High School Marine Science

A rope five miles long

On 23 March 1875, in the western Pacific near the Mariana Islands, sailors on HMS Challenger lowered a lead weight of about 450 pounds on a hemp line. The line ran out over a wheel, and kept running, until the weight struck bottom at 4,475 fathoms. A fathom is six feet, about 1.83 metres, so the bottom lay roughly 8,180 metres down: five miles of rope. Nobody had measured water that deep before.

That number came from one rope at one spot. It said nothing about the seafloor a kilometre away. This lesson follows the people and machines that turned single points into maps of whole ocean basins, and it ends with an uncomfortable fact: even now, less than a third of the seafloor has been measured in detail.

What a weighted rope can and cannot tell you

Challenger was a wooden warship about 200 feet long, with 15 of its 17 guns taken out to make room for laboratories. It left Portsmouth in December 1872 and spent 1,250 days on a voyage of 68,890 nautical miles through the Atlantic and Pacific. At 362 stations the crew stopped the ship and recorded the depth, the water temperature at several depths, the speed and direction of the surface current, and samples of the mud and animals on the bottom. The results filled 50 volumes that took 23 years to publish, and they described more than 4,700 new species. The expedition's naturalist John Murray called the work "the greatest advance in the knowledge of our planet."

The deep soundings were feats of engineering. The hemp line had a breaking strain of about 700 kilograms, and a rolling ship can jerk a line harder than its load alone. So the crew hung forty rubber "accumulators" on it, stretchy bands that soaked up the jerks. The sounding devices were built to drop their heavy weights when they struck bottom, so that the crew hauled back only a metal tube, and that tube had punched into the bottom and held a short core of sediment. A small steam engine wound the line back in.

Now look at what the method could not do. One sounding gave one depth, at one place, with the ship stopped. If a current bent the line into a curve, more rope ran out than the true depth, and the number came out too large. Over three and a half years, Challenger stopped at 362 stations across two oceans. Imagine trying to draw the mountains of North America from 362 measured heights scattered from coast to coast, a few hundred kilometres apart. You would know the continent is high in places and low in others, but you could not draw a single mountain range.

Key idea: A weighted line measures depth at a single point. Mapping the shape of the seafloor needed a way to measure depth quickly and continuously while the ship kept moving.

Listening for the bottom

The way to do that was sound. An echo sounder sends a short pulse of sound down from the hull and listens for the echo off the bottom. Sound travels through seawater at about 1,500 metres per second, more than four times faster than in air. The pulse goes down and comes back, so the depth is half the distance it travels:

depth = (speed of sound x two-way travel time) / 2

Work one. The echo returns 5.0 seconds after the pulse leaves. The sound travelled 1,500 x 5.0 = 7,500 metres in total, down and back up, so the water is 7,500 / 2 = 3,750 metres deep. The most common mistake is to forget the halving and report 7,500 metres.

Now change one input. The speed of sound is not exactly 1,500 metres per second: it rises with temperature, salinity and pressure. Suppose the water you are surveying actually carries sound at 1,530 metres per second. The same 5.0 second echo then means 1,530 x 5.0 / 2 = 3,825 metres, 75 metres deeper than your first answer. That is why survey ships still measure the speed of sound in the water they are working in before they trust their depths.

The German research ship Meteor showed what echo sounding could do. Between April 1925 and June 1927 it crossed the South Atlantic again and again, fourteen cross-sections in all, and made about 67,000 echo soundings. Its records showed the Mid-Atlantic Ridge clearly for the first time: not a gentle swell in the middle of the ocean but a rugged mountain range.

Marie Tharp sees a valley

In 1948 Marie Tharp, who had studied geology and mathematics, went looking for work at Columbia University. The geophysicist Maurice Ewing asked her one question, "Can you draft?", and hired her. Women were not then allowed on most research cruises, so her colleague Bruce Heezen collected echo-sounding records at sea while Tharp turned them into pictures of the bottom on land.

In 1952 she plotted six roughly parallel profiles across the North Atlantic, each a long line of depths from one side of the ocean to the other. The mountains did not match from profile to profile, but one thing did. Near the crest of the ridge, each profile dipped in a V-shaped notch. Tharp thought the notches were one continuous valley running down the middle of the ridge, a rift valley like the one in East Africa, where the land is being pulled apart.

Heezen's first reaction, in her own account, was "It cannot be. It looks too much like continental drift." Continental drift was then an unpopular idea, and he dismissed her reading as "girl talk." What changed his mind was a second, independent dataset. A colleague, Howard Foster, had hand-plotted the locations of tens of thousands of undersea earthquakes. When the two maps were laid over each other on a light table, the earthquake centres fell inside the valley. By about mid-1953 Heezen accepted it.

Tharp drew the North Atlantic as a physiographic diagram, a shaded drawing of the bottom as it would look from a low-flying plane if the water were drained away. Part of the reason was secrecy: during the Cold War, detailed depth charts of the seafloor were restricted, and a picture could show the shape of the bottom without publishing classified numbers. The North Atlantic map was first published in 1956. Heezen and Tharp went on to the South Atlantic and Indian Oceans, and in 1977, with the painter Heinrich Berann, they published the World Ocean Floor panorama, which showed a ridge system winding through every ocean like the seam on a baseball.

That rift valley is where new seafloor forms as plates move apart, one of the observations that made plate tectonics the accepted theory of Earth's surface. If you want the full argument, it is taught as a case that was won in High School Earth & Space Science. In this course you need only the shape it left behind, which is Lesson 2.

Key idea: Tharp's discovery came from comparing several independent profiles and then testing the pattern against a second kind of data, the earthquake locations. One odd profile is a curiosity; the same notch in six profiles, lined with earthquakes, is evidence.

From one beam to a fan

An older single-beam echo sounder measures one depth at a time, straight down, so a ship's record is a line of depths along its track. A multibeam sonar sends out many beams at once in a fan, covering the bottom directly beneath the ship and a strip out to each side. A survey ship mows the ocean like a lawn, overlapping strips until an area is covered completely.

Multibeam systems record two things. One is depth, from the travel time of each beam's echo. The other is backscatter, the strength of the echo, which tells hard rock (a strong echo) from soft mud (a weaker one). There is a catch you can predict from the geometry: the deeper the water, the wider each beam's patch of seafloor, so a multibeam map of deep water is less detailed than one of shallow water.

Mapping the seafloor from space

Radar cannot see through kilometres of water, yet many world maps show the whole seafloor. How? The sea surface is not perfectly level. A seamount, an undersea mountain, adds a little extra mass, and its extra pull draws water toward it. NOAA's altimetry laboratory explains that a mountain several kilometres high raises the sea surface above it by only a few metres, tilting it by about 30 microradians, a slope far too gentle to see. Satellites carrying radar altimeters measure the height of the sea surface precisely enough to find those bumps and, by the same reasoning, the dips over trenches.

NOAA is frank about the limits. An undersea mountain has to be about a mile high and several miles wide to raise a bump that today's altimeters can pick out, and depths worked out this way are "not as accurate or detailed as echo sounding from ships." A satellite map is an estimate of the seafloor, very good for finding big features and poor for anything small.

How much has been measured

In 2017 the Nippon Foundation and GEBCO, the international project that compiles the world's seafloor map, launched Seabed 2030, with the goal of a complete map of the seabed by 2030. At the launch, only about 6 percent of the seabed had been mapped to modern standards. In April 2026 the project announced that 28.7 percent had been mapped, about 104 million square kilometres, with almost five million square kilometres added in the previous year. The remaining seven tenths of the seafloor is still known mostly from satellite estimates and scattered old soundings.

Why care? Tharp's own story gives one answer. In 1929 an earthquake off the Grand Banks of Newfoundland set off underwater avalanches of sediment that snapped transatlantic telegraph cables, and Bell Laboratories later asked Heezen's group where new cables could be laid safely. Cables on the seafloor still carry telephone and internet traffic between continents, and they still have to avoid slopes where sediment slides. The shape of the bottom also steers currents and tsunamis, decides where fish gather and where corals grow on seamounts, and keeps ships and submarines off uncharted rocks.

Common misconceptions

• "Satellites have mapped the whole seafloor." They have estimated it, from bumps in the sea surface. Those estimates miss features smaller than about a mile high and several miles wide; measured, detailed depths still cover less than a third of the seafloor.
• "Sonar sees the bottom like a camera." An echo sounder times echoes. The depth comes from a calculation that depends on the speed of sound, which has to be measured, and the echo strength says something about the bottom's hardness, not what it looks like.
• "The deep trenches were found by modern submarines." The Mariana Trench was first sounded with a rope in 1875, which is why its deepest point is called the Challenger Deep.
• "Tharp simply drew what the data showed." The data were a few lines of depths with wide gaps between them. Tharp had to interpret, using geology to connect features across the gaps, and the interpretation was tested against independent earthquake data before it was accepted.

What to carry forward

• Challenger's rope soundings (1872-1876) gave single depths at 362 stations, including about 8,180 m near the Mariana Islands, but no map.
• Echo sounding measures depth as speed of sound x two-way time / 2, with the speed taken as about 1,500 m/s unless measured; the Meteor's 67,000 soundings first showed the Mid-Atlantic Ridge clearly.
• Marie Tharp found a V-shaped rift valley along the ridge crest in 1952 profiles, confirmed by earthquake locations, and mapped the ocean floors with Heezen through the 1977 world panorama.
• Multibeam sonar maps a swath of depths and backscatter; detail falls as water deepens.
• Satellite altimetry estimates the seafloor from tiny bumps in the sea surface, good for large features and poor for small ones.
• By April 2026, 28.7 percent of the seafloor had been mapped to modern standards (Seabed 2030).

Sources

1. Tharp, M. (1999). Connect the dots: Mapping the seafloor and discovering the mid-ocean ridge. In L. Lippsett (Ed.), Lamont-Doherty Earth Observatory of Columbia: Twelve perspectives on the first fifty years, 1949-1999. Columbia University. Republished by the Columbia Climate School (2020). news.climate.columbia.edu
2. National Oceanography Centre. (n.d.). The Challenger expedition. noc.ac.uk
3. University of Edinburgh. (n.d.). Measuring the depths. In Sea change: The Challenger expedition [Online exhibition]. exhibitions.ed.ac.uk
4. GEOMAR Helmholtz Centre for Ocean Research Kiel. (2025). 100 years of the Meteor expedition. geomar.de
5. NOAA Ocean Exploration. (2020, updated 2023). What is multibeam sonar? oceanexplorer.noaa.gov
6. NOAA NESDIS Laboratory for Satellite Altimetry. (n.d.). Altimetric bathymetry. star.nesdis.noaa.gov
7. International Hydrographic Organization. (2026, April 28). Global seabed mapping reaches new milestone as five million square kilometres added in a year. iho.int
8. NOAA National Ocean Service. (2026). How deep is the ocean? oceanservice.noaa.gov
9. University of Rhode Island and Inner Space Center. (n.d.). How fast does sound travel? Discovery of Sound in the Sea (dosits.org). Used for the speed of sound in seawater, about 1,500 m/s.

Cables that broke in order

On 18 November 1929, a magnitude 7.2 earthquake shook the seafloor south of Newfoundland. Twelve telegraph cables crossed that stretch of ocean, carrying messages between North America and Europe. Six of them broke at the moment of the earthquake. The other six broke one at a time over the next 13 hours and 17 minutes, each one farther from shore and in deeper water than the one before, the last almost 500 km away.

An earthquake lasts seconds, not thirteen hours, so something else was travelling down the seafloor and snapping each cable as it arrived. In 1952 Maurice Ewing and Bruce Heezen, the same Heezen you met in Lesson 1, worked out what: the shaking had set off an underwater landslide that turned into a turbidity current, a dense flow of mud and sand racing down the slope. From the times and the cable positions, it moved at an estimated 60 to 100 km per hour and carried 100 to 150 cubic kilometres of sediment, spreading it as far as 1,000 km out onto the deep seafloor.

