Chapter A · Radar altimetry, bathymetric LiDAR, multispectral bathymetry, multibeam sonar and global bathymetric models
Water stops radar completely and stops light within metres to a few tens of metres. Almost everything we “see” of the deep ocean floor from space is therefore inferred, not observed: satellites measure the shape of the sea surface, and the relief beneath is estimated from it. Only acoustic surveys from ships and underwater vehicles measure deep seafloor depth directly — and in April 2026 GEBCO reported that about 28.7% of the ocean floor had been mapped to modern standards. This chapter explains each method, what it measures, what it can only infer, and where it stops.
Status labels.OPERATIONAL — the method is in routine professional use; that is a statement about the method, never a promise about any particular site. EXPERIMENTAL — demonstrated in research or in emerging products, not routine for this purpose. GEASPIRIT PROPOSAL — an idea for GeaSpirit that has not been built.
Platform labels show where the instrument physically is. The platform decides what is possible: an instrument in orbit cannot see what a ground crew sees, and a ground crew cannot cover a continent.
Why water is the problem
Every remote method relies on energy that travels to the target and back. Seawater is electrically conductive, so a radar pulse is absorbed in the top millimetres and never reaches the bottom. Visible light does penetrate, but water, suspended sediment and plankton absorb and scatter it; blue-green light reaches furthest, and even in very clear water optical methods work only in the shallowest tens of metres. Sound, by contrast, travels thousands of kilometres in water — which is why depth in the open ocean is measured acoustically, from ships and underwater vehicles.
The practical consequence. In this chapter, every depth that comes from a satellite is either an inference from the sea surface (altimetry) or a shallow-water estimate (LiDAR, multispectral). Direct depth measurement of the deep ocean is acoustic.
1
Radar altimetry and marine gravity
Sentinel-6 Michael Freilich, SWOT and earlier altimeters
SATELLITEOPERATIONAL
Measures the height of the sea surface, not the seafloor. Relief is inferred from the faint bumps and dips that the seafloor's gravity imprints on the ocean surface.
SCHEMATIC — NOT TO SCALE
Radar pulse to the sea surface and back.
The time-averaged sea surface bulges over extra mass (a seamount) and dips over a trench.
Surface slopes are converted into a marine gravity anomaly.
Relief is inferred from gravity and calibrated with ship soundings: smoother and less certain than the real seafloor.
Figure A1. From radar echo to inferred relief. The altimeter measures the sea surface; the seafloor is inferred from the surface's shape. The bulge is drawn hugely exaggerated: in reality it is far too gentle to see.Technical sheet — principle, instrument, reach, resolution, limits
Physical principle
A radar altimeter sends microwave pulses straight down and times the echo from the sea surface; with a precisely known orbit this gives sea-surface height. Averaged over time to remove waves, tides and currents, the sea surface follows the geoid, the equipotential surface of Earth's gravity. Extra mass on the seafloor, such as a seamount, pulls water towards it and raises the surface above it; a trench produces a dip. Sea-surface slopes are converted into a marine gravity anomaly, and depth is estimated from gravity, calibrated with ship soundings.
Instrument and platform
Sentinel-6 Michael Freilich (launched 21 November 2020; Copernicus programme with ESA, EUMETSAT, NASA, NOAA and CNES) carries the Poseidon-4 nadir altimeter: Ku band (13.575 GHz) for range and C band (5.41 GHz) for ionospheric correction. SWOT (NASA and CNES, launched December 2022) carries KaRIn, a Ka-band interferometer that maps sea-surface height in two dimensions across two 50 km swaths separated by a 20 km gap below the satellite. Earlier missions — Geosat, ERS-1, CryoSat-2, the Jason series, SARAL/AltiKa — built the existing marine gravity grids.
What it measures directly
The two-way travel time of a radar pulse to the sea surface, converted into sea-surface height.
What it allows us to infer
Marine gravity anomalies and, from them, estimated (“predicted”) seafloor depth: ridges, fracture zones, seamounts, trenches and the fabric of abyssal hills.
Reach and depth (conditional)
All ice-free oceans, whatever the depth, because the signal never has to reach the seafloor. Gravity and topography are well correlated only in a band of wavelengths — roughly 15 to 160 km in the original method of Smith and Sandwell (1994) — and thick sediment or isostatic compensation weakens the link.
