Measuring underwater topography with light and sound
2026-02-27
On 25 July 2020, the bulk carrier MV Wakashio ran aground on a coral reef off the south-east coast of Mauritius, eventually spilling more than 1,000 tonnes of fuel oil across one of the Indian Ocean’s most ecologically sensitive coastlines. The reef had been charted — but not at the decimetre resolution that would have flagged the hazard to the vessel’s automated navigation system. That data gap is exactly what modern bathymetric mapping exists to close.
Measuring water depth sounds deceptively simple. Drop a weighted line, read the rope. Nineteenth-century hydrographers did exactly that, sounding harbour floors one point at a time. Today two technologies dominate. In clear, shallow water — coastal reefs, river channels, lake beds — green-wavelength laser light (532 nm) penetrates the water column, reflects off the bottom, and returns a depth with centimetre precision. In deeper or turbid water, acoustic pulses do the same job: sonar calculates depth from the travel time of a sound wave bouncing off the seabed, mapping swaths hundreds of metres wide in a single pass. Neither method is universal; each has its physics, its geometric corrections, and its characteristic errors.
This model works through those physics from first principles. The core challenge in both cases is the same: the signal doesn’t travel in a straight line at constant speed. Light refracts at the air-water interface and slows as it enters denser water. Sound bends with temperature and salinity gradients. Getting the geometry right — converting raw travel times and angles into accurate XYZ coordinates on a map — is the craft that separates a chart you can navigate by from one that gets ships grounded.
What’s the depth and bottom topography of this shallow coastal bay?
Two technologies for underwater elevation:
Bathymetric LiDAR (ALB - Airborne Lidar Bathymetry): - Green laser (532 nm) penetrates water - Measures from aircraft - Shallow water (<50m, clear conditions) - Simultaneous topography (infrared) + bathymetry (green)
Sonar (Sound Navigation and Ranging): - Acoustic pulses (10-500 kHz) - Boat/ship-mounted or autonomous underwater vehicles - Deep water (thousands of meters) - Works in turbid water
Applications: - Nautical chart production - Coastal hazard assessment (tsunami inundation) - Habitat mapping (coral reefs, seagrass) - Dredging and navigation - Underwater archaeology - River bathymetry - Reservoir volume calculation
Dual-wavelength system:
Infrared (1064 nm): - Reflects off water surface - Measures water surface elevation
Green (532 nm): - Penetrates water column - Reflects off bottom - Return delayed by water travel time
Water column correction:
d_{\text{water}} = \frac{c_{\text{air}} t_{\text{total}} - c_{\text{water}} t_{\text{water}}}{2}
More precisely:
z_{\text{bottom}} = z_{\text{surface}} - \frac{c_{\text{water}}}{2} (t_{\text{green}} - t_{\text{IR}})
Where: - c_{\text{water}} = c_{\text{air}}/n (n = refractive index ≈ 1.33) - Speed of light in water: ~2.25 × 10⁸ m/s
Limitations:
Turbidity (Secchi depth):
Beer’s Law attenuation:
I(z) = I_0 e^{-Kz}
Where: - K = attenuation coefficient (m⁻¹) - Typical: K = 0.05-0.5 m⁻¹
Maximum depth:
Approximately 2-3 × Secchi depth.
Clear water: 40-50 m
Turbid water: 5-10 m
Very turbid: No penetration
Principle:
Transmit acoustic pulse, receive echoes from multiple angles simultaneously.
Sound speed in water:
c = 1449 + 4.6T - 0.055T^2 + 0.00029T^3 + 1.34(S - 35) + 0.016z
Where: - c = sound speed (m/s) - T = temperature (°C) - S = salinity (PSU) - z = depth (m)
Typical: c ≈ 1500 m/s (much slower than light!)
Two-way travel time:
d = \frac{ct}{2}
Multibeam geometry:
Example: 100m depth → 300-400m swath width
Snell’s Law at water surface:
n_1 \sin\theta_1 = n_2 \sin\theta_2
For air (n_1 = 1) to water (n_2 = 1.33):
\sin\theta_{\text{water}} = \frac{\sin\theta_{\text{air}}}{1.33}
Off-nadir beams refract toward normal.
Effect: Horizontal position shifts inward.
Correction essential for accurate positioning.
Surface return time: t_{\text{IR}}
Bottom return time: t_{\text{green}}
Water travel time:
\Delta t = t_{\text{green}} - t_{\text{IR}}
Depth below surface:
d = \frac{c_{\text{water}} \Delta t}{2} = \frac{c_{\text{air}} \Delta t}{2n}
Where n = 1.33 (refractive index of water).