That story only makes sense once you know the shape of the bottom: a shallow shelf, a steep slope, and a long gentle apron leading to flat plains far below. This lesson builds that shape feature by feature, compares the features in one table, and then reads what covers them.

The features side by side

Picture walking away from a beach and never stopping. The table below lists what you would cross, in the order you would cross it on a typical Atlantic coast, plus three features you meet elsewhere. The depths are typical values; real features vary a lot.

A diagram helps here. NOAA's ocean floor features collection has a labelled illustration of a margin, a plain, a seamount and a trench; keep it open beside the table.

Reading the table: two families of features

Read down the "What builds it" column and the features sort themselves into two families.

The first family, the shelf, slope and rise, is the edge of a continent. The shelf is continental crust that happens to be under water. Its outer edge, the shelf break, sits at a strikingly similar depth around the world, about 135 m on average. The slope is where continental crust ends and the seafloor drops toward the deep. And the rise is the debris pile at the bottom of that drop, built from material that slid or flowed off the shelf, the way the 1929 flow did. These features are made mostly by sediment from land moving downhill.

The second family, the ridges, trenches and seamounts, is made by plate motion and volcanoes. At a mid-ocean ridge, plates move apart and magma rises to fill the gap. New crust is hot, so it is less dense and floats higher on the mantle beneath, which is why the ridge stands two or three kilometres above the plains on either side. As crust moves away from the ridge it cools, shrinks and sinks. That is also why the deepest water is not in the middle of the ocean but often near its edges, in trenches, where old, cold, dense plate bends down and slides beneath another plate. The Mariana Trench marks where the Pacific Plate dives beneath the Philippine Sea Plate.

Seamounts are volcanoes that never reached the surface, or that did and then sank. A volcano that breaks the surface becomes an island, and waves can plane its top flat. If the crust beneath it then subsides, the flat-topped mountain drops below the waves and becomes a guyot. Over a hot spot, a fixed plume of hot mantle, the moving plate carries a line of volcanoes away like a conveyor belt. The rock of the island of Hawaii, the youngest island, is almost all less than a million years old; the oldest of the Hawaiian-Emperor seamounts, far to the northwest, are around 80 million years old.

How fast do plates move apart? Slowly, and at different speeds. The Atlantic ridge spreads at less than 2 cm a year on each side in places, slower than your fingernails grow, and the fast Pacific ridges approach 10 cm a year on each side. The oldest crust on either side of the North Atlantic is around 180 million years old. The full evidence for plate tectonics is taught in High School Earth & Space Science; here you need its consequences for the shape of the floor.

Remember: The edges of the ocean are built mainly by sediment moving off the continents. The deep features, ridges, trenches and seamounts, are built by plates moving apart, plates sinking, and volcanoes.

A worked reading: how steep is "steep"?

Profiles of the seafloor, including the one you will draw in the investigation, are almost always vertically exaggerated: the depth axis is stretched compared with the distance axis, or nothing would be visible. Tharp plotted her profiles at 40 times vertical exaggeration. So a slope that looks like a cliff on a profile is usually gentle in real life.

Work it out. The continental slope averages about 4 degrees. The tangent of 4 degrees is about 0.07, so for every kilometre you travel outward, the floor drops about 70 m. To fall from the shelf break at 135 m to a rise at 3,500 m, a drop of about 3,365 m, takes about 3,365 / 70 = 48 km of travel. A 4 degree slope is a 7 percent grade, like a steep stretch of highway, yet it carries the seafloor down more than three kilometres in less than 50 km.

Now change the input. Suppose a canyon wall sloped at 20 degrees, where the tangent is about 0.36. The same 3,365 m drop would then take only about 9 km. Slopes like that are where sediment piles up and then fails, which is how turbidity currents start.

What covers the floor

Almost everywhere, the volcanic crust of the deep seafloor lies buried under sediment. Oceanographers sort sediment by where it came from.

A sediment counts as an ooze when at least 30 percent of it is the remains of microscopic organisms. If you have seen the White Cliffs of Dover, you have seen coccolithophore ooze that hardened into chalk and was later lifted above the sea.

Why does calcareous ooze stop at about 4 km? Calcium carbonate dissolves more readily in more acidic water, and deep water is cold and holds more dissolved carbon dioxide, which makes it more acidic. The depth at which carbonate dissolves as fast as it rains down is called the carbonate compensation depth, usually about 4-4.5 km, and shallower near the poles. Think of it as a snow line upside down: ridge crests and seamount tops stand above it and are dusted with white carbonate ooze, while deep basins below it collect only clay. Keep the carbonate compensation depth in mind: Lesson 24 shows why adding carbon dioxide to the ocean matters for shells.

Sediment also keeps a clock. Near a ridge the crust is brand new, so it has had little time to collect sediment, and the cover is thin. Farther away, the crust is older and the sediment thicker, by roughly 100-200 m for every 1,000 km from the ridge. Thin sediment at the ridge and thicker sediment away from it is exactly what seafloor spreading predicts.

Try the arithmetic. Deep-sea clay builds up at about 1 mm per 1,000 years. A 50 m layer is 50,000 mm, so it took about 50,000 x 1,000 = 50 million years to accumulate. The same thickness of calcareous ooze, at about 1 cm per 1,000 years, would take about 5 million years.

Remember: Sediment type depends on what reaches the bottom and survives there: land debris near the coasts, plankton shells under productive water and above the carbonate compensation depth, and slow-growing clays and nodules in the remote deep.

Common misconceptions

• "The deepest water is in the middle of the ocean." The middle of the Atlantic is a mountain range. The deepest water is in trenches, which lie mostly near the edges of the Pacific, where plates sink.
• "The continental shelf is part of the deep ocean." It is part of the continent, flooded to a depth of about 150 m or less. The deep ocean begins at the foot of the slope.
• "Mid-ocean ridges are folded mountains like the Himalayas." The Himalayas formed where continents collided. Ridges are volcanic, built where plates pull apart, and they stand high because the new crust is hot.
• "A seafloor profile shows the real steepness." Profiles are stretched vertically, often by tens of times (Tharp used 40). A 4 degree slope can look like a cliff.
• "The seafloor is bare rock." Most of it is under sediment, from a thin skin at the ridges to kilometres thick along some margins.

The short version

• From the coast outward: shelf (to about 135-150 m), slope (about 4 degrees, to 3,000-5,000 m), rise (a sediment apron), abyssal plain (about 4,500-6,000 m, roughly 70 percent of the ocean floor).
• Mid-ocean ridges, more than 65,000 km long, stand high because new crust is hot; trenches, the deepest places, form where plates sink; seamounts and guyots are volcanoes, many over hot spots.
• Turbidity currents carry sediment down slopes fast; the 1929 Grand Banks flow broke cables in order over 13 hours.
• Sediments are lithogenous, biogenous, hydrogenous or cosmogenous. Calcareous ooze collects above the carbonate compensation depth, about 4-4.5 km; deep basins collect clay at about 1 mm per 1,000 years.
• Sediment thickens with distance from a ridge, because the crust there is older.

Sources

1. Derouin, S. (2017, October 27). Benchmarks: November 18, 1929: Turbidity currents snap trans-Atlantic cables. EARTH Magazine, American Geosciences Institute. earthmagazine.org
2. Gownaris, T. (n.d.). 1.2: Continental margins. In Oceans of data. Geosciences LibreTexts, CC BY 4.0 (adapted from S. Earle, Physical geology). geo.libretexts.org
3. Webb, P. (2019). Introduction to oceanography. Roger Williams University Open Publishing; Geosciences LibreTexts edition, CC BY 4.0. Sections 4.5 (divergent boundaries), 4.6 (convergent boundaries), 4.9 (seamounts and hot spots), 12.3-12.4 (biogenous and hydrogenous sediments) and 12.6 (sediment distribution). geo.libretexts.org
4. NOAA Office of Education. (2026). Ocean floor features [Resource collection]. noaa.gov
5. NOAA National Ocean Service. (2026). What is the longest mountain range on Earth? oceanservice.noaa.gov
6. Lamont-Doherty Earth Observatory. (n.d.). About GMRT: Global Multi-Resolution Topography synthesis. Columbia University. gmrt.org

A clock made of salt

In November 1899 the Irish physicist John Joly published an estimate of the age of the Earth worked out from salt.^[4] Rivers carry sodium into the sea every year. Joly estimated how much, compared it with the sodium already dissolved in the ocean, and assumed the rate had never changed. His answer was 90 to 100 million years. The idea was not new: Edmond Halley had suggested timing the ocean by its salt in a paper read to the Royal Society in 1715.

Joly was badly wrong. Within a few years, dating rocks by radioactive decay gave ages in the billions of years, and today rocks and meteorites put the Earth at about 4.5 billion years old (the method is taught in High School Earth & Space Science). So here is the problem for this lesson. Rivers really do pour salt into the sea every year, and the sea really is old, yet the ocean is not getting steadily saltier. Where is the salt going? The answer explains what seawater is made of, and why its saltiness matters to everything living in it.

Working the salt clock

Take two figures from a U.S. Geological Survey booklet on exactly this question. The ocean holds roughly 50 quadrillion tons of dissolved salts: 50,000,000,000,000,000 tons. Rivers carry about 4 billion tons of dissolved salts to the sea each year.

If salt only ever came in, how long would rivers take to deliver what is there now? Divide: 50,000,000,000,000,000 / 4,000,000,000 = 12,500,000. About 12.5 million years. Yet the rocks hold fossils of sea creatures hundreds of millions of years old, and those creatures needed salty water too. A sea filling with salt for hundreds of millions of years would have become far saltier than it is.

So salt must be leaving about as fast as it arrives. The USGS booklet says so directly: "About the same tonnage of salt from the ocean water probably is deposited as sediment on the ocean bottom, and thus, yearly gains may offset yearly losses." Think of a bathtub with the tap running and the drain open. When water drains out as fast as it pours in, the level stays put, however long the tap runs. Oceanographers call that a steady state. Joly's clock assumed a bathtub with no drain.

The point: The ocean's saltiness stays roughly steady because salt is removed about as fast as rivers supply it. A steady level tells you nothing about how long the tap has been running.

Seawater is not concentrated river water

If the sea were simply river water that had been piling up, it should contain the same salts in the same proportions. It does not. The table compares the share of each dissolved substance in seawater (from William Dittmar's analysis of the Challenger samples) and in average river water, as the USGS booklet gives them.

Read the table from both ends. Sodium and chloride, the two halves of table salt, are a little more than 85 percent of the salt in the sea but less than 16 percent of the salt in rivers. Calcium and bicarbonate are nearly half of what rivers carry and less than 2 percent of seawater. Rivers bring more calcium than chloride, yet the ocean holds about 46 times more chloride than calcium.

The explanation is that different substances leave at very different speeds. Living things pull some of them out fast. Oysters, clams, corals, crabs and tiny foraminifera take calcium to build shells and skeletons, and diatoms take silica for theirs, as Lesson 2's oozes showed. But as the USGS booklet notes, "no known biological process removes the element sodium from the sea." Sodium and chloride leave only by slow routes, for example when seawater is trapped and evaporated in a nearly enclosed basin and leaves a bed of salt behind. The Mediterranean did that about 6 million years ago, when it nearly dried out, and thick salt beds still lie buried beneath its floor. Hydrothermal vents and undersea eruptions also add and remove dissolved substances.

Residence time: how long a salt stays

The idea that ties this together is residence time, the average time an atom of a substance spends in the ocean before it is removed. In a steady state you can estimate it by dividing the amount in the ocean by the amount arriving each year. Using the rounded USGS figures from the table:

Calcium passes through the ocean in under a million years; sodium and chloride linger for tens of millions. Substances with long residence times pile up until they dominate, which is why the sea tastes of table salt and not of the calcium bicarbonate that rivers mostly carry. These are rough numbers, since the inputs are estimates and chloride also comes from volcanic sources, but the pattern is robust.