Resolution and error sources
Predicted bathymetry from conventional altimeters resolves features of a few kilometres at best, with depth errors that can reach hundreds of metres; NOAA's Laboratory for Satellite Altimetry notes that a seamount has had to be roughly a mile high and several miles wide to be detected. SWOT's two-dimensional measurements produced a marine gravity map at about 8 km resolution that reveals seamounts less than half the height earlier satellites could detect (Yu, Sandwell and Dibarboure, Science, 2024). Errors come from ocean dynamics, sea state, sediment cover, density assumptions and the sparse ship data used for calibration.
Real mining application
Regional marine-geology context: ridge segments, transform faults, back-arc basins and seamount chains — the settings in which seafloor massive sulphides, cobalt-rich crusts and nodules occur. A first-pass map for planning ship surveys, not a resource tool.
Limitations
It cannot see the seabed, cannot measure composition and cannot detect a sulphide mound, a nodule field or any metal. Features of hundreds of metres are below its resolution. In GEBCO, depths estimated this way are flagged separately from measured ones (TID code 40).
Status
OPERATIONAL for sea-surface height and marine gravity. Depth estimation from gravity is an established product, but indirect and of low resolution.
Green-laser ranging from aircraft and drones; ICESat-2 from orbit
AIRCRAFT / DRONESATELLITEOPERATIONALEXPERIMENTAL
Measures depth directly — but only in shallow, clear water, typically to about two to three times the Secchi depth.
SCHEMATIC — NOT TO SCALE
Figure A2. Bathymetric LiDAR. Green light reaches the bottom only where the water is shallow and clear; beyond that, no bottom echo returns.Technical sheet — principle, instrument, reach, resolution, limits
Physical principle
A pulsed laser at 532 nm — green light, which seawater absorbs relatively little — is fired at the water. Part of each pulse returns from the surface and part from the bottom; the time difference, corrected for refraction at the surface and for the slower speed of light in water, gives the depth. Many airborne systems add an infrared pulse (1064 nm), which does not penetrate water, to fix the surface.
Instrument and platform
Airborne topo-bathymetric LiDAR on aircraft and, at smaller scale, on drones. From orbit, NASA's ICESat-2 carries ATLAS, a photon-counting 532 nm laser altimeter designed for ice and land; its bottom returns in clear coastal water are now distributed as a global along-track bathymetry product (ATL24, at NSIDC).
What it measures directly
The travel time of laser light to the water surface and to the bottom: a direct depth measurement after refraction correction.
What it allows us to infer
Detailed shallow seabed relief, coastline, reef and channel morphology; from the return intensity, a rough indication of seabed reflectance.
Reach and depth (conditional)
Set by water clarity rather than by the instrument alone. NOAA documentation for the SHOALS system gives the rule of thumb of two to three Secchi depths: a turbid harbour may allow a few metres, very clear water several tens of metres. In early validation ICESat-2 recorded bottom returns down to about 38 m in clear water (Parrish et al., 2019).
Resolution and error sources
Airborne surveys sample the seabed densely, at metre scale or finer depending on system and flight plan. ICESat-2 depths agreed with airborne LiDAR to roughly 0.4–0.6 m RMSE in validation, but only along narrow tracks kilometres apart. Errors come from turbidity, waves, bubbles, seabed vegetation, refraction modelling and dark bottoms that return little light.
Real mining application
Coastal and nearshore work: heavy-mineral sand and placer environments, dredging and port engineering, shallow-water geohazards, and the land–sea transition next to coastal mines and tailings outfalls.
Limitations
Useless in deep or turbid water; gives no composition. Aircraft campaigns are costly and local; ICESat-2 is a sparse profiler, not a mapping instrument.
Status
OPERATIONAL from aircraft. EXPERIMENTAL from orbit: ICESat-2 bathymetry (ATL24) is an emerging along-track product.
Water attenuates light at different rates for different wavelengths: blue and green reach deeper than red and near-infrared. Over a uniform bottom, reflectance in the blue and green bands — or their log ratio (Stumpf et al., 2003) — changes systematically with depth. An empirical relation is fitted with known depths from LiDAR, sonar or charts and then applied to the whole image; physics-based inversion models are an alternative.
Instrument and platform
The Copernicus Sentinel-2 MultiSpectral Instrument (10 m pixels in the blue, green, red and near-infrared bands); Landsat 8 and 9 OLI (30 m); commercial high-resolution satellites.
