Real Ocean Winds Undercut the Design Assumptions Behind Wind-Assisted Ship Propulsion

Key Findings · At a Glance

• ▸Commonly used power-law exponents of 1/7 or 1/9 may not represent the flow actually reaching the vessel and its sails.
• ▸Winds from the side, the most useful for sails, occurred roughly 69% of the time across the four crossings.
• ▸Starboard inflow gave fuller profiles than port, linked to flow acceleration at the railing where the lidar sat.
• ▸Gusts typically lasted about 10 s, were roughly symmetric and rarely exceeded a gust factor of 1.5.
• ▸Heavy-tailed velocity statistics persisted up to scales of about 15 km, far beyond typical laboratory turbulence.
• ▸The spectrum followed f−1 scaling over almost two decades, implying friction Reynolds numbers of order 107.
• ▸Ten representative 1 Hz time series have been released publicly in apparent and true wind form.

Opportunity

Wind is returning to commercial shipping as a way to cut fuel burn, most often through retrofitted rotor, suction, wing or kite sails. The authors note that most current installations are designed for fuel savings of 5% to 20%. Where a vessel lands in that range depends on the wind its sails actually see, the quantity measured least.

Data at the heights that matter, the lowest 100 m above the sea, exist mainly near coasts. On the open ocean, measurements are scarce, and time-resolved measurements scarcer still. Yet they are what is needed to predict savings, design against gust loads and build sail control systems.

The team, led from the Norwegian University of Science and Technology with Fugro Norway and SINTEF Energy Research, mounted two upward-facing continuous-wave lidars on the bow of the 182.8 m chemical tanker Bow Olympus during four trans-Atlantic crossings. The first two voyages ran in the ship's original configuration; for the last two, four 22 m suction sails were installed in pairs on deck. One lidar profiled the apparent wind from 10 m to 100 m above deck; the second sampled continuously at 22 m, roughly sail height, at about 1 Hz.

Why Wind Propulsion Has Not Yet Delivered

The paper points to a confidence problem rather than a hardware one. The authors report that wind has slipped in an industry ranking of viable propulsion options, with a lack of regulatory clarity cited as one reason, including how wind should be specified as a boundary condition. Guidance currently in circulation suggests power-law profiles with exponents of 1/9 or 1/7, but the bodies issuing such guidance need reliable data at relevant heights to go further. Without it, performance predictions and load cases rest on assumptions unchecked at sea.

Results

The Practical Applications

Assessment

The authors are candid about scope. The campaign covered only four crossings over a short period relative to seasonal variability, so broader measurements across seasons and atmospheric conditions are needed before the findings can be generalised. The limited vertical resolution made fitting boundary-layer profiles uncertain, lidar turbulence estimates are inherently less accurate than mean-speed estimates, and the flow around the hull affects the readings. The mechanics of the acceleration zones, and how the sails alter turbulence upstream, remain to be investigated.

Readers should also weigh the evidence behind the headline savings. The up-to-20% fuel figure comes from one vessel with one sail type used continuously, and the reported stabilising effect of the sails rests only on the crew's subjective observations. The port-starboard contrast may partly reflect the single asymmetric lidar position. Still, the core message is consistent with earlier open-Atlantic measurements – the inflow used in design, simulation and regulation of wind-assisted ships deserves reconsideration, and there is now an open dataset against which to test it.