Example:
\Delta t = 200 ns
d = \frac{3 \times 10^8 \times 200 \times 10^{-9}}{2 \times 1.33} = \frac{60}{2.66} = 22.6 \text{ m}
Total elevation:
z_{\text{bottom}} = z_{\text{surface}} - d
Signal-to-noise ratio:
SNR = \frac{I_{\text{bottom}}}{I_{\text{noise}}}
Bottom return intensity:
I_{\text{bottom}} = I_0 R_{\text{bottom}} e^{-2Kd}
Where: - R_{\text{bottom}} = bottom reflectance (~0.1-0.4) - Factor of 2: Down and back through water column
Detection threshold: SNR > 3 (minimum)
Maximum depth:
d_{\max} = \frac{\ln(I_0 R_{\text{bottom}} / I_{\text{threshold}})}{2K}
Example: K = 0.1 m⁻¹ (moderate clarity)
d_{\max} \approx \frac{\ln(1000)}{0.2} = \frac{6.9}{0.2} = 34.5 \text{ m}
Range to bottom:
R = \frac{ct}{2}
Beam angle: \theta (from vertical)
Horizontal offset:
x = R \sin\theta
Vertical depth:
z = R \cos\theta
Sound speed correction:
If c varies with depth (common):
Ray tracing through water column needed.
\frac{dx}{dz} = \frac{\sin\theta}{\cos\theta} = \tan\theta
With \theta(z) from Snell’s Law:
c(z) \sin\theta(z) = c_0 \sin\theta_0
Iterative solution required for accurate positioning.
Swath width:
For flat bottom at depth D, beam angle \pm \theta_{\max}:
W = 2D \tan\theta_{\max}
Example: D = 100 m, \theta_{\max} = 60°
W = 2(100) \tan(60°) = 200 \times 1.73 = 346 \text{ m}
Survey efficiency:
Line spacing = swath width → 100% coverage
Tighter spacing → overlap for quality assurance
Problem: Calculate bathymetric LiDAR depth measurement.
Data: - Water surface elevation: z_{\text{surface}} = 2.5 m (above datum) - Infrared return time: t_{\text{IR}} = 1000 ns (round-trip to surface) - Green return time: t_{\text{green}} = 1400 ns (round-trip to bottom) - Refractive index: n = 1.33
Calculate bottom elevation.
Step 1: Water column travel time
\Delta t = t_{\text{green}} - t_{\text{IR}} = 1400 - 1000 = 400 \text{ ns}
Step 2: Speed of light in water
c_{\text{water}} = \frac{c_{\text{air}}}{n} = \frac{3 \times 10^8}{1.33} = 2.26 \times 10^8 \text{ m/s}
Step 3: Water depth
d = \frac{c_{\text{water}} \Delta t}{2} = \frac{2.26 \times 10^8 \times 400 \times 10^{-9}}{2}
= \frac{90.4}{2} = 45.2 \text{ m}
Step 4: Bottom elevation
z_{\text{bottom}} = z_{\text{surface}} - d = 2.5 - 45.2 = -42.7 \text{ m}
(Negative: below datum)
Summary: - Water depth: 45.2 m - Bottom elevation: -42.7 m (relative to datum)
Note: This is near maximum depth for bathymetric LiDAR in clear water.
Below is an interactive bathymetric mapping simulator.
<label>
Technology:
<select id="tech-type">
<option value="lidar">Bathymetric LiDAR</option>
<option value="sonar">Multibeam Sonar</option>
</select>
</label>
<label>
Water clarity (Secchi depth, m):
<input type="range" id="clarity" min="1" max="30" step="1" value="15">
<span id="clarity-val">15</span>
</label>
<label>
Bottom type:
<select id="bottom-type">
<option value="sand">Sand (bright)</option>
<option value="mud">Mud (dark)</option>
<option value="rock">Rock (variable)</option>
<option value="seagrass">Seagrass</option>
</select>
</label>
<div class="bathy-info">
<p><strong>Max measurable depth:</strong> <span id="max-depth">--</span> m</p>
<p><strong>Swath width:</strong> <span id="swath-width">--</span> m</p>
<p><strong>Points measured:</strong> <span id="point-count">--</span></p>
<p><strong>Mean depth:</strong> <span id="mean-depth">--</span> m</p>
</div><canvas id="bathy-canvas" width="700" height="400" style="border: 1px solid #ddd;"></canvas>Observations: - Blue area represents water column - Brown area shows measured seafloor - Orange dotted line marks maximum measurable depth - LiDAR limited by water clarity (turbidity) - Sonar works in deep/turbid water but from vessel - Clear water (high Secchi) enables deeper LiDAR penetration - Bright bottoms (sand) return stronger signal than dark (mud) - Data gaps occur where depth exceeds technology limits
Key findings: - Bathymetric LiDAR optimal for shallow clear water (<50m) - Sonar required for deep or turbid conditions - Water clarity and bottom reflectance control maximum LiDAR depth - Complementary technologies cover full range of conditions
NOAA bathymetric LiDAR program:
Systematic coastal mapping for: - Nautical chart updates - Tsunami inundation modelling - Habitat assessment - Emergency response
Example - Hurricane Sandy (2012):
Post-storm bathymetric LiDAR: - Documented massive sediment redistribution - Inlet breaches - Dune erosion - Critical for recovery planning
Challenges:
Complex 3D structure at fine spatial scale.