What it measures directly
Reflected sunlight in several spectral bands at the top of the atmosphere.
What it allows us to infer
Approximate depth in shallow, clear water; to some extent, bottom type (sand, rock, seagrass).
Reach and depth (conditional)
Depends on clarity and bottom brightness. A NOAA study in South Florida validated Sentinel-2 depths down to 18 m — the limit of the LiDAR used for validation — with a median absolute error of about 0.5 m (Caballero and Stumpf). Other studies show river plumes and suspended sediment reducing usable depth to a few metres.
Resolution and error sources
Pixels of 10–30 m. Errors from atmospheric correction, sun glint, waves, turbidity, variable bottom reflectance (dark seagrass looks “deep”) and the quality of the calibration soundings; each scene usually needs its own calibration.
Real mining application
Low-cost reconnaissance of shallow coasts, lagoons and nearshore shelves; monitoring sediment plumes and shoreline change near coastal mines, tailings outfalls and dredging.
Limitations
Useless in deep or turbid water. A statistical estimate, not a hydrographic survey; not suitable for navigation without validation.
Status
OPERATIONAL in clear shallow water with local calibration. Outside those conditions it should not be used.
Swath sonar on ships and on autonomous underwater vehicles
SHIP / AUVOPERATIONAL
The reference method for measuring seafloor depth directly, at any ocean depth — but only where a vessel has actually passed.
SCHEMATIC — NOT TO SCALE
Figure A3. Multibeam sonar. A fan of acoustic beams measures a swath of depths beneath the vessel; overlapping swaths become a 3D model.Technical sheet — principle, instrument, reach, resolution, limits
Physical principle
A transducer array emits a fan-shaped acoustic pulse (a “ping”) across the ship's track and forms many narrow receive beams. The travel time of each echo, combined with the sound-speed profile of the water column and the ship's position and attitude, gives the depth of a line of points across the seafloor. Successive pings build a swath; overlapping swaths build a map. The strength of each echo (backscatter) carries information on seabed texture and hardness.
Instrument and platform
Hull-mounted systems on research and survey ships — low frequencies, around 12 kHz, for full ocean depth; hundreds of kHz for shallow, high-resolution work — and compact systems on autonomous underwater vehicles (AUVs) and remotely operated vehicles (ROVs) that fly close to the bottom.
What it measures directly
Acoustic two-way travel time for each beam, and the echo intensity (backscatter).
What it allows us to infer
A high-resolution depth grid (digital bathymetric model); seabed character — sediment or rock, mounds, chimneys — from backscatter; from water-column data, gas or hydrothermal plumes.
Reach and depth (conditional)
Full ocean depth with low-frequency systems. Swath width grows with depth, typically several times the water depth: NIWA reports a Kongsberg EM302 swath of up to about 7 km at about 2,500 m depth.
Resolution and error sources
Each beam's footprint widens with distance, so detail decreases with depth: ship-based deep-ocean grids are commonly tens of metres per cell, while an AUV flying tens of metres above the bottom resolves features at metre to sub-metre scale. Errors come from the sound-speed profile, vessel motion and positioning, refraction at the swath edges and noise.
Real mining application
The basis of every serious seafloor-minerals survey: mapping hydrothermal mounds and chimneys, nodule-field terrain and crust-covered seamount flanks, and supporting dredging and offshore engineering. AUV multibeam is how individual sulphide mounds are resolved.
Limitations
Slow and expensive: it maps only the strip beneath the vessel. Backscatter is not a chemical assay; it cannot say how much metal a mound contains.
Global models: GEBCO, EMODnet, NOAA and Seabed 2030
Most seafloor maps in circulation are compilations. They merge multibeam and single-beam surveys, LiDAR, satellite-derived bathymetry and — to fill the gaps — depths predicted from satellite gravity. The resulting grids look uniformly detailed, which hides how different their sources are.
GEBCO
The General Bathymetric Chart of the Oceans (IHO and IOC-UNESCO) publishes a global grid at 15 arc-second spacing; the GEBCO_2026 grid was released in April 2026. Values are elevations in metres at the centre of each cell.
