Wind blows, water turns ninety degrees. The gentle physics of Ekman transport — the spiral beneath the surface, coastal upwelling, gyre hills and spills.
Here is a fact that breaks most people's intuition about the ocean: when wind blows steadily across open water, the surface layer it drives does not simply flow downwind. Averaged through its depth, that wind-driven layer slides off at a full ninety degrees to the wind's direction — to the right in the northern hemisphere, to the left in the southern. This is Ekman transport, and it is quietly one of the most consequential ideas in oceanography. It explains why cold nutrient-rich water rises along the great fishing coasts, why the centres of the ocean gyres are biological deserts, where spilt oil will drift, and why a restoration planner cares which way the wind blows in March. This article builds the physics gently, step by step.
In 1893, the Norwegian explorer Fridtjof Nansen deliberately froze the ship Fram into the Arctic pack ice, gambling that the ice's drift would carry it across the polar sea. It did — but not downwind. Nansen watched, month after month, as the ice moved twenty to forty degrees to the right of the wind, never along it. He reasoned that Earth's rotation must be deflecting the moving ice, and that layers of water beneath must drag on one another in some spiralling fashion. Being an explorer rather than a mathematician, he handed the problem to a young Swedish student, Vagn Walfrid Ekman. Ekman's 1905 solution became one of the founding results of physical oceanography.
Work through it in three steps. (We skim the same outline in our upwelling explainer; here we go slower and deeper.)
Step one — the surface layer. Wind drags the top skin of water with it through friction. In the northern hemisphere, Earth's rotation — the Coriolis effect — immediately begins deflecting that moving water to the right. In the classic Ekman solution, the equilibrium surface flow settles at about 45° to the right of the wind (real oceans usually show a smaller angle, since the assumptions are idealised).
Step two — the layers below. The moving surface layer drags the layer beneath it by internal friction; that second layer feels Coriolis deflection too and turns a little further right, moving a little slower; and so on downward. The result is the Ekman spiral: current direction rotating smoothly with depth while speed decays, the whole structure fading out over tens of metres — typically a few tens to around a hundred metres down, depending on latitude and how turbulent the water is.
Step three — add up the spiral. Integrate all those rotated, weakening velocity vectors from surface to the bottom of the spiral and a beautiful simplification appears: the net water transport points exactly 90° to the right of the wind (northern hemisphere; mirror it for the southern). Better still, the magnitude of that transport depends only on wind stress, water density and the local strength of Earth's rotational effect — the messy internal friction cancels out of the final sum. Oceanographers compute Ekman transport from wind data alone, hundreds of times a day, using that tidy result.
One nuance matters: the ice Nansen watched moved only 20–40° right of the wind, not 90°, because ice is a solid raft moved by wind stress directly at its surface, rather than a water column integrating a spiral. The discrepancy isn't a failure of the theory — it's the observation that inspired it.
Give a whole slab of ocean a sideways shove and the geometry of coasts, equators and gyres turns that shove into vertical motion — seawater's great conveyor switch.
Coastal upwelling. Where winds blow parallel to a coast with the coast on their left in the northern hemisphere — equatorward winds along California, Peru, Namibia — Ekman transport drives the surface layer offshore. Surface water leaving must be replaced, so cold, deep, nutrient-loaded water wells up along the shore. This single mechanism underwrites the planet's great eastern boundary current ecosystems: sardines, anchoveta, squid — and the fastest kelp growth on Earth. Our atlas guide to Point Reyes walks a headland where you can watch the fog and cold water arrive together when the northwesterlies set in.
Coastal downwelling. Reverse the wind and Ekman transport drives surface water toward the coast, where it piles up and is forced downward. Downwelling coasts lose their nutrient tap: surface water warms, stratifies and starves. Many coasts seasonally alternate between the two regimes — a detail that decides when kelp gets fed and when it merely endures.
Equatorial divergence. At the equator the rotational effect flips sign across the line. Westward-blowing trade winds therefore push surface water northward just north of the equator and southward just south of it: the surface parts, and deep water rises along the seam — the cold tongue visible on any global temperature map.