Now look back at Joly. His 90 to 100 million years is the same kind of number as the sodium row. What he had really measured was roughly how long sodium stays in the sea, not how long the sea has existed.

Constant proportions, and how salinity is measured

Dittmar spent nine years analysing 77 seawater samples collected by Challenger from all over the world. He found that six substances, chloride, sodium, sulfate, magnesium, calcium and potassium, make up 99 percent of the dissolved solids, and that their ratios are nearly the same everywhere, even where the total saltiness differs. Ocean water can be more or less concentrated, but the recipe stays the same. This is the principle of constant proportions, and it follows from the long residence times: the major ions stay long enough for currents to mix them through the whole ocean many times over.

Salinity is the total mass of dissolved salts in a kilogram of seawater, so the average, about 35, means about 35 grams of salt per kilogram, or 3.5 percent. Constant proportions made measurement easy: measure one ion, usually chloride, and you can calculate the rest. Today oceanographers measure electrical conductivity instead, since saltier water conducts electricity better, and report practical salinity, where 1 unit is about one part per thousand.

The open ocean is rarely below 33 or above 38. The USGS booklet gives a maximum of about 36 near latitudes 20 degrees north and south, where hot, dry, windy air evaporates more water than rain replaces; about 35 at the equator, where heavy rain offsets evaporation; and a minimum of about 31 near 60 degrees north, where rain and melting ice dilute the sea. Enclosed seas go further: about 40 in the Red Sea and the Persian Gulf, where evaporation is intense, and only about 5 to 15 in the Baltic, which is fed by many rivers.

What salt and heat do to density

Salinity matters to sea life directly, as Lesson 14 shows for fish, but it also does something to the water itself: it changes density, how much mass is packed into each litre. Fresh water at 4 °C has a density of 1.000 grams per cubic centimetre. Dissolved salt raises surface seawater to between 1.02 and 1.03. Temperature works the other way: warming fresh water from 0 °C to about 32 °C lowers its density from 0.99987 to 0.99510 grams per cubic centimetre, about half a percent.

Which matters more in the real ocean? Compare the ranges. Open-ocean salinity varies only from about 33 to 38, a change worth a few tenths of a percent in density, while surface temperatures run from about -2 °C to 30 °C. Across most of the open ocean, temperature has the greater effect. Salinity takes over where the temperature is nearly uniform and fresh water arrives, in polar seas, near ice and off river mouths.

Because light water floats on dense water, the ocean stacks itself in layers:

• The mixed layer, the top 100-200 m or so, is stirred by wind and waves and has nearly uniform temperature and salinity.
• Below it, the thermocline is a zone where temperature drops quickly with depth. The halocline is a zone where salinity changes quickly, and the pycnocline is where density increases quickly. In most of the ocean the pycnocline coincides with the thermocline.
• Beneath that, the deep ocean is cold, about 2 °C, all the way to the bottom.

Warm, light water sitting on cold, dense water is stable: it resists mixing, the way oil sits on vinegar. That stability decides a great deal in later lessons. It keeps nutrients locked in the deep, away from the sunlit plants that need them (Lesson 9), and it limits how far oxygen from the air reaches down (Lesson 4). In polar seas there is little difference between surface and deep temperatures, the layering is weak, and the water can mix deeply.

Ice, brine and the densest water

Salt also lowers the temperature at which water freezes. The National Snow and Ice Data Center gives -1.8 °C (28.8 °F) for ordinary seawater, and each increase of 5 in salinity lowers the freezing point by another 0.28 °C. When seawater does freeze, the ice crystals push most of the salt out. Some is trapped as pockets of brine that slowly drain away, so new sea ice is salty but ice that survives four years or more is nearly fresh.

The salt that the ice rejects goes into the water just below. That water becomes saltier, and it is already at its freezing point, so it is both very cold and salty: the densest combination there is. It sinks. Where winds keep pushing new ice away from a coast, in openings called coastal polynyas, ice forms over and over, and NSIDC describes those polynyas as a major source of the world's bottom waters. Lesson 7 follows that water as it spreads along the seafloor.

The point: Cold and salt both make seawater denser. The ocean sorts itself into a light surface layer over dense deep water, and the densest water of all is made where sea ice forms and leaves its salt behind.

Measuring the layers today: Argo floats

For most of history, a temperature and salinity profile meant a ship stopping to lower instruments on a wire. Since 1999 the Argo program has done it with robots. In 2019 NOAA described an array of nearly 4,000 floats maintained by 26 countries. Every 10 days each float sinks to 2,000 m, measures temperature and salinity as it rises (salinity is calculated from conductivity), then surfaces and sends its data by satellite. The data are free to anyone, which is how you will read the ocean's layers yourself in this module's investigation.

Common misconceptions

• "The sea is salty because salt rocks on the seafloor dissolve into it." NOAA identifies rocks on land as the major source: slightly acidic rain weathers them, and rivers carry the ions to the sea. Hydrothermal vents and undersea eruptions add more.
• "The ocean keeps getting saltier." Over long spans, salt is removed about as fast as it arrives, so the saltiness stays roughly steady.
• "Seawater is just table salt dissolved in water." Sodium and chloride are about 85 percent of the dissolved salts; sulfate, magnesium, calcium, potassium and others make up the rest, in nearly constant proportions.
• "Salt water freezes at 0 °C, like fresh water." Ordinary seawater freezes at about -1.8 °C.
• "Sea ice is frozen seawater, so it is as salty as the sea." Ice pushes salt out as it forms, and old sea ice is nearly fresh.

Where this leaves us

• Rivers deliver about 4 billion tons of dissolved salts a year to an ocean holding about 50 quadrillion tons; the saltiness stays roughly steady because removal keeps pace.
• Residence time = amount in the ocean / yearly input: under a million years for calcium, tens of millions for sodium and chloride, which is why those two dominate.
• Dittmar's 77 Challenger samples showed constant proportions: six ions make up 99 percent of the dissolved salts, in nearly the same ratios everywhere. Average salinity is about 35 g/kg.
• Salinity is highest in the subtropics, lower at the equator and in polar and river-fed seas.
• Cold, salty water is dense. The ocean layers into a mixed layer, a thermocline or pycnocline, and cold deep water at about 2 °C; the layering resists mixing.
• Seawater freezes at about -1.8 °C; rejected brine makes the densest water, which sinks to the bottom.

Sources

1. Swenson, H. (1994). Why is the ocean salty? U.S. Geological Survey General Interest Publication. Source of the salt budget, the seawater and river water composition tables and Dittmar's data. pubs.usgs.gov
2. NOAA National Ocean Service. (2026). Why is the ocean salty? oceanservice.noaa.gov
3. Trinity College Dublin School of Physics. (n.d.). John Joly (1857-1933). tcd.ie
4. Joly, J. (1899). An estimate of the geological age of the earth. Scientific Transactions of the Royal Dublin Society, Series 2, 7, 23-66.
5. Webb, P. (2019). Introduction to oceanography. Roger Williams University Open Publishing; Geosciences LibreTexts edition, CC BY 4.0. Sections 5.3 (salinity patterns), 6.2 (temperature), 6.3 (density) and 12.4 (evaporites). geo.libretexts.org
6. National Snow and Ice Data Center. (n.d.). Science of sea ice. nsidc.org
7. U.S. Geological Survey Water Science School. (n.d.). Water density. usgs.gov
8. NOAA Research. (2019, October 9). NOAA invests in new tools to measure the ocean. research.noaa.gov
9. Euro-Argo ERIC. (n.d.). European fleet monitoring. euro-argo.eu

A fish at 8,336 metres

In 2022 a team of Japanese and Australian scientists lowered a baited camera into the Izu-Ogasawara Trench, south of Japan. At 8,336 metres it filmed a small, pale snailfish nosing around the bait, the deepest a fish has ever been recorded. The animal was soft, with a gelatinous covering instead of scales and a flexible skeleton.

Most people's first reaction runs something like this: "The pressure down there must be enormous. Submarines need thick steel hulls to survive it. So how can a soft little fish not be squashed flat?" That reasoning sounds airtight. It is wrong in an instructive way, and tracing exactly where it breaks teaches you how pressure, and then light and dissolved gases, change with depth.

The wrong answer, link by link

Lay the reasoning out as a chain:

1. The pressure in the deep sea is enormous.
2. Enormous pressure crushes things.
3. So a deep-sea animal must be built like a submarine, or it would be crushed.

The first link is true. The second is where it breaks, and the third falls with it. But there is a real limit to how deep fish live, so pressure does matter, just not in the way the chain says.

Link 1 holds: working out the pressure

At sea level the air presses on you at 14.7 pounds per square inch, a pressure called one atmosphere. NOAA gives the rule for water: for every 10.06 metres (33 feet) you descend, pressure increases by one more atmosphere. The total, or absolute, pressure at a depth is the air's one atmosphere plus the water's share.

At the snailfish's depth, every square inch of its body carries about 830 x 14.7, roughly 12,000 pounds. That is a fact, not an exaggeration. So why is the fish fine?

Link 2 breaks: pressure crushes differences, not things

A submarine is crushed by the difference between the water pressure outside and the air at one atmosphere inside, where the crew breathes. The hull has to hold back hundreds of atmospheres of difference. Take away the air inside and there is no difference to resist.

A fish's body is mostly water, and water is very hard to compress. Webb's oceanography text points out how little it squeezes: even under the full weight of the ocean, water compresses so slightly that sea level is only about 50 m lower than it would be if water could not be compressed at all, over an average depth of nearly 3,700 m. The water inside the fish is at the same pressure as the water outside, pushing back equally in every direction. Nothing is being squeezed into a smaller space, so nothing collapses.

What pressure does crush is gas. Gas compresses easily: double the pressure and its volume halves. Research crews show this with a tradition you can find in many expedition photos: they decorate foam cups, which are full of tiny air bubbles, tie them to an instrument and send them down. The cups come back shrunken but in perfect shape, because the pressure pushed evenly from every side and squeezed the air out of the foam. The snailfish has no gas-filled swim bladder, which is one reason it does well where gas-filled spaces would be a problem. Diving mammals do carry air, in their lungs, and NOAA explains that whales handle it with rib cages bound by flexible cartilage that let the lungs collapse without rupturing. Lesson 18 follows them down.

What matters here: Pressure crushes things only where there is a difference between inside and outside, which means gas spaces. Water-filled bodies feel the same pressure inside and out and are not crushed.

Link 3 rebuilt: the real limit is chemistry

So is pressure harmless? No. It acts on something smaller than bodies: molecules. High pressure distorts proteins, the folded molecules that do nearly all the work in a cell, so that they lose their shape and stop working. Deep-sea fish counter this by packing their cells with a small molecule called trimethylamine N-oxide, TMAO, which helps proteins keep their shape. The deeper a fish lives, the more TMAO it carries.

That pattern led researchers to a prediction. If TMAO rises steadily with depth, there must be a depth at which a fish cannot pack in any more without its cells running into other problems. Working from the measured trend, Paul Yancey and colleagues estimated in 2014 that fish should run out of room for TMAO between about 8,200 and 8,400 metres.^[4] The 2022 snailfish at 8,336 metres sits right inside that window. No fish has been reliably recorded below it, although animals without backbones, which do not depend on the same chemistry, live deeper still.

Now the corrected chain reads: pressure is enormous; it crushes gas spaces and strains molecules; so deep-sea animals avoid gas spaces and protect their proteins chemically, and fish meet a chemical depth limit near 8,200-8,400 m.

A second gradient: light

Pressure rises steadily with depth. Light falls away fast. Webb gives the numbers for solar energy entering clear ocean water: 45 percent remains at 1 m, 16 percent at 10 m, and about 1 percent at 100 m.

The colours do not fade together. Water absorbs long wavelengths first: red is gone in the top 10 m or so, orange by about 40 m, yellow before 100 m. Blue and green light reach deepest. One consequence surprises divers. A red fish photographed without a flash at 30 m looks grey or black, because there is no red light left to reflect off it. That is why so many deep-water animals are red: in the dark, red is a way to be invisible.

