Work

MARS lander: Georeferencing landing and pop points of untethered ocean monitoring systems using fundamental physics

Year: 2026

Type: article

Abstract: Subsurface observations are crucial for understanding the ocean's role in Earth’s climate and for refining climate models. However, existing aquatic monitoring systems that allow such insights remain complex and costly due to their high demands for deployment, sampling, and recapture. Since low-cost, easy-to-deploy deep-sea landers are scarce, and with the aim of facilitating more subsurface observations, this study provides a simple method for georeferencing small-sized untethered landers. Their underwater trajectories are modelled with fundamental physics, dead reckoning, lander geometry, and numerical simulations. Using free fall, upthrust, and ocean current dynamics, the proposed approach estimates their underwater trajectories, including landing (at the seabed) and pop (at the sea surface) points. The method relies on the lander's physical characteristics, including its vertical and horizontal cross-sectional areas, to calculate the drag force coefficients used to determine its trajectories during descent and ascent through the water column. Ocean currents' magnitudes are modelled using Ekman’s exponential decay down to 90 m of the water column, while the depths until 900 m are modelled from prior ADCP surveys by varying ocean current headings with depth between -20 and 20 degrees. Surface ocean and wind current headings are modelled with open datasets from satellite telemetry. Lander's velocity, displacement, and dive time to the landing and pop points, including the total radial excursion and uncertainty in heading, are analytically derived, numerically calculated, and empirically assessed a-posteriori until 90 m, yielding a ∼38 m radial excursion (40% error) against the obtained GNSS coordinates in field deployment, and 33 degrees in heading uncertainty during a 138-second excursion. Additional random walk simulations are shown for full ocean depth obtaining radial excursion of 1,038 m with 278 min total dive time. This approach is generalizable to any subsurface aquatic monitoring systems targeting deployments with diverse payloads from smaller sea vessels, not requiring cranes, radio, GNSS, or acoustic telemetry. Since it accounts for key nature factors, our method provides special benefits in planning and optimizing deployments. Additional discussion focuses on the method's practicality for full ocean depth deployments. Graphical abstract highlighting the challenge in georeferencing the untethered low-cost landers without costly satellite uplink telemetry. The lander is deployed from smaller sea vessels at the desired (drop point) location, where the required depth is obtainable from the sea vessel's sonar. The ocean currents cause the lander to drift during descent and ascent with radial excursion relative to the drop point. Heading uncertainties are depicted as estimated circular areas of probabilities where the lander may touch the seabed (landing point). The same approach also applies to estimating the lander’s position at the sea surface (pop point). • Predicting landing and pop points of low-cost deep-sea landers a-posteriori. • Providing underwater trajectories based on the lander’s geometry, dead reckoning, numerical simulations, and hydrodynamics. • Optimizing planning, deployment, sampling, and recapture of untethered aquatic observatories. • Obtained 40% radial excursion error with in-situ deployment at 90 m depth. • Obtained radial excursion of 1,038 m with 278 min total dive time, including 95 degrees heading uncertainty using random walk simulation for the case of full ocean depth (11,000 m).

Source: Deep Sea Research Part I Oceanographic Research Papers

DOI: 10.1016/j.dsr.2026.104650