Debris Field Mapping After a Storm
By Gus Marchetti, Salvage master. Reviewed by Amanda McCallister, editor.
Marine Salvage & Recovery | Underwater Inspection The morning after Hurricane Michael, I was watching a Klein 3000 side scan echo through a 45-foot workboat, off Panama City, Florida. The water was coffee-on-cream, and the screen was showing a shadow that could have been a wrecked sailboat or a pile of tires. I marked it and kept going. That's how debris mapping starts - cover the grid first, then go back and look. After a storm, we don't know what's down there. We know what was on the docks, but not what the surge ripped loose and rolled along the bottom. A debris field survey is the act of turning that unknown into a chart, so a diver doesn't get fouled and a dredge doesn't chew through a cable. It's not glamorous, but it's essential. By now I've done this for twenty-two years, and I can tell you the secret: nothing replaces a good sonar run with a measured plan. You can't guess. You make a grid, you keep the sensor at the right height, and you record everything.
Before you start, collect the breadcrumbs
First thing I do is hit the charts and the harbor master's office. If there were moorings, dock lines, boats, traps - any of that - I want to know where they were, because they're not there anymore. We also look at the last few days of wind and current models. If the storm came from the southeast, debris is likely scattered in the west-northwest quadrant of the harbor. That's not a hard rule, but it tells you where to focus. Then I deploy a current meter at the survey site, a simple in-situ unit like the Nortek Aquadopp. Debris moves with the tide, so you need to know if the bottom sand is shifting. If the water's still roaring, you might not be able to keep your towfish at altitude. Unless it's dead calm, we tow at a slower speed, call it 2.5 knots, to keep the line from dragging. It's tedious, but it's the difference between a clean image and a smudge.
Pick your sensor: side scan is the workhorse
The tool I reach for first is a side scan sonar, something like an EdgeTech 4200MP. It fires a fan-shaped pulse to both sides and paints a picture of the seafloor. At 300 kHz you can cover a swath of 100 meters per side, but you lose the small stuff. At 600 kHz you get better resolution but less coverage - call it 50 meters per side. For storm debris, I usually run 600 kHz to catch the six-foot kayak and 55-gallon drums. But side scan responds to anything with an acoustic reflectivity. A plastic pipe will show up, a tire will too. If you're looking for steel - a lost anchor, a steel hull, a shipping container - run a magnetometer alongside. I've used a Marine Magnetics SeaSPY and it's a simple thing: tow it low, it reads total magnetic field, and a huge spike tells you there's ferrous metal under the line. Neither is perfect. Side scan can miss a cleanly buried cable, magnetometer can't tell you shape. Run them together and you get the best shot.
Lay out your lines like you're mowing a lawn
I keep a grid in my head. You start with a baseline along the main axis of the area - usually parallel to the channel or shoreline. Then you run lines at spacing that depends on the plant - if you're surveying around a power plant's discharge, you want tighter lines near the structure to catch everything in the thermal plume. But for a general debris field, I'll set the line spacing at 20 meters with a 50-meter swath per side. That gives you a 100-meter coverage, so you've got about 80% overlap. That's a lot of data, but you'd rather have it. Speed matters too. We stay around 3 knots, give or take half a knot depending on the bottom. Too fast and the sonar image gets smeared; too slow and you're wasting fuel and not covering ground. One digression: I remember a survey off the coast of Mobile after a big blow, where we spent a whole afternoon scanning an area that turned out to be full of artificial reefs - old cars, tires, a water heater. The contact list looked like a car crusher dump. But because we had them charted, the client could avoid them or remove them properly. The point is, you don't know what's down there until you look.
Process the data and make a target list
When the survey's done, I pull the raw files into SonarWiz and run through the whole mosaic. The software stitches the track lines together, and I mark every contact that's bigger than, say, a bicycle. I measure its height from the shadow length. That gives me a report: each contact has a position, a size estimate, and an acoustic shadow. I put that in a shapefile for the client's GIS system. If it's a navigation hazard, it goes on the chart as an obstruction. But you can't trust everything, especially in sediment or weeds. A patch of kelp can look like a wreck, and a natural rock formation can be mistaken for a dumpster. So I run through the mosaic twice, once for the big obvious things and once for the little blips. Then I rank them: what's a real hazard, what's maybe a hazard, and what's probably natural. You have to be honest about uncertainty. The report says 'contact estimated height 1.2m' not 'this is a car.'
Verify with divers or a small ROV
Nobody in the salvage business trusts a sonar target until someone has laid eyes on it. For the top three or four hazards on my list, I'll send a diver with a dive torch and a line. If the water's too deep or too dirty, a VideoRay ROV does the job. Put the ROV at the marked position, fly a small circle, and get a video of what's actually there. That's the ground-truth step, and it's non-negotiable. A good diver can also attach a lift bag, if that's the plan, but that's for a different article. The point is you don't do the removal until everything on your map is verified and the coordinates are tagged. I've seen too many jobs skip this step and end up with a diver wrestling an anchor chain. Save yourself the trouble. Map it, then verify.
Frequently Asked Questions
How long does a storm debris survey take? For a typical harbor channel of about half a square mile, with a side scan at 4 knots and 20m line spacing, you could do it in a day of steaming. Add a day for processing and an extra day for ground truthing. Give or take. What's the smallest thing you'll catch on a side scan? At 600 kHz with a 50m swath, you can see objects 10-20 cm across if they stand up enough to cast a shadow. A tire, a bicycle, a small anchor. For smaller stuff, you'd need a higher frequency, but you don't usually need to find every beer can. Can you use a fishfinder for this? No. A single-beam only gives you the depth directly under the boat. You'd miss 99% of the debris. You need something that fans out to the sides - that's why side scan sonar is the standard. Do I need an ROV to verify targets? Depends on the size of the target and whether it's steel or not. A magnetometer can tell you if it's ferrous, but it won't tell you if it's a boat or a car. A diver is cheap for the first inspection. For deep water or hazardous conditions, an ROV is worth the money.