Those depths are for clear open ocean. Coastal water carries silt, plankton and dissolved material that block light far sooner, and plankton make it look green. A simple tool measures that clarity. In 1865, at the request of Pope Pius IX, the astronomer Angelo Secchi and the naval commander Alessandro Cialdi tested a 30 cm white disk lowered on a rope from a papal ship; the depth at which it disappears from view measures how clear the water is. The Secchi disk, now usually painted in black and white quarters, is still used today, including by volunteers.

Think about the size of the sunlit layer. The ocean averages about 3,700 m deep, and photosynthesis happens only in roughly the top 200 m, about one twentieth of the depth. Everything that lives below depends, almost entirely, on food sinking or swimming down from that thin skin.

A third gradient: the gases

Oxygen enters seawater in two places, both at the top: it dissolves in from the air, and phytoplankton release it by photosynthesis. How much the water can hold depends on conditions. Gases dissolve better in colder water, in less salty water and under higher pressure. Even so, well-oxygenated surface seawater holds only about 8 milligrams of oxygen per litre, far less than the same volume of air.

Below the surface, oxygen is used up. Dead plankton, fecal pellets and other organic matter sink, and bacteria break them down, using oxygen to do it. Below the sunlit zone nothing replaces that oxygen by photosynthesis, and the pycnocline from Lesson 3 keeps the layer from mixing with the oxygen-rich water above. So oxygen falls to a minimum somewhere between a few hundred metres and 1,000 m: the oxygen minimum zone. Deeper still, oxygen often rises again. That deep water sank long ago from cold polar seas, carrying oxygen absorbed at the surface, and by then most of the sinking food has already been eaten higher up, so less of its oxygen is used.

Biologists have two thresholds for low oxygen. Hypoxia is usually defined as below 2 mg per litre, where many animals struggle; anoxia, below 0.5 mg per litre, means almost none at all. Lesson 10 shows how a coastal dead zone forms, and Lesson 23 how a warming ocean is losing oxygen.

Carbon dioxide behaves almost as a mirror image of oxygen. It is used up at the surface by photosynthesis and released at depth by decay, so its profile is roughly the opposite of oxygen's. Unlike oxygen, it also reacts with water:

CO₂ + H₂O ↔ H₂CO₃ ↔ H⁺ + HCO₃^- ↔ 2H⁺ + CO₃^2-

Read it left to right: carbon dioxide and water form carbonic acid, which gives up a hydrogen ion to become bicarbonate, which can give up another to become carbonate. At the ocean's average pH of about 8.1, slightly basic, about 92 percent of the dissolved inorganic carbon is bicarbonate, about 7 percent carbonate and about 1 percent carbon dioxide. Because the reactions can run either way, they soak up added hydrogen ions or release them, which buffers the ocean's pH against sudden change. But the buffer has limits: researchers have measured a fall in average surface pH from about 8.2 to 8.1 over the last century. Lesson 24 works through what that small-looking number means.

Common misconceptions

• "Deep-sea pressure would crush any animal that is not armoured." Pressure crushes only where gas spaces create a difference between inside and outside. Water-filled bodies are not crushed; the real limit is the strain on molecules.
• "Clear ocean water lets sunlight reach the seafloor." Only about 1 percent of the sunlight remains at 100 m in clear water, there is rarely any significant light beyond 200 m, and below 1,000 m there is none.
• "Warm water holds more oxygen, like warm water dissolves more sugar." Gases do the opposite of most solids: cold water holds more dissolved oxygen.
• "The deeper you go, the less oxygen there is." Oxygen usually falls to a minimum at a few hundred to 1,000 m and then rises again in cold deep water.
• "Ocean acidification means the sea is acidic." Seawater is slightly basic, about pH 8.1. Acidification means it is moving toward acid.

Pulling it together

• Pressure rises by one atmosphere for every 10.06 m, to about 400 atmospheres on an abyssal plain and about 1,088 at the Challenger Deep.
• Pressure crushes gas spaces, not water-filled bodies. Its real effect on deep-sea life is on molecules, which fish protect with TMAO; that chemistry probably sets the fish depth limit near 8,200-8,400 m, and the deepest fish seen was at 8,336 m.
• Light fades fast: about 1 percent at 100 m in clear water, red first and blue last. The sunlit zone is 0-200 m, the twilight zone 200-1,000 m, and the midnight zone below.
• Oxygen enters at the surface, is used by decay below, falls to a minimum between a few hundred metres and 1,000 m, and often rises in cold deep water. Cold water holds more.
• Carbon dioxide reacts with water; at pH 8.1 most of it is bicarbonate, and the reactions buffer pH.

Sources

1. NOAA National Ocean Service. (2026). How does pressure change with ocean depth? oceanservice.noaa.gov
2. NOAA National Ocean Service. (2026). How far does light travel in the ocean? oceanservice.noaa.gov
3. Natural History Museum, London. (2023, April 5). Deepest ever fish filmed at a depth of 8,336 metres. nhm.ac.uk
4. Yancey, P. H., Gerringer, M. E., Drazen, J. C., Rowden, A. A., and Jamieson, A. (2014). Marine fish may be biochemically constrained from inhabiting the deepest ocean depths. Proceedings of the National Academy of Sciences, 111(12), 4461-4465.
5. Webb, P. (2019). Introduction to oceanography. Roger Williams University Open Publishing; Geosciences LibreTexts edition, CC BY 4.0. Sections 5.4 (oxygen), 5.5 (carbon dioxide and pH), 6.3 (density and compressibility) and 6.5 (light). geo.libretexts.org
6. Reid, M. (2022, April 5). The simple usefulness of the Secchi disk. Distillations, Science History Institute. sciencehistory.org

Seven-tenths of a foot

At Cabrillo National Monument, on the tip of Point Loma in San Diego, the National Park Service gives tide-pool visitors one number to look for. "A tide of 0.7 or below," its advice says, "should expose the tidepools." At that level the sea pulls back far enough to uncover rocks that are underwater most of the day, crowded with anemones, sea stars, limpets and crabs. A few hours later the same rocks are under a metre or more of water.

To use that number you need to know what it measures, where to find it, and when it will happen. This lesson works through all three with real NOAA predictions, then looks at how the life on a rocky shore is arranged by the same daily clock.

Why the water rises and falls twice a day

The Moon's gravity pulls on the whole Earth, but it pulls harder on the side facing the Moon than on the far side. That difference stretches the oceans into two bulges, one toward the Moon and one away from it, and the Earth turns beneath them. A spot on the coast passes through both bulges, so most coasts get two high tides and two low tides a day. Why there is a second bulge on the far side is a beautiful piece of physics, worked through in the tides lesson of High School Earth & Space Science; here you need its consequences.

The first consequence is timing. While the Earth turns once, the Moon moves along its orbit in the same direction, so the Earth needs an extra 50 minutes to catch up with it. NOAA gives the numbers: a lunar day is 24 hours and 50 minutes, and high tides come 12 hours and 25 minutes apart. So each day's tides arrive about 50 minutes later than the day before.

The Sun raises tides too. It is far more massive than the Moon but about 390 times farther away, and NOAA puts its tide-generating force at about half the Moon's. How the two combine sets the size of the tides through the month, which you will see in the data below.

Three rhythms of tide

Not every coast follows the simple two-a-day pattern. NOAA sorts tides into three types, and real predictions for three U.S. stations on 3 November 2026 show all three.

Notice the ranges too. Boston's water rose and fell about 9 feet that day, San Francisco's about 5, Pensacola's less than 1. The shape and depth of each coast and basin decide how the ocean responds to the same Moon. The extremes are large: NOAA lists the highest tides in the world at Burntcoat Head in Canada's Bay of Fundy, and the highest in the United States near Anchorage, Alaska, where ranges average around 30 feet.

Spring and neap tides

Twice a month, at new moon and full moon, the Sun, Moon and Earth line up and their tidal pulls add together. The result is a spring tide: higher highs and lower lows. The name has nothing to do with the season; NOAA describes it as the tide "springing forth," and it happens all year. At first and third quarter moon, when the Sun and Moon pull at right angles, their effects partly cancel and the range is smaller: a neap tide.

You can watch it in the predictions. The U.S. Naval Observatory lists a last-quarter moon on 1 November 2026 and a new moon on 9 November 2026. Compare the same two stations a week apart.

The spring tide's range is wider at both stations, and both get a low below zero. What does a tide below zero mean?

What zero means: tidal datums

Tide heights are measured from a reference level called a datum, and the zero on a U.S. tide prediction is not sea level. It is Mean Lower Low Water (MLLW): the average height of the lower of each day's two low tides, measured over a 19-year period called the National Tidal Datum Epoch (currently 1983 to 2001). NOAA uses MLLW as the reference for tide predictions and nautical charts, so a chart's depths show the water you can count on most days.

A minus tide, like San Francisco's -0.54 ft on 9 November, means the water will drop below that average lower low. Rocks that are covered on an ordinary low tide come out of the water. That is why tide-poolers hunt for minus tides, and why Olympic National Park recommends tides lower than 1 foot and calls a 3-foot low tide "not great for tidepooling."

Why this matters: Tides come 50 minutes later each day, swing between springs and neaps twice a month, and are measured from a zero that sits below most low tides. Read all three and you can predict when the shore will open, and when it will close again.

Worked example: planning a visit from a prediction

Suppose you want to visit a rocky shore near San Francisco Bay. Here is the reasoning, step by step, with NOAA's predictions for 9 November 2026 at the San Francisco station (feet above MLLW): low 2.68 at 04:25, high 6.19 at 10:35, low -0.54 at 17:24, high 4.70 at 00:39 the next morning.

1. Identify the pattern. Two highs and two lows of very different heights: a mixed tide, as expected on the West Coast.
2. Find the lowest low. -0.54 ft at 17:24. It is a minus tide, and well below the Cabrillo guideline of 0.7.
3. Set the window. The Park Service at Cabrillo advises exploring from an hour before to an hour after the low-tide peak, so about 16:24 to 18:24.
4. Check the light. In early November the sun sets around the time of this low, so the second half of the window would be in the dark. Plan to arrive at the start of the window and leave before dusk, or pick another day.
5. Know when the water returns. The tide turns and rises toward the 00:39 high. Watch it, and leave the outer rocks as soon as it starts to come back.

Now change one input: plan the same trip a week earlier, on 3 November, two days after the last-quarter moon. The lows that day are 0.14 ft at 00:06, in the middle of the night, and 2.07 ft at 13:01. Neither is a daylight minus tide, and the afternoon low is nearly three feet higher than the spring low a week later. The answer flips: do not go. Same place, same Moon, a week apart, and the plan reverses because neap tides keep the shore covered.

Life arranged by the clock

The intertidal zone is the strip of shore between high and low tide, and it is one of the harshest places on Earth to live. Twice a day its residents are pounded by waves and then left in the air to dry and bake or freeze. NOAA divides it into bands:

Why are the bands so sharp? On the Isle of Cumbrae in Scotland in the 1950s, the ecologist Joseph Connell studied two barnacles. A small species, Chthamalus, lived high on the rocks; a larger one, Balanus, lived lower down. When he cleared Balanus from rocks, young Chthamalus survived lower on the shore than they ever normally lived; where Balanus was present, it grew faster and smothered, crushed and undercut them. But Balanus could not survive high on the shore, where the rocks dried out for too long.

Connell's conclusion, published in 1961, has become a classic: the upper limit of a species on the shore tends to be set by physical stress such as drying, and the lower limit by other organisms, through competition and predation. Chthamalus could live in a wider band than it actually occupied, a difference ecologists describe as the fundamental niche versus the realized niche.

