I've seen fail came down to skipping the survey and guessing. Tides and currents matter too. On the Gulf, a river may run slack at high tide, but 30 minutes later it's moving 4 knots. That changes your pulling plan. I always rig with a current meter and a GPS, and I mark the channel with temporary buoys. We once lost a pull because a tug's wake caught our slack line and wrapped it on a snag. That was a long afternoon.

Pick Your Method: HDD vs. Open-Cut

For a river crossing, it's usually horizontal directional drilling (HDD) or a jet trench. HDD is cleaner if you've got the lay length. You bore a pilot hole from one bank to the other, ream it out, and pull the cable back through. I've used Ditch Witch and Vermeer rigs, and they'll do the job if the ground cooperates. But here's the thing: HDD isn't magic. You need a geotech report and a rig with enough pullback — I've seen jobs need 200,000 pounds. And if you hit cobble or a buried boulder, you can stall out. Then you're fishing a drill string. Open-cut is more direct. You dig or jet a trench across the bed, lay the cable, and cover it. For a small river, we'll use a long-reach excavator on a barge. For a big crossing, we might use a jet plow pulled by a winch. The catch is that you're disturbing the bottom, which means permits and, sometimes, a time window when the fish aren't spawning. On the St. Lucie we had to be out before manatee season. That's not a joke.

Pulling the Cable: The Part That Hurts

This is where the money gets made or lost. You pull from the opposite shore, not the reel side. The reel sits on the bank, the cable feeds through a chute, over rollers, and then a winch line — usually 1-inch wire or a synthetic rope — pulls it through the trench or duct. I always use a swivel and a pulling eye rated for the cable's maximum tension. The cable manufacturer gives you a number. For a typical fiber optic cable, it's around 600 pounds per inch of cable diameter. A 2-inch cable, you're looking at 1,200 pounds. But that's static. When the current grabs the cable, the tension spikes. I watch the tension meter like a hawk. We set a limit at 80% of the rated pulling tension. If it hits that, we stop, back off, and reassess. Too many guys try to haul through a problem and end up with a stretch in the cable that turns into a break later. I also keep a spotting boat with a diver dressed out, not to work but to be ready. Last thing: don't forget the lubricant. Coconut-based pulling lube, applied every 10 feet as the cable leaves the reel. That stuff cuts friction by a third, easy. One race I never win is when the cable hangs on a clam shell or a snag. You feel it in the winch. The line goes tight, then the tension spikes, then it goes slack. That's a break. I've seen it happen twice in my career. Both times, no one died, but the project cost doubled. Slow is smooth, smooth is fast.

Burial and Protection: Don't Get Cheap

After the cable is in place, you've got to bury it. For a navigable river, the spec usually calls for 4 to 6 feet of cover below the bed. But I've also used concrete mattresses on top of the backfill, especially near bridge piers and approaches. Those mattresses — 4 feet by 10 feet, filled with grout — weigh in at about one ton each. We lower them with a crane barge, and they keep the cable from being exposed by scour. In faster currents, I've had to lay rock riprap over the mattress. That's an extra cost, but you don't want to be the guy explaining why a barge anchor cut the town's internet. Depth of burial is a judgment call. In sandy bottoms, a jet plow can take a cable down 8 feet without a problem. In hard clay, you might get 3 feet and then you're just flushing. That's when you need to either change the route or add protection. Call it 6 feet in a high-traffic river, but I've done crossings in canals where 3 feet was fine. Depends on the plant. I always leave a measured as-built with coordinates and depth readings every 25 feet. You might think that's paperwork, but when a dredge comes through in 10 years, that document is what saves your cable from a cutter arm.

Testing and Handover: No Guesswork

Before I hand over a job, the cable gets testes — mech... Tested. For fiber, that's an OTDR trace and a power meter reading. For power cable, we do a hi-pot test and measure insulation resistance. The numbers have to match the factory sheet. I've seen installations where the cable passed the pull but failed testing because the jacket was nicked during handling. That's on the crew, and we fix it with a joint. Joints are the weak point, so I try to avoid them on a crossing. If you need a joint, you better order the proper kit from the cable maker — Prysmian, Nexans, whoever — and have a certified splicer do the work. A joint fatter than the cable can hang on a roller or get damaged during pullback. That's why we test every joint after installation. And we leave spare cable at both ends, coiled up in a manhole or hand hole, so a future repair isn't a nightmare. Documentation isn't glamorous. But that's the difference between a professional crossing and a hobby job. I've seen too many cables laid with zero as-built, and then when a problem shows up, the owner has to run a fishing tool and guess. Don't be that guy.

Frequently Asked Questions

How deep do you bury a submarine cable under a river? For a navigable river, I aim for 4 to 6 feet of cover below the natural bottom. In high-traffic anchorages or near bridge piers, I may go deeper or add concrete mattresses. It really depends on the bottom type and what the buried hazards are. What's better for river crossings: HDD or open trench? They both work. HDD is cleaner — no open cut, no disturbance of the riverbed. But it needs long lay lengths and good ground conditions. Open trenching or jetting is simpler and works in almost any bottom, but you may need environmental permits and more surface equipment. Can you repair a submarine cable if it gets damaged later? Yes, but it's expensive and takes time. You need a cable joint from the original manufacturer and a certified splice team. That's why we over-protect crossings with deep burial and concrete mattresses. A repair on a river crossing can cost six figures and take weeks. What is the pulling tension limit for a submarine cable? Cable makers give you a maximum pulling tension, usually in pounds per square inch of cable cross-section. I use the rule of thumb around 600 pounds per inch of cable OD, but I set the winch limit at 80% of that. Anything higher risks stretch damage that won't show up until later.