Sound Velocity Profile Mismatch in a Harbour Channel
You are running a multibeam survey in a busy harbour channel with depths from 4 m to 28 m. The sound velocity probe at the transducer reads 1,502 m/s, while a cast taken at the channel entrance reads 1,487 m/s at the surface and 1,495 m/s at depth. The harbour has strong freshwater input after recent rain, creating a stratified surface layer. Your multibeam data show 'smile' and 'frown' artefacts across the outer swaths, and the inner beams appear consistent. The survey must meet IHO S-44 Order 1a vertical accuracy of ±0.5 m at 95% confidence. You have a calibrated bar-check available but limited time.
1. Diagnose the likely cause of the swath artefacts and decide how to handle the sound velocity data.
Read the worked answer
Diagnosis: The mismatch between the transducer-mounted probe (1,502 m/s) and the cast (1,487 m/s surface, 1,495 m/s at depth) indicates a stratified water column, likely a freshwater lens at the surface. The probe is measuring at a single depth and cannot capture the vertical gradient. Outer-beam artefacts (smile/frown) are classic symptoms of an incorrect sound velocity profile (SVP), because outer beams have longer ray paths and are more sensitive to refraction errors.
Handling:
- Take casts at multiple locations along the channel (entrance, mid-channel, inner) and at intervals during the survey, especially after rain.
- Apply the cast nearest in space and time to each line, not a single cast for the whole survey.
- Use the bar-check to validate the probe's absolute accuracy; if the probe is biased, correct or replace it.
- Re-run lines in the affected areas if artefacts persist after correct SVP application.
Document the SVP strategy and any re-runs in the survey report.
2. How do you verify that the final data meet the ±0.5 m Order 1a requirement despite the stratification?
Read the worked answer
Verification approach:
- Compute a total propagated uncertainty (TPU) budget including SVP uncertainty, tide/datum uncertainty, positioning, and motion sensor errors. Show that the combined vertical uncertainty is within ±0.5 m at 95% confidence.
- Compare overlapping swaths and cross-lines; the standard deviation of depth differences at the same location is a direct check on refraction and motion errors.
- Use the bar-check and independent check lines (e.g., single-beam or lead-line) to confirm absolute depths.
- Assess the inner vs outer beam agreement; if outer beams remain biased, either reduce swath width or flag the outer beams as lower confidence.
Decision: If TPU exceeds the requirement in the stratified zone, re-survey with tighter SVP control or narrower swaths before certifying the data. Do not certify data that cannot demonstrate compliance.