Common misconceptions

• "High tide happens once a day, when the Moon is overhead." Most coasts have two highs a day, one for each bulge, and local coastlines delay the timing. Some coasts, such as much of the Gulf of Mexico, have one.
• "Spring tides happen in spring." They happen twice every month, at new and full moon.
• "Zero on a tide table is sea level." In the U.S. it is Mean Lower Low Water, so most low tides are above zero and minus tides are unusually low.
• "Tides are about the same everywhere." On 3 November 2026 the range was about 9 feet at Boston and less than 1 foot at Pensacola.
• "Taking one shell or one crab cannot hurt." Thousands of visitors make the same choice. Parks ask you to look, not touch, and to take only photographs; many shores are protected areas where collecting is illegal.

Summing up

• Two tidal bulges give most coasts two highs and two lows a day, 12 hours 25 minutes apart, arriving about 50 minutes later each day.
• Tides are semidiurnal (East Coast), mixed (West Coast) or diurnal (much of the Gulf of Mexico), and ranges vary from under a foot to tens of feet.
• Spring tides, at new and full moon, have the widest range; neap tides, at the quarter moons, the narrowest.
• U.S. tide heights are measured from Mean Lower Low Water; a minus tide uncovers the low intertidal.
• Plan visits around a daylight minus tide, arrive about an hour before the low, and leave as the tide turns.
• Shore life forms bands; Connell's barnacles showed that drying sets upper limits and competition sets lower ones.

Sources

1. NOAA National Ocean Service. (n.d.). Tides and water levels [Tutorial]: Gravity, inertia and the two bulges; Changing angles and the lunar day; Types and causes of tidal cycles. oceanservice.noaa.gov
2. NOAA National Ocean Service. (2026). What are spring and neap tides? oceanservice.noaa.gov
3. NOAA Center for Operational Oceanographic Products and Services. (2026). Tide predictions for San Francisco (9414290), Boston (8443970) and Pensacola (8729840), 3 and 9 November 2026, retrieved 2 November 2026. tidesandcurrents.noaa.gov
4. NOAA Center for Operational Oceanographic Products and Services. (n.d.). Datum options. tidesandcurrents.noaa.gov
5. U.S. Naval Observatory. (2026). Phases of the Moon. aa.usno.navy.mil
6. NOAA National Ocean Service. (2026). What is the intertidal zone? and Where is the highest tide? oceanservice.noaa.gov
7. National Park Service. (n.d.). Cabrillo tidepools: Top 10 tips. nps.gov
8. National Park Service. (n.d.). Tidepool activities on the coast of Olympic National Park. nps.gov
9. Connell, J. H. (1961). The influence of interspecific competition and other factors on the distribution of the barnacle Chthamalus stellatus. Ecology, 42(4), 710-723.

Eighty-two percent

In 2019 three researchers, B. Chris Brewster, Richard Gould and Robert Brander, analysed twenty years of rescue reports sent in by lifeguards at U.S. surf beaches, from 1997 to 2016. They found that rip currents were the main cause of 81.9 percent of the rescues. NOAA estimates that rip currents kill about 100 people a year in the United States.

That is a puzzle if you hold a common belief about waves: that they carry water toward the shore, the way a river carries water downstream. If waves only push water in, how does a swimmer end up dragged out to sea? The belief is wrong in deep water and partly right at the beach, and following exactly where it fails explains both the waves and the danger.

A test the belief fails

Watch a gull sitting on the sea beyond the surf as swells roll past. If waves carried water along, the gull would ride toward the beach at the speed of the waves. It does not. It rises as each crest passes, sinks into each trough, and after a dozen waves it is almost exactly where it started. Webb's oceanography text puts it plainly: a passing wave moves energy through the water, while the water itself shows little horizontal movement.

The same is true on the scale of oceans. In a famous experiment published in 1966, Walter Munk and his colleagues set up six recording stations along a line across the Pacific and followed swell from storms in the far south as it travelled northward across the entire ocean. The energy made that trip of thousands of kilometres. The southern water did not.

What actually moves: energy, in circles

First the vocabulary. A wave's crest is its highest point and its trough its lowest. Wave height is the vertical distance from trough to crest, wavelength is the distance from one crest to the next, and period is the time between two crests passing the same point.

As a wave passes, each bit of water near the surface travels around a circle and comes back close to where it began. At the surface, the circle's diameter equals the wave height. Below the surface the circles shrink, until at a depth of half the wavelength, called the wave base, the motion has almost died away.

Work an example. A swell with a wavelength of 100 m has a wave base of 50 m. A diver at 60 m would barely feel it. Now change the input: in a storm, waves with wavelengths of 200 m or more stir the water to 100 m and deeper, deep enough to churn the sediment on much of the continental shelf, which in Lesson 2 you met at less than about 150 m.

In short: In deep water a wave is energy travelling through water that mostly goes round in circles. The motion fades with depth and is nearly gone at half a wavelength.

How wind builds waves

Most ocean waves are made by wind. Three things decide how much energy the wind hands to the sea: how hard it blows (wind speed), for how long (duration), and over what distance of open water (fetch). Double any of them and the waves grow. When all three are large enough that the waves cannot grow any more, the result is called a fully developed sea. Webb gives a sense of scale: about half the waves in the open sea are less than 2 m high, and only 10-15 percent exceed 6 m.^[5]

Waves also limit their own height. Wave steepness is height divided by wavelength, and when the height exceeds one-seventh of the wavelength, the wave becomes unstable and its crest collapses into a whitecap. A wave 10 m long can be at most about 1.4 m high before it breaks; a wave 140 m long can stand up to about 20 m.

A storm makes a jumble of waves of many sizes, called sea. As they travel away from the storm, they sort themselves out, because in deep water longer waves travel faster. Webb gives deep-water wave speed as C = 1.25 times the square root of the wavelength (C in metres per second, wavelength in metres):

So the long waves race ahead and arrive at a distant coast first, as regular, long-period swell. When you see smooth, evenly spaced waves on a calm day, you are usually looking at energy from a storm that may be a thousand kilometres or more away.

Sometimes waves from different directions meet in step and add together, which is called constructive interference. Oceanographers usually define a rogue wave as one at least twice the significant wave height, a measure you will meet in this lesson's buoy activity.

Where the belief becomes true: the shore

When a wave reaches water shallower than half its wavelength, its circles start to drag on the bottom. The wave slows, its wavelength shortens and its height grows. Once the height passes about one-seventh of the wavelength, the crest topples forward and the wave breaks. How it breaks depends on the beach: gentle slopes make spilling breakers that crumble slowly; steeper slopes make plunging breakers whose crests curl over; the steepest shores make surging breakers that rush straight up the beach.

Waves also bend as they reach shallow water, because the part of a wave in shallower water slows first. This refraction swings wave crests nearly parallel to the shore and focuses energy on headlands, which is why waves are usually bigger off points and smaller in bays.

Here, finally, the common belief has some truth in it. Breaking waves do push water up the beach and toward the shore. That water cannot pile up forever. It flows along the beach, as a longshore current, and it has to get back out to sea somewhere.

The way back out: rip currents

Where the returning water finds an easier way back, it pours seaward in a narrow stream: a rip current. NOAA describes rip currents as "powerful, narrow channels of fast-moving water." They can move at up to eight feet per second, faster than an Olympic swimmer. They often form on calm, sunny days, so good beach weather does not mean safe water.

The danger is panic, not suction. NOAA's National Hurricane Center is specific: "Rip currents do not pull you under; they pull you away from shore." People drown when they exhaust themselves swimming straight back against the current. The advice:

1. Relax and float. Do not fight the current; it will not pull you under.
2. Swim parallel to the shoreline to get out of the narrow channel, then angle back toward the beach with the waves.
3. Wave and call for help, and keep floating.

If you see someone else caught, the advice is just as specific: alert a lifeguard, throw them something that floats and call 911, but do not go in after them, because many rip current deaths are of would-be rescuers. The single best precaution is to swim at a beach with lifeguards.

In short: Waves carry energy, not water, until they break. At the shore they do push water landward, and the return flow, the rip current, is the most dangerous water on a surf beach. Float, then swim parallel.

When the whole ocean is shallow: tsunamis

A tsunami is a wave set off when an earthquake, landslide or eruption suddenly moves the seafloor and the water above it. Its wavelength is enormous, 100-200 km. Since even the deepest ocean is shallower than half of that, a tsunami behaves everywhere as a shallow-water wave, and its speed depends only on depth. Webb gives C = 3.13 times the square root of the depth (C in metres per second, depth in metres).

Work it at the depth of a typical abyssal plain, 4,000 m. The square root of 4,000 is about 63, and 3.13 x 63 = about 198 m/s. That is roughly 710 km per hour, as fast as a jet airliner. Now change the input to 40 m of water near the coast. The square root of 40 is about 6.3, and 3.13 x 6.3 = about 20 m/s, or about 71 km per hour. The wave slows tenfold, and the energy that was spread through four kilometres of water piles up into a surge.

Out at sea a tsunami may be no taller than the seafloor movement that made it, often less than 10 m. At the coast the sea can rise by tens of metres in minutes. In 2004 a magnitude 9 earthquake off Indonesia sent waves up to 33 m high that killed about 230,000 people. In 2011 a magnitude 9.0 earthquake off Japan sent a tsunami up to 40.5 m high.

The National Weather Service lists the natural warnings: a strong or long earthquake, the ocean acting strangely, rising fast or pulling far out, or a loud roar from the sea. If you are on a coast and notice any of them, do not wait for an official warning. Move off the beach to high ground or inland at once.

Common misconceptions

• "Waves carry water across the ocean." They carry energy. In deep water the water mostly moves in circles and stays close to where it started.
• "Undertow pulls swimmers under." Rip currents pull swimmers away from shore, not down. The danger is exhaustion from fighting them.
• "If you are caught in a rip, swim hard for the beach." Swim parallel to the shore to leave the channel, or float until it releases you, and signal for help.
• "A tsunami is a giant version of a surfing wave." It is a wave 100-200 km long that arrives as a rapid surge of the whole sea, often more like a fast flood than a curling breaker.
• "Calm weather means safe water." Rip currents often form on calm, sunny days, fed by swell from distant storms.

What you now know

• Wave height, wavelength and period describe a wave; in deep water, water moves in circles that fade to almost nothing at the wave base, half a wavelength down.
• Wind speed, duration and fetch set wave size. Waves break when height exceeds one-seventh of wavelength. Longer waves travel faster in deep water (C = 1.25 x square root of wavelength), so swell arrives sorted.
• In shallow water waves slow, steepen and break; refraction focuses them on headlands.
• Rip currents return water seaward at up to eight feet per second and cause about 82 percent of surf-beach rescues. Relax, float, swim parallel, signal for help.
• Tsunamis are always shallow-water waves (C = 3.13 x square root of depth), about 700 km per hour over the deep ocean. A strong quake or a strange sea means: go to high ground now.

Sources

1. Brewster, B. C., Gould, R. E., and Brander, R. W. (2019). Estimations of rip current rescues and drowning in the United States. Natural Hazards and Earth System Sciences, 19, 389-397. nhess.copernicus.org
2. NOAA National Ocean Service. (2026). What is a rip current? oceanservice.noaa.gov
3. NOAA National Hurricane Center. (n.d.). Rip currents: How to survive. nhc.noaa.gov
4. National Weather Service. (n.d.). Rip current safety and Tsunami safety: During a tsunami. weather.gov
5. Webb, P. (2019). Introduction to oceanography. Roger Williams University Open Publishing; Geosciences LibreTexts edition, CC BY 4.0. Sections 10.1-10.4 (wave basics, waves at sea, waves on the shore, tsunamis). geo.libretexts.org
6. NOAA National Data Buoy Center. (n.d.). Measurement descriptions and units. ndbc.noaa.gov
7. Snodgrass, F. E., Groves, G. W., Hasselmann, K. F., Miller, G. R., Munk, W. H., and Powers, W. H. (1966). Propagation of ocean swell across the Pacific. Philosophical Transactions of the Royal Society of London, Series A, 259, 431-497.

Two weeks lost in the Atlantic

In 1768 the new customs commissioners in Boston sent a complaint to London. The mail packets that sailed from Falmouth in England to New York were taking about two weeks longer on the crossing than merchant ships sailing from London to Rhode Island. The merchant ships were slower, and Boston and Newport were barely a day's sail apart. Why should the faster ships arrive so late?