GEBCO TID grid
The Type Identifier (TID) grid states, cell by cell, where each depth came from: direct measurements (for example 10 single-beam, 11 multibeam, 15 LiDAR, 16 optical light sensor) or indirect ones (for example 40, predicted from satellite-derived gravity). Reading the TID grid is the honest way to know whether a seamount in the grid was measured or estimated.
EMODnet Bathymetry
The European Marine Observation and Data Network's DTM covers the European seas at 1/16 arc-minute (about 115 m), integrating acoustic surveys, airborne LiDAR and satellite-derived bathymetry; the 2024 release also draws on GEBCO for gaps.
NOAA
NOAA's National Centers for Environmental Information (NCEI) archive and serve multibeam, single-beam and coastal elevation data; NOAA's Laboratory for Satellite Altimetry works on altimetric bathymetry.
Seabed 2030
The Nippon Foundation–GEBCO Seabed 2030 Project aims at a complete map of the ocean floor by 2030. With the GEBCO_2026 grid it reported 28.7% of the seafloor mapped to modern standards, up from 27.3% a year earlier.
Figure A4. Share of the global seafloor mapped to modern standards, as reported with the GEBCO_2026 grid (April 2026). Most of the remainder is estimated — from satellite gravity or by interpolation — not measured.
Measured versus estimated. A depth in a global grid is a measurement only where the TID grid says so. Everywhere else — still most of the ocean — it is an estimate guided by gravity or interpolation, with errors that can reach hundreds of metres.
Ridges, seamounts, volcanoes, trenches and hydrothermal systems
What remote methods can and cannot contribute for each kind of seafloor feature.
Feature
What it is
What remote methods show
What needs ships, vehicles or samples
Mid-ocean ridges
Spreading centres where new oceanic crust forms; host most known high-temperature vent fields.
Gravity from altimetry maps ridge axes, segments and transform faults globally.
The axial valley, fissures and young lava flows need multibeam; vents need near-bottom surveys.
Seamounts
Underwater mountains, mostly extinct volcanoes; their flanks can carry cobalt-rich ferromanganese crusts.
Large seamounts appear in gravity-derived bathymetry; SWOT detects smaller ones than earlier satellites.
Shape, summit depth and slopes need multibeam; crust thickness and grade need sampling.
Submarine volcanoes
Active volcanoes on arcs, hotspots and ridges.
Large edifices in gravity grids; when shallow, eruption plumes or discoloured water may be visible to optical satellites.
Changes in shape need repeated multibeam surveys; activity needs hydrophones, sampling and vehicles.
Trenches
Subduction zones; the deepest parts of the ocean.
Their large gravity lows are among the clearest altimetry signals.
True depth and slope features below altimetry resolution need multibeam.
Hydrothermal systems
Vents where seawater heated by magma leaches metals and deposits sulphides, sometimes building seafloor massive sulphide (SMS) mounds.
No satellite detects a vent field. Altimetry maps only the tectonic setting — a ridge or a back-arc basin.
Found by ship and vehicle work: water-column plume sensors (temperature, turbidity, redox), near-bottom multibeam and magnetics, then cameras, dives and sampling.
Case: the 2026 Xiangyanghong 10 hydrothermal field
REPORTED — PRELIMINARY
On 11 and 12 September 2026, Chinese state media (CGTN, and Global Times citing CCTV) reported the discovery of a previously unknown, active high-temperature hydrothermal field by an expedition aboard the research vessel Xiangyanghong 10, with Tsinghua University, the Qingdao Institute of Marine Geology (China Geological Survey) and Shanghai Jiao Tong University. The ship left Shenzhen on 10 August for an 18-day voyage with seven submersible dives.
Location. Described only as a back-arc basin in the western Pacific within China's exclusive economic zone, near the boundary between the Eurasian and Philippine Sea plates. No coordinates were published. That description is consistent with the Okinawa Trough back-arc basin in the East China Sea — our inference, not a published fact.
Reported values. Some samples contained up to 15.4 g/t gold and up to 1,271 g/t silver, according to preliminary analysis by the shipboard laboratory. These are maxima in some samples, not averages.
Context. Earlier surveys in the area had found dozens of high-temperature vents at about 700–1,500 m depth.
Resource. State media quoted preliminary estimates that the field “could reach the level of a world-class seafloor deposit”, and also that it is not ready for mining. No resource estimate under any public reporting code has been published.
