Underwater Mapping and What It Is For

As of 2026, 28.7 percent of the world's ocean floor has been mapped to modern standards. The Nippon Foundation-GEBCO Seabed 2030 project added almost five million square kilometres in the preceding year, with data now coming from more than 185 organisations.

That leaves roughly seventy percent of the planet's solid surface unmeasured at any useful resolution.

What "mapped" means here

There is a global depth grid covering the whole ocean already, and it is mostly not measurements. Much of it is derived from satellite altimetry, which infers seabed shape from tiny variations in sea surface height caused by the gravitational pull of underwater terrain. A seamount pulls water toward it and raises the surface by a few centimetres.

It works, at kilometre resolution. A shipwreck, a cable route hazard, a hydrothermal vent field: none of them show up at that scale, and direct acoustic measurement is the only way to see them.

How multibeam works

A multibeam echosounder fires a fan of acoustic beams across the vessel track and times each return. Instead of one depth under the hull, you get a swath.

Swath width scales with water depth, typically two to twelve times depth depending on the system and the beam angle. Some Kongsberg systems reach 5.5 times water depth, giving coverage over 4,400 metres wide in deep water. Deep water is therefore mapped faster per line, which is a pleasant surprise the first time you encounter it.

Frequency sets the trade. Lower frequencies around 200 kHz reach deeper and cut wider swaths at coarser resolution, while higher frequencies above 400 kHz give denser soundings and sharper detail across a narrower swath at shallower range. Resolution lands between 0.5 and 5 metres per sounding.

Shallow water is the slow, expensive problem. A survey in 20 metres of water gets a swath perhaps 100 metres wide, so covering a coastal area means running a great many closely spaced lines at survey speed. Coastal mapping costs far more per square kilometre than deep ocean does.

Why altitude matters more than depth

Resolution depends on the distance between the sensor and the seabed, not on the depth of the water. A hull-mounted system in 3,000 metres of water is 3,000 metres from the target.

Put the same sonar on an AUV flying 50 metres off the bottom and the geometry changes completely. This is the entire argument for deep-water autonomous survey, and it is why detailed mapping of abyssal terrain waited for vehicles that could get down there and stay.

What the maps get used for

Habitat mapping is the conservation application people usually mean. Cold-water coral, seagrass, kelp and reef structure all carry acoustic signatures, and backscatter alongside bathymetry gives a workable picture of substrate across an area. Protected area boundaries drawn without it are drawn on guesses.

Baseline monitoring matters as much. Repeat surveys over the same ground show change: scour around infrastructure, sediment transport, trawl damage on the seabed, recovery inside a protected area. A single survey is a snapshot and a hypothesis. Two surveys are evidence.

Then there is the infrastructure work that pays for most of it. Cable routes, wind farm foundations, pipeline corridors and port approaches all require detailed bathymetry before anything gets built, and commercial survey generates far more data than science budgets ever could. A significant part of the Seabed 2030 contribution comes from industry data released after its commercial value has expired.

Why the gap persists

Ships. Multibeam survey requires a vessel on station moving at survey speed, and vessel time is the constraint on everything.

Uncrewed surface vessels are changing that arithmetic by removing the crew and much of the fuel, and satellite-derived bathymetry handles clear shallow water reasonably well. Neither replaces a hull-mounted multibeam over deep ocean. At current rates the 2030 target will be missed, which does not make the effort less worthwhile.