The question reached Benjamin Franklin, then a deputy postmaster for the colonies, who was living in London. His cousin Timothy Folger, a ship captain from Nantucket, happened to be in town, and Franklin asked him. Folger's answer, which Franklin wrote down in a letter to the Post Office on 29 October 1768, was a lesson in oceanography learned from whales.

Nantucket, Folger explained, sent out nearly 150 whaling vessels, and the whales were found "generally near the Edges of the Gulph Stream, a strong Current so called which comes out of the Gulph of Florida." Chasing whales back and forth across its edges, the whalers had learned its course and strength better than any captain who only crossed it. A whaleboat that rowed into the stream while its ship stayed outside would be swept apart from the ship.

The current ran, Folger said, at 4, 3.5, 3 and 2.5 miles an hour, and a ship heading west in it lost "60, or 70 Miles a day." The whalers sometimes met westbound ships still far from land after 8, 9 and 10 weeks at sea, stuck in the part of the stream that ran straight against them. The merchant captains bound for Rhode Island knew to keep out of it; the packet captains did not. When the whalers told them so, by Franklin's later account, they were "too wise to be counselled by simple American fishermen."

At Franklin's request, Folger marked the stream's course on a chart, with written directions for avoiding it, and the Post Office had it printed and sent to the packet captains. They ignored it. You can see a surviving copy of the Franklin-Folger chart in the Newberry Library's Mapping Movement collection, whose digitized images are free to use. The path it shows is remarkably close to the one satellites trace today.

A river in the sea

The whalers were right about its strength. Webb's oceanography text describes the Gulf Stream as the fastest current in the world ocean, averaging about 6.4 km per hour with a maximum of about 9 km per hour. Folger's 2.5 to 4 miles an hour is 4 to 6.4 km per hour, close to the modern average. The stream is narrow, 50-100 km wide, and reaches depths of about 1.5 km, and it carries more than 100 times the combined flow of all the rivers on Earth. As it meanders, loops pinch off as separate rings of warm or cold water.

The Gulf Stream is still being measured closely. In 2023 Christopher Piecuch and Lisa Beal combined forty years of observations from the Florida Straits and found that the flow had weakened by about 4 percent since 1982, a decline they judged virtually certain. They were careful to add that they could not yet say how much of it was caused by climate change rather than natural variation.

So the stream is real, fast and enormous. The rest of this lesson follows the questions it raises: what drives it, why it hugs the American coast, and where its water finally goes.

The first push: wind

Surface currents generally move in the direction of the winds that create them. Near the equator, the trade winds blow from east to west and drive equatorial currents westward. Between about 30 and 60 degrees of latitude, the westerlies blow from west to east and push surface water east. Wind can only stir the upper part of the ocean, so these wind-driven currents affect the top 100-200 m, about 10 percent of the ocean's volume.

The twist: Earth's spin

On a spinning planet, anything moving over the surface for a long time and a long distance appears to curve: to the right in the Northern Hemisphere and to the left in the Southern Hemisphere. This is the Coriolis effect. It means surface water does not move straight downwind. It moves about 45 degrees to the right of the wind in the Northern Hemisphere.

Each layer of water drags the layer beneath it, and each is deflected a little further and moves a little slower, so the direction of flow twists with depth in a pattern called the Ekman spiral, fading out at about 100 m. Add up the motion of that whole layer and the net result is surprising: the water moves 90 degrees to the right of the wind in the Northern Hemisphere (90 degrees to the left in the Southern). This net flow is called Ekman transport.

Now put the winds and the twist together. The trade winds push water west near the equator; the westerlies push it east farther north. Each push is turned to the right, so water piles up toward the middle of the ocean, and the currents settle into great loops around that central hill of water. These loops are gyres. There are five major ones: the North Atlantic, South Atlantic, North Pacific, South Pacific and Indian Ocean gyres. Those in the north turn clockwise and those in the south anticlockwise.

The North Atlantic gyre is the loop Folger was describing. The North Equatorial Current flows west toward the Caribbean, feeds the Gulf Stream, which runs north along the American coast and becomes the North Atlantic Current heading toward Europe, which turns south as the Canary Current off Africa and closes the loop.

The upshot: Wind starts surface currents, Earth's spin turns them, and together they build five great gyres, turning clockwise in the north and anticlockwise in the south.

Why the Gulf Stream hugs the west

Look at any map of the gyres and you will notice a lopsidedness. The currents on the western sides of the oceans, the Gulf Stream and Japan's Kuroshio, are narrow, deep and fast. The currents on the eastern sides, such as the California Current and the Canary Current, are broad, shallow and slow. Webb gives a comparison: the Kuroshio is about 15 times faster, 20 times narrower and 5 times deeper than the California Current.

This is called western intensification, and it also comes from Earth's spin. The Coriolis effect is weak near the equator and strong near the poles, so it squeezes the gyre's flow toward the western side of the basin. The same amount of water has to travel around the whole loop, so where the path is narrow the water must move faster, like water from a garden hose when you put your thumb over the end.

Where the deep comes up

Ekman transport has one more consequence, and it feeds much of the world. Along the coast of California the prevailing winds blow toward the south. Ekman transport turns the surface water 90 degrees to the right, which is away from the shore. Water from below rises to replace it, and that deeper water is cold and rich in nutrients. This is coastal upwelling. Off Peru the winds blow north, but in the Southern Hemisphere the turn is to the left, which again carries surface water offshore, so Peru has upwelling too. Along the equator, the trade winds pull surface water apart to the north and south, and water wells up between. Many of the most productive regions on Earth are upwelling zones, for reasons Lesson 9 explains and Lesson 20 shows breaking down.

The slow conveyor in the deep

Below the reach of the wind lies the other 90 percent of the ocean, and it moves too, driven not by wind but by density. In Lesson 3 you saw where the densest water is made: where surface water is chilled and made saltier, often as sea ice forms. Webb names two great sources. North Atlantic Deep Water forms in the Greenland Sea, and the densest water mass of all, Antarctic Bottom Water, forms in the Weddell Sea off Antarctica. These waters sink and spread along the ocean floor, slowly rising elsewhere and returning at the surface.

Because it depends on temperature and salinity, this is called the thermohaline circulation, and the whole loop is often called the global conveyor belt. It is slow. Deep currents move on the order of 10-20 km per year, compared with several kilometres per hour at the surface, and a complete circuit may take 1,000 to 2,000 years. The water sinking near Greenland today will not see the surface again for centuries. That slowness matters for climate, which Lesson 19 takes up.

Following the water today

Folger tracked the Gulf Stream by whales and boats. Modern oceanographers use drifting buoys, Argo floats and satellites that measure the height of the sea surface, whose slopes reveal the currents. Sometimes they use accidents. In January 1992 a container of plastic bath toys washed off a ship in the North Pacific, and oceanographers Curtis Ebbesmeyer and James Ingraham used reports from beachcombers who found the toys to test a computer model of surface currents. Today ocean models such as ECCO2, a joint project of MIT and NASA's Jet Propulsion Laboratory, combine satellite and ship measurements into moving pictures of the whole ocean, which you will watch in the activity.

Common misconceptions

• "Ocean currents are caused by the tides." Tides move water back and forth twice a day. The great surface currents are driven by wind and turned by Earth's spin, and the deep circulation is driven by density.
• "Currents flow straight downwind." In the Northern Hemisphere the surface water moves about 45 degrees to the right of the wind, and the whole wind-driven layer moves 90 degrees to the right.
• "The deep ocean is still." It moves slowly, 10-20 km a year, in a circulation that takes 1,000-2,000 years to complete.
• "The Gulf Stream flows from the Gulf of Mexico straight to Europe." It is one side of a wind-driven gyre that loops back south along Africa, and only part of its water carries on north. How much it warms Europe is argued over in Lesson 19.

Looking back

• In 1768 Nantucket whalers knew the Gulf Stream well enough to explain why westbound mail packets lost two weeks; Franklin and Folger charted it, and the packet captains ignored them.
• The Gulf Stream averages about 6.4 km per hour, is 50-100 km wide and carries more than 100 times the flow of all rivers; it weakened about 4 percent between 1982 and the early 2020s.
• Wind drives surface currents; the Coriolis effect turns them right in the north and left in the south; Ekman transport is 90 degrees to the wind; five major gyres result.
• Western boundary currents are narrow, deep and fast because of how Earth's spin varies with latitude.
• Winds along a coast can carry surface water offshore so that cold, nutrient-rich water upwells, as off California and Peru.
• Below the wind-stirred layer, dense water formed near Greenland and Antarctica drives a slow conveyor taking 1,000-2,000 years per circuit.

Sources

1. Franklin, B. (1768, October 29). Letter to Anthony Todd. In W. B. Willcox (Ed.), The papers of Benjamin Franklin (Vol. 15, pp. 246-248). Yale University Press, 1972. Founders Online, National Archives. founders.archives.gov
2. Newberry Library. (n.d.). Benjamin Franklin's chart of the Gulf Stream, 1768-1770. Mapping Movement in American History and Culture. newberry.org
3. Webb, P. (2019). Introduction to oceanography. Roger Williams University Open Publishing; Geosciences LibreTexts edition, CC BY 4.0. Sections 9.1-9.5 and 9.8. geo.libretexts.org
4. Piecuch, C. G., and Beal, L. M. (2023). Robust weakening of the Gulf Stream during the past four decades observed in the Florida Straits. Geophysical Research Letters, 50, e2023GL105170.
5. Ebbesmeyer, C. C., and Ingraham, W. J., Jr. (1994). Pacific toy spill fuels ocean current pathways research. Eos, Transactions American Geophysical Union, 75(37), 425-432.
6. NASA Goddard Space Flight Center Scientific Visualization Studio. (2011). Perpetual Ocean [Visualization of ECCO2 ocean model currents]. svs.gsfc.nasa.gov

A cell nobody had counted

Until 1988, nobody knew that the most abundant photosynthetic organism on Earth existed. That year Sallie (Penny) Chisholm of MIT and her colleagues described Prochlorococcus, a single-celled bacterium about a thousandth of a millimetre across, small enough to have slipped through every earlier survey of the sea. Later estimates put its numbers at around a billion billion billion cells in the world's oceans. NOAA credits it with producing up to 20 percent of the oxygen in the entire biosphere; MIT's version is that it is responsible for "every fifth breath we take."

If something that important could hide until 1988, plankton deserve a closer look. This lesson lines up the main kinds side by side and then reads the comparison for what it says about where the ocean is rich and where it is poor.

What makes something plankton

NOAA defines plankton as marine drifters, organisms carried along by tides and currents; the word comes from the Greek for "drifter" or "wanderer." The definition is about movement, not size. Most plankton are microscopic, but a jellyfish half a metre wide is plankton too, because it cannot swim against a current. Animals that can, such as fish, squid and whales, are called nekton.

Two pairs of labels sort plankton further:

• Phytoplankton make their own food by photosynthesis, using sunlight, carbon dioxide and nutrients. Zooplankton are animals and animal-like microbes that eat other plankton.
• Holoplankton drift their whole lives, like copepods and krill. Meroplankton drift for only part of their lives, usually as larvae, before settling down or swimming free: the young of crabs, barnacles, sea stars, corals and many fish begin as plankton.

NOAA calls phytoplankton "microscopic plants," which is a useful shorthand but not quite right: many are bacteria or single-celled protists, not plants at all. What they share with plants is photosynthesis.

The cast, side by side

A picture helps with the shapes. The WHOI Imaging FlowCytobot dashboard, which you will use in the investigation, shows thousands of real phytoplankton photographed in seawater off Massachusetts, from glassy chains of diatoms to armoured dinoflagellates.