How to read it. A shipboard result on selected samples is a first observation, not a grade. Sulphide chimneys are notoriously heterogeneous; maximum values from a few samples say nothing about tonnage or average grade. Independent laboratory assays, systematic sampling and drilling would be needed before anything could be called a resource.
Why no satellite saw it. No satellite detected these metals, and none could. Under hundreds of metres of water, radar is stopped at the surface and light is extinguished in the upper tens of metres. Altimetry can outline a back-arc basin; it cannot see a vent, a chimney or gold. The field was found the way hydrothermal fields are found: by ship, with sensors in the water column and submersible dives.
Marine-minerals research deals with three kinds of seafloor occurrence: seafloor massive sulphides at hydrothermal systems (copper, zinc, gold, silver), polymetallic nodules on abyssal plains (manganese, nickel, copper, cobalt), and cobalt-rich ferromanganese crusts on seamount flanks. None of them can be measured from orbit.
Shallow coastal seabed only; irrelevant at the depths of vents, nodules and crusts.
Ship multibeam and backscatter
No — indirect clues only
Shapes of mounds and chimneys; hard versus soft seabed; nodule-field terrain.
Water-column sensors
No — indirect clues only
Plumes that point to active venting nearby.
Near-bottom magnetics (AUV)
No — indirect clues only
Hydrothermal alteration can demagnetise the rock beneath vent fields; the resulting anomalies are a few hundred metres across or less and need near-bottom surveys (Tivey and Dyment, 2010).
Sampling, coring, drilling and laboratory assay
Yes
The only route to grade and — with enough systematic data — to tonnage.
Seabed mining is also a matter of law: beyond national jurisdiction it falls under the International Seabed Authority; inside an exclusive economic zone, under national law. This chapter takes no position on exploitation.
A GeaSpirit proposal, not built and not scheduled: show the ocean floor as 3D relief beneath the Asset Map, with layers that keep measured and estimated depth visibly apart.
3D relief from GEBCO (global) and EMODnet (European seas), within their licence terms and with their attribution.
A “measured versus estimated” overlay from the GEBCO TID grid, so that a seamount predicted from gravity is never shown as if it had been surveyed.
Context layers: ridges and transform faults, vent fields from public databases such as the InterRidge Vents Database, and areas under International Seabed Authority contracts.
No grades, resource figures or targets would be displayed; every vent-field location would carry its published source.
Status: GEASPIRIT PROPOSAL. Nothing in this section exists in the Asset Map today.
NASA PO.DAAC. SWOT Level 2 KaRIn Low Rate Sea Surface Height Data Product, Version D (swath geometry: 10–60 km each side of nadir). DOI 10.5067/SWOT-SSH-D. doi.org/10.5067/SWOT-SSH-D
Smith, W. H. F. and Sandwell, D. T. (1994). Bathymetric prediction from dense satellite altimetry and sparse shipboard bathymetry. Journal of Geophysical Research 99(B11), 21803–21824. doi.org/10.1029/94JB00988
Parrish, C. E. et al. (2019). Validation of ICESat-2 ATLAS bathymetry and analysis of ATLAS's bathymetric mapping performance. Remote Sensing 11(14), 1634. mdpi.com/2072-4292/11/14/1634
NSIDC. ATL24 — ICESat-2 along-track coastal and nearshore bathymetry. nsidc.org/data/atl24
Stumpf, R. P., Holderied, K. and Sinclair, M. (2003). Determination of water depth with high-resolution satellite imagery over variable bottom types. Limnology and Oceanography 48(1, part 2), 547–556. doi.org/10.4319/lo.2003.48.1_part_2.0547
Caballero, I. and Stumpf, R. P. Retrieval of nearshore bathymetry from Sentinel-2A and 2B satellites in South Florida coastal waters. Estuarine, Coastal and Shelf Science (NOAA repository). repository.library.noaa.gov/view/noaa/22109
Tivey, M. A. and Dyment, J. (2010). The magnetic signature of hydrothermal systems in slow spreading environments. In: Diversity of Hydrothermal Systems on Slow Spreading Ocean Ridges, AGU Geophysical Monograph 188. hal-insu.archives-ouvertes.fr/insu-03605354
This chapter describes measurement methods. It is not an exploration service, not investment advice, and contains no detection, resource, reserve or grade claim by GeaSpirit. Third-party figures are quoted with their sources; schematic figures are drawn by GeaSpirit and are not to scale.