LiDAR advantages: - Complete coverage (vs. transects) - Rugosity measurement (surface complexity) - Change detection (growth, bleaching, storm damage)
Example - Florida Keys:
Repeated surveys detect: - Coral decline - Algal overgrowth - Storm impacts
Management application:
Prioritize restoration sites, track recovery.
Traditional surveys:
Boat-based sonar in navigable reaches only.
Bathymetric LiDAR:
Maps entire river including: - Shallow riffles (boats can’t access) - Bank-to-bank coverage - Floodplain + channel in one dataset
Applications: - Salmon habitat (pool-riffle structure) - Flood modelling (channel capacity) - Sediment transport (bedform mapping)
Underwater sites:
Shipwrecks, submerged settlements, harbor structures.
LiDAR detection:
Subtle features on seafloor visible in high-resolution bathymetry.
Example - Submerged prehistoric landscapes:
8000-year-old settlement sites now offshore mapped with bathymetric LiDAR.
Backscatter from water column:
Sediment particles reflect laser/acoustic signal.
LiDAR: False bottom returns from turbid layers.
Sonar: Reduced penetration, noise.
Worst case: Impossible to detect true bottom.
Solution: - Post-processing filters - Survey during low-turbidity conditions - Combined interpretation with water column imagery
Rough sea state:
Water surface moves during measurement.
LiDAR error: ±0.5-1m in moderate waves.
Sonar: Motion compensation systems on vessel.
Solution: - Survey in calm conditions - Average multiple measurements - Motion reference unit (MRU) on aircraft
LiDAR off-nadir beams:
Snell’s Law correction requires knowing: - Beam angle - Water surface orientation - Refractive index (varies with salinity, temperature)
Uncertainty propagates to horizontal position.
Solution: - Minimize scan angle - Measure water properties - Cross-line overlap for validation
Sonar backscatter varies with: - Grain size (sand vs. mud) - Roughness - Biology (seagrass, coral)
Challenge: Distinguishing features from noise.
Solution: - Backscatter mosaics - Ground-truth sampling - Multi-frequency analysis
Different technology: Measures backscatter, not just bathymetry.
Geometry:
Transducers aimed sideways (perpendicular to track).
Creates acoustic image of seafloor.
Applications: - Object detection (shipwrecks, boulders, debris) - Seafloor classification (sediment type, bedforms) - Pipeline/cable inspection - Search and recovery
Advantages: - High-resolution imagery (cm-scale) - Wide swath coverage
Disadvantages: - Doesn’t directly measure depth - Geometric distortion - Interpretation requires expertise
Often combined with multibeam sonar: - Bathymetry from multibeam - Texture from side-scan - Comprehensive seafloor characterization
Exponential decay:
I(z) = I_0 e^{-\alpha z}
Where: - I(z) = intensity at depth z - I_0 = incident intensity - \alpha = attenuation coefficient (m⁻¹)
Applicable to both light and sound in water.
Light:
\alpha_{\text{light}} = \alpha_{\text{absorption}} + \alpha_{\text{scattering}}
Dominated by: - Pure water absorption (increases with wavelength) - Chlorophyll absorption (green algae) - Suspended sediment scattering
Sound:
\alpha_{\text{sound}} = \alpha_0 f^2
Where f = frequency.
Higher frequency → more attenuation → shorter range
Trade-off: Resolution vs range.
Laser/sonar pulse travels: - Down to bottom: attenuated by e^{-\alpha z} - Back to sensor: attenuated again by e^{-\alpha z}
Total: I_{\text{return}} = I_0 R e^{-2\alpha z}
Factor of 2 in exponent critical for range calculations.