Reading the table

The producers are tiny, and that is no accident. Phytoplankton must take up dissolved nutrients through their surfaces, and small cells have more surface for their volume. Work it with cubes: a cube 1 unit on a side has a surface of 6 square units and a volume of 1, a ratio of 6. A cube 10 units on a side has a surface of 600 and a volume of 1,000, a ratio of 0.6. Shrinking a cell tenfold gives it ten times more surface for each bit of its body, which is a large advantage in water where nutrients are scarce.

Coverings decide what is left behind. Diatoms build in silica and coccolithophores in calcium carbonate, which is why, as Lesson 2 showed, the seafloor collects siliceous ooze under some waters and chalky ooze under others.

Different water favours different producers. Diatoms dominate where the water is cool, mixed and rich in nutrients, as in spring and in upwelling zones. Where warm water sits in a stable layer and nutrients run out, as in the open subtropics, the tiny cyanobacteria come into their own.

Worth holding on to: Plankton are defined by drifting. The producers among them are mostly microscopic bacteria and single-celled algae, and which ones thrive depends on light, nutrients and how well the water is mixed.

Half of the planet's production

In 1998 Christopher Field and colleagues combined satellite measurements of ocean colour with models of land vegetation to estimate how much new plant material the whole Earth makes in a year. Their answer was 104.9 billion tonnes of carbon, of which 48.5 billion tonnes, about 46 percent, came from the ocean. Phytoplankton account for about 95 percent of that marine production; seaweeds and seagrasses make the rest. NOAA states that about half of Earth's oxygen comes from the ocean, and adds a point that is often left out: about the same amount is used up by marine life, which breathes too.

That production is very unevenly spread. NASA's satellite maps of chlorophyll, the green pigment of photosynthesis, show high values in cold polar waters, along the equator and off the west coasts of North and South America and southern Africa. NASA is careful to say that cold water itself is not the cause. Cold surface water is a sign of upwelling, which brings nutrients up from the deep, as you saw in Lesson 7. Webb's oceanography text gives typical numbers, in grams of carbon fixed per square metre per year:

The seasons matter too. In temperate seas, winter storms mix nutrients up from below, but there is too little light to use them. In spring the light returns, the water still holds its nutrients, and phytoplankton bloom; a smaller bloom can follow in autumn when storms stir up nutrients again. Polar seas have a single spring and summer bloom. Tropical open water stays layered all year, so it is nutrient-limited and productive at a low rate year-round. That is why clear blue tropical water, beautiful as it is, holds little phytoplankton.

When blooms turn harmful

Most blooms feed the food web. Some do damage. NOAA describes harmful algal blooms as colonies of algae that grow out of control and harm people or wildlife, some by producing toxins and others by using up oxygen or smothering corals and seagrass.^[8] Two examples from U.S. waters:

• Karenia brevis, a dinoflagellate, is the most frequent cause of harmful blooms along Florida's west coast and the Texas coast. Its toxin gets into the air when waves break and causes severe respiratory irritation. NOAA estimates that a 2011 red tide cut Texas oyster landings by $10.3 million.
• Pseudo-nitzschia, a diatom, may produce domoic acid, a nerve toxin that can cause amnesic shellfish poisoning in people who eat contaminated shellfish. In 2015 Washington closed its recreational razor clam harvest, at an estimated cost of $40 million in lost tourism spending.

NOAA lists the conditions that favour harmful blooms: extra nutrients, mainly nitrogen and phosphorus from lawns and farms washing into bays and rivers, unusually warm water, sluggish circulation and extreme weather. Lesson 10 follows those nutrients downstream into an estuary.

Common misconceptions

• "Plankton means microscopic." Plankton means drifting. Large jellyfish are plankton; a tiny fish larva that can hold its position against a current is starting to become nekton.
• "Phytoplankton are tiny plants." They photosynthesize like plants, but many are bacteria or single-celled protists.
• "Rainforests make most of the world's oxygen." The ocean produces about half of Earth's oxygen, mostly through phytoplankton, though marine life also uses about as much as the ocean makes.
• "The open ocean is full of plankton everywhere." The central ocean produces less than 50 g of carbon per square metre per year, a fraction of what upwelling coasts produce.
• "Algal blooms are always pollution." Most blooms are natural and feed the food web. Harmful blooms are the subset that produce toxins or cause other damage, and extra nutrients make them more likely.

The takeaway

• Plankton are drifters; phytoplankton photosynthesize, zooplankton eat; holoplankton drift for life, meroplankton only as young.
• The producers are mostly microscopic: cyanobacteria such as Prochlorococcus (discovered in 1988), diatoms in silica, dinoflagellates with two flagella, and coccolithophores in calcium carbonate.
• The ocean makes about 46 percent of Earth's net primary production (Field and colleagues, 1998), about 95 percent of it by phytoplankton.
• Production is high where nutrients reach sunlit water, along upwelling coasts, the equator and polar seas, and low in the layered central ocean; temperate seas bloom in spring.
• Some blooms are harmful, such as Karenia brevis red tides and toxic Pseudo-nitzschia; extra nutrients and warm water make them more likely.

Sources

1. NOAA National Ocean Service. (2026). What are plankton? and How much oxygen comes from the ocean? oceanservice.noaa.gov
2. MIT. (n.d.). The most abundant species on Earth. Better World. betterworld.mit.edu
3. Biller, S. J., Berube, P. M., Lindell, D., and Chisholm, S. W. (2015). Prochlorococcus: The structure and function of collective diversity. Nature Reviews Microbiology, 13, 13-27.
4. Field, C. B., Behrenfeld, M. J., Randerson, J. T., and Falkowski, P. (1998). Primary production of the biosphere: Integrating terrestrial and oceanic components. Science, 281(5374), 237-240.
5. Armbrust, E. V. (2009). The life of diatoms in the world's oceans. Nature, 459, 185-192.
6. Webb, P. (2019). Introduction to oceanography. Roger Williams University Open Publishing; Geosciences LibreTexts edition, CC BY 4.0. Sections 7.2-7.4 (the producers; factors influencing production; patterns of primary production). geo.libretexts.org
7. NASA Earth Observatory. (n.d.). Chlorophyll [Global maps from MODIS on Aqua]. science.nasa.gov
8. NOAA. (n.d.). What is a harmful algal bloom? noaa.gov
9. Atkinson, A., Siegel, V., Pakhomov, E. A., Jessopp, M. J., and Loeb, V. (2009). A re-appraisal of the total biomass and annual production of Antarctic krill. Deep-Sea Research I, 56, 727-740. nora.nerc.ac.uk
10. Woods Hole Oceanographic Institution. (n.d.). About IFCB dashboard. ifcb-data.whoi.edu

Half the fish from a thousandth of the sea

In 1969 the biologist John Ryther published a short paper in Science with an arresting estimate.^[1] The world's upwelling regions, the places where winds pull cold, deep water to the surface, cover only about 0.1 percent of the ocean's area. Yet by his reckoning they produce about half of the world's fish. NOAA says the same thing in plainer words today: good fishing grounds are typically found where upwelling is common.

Lesson 8 gives you a reason to be puzzled. Upwelling coasts make roughly 200-400 grams of carbon per square metre per year by photosynthesis, while the central ocean makes less than 50. That is a difference of four to eight times, not a difference of thousands. So how can a sliver of the sea yield as many fish as all the rest put together? Solving that problem takes the two ideas that run through every food web: energy is lost at every step, and the number of steps varies.

Energy lost at every step

Each feeding level in a food web is a trophic level: producers first, then grazers, then the animals that eat grazers, and so on. Energy passing from one level to the next mostly does not arrive. An animal burns most of what it eats to stay alive, and much of its food is never eaten by the next level at all. If you have taken High School Biology, you have met the energy pyramid; here is the marine version with numbers.

How much gets through? Ryther had to guess. Decades later, Daniel Pauly and Villy Christensen compiled food-web models of dozens of aquatic ecosystems and found that the transfer efficiency between trophic levels, though highly variable, has a mean of about 10 percent. Work it once: start with 10,000 units of energy in phytoplankton.

The lesson of the table is that the length of a food chain matters enormously. Each extra step divides what reaches the top by about ten.

Working Ryther's problem

Ryther split the ocean into three provinces and gave each a food-chain length and an efficiency. In his words, quoted later by Pauly: "I assign efficiencies of 10, 15, and 20 percent, respectively, to the oceanic, the coastal, and the upwelling provinces." He admitted that the true values could be quite different. For the number of steps between producers and the fish people catch, he used about 5 in the open ocean, 3 on coastal shelves and 1.5 in upwelling zones. Combine those with production figures in the ranges Webb gives (Lesson 8) and do the arithmetic per square metre:

Per square metre, an upwelling zone in this model yields about 50,000 times as much fish as the open ocean. Now multiply by area. The open ocean has about 900 times the area of the upwelling zones, so in total it yields 0.0005 x 900 = about 0.45 units against the upwelling zones' 27. The coastal shelves, with about 100 times the upwelling area, yield about 0.34 x 100 = 34. Coastal shelves and upwelling zones split nearly all the fish between them, and the vast open ocean adds almost nothing.

Three multipliers stack up: a few times more production, a shorter chain and a higher efficiency. The shorter chain does most of the work, because it changes an exponent. Ryther's numbers were rough, and other scientists soon argued with his conclusion that the open ocean is a biological desert. But the structure of the argument has held up, and it explains why so many of the world's great fishing grounds lie off upwelling coasts such as Peru, California and southern Africa.

The core of it: Fish production depends on the energy at the base and on how many steps it must pass through, at roughly 10 percent per step. Upwelling zones win on all three counts, but above all on short food chains.

Why chains are short in some seas and long in others

Lesson 8's table of producers explains the chain lengths. In nutrient-rich upwelling water the producers include large cells such as chain-forming diatoms, up to about a millimetre long. Those are big enough for krill and small schooling fish such as anchovies to eat almost directly. In the nutrient-poor open ocean, much of the production comes from cells around a thousandth of a millimetre across, far too small for a fish or even most copepods to catch. Those cells are eaten by single-celled grazers, which are eaten by slightly larger ones, which are eaten by copepods, and only then by fish. Every extra link costs about 90 percent of the energy.

There is a further loop. Phytoplankton leak dissolved organic matter, and grazers spill more when they feed sloppily. In 1983 Farooq Azam and colleagues described how bacteria take up this dissolved material and are then eaten by tiny single-celled predators, returning some of the energy to the food web.^[5] They called this pathway the microbial loop. It recycles material that would otherwise be lost, but it adds steps, and each step costs energy, which is one more reason the open ocean supports so few fish.

One predator, a whole shore

Energy flows up a food web, but control can flow down. In 1963 the ecologist Robert Paine began prying ochre sea stars, Pisaster ochraceus, off a stretch of rocky shore at Mukkaw Bay, Washington, and throwing them back into the sea. He left a neighbouring stretch untouched for comparison. Without the sea stars, the mussels they ate spread across the rock and crowded out other species. In his 1966 paper Paine reported that the number of species on the cleared shore fell from 15 to 8.^[6] In 1969 he gave the idea a name: Pisaster was a keystone species, one whose effect on its community is far larger than its abundance would suggest, like the stone at the top of an arch.

When a predator's effect passes down through several levels, ecologists call it a trophic cascade. The best-known marine example runs from sea otters to sea urchins to kelp, and it is the story of Lesson 11.

The keystone idea has been refined since. As Hakai Magazine reported in 2016, Bruce Menge found over decades of work on the Oregon coast that Pisaster rules most strongly on wave-exposed shores like Paine's site and has less effect in sheltered ones, and that its control depends on there being enough sea stars and enough mussels. His colleague Sarah Gravem found that the black turban snails thought to be controlled by Pisaster were actually controlled by a small, overlooked sea star, Leptasterias. As she put it, Paine's real legacy is not that Pisaster matters everywhere, but that "predators can shape the world around them."

The biological pump

Not everything made in the sunlit layer is eaten there. Dead plankton, fecal pellets, mucus, sand and dust clump together and sink as marine snow. NOAA describes flakes that grow as they fall, some reaching several centimetres across, and some falling for weeks before reaching the bottom. For the animals of the deep, which live below the reach of sunlight (Lesson 13), this rain is the main source of food.