Ekman pumping inside the gyres. Over the open ocean, rotating wind systems spin up convergences and divergences through their curl. In the subtropical gyres, Ekman transport pushes surface water gently toward the gyre centres, piling the sea surface up by roughly a metre (measured by satellite altimetry) and forcing slow downwelling — which is why gyre centres are the ocean's starved deserts, a story we tell in full in Ocean Gyres Explained.
Ice, oil and drifters. Any floating thing that rides the wind — sea ice, oil slicks, debris, life rafts — inherits some combination of wind drag and Ekman flow, drifting at an angle downwind rather than straight downwind. Search-and-rescue leeway models and oil-spill trajectory forecasts have Ekman's physics baked into their cores. Even the larvae in our piece on how marine life colonises new reefs are playing against this current geometry.
For anyone who grows or restores kelp, Ekman transport is not trivia; it is plumbing. Whether a reef gets fed each spring depends on whether the seasonal winds are oriented to haul surface water away from that coast. Marine heatwaves — covered in our marine heatwaves explainer — often coincide with the winds failing: when the upwelling-favourable flow slackens, warm surface water lingers and the nutrient tap closes at once. Conversely, a well-placed cape that concentrates Ekman-driven upwelling can buffer a forest through an otherwise brutal summer.
This is why restoration siting — including the illustrative scoring we use at Tidewright — starts with wind climatology: not just "how warm is this coast" but "which direction does the water leave this coast, month by month, and what arrives from below to replace it." A reef that current geometry feeds is infrastructure; a reef it starves is a gamble.
Ekman transport versus the Ekman spiral. The spiral is the depth structure of the flow; the transport is its integrated, net sideways shove. You can measure one and infer the other from wind.
Ekman flow versus geostrophic currents. Ekman flow is wind's direct, friction-driven response in the surface layer. Below and beyond it, most steady ocean currents are geostrophic — set up by pressure gradients (like the piled-up gyre centres Ekman pumping creates) and steered by rotation. The two stack: wind makes the pile, the pile drives the great boundary currents, and those currents shape climate and kelp distributions in turn.
Ninety degrees is net, not surface. The surface water itself moves at a fraction of the wind speed at an angle well under 90°; the pure right angle belongs to the column as a whole.
For the net transport through the whole wind-driven layer, theory is robust and observations support it well in steady conditions. The surface current angle is smaller and variable — often just 10–45° — because real turbulence and stratification differ from the idealised assumptions.
The Coriolis deflection reverses sign across the equator: moving water veers left instead of right. So southern-hemisphere Ekman transport runs 90° to the left of the wind — which is why equatorward winds cause upwelling on South America's west coast just as they do on North America's.
The surface Ekman current typically runs at about one to a few percent of the wind speed — slow-looking, but applied across thousands of kilometres of coastline and months of winds, it moves enormous volumes and sets the vertical exchange that fertilises whole ecosystems.
Yes. Upwelled deep water is naturally low in oxygen and rich in dissolved CO₂, and in some systems hypoxic or acidified upwelling has reached shallow shelf waters, stressing shellfish and fisheries. The same valve that opens the nutrient tap can deliver these stressors.
Indirectly. The 90° rule needs long, steady winds and open fetch to develop fully, so small enclosed basins are dominated by tides and local topography. But the bay's background temperature and nutrients are still set by the shelf-scale Ekman regime outside its mouth.
A deep dive into upwelling physics: Ekman transport, the world's great upwelling systems, how scientists track them from space, and what climate change does.
Dr. Lena Fischer ·
Five vast gyres turn like slow wheels in every ocean basin, shaping currents, climate and where plastic gathers. Here's how ocean gyres actually work.
Dr. Lena Fischer ·
Every great kelp forest sits in the shadow of a great current. Here's how western and eastern boundary currents shape temperature, nutrients and the world's temperate reefs.
Dr. Lena Fischer ·