It also carries carbon. Phytoplankton take carbon dioxide out of surface water to build their cells. When that carbon sinks below the surface layer, it is cut off from the air, and if it reaches the deep circulation of Lesson 7, it can stay there for centuries. This transfer is called the biological pump. The scientists who planned NASA's EXPORTS field campaign noted in 2016 that estimates of how much carbon leaves the sunlit layer each year range from about 5 to more than 12 billion tonnes, an uncertainty as large as the yearly changes humans cause in the global carbon cycle. They described three routes down: sinking particles, dissolved and suspended carbon mixed downward, and zooplankton that feed near the surface at night and release carbon at depth by day.

What reaches the seafloor feeds bottom life and becomes sediment. NOAA notes that about three quarters of the deep ocean floor is covered with this fine ooze, building up as much as six metres every million years.

Common misconceptions

• "Most of what plankton make ends up as fish." At about 10 percent per step, only a tiny fraction reaches fish, and how tiny depends on how many steps there are.
• "Food chains are simple lines." Real food webs branch and loop; bacteria recycle dissolved material back into the web through the microbial loop.
• "The most numerous species is always the most important." A keystone predator can shape a whole community while being relatively rare, although, as later work showed, its influence depends on conditions.
• "Dead plankton simply dissolve." Much of it clumps into marine snow that feeds deep-sea life and carries carbon away from the surface.

Recap

• Ryther estimated that upwelling zones, about 0.1 percent of the ocean, produce about half of its fish.
• Energy passes between trophic levels at a mean efficiency of about 10 percent (Pauly and Christensen), so each extra step divides the top by about ten.
• Upwelling zones have more production, higher efficiency and, above all, short chains, because large diatoms can be eaten almost directly by small fish; tiny open-ocean producers need extra steps, including the microbial loop.
• Paine's removal of Pisaster cut species on the shore from 15 to 8, founding the keystone idea; later work shows keystone effects depend on context.
• Marine snow carries food and carbon down; the biological pump exports an estimated 5 to more than 12 billion tonnes of carbon a year from the sunlit layer.

Sources

1. Ryther, J. H. (1969). Photosynthesis and fish production in the sea. Science, 166(3901), 72-76.
2. Pauly, D., and Christensen, V. (1995). Primary production required to sustain global fisheries. Nature, 374, 255-257.
3. FishBase. (n.d.). Ecopath modeling and some key results, quoting Ryther (1969) and reporting the 10 percent mean transfer efficiency. fishbase.se
4. NOAA National Ocean Service. (n.d.). What is upwelling? and What is marine snow? oceanservice.noaa.gov
5. Azam, F., Fenchel, T., Field, J. G., Gray, J. S., Meyer-Reil, L. A., and Thingstad, F. (1983). The ecological role of water-column microbes in the sea. Marine Ecology Progress Series, 10, 257-263. int-res.com
6. Paine, R. T. (1966). Food web complexity and species diversity. The American Naturalist, 100(910), 65-75.
7. Simons, E. (2016, July 7). Shrinking the keystone. Hakai Magazine. hakaimagazine.com
8. Siegel, D. A., Buesseler, K. O., Behrenfeld, M. J., and others. (2016). Prediction of the export and fate of global ocean net primary production: The EXPORTS science plan. Frontiers in Marine Science, 3, 22. frontiersin.org

A jubilee on Mobile Bay

Along the eastern shore of Mobile Bay, Alabama, there are summer days when crabs and flounder crowd into the shallow water at the edge of the beach, so many and so sluggish that people walk along the shore filling ice chests with them. Local communities call it a jubilee. NOAA explains what is really happening: the water offshore has run low on dissolved oxygen, and the animals have moved into the shallows trying to get more.

A jubilee is a feast for the people on the shore and an emergency for the animals. It is also a sign that an estuary, one of the most productive habitats on Earth, can run out of breath. This lesson explains what estuaries are and why so much sea life depends on them, then works step by step through the measurements that tell scientists when their oxygen is running out.

Where rivers meet the sea

An estuary is a body of water where fresh water draining from the land mixes with seawater, usually where a river meets the sea. The mixture is called brackish water. NOAA gives the range: fresh water has a salinity near zero, the ocean averages about 35, and brackish water can be anywhere from 0.5 to 35, changing from place to place and from hour to hour with the tides, the weather and the river's flow. Bays, sounds, lagoons and tidal rivers can all be estuaries. Many are sheltered from the full force of ocean waves by barrier islands, reefs or spits of sand.

Estuaries matter far beyond their size. NOAA estimates that they provide habitat for more than 75 percent of the U.S. commercial fish catch, and for an even greater share of the recreational catch. Some animals, such as oysters, live there all their lives. Others visit: American shad, Atlantic menhaden and striped bass spend most of their lives in the ocean but return to brackish water to spawn, and the young of many fish and shellfish grow up in the shelter of marsh creeks and mangrove roots. That is why estuaries are called the nurseries of the sea.

Four ways to mix fresh and salt water

Fresh water is lighter than seawater, as Lesson 3 showed, so it tends to slide out over the top. How much the two mix depends on a contest between the river's flow and the stirring of the tides. NOAA classifies estuaries by the result:

The arrangement matters to life in the estuary. A strongly layered estuary behaves like the layered ocean of Lesson 3: the bottom water is cut off from the air and can lose its oxygen. A well-mixed estuary keeps stirring oxygen down. NOAA notes that shallow, well-mixed estuaries are less likely to become hypoxic for exactly that reason.

Habitats on the edge: marsh and mangrove

Along temperate coasts, the edges of estuaries are salt marshes: flat, grassy wetlands flooded by the tide. Like the rocky shore in Lesson 5, a marsh is arranged in bands by how often the sea covers it.

The marsh ground is peat, a spongy, waterlogged mat of roots and decaying plants that can be several feet thick. Because it is waterlogged and full of rotting material, its oxygen is extremely low, and the bacteria that thrive there make the rotten-egg smell of marsh mud. That smell is natural, not a sign of pollution. Mud flats between the creeks hold clams, oysters, fiddler crabs and worms.

In the tropics, mangrove forests take the place of marshes. NOAA says they grow where the sea surface never falls below 16 °C, line about two-thirds of tropical coastlines, and include about 80 species of mangrove tree. In Florida, red mangroves stand on tangled prop roots at the water's edge, black mangroves grow a little higher, and white mangroves occupy the highest ground. The roots slow the water, trap sediment, shelter young snapper, tarpon, oysters, crabs and shrimp, and reduce erosion from storm surges.

Reading the water: what a monitoring station records

NOAA and the coastal states run a network of 30 protected estuaries, the National Estuarine Research Reserves, set aside for research and education. In each one, the System-Wide Monitoring Program keeps instruments in the water at four or more fixed stations, recording every 15 minutes: water temperature, specific conductivity and salinity, dissolved oxygen (as milligrams per litre and as percent saturation), depth, pH and turbidity. Nutrients and chlorophyll are sampled monthly, and a weather station logs rain, wind and sunlight.

Here is the procedure for reading such a record, worked on a constructed example of the kind of late-summer day that a station in a shallow, nutrient-rich creek can log. The numbers are illustrative, chosen to show the patterns NOAA describes; in the investigation you will read real ones.

1. Find the tide first. Depth rises and falls: high water around midnight and noon, low water around 06:00 and 20:00. Read every other column against it.
2. Match salinity to the tide. Salinity climbs to 24-25 when the tide is high and drops to 16-17 at low tide. Seawater pushes in on the flood tide; river water dominates on the ebb. That is a healthy estuary breathing in and out.
3. Read the oxygen against the sun, not the tide. Oxygen climbs from mid-morning to a peak of 9.0 mg/L at 16:00 and falls all night to 1.8 mg/L at 06:00. Photosynthesis adds oxygen only in daylight; respiration and decay use it up around the clock. In nutrient-rich water, with a lot of algae, both effects are large, so the daily swing is large.
4. Check against the thresholds. Hypoxia is usually defined as below 2 mg/L. The 06:00 reading is hypoxic. For a few hours around dawn, fish and crabs in this creek would be struggling for oxygen.
5. Think about temperature. The water is near 28-30 °C, and warm water holds less oxygen. NOAA gives fresh water's capacity as 14.6 mg/L at 0 °C but only 9.2 mg/L at 20 °C, and seawater holds about 20 percent less than fresh water at the same temperature. That is why the lowest oxygen levels in estuaries usually come in late summer.

Now change one input. Suppose a cool front brings a windy, cloudy day. Wind mixes oxygen in from the air, and clouds cut photosynthesis. The afternoon peak would be lower, but the wind would keep the dawn minimum from dropping so far, and the creek might never become hypoxic. The same nutrients, different weather, a different result: keep that in mind for the next section.

Bottom line: Read an estuary record in order: tide, then salinity, then oxygen against day and night, then thresholds and temperature. In nutrient-rich water, the most dangerous moment is often just before dawn in late summer.

Scaling up: the Gulf dead zone

The same chemistry, on a vast scale, makes one of the largest dead zones in the world. The Mississippi and Atchafalaya Rivers drain a watershed covering more than 40 percent of the continental United States, crossing 22 states, much of it farmland. Nitrogen and phosphorus washing off fields and out of cities feed huge blooms of algae in the northern Gulf. When the algae die and sink, their decay uses up the oxygen near the bottom, and a layer of water with 2 mg/L of oxygen or less spreads across the seafloor every summer. Fish and shrimp leave or die. (Since 2025, U.S. federal agencies such as NOAA have called this body of water the Gulf of America; it is still the Gulf of Mexico on most international maps.)

Scientists have mapped the zone every summer since 1985, and NOAA forecasts its size each June from the river's flow and nutrient load. Look at what happened in two recent years:

Does the tiny 2026 zone mean the problem is solved? Read it the way you read the creek. The nutrients were still there; the weather changed the result, exactly as the windy day did in the worked example. That is why the Mississippi River/Gulf Hypoxia Task Force, a partnership of states and federal agencies, measures progress with a five-year average. After the 2026 survey, NOAA reported that average at 3,745 square miles, almost twice the task force's goal of fewer than 1,900 square miles by 2035. Reaching it depends on reducing the nutrients that leave farms and cities more than a thousand kilometres upstream, which is one reason marine science matters to students who live nowhere near the sea.

Common misconceptions

• "An estuary is just the mouth of a river." Estuaries include bays, sounds, lagoons and tidal rivers, and in a salt-wedge estuary seawater can creep far upstream along the bottom.
• "The rotten-egg smell of a marsh means it is polluted." Waterlogged peat is naturally low in oxygen, and the bacteria living there make that smell.
• "Oxygen in the water stays about the same all day." In productive water it swings widely, peaking in the afternoon and bottoming out before dawn.
• "Dead zones are caused by poisons." They are caused by too many nutrients: the algae they feed die, and their decay uses up the oxygen.
• "A small dead zone in one year means the problem is fixed." One summer's size depends on weather; the five-year average is still about twice the goal.

What to remember

• Estuaries hold brackish water, salinity 0.5 to 35, and provide habitat for more than 75 percent of the U.S. commercial fish catch.
• Mixing ranges from salt-wedge estuaries, where river water slides over seawater, to vertically mixed ones stirred by strong tides; layered water is more likely to lose its oxygen.
• Salt marshes are zoned by flooding: smooth cordgrass in the daily-flooded low marsh, salt hay in the high marsh flooded at spring tides. Mangroves replace marshes where the sea stays above 16 °C.
• Monitoring stations record salinity, temperature and dissolved oxygen every 15 minutes. Read tide first, then salinity, then oxygen against day and night; hypoxia is below 2 mg/L.
• Nutrients from the Mississippi watershed create a summer dead zone in the Gulf; weather can shrink one year's zone, as Tropical Storm Bertha did in 2026, but the five-year average remains about twice the 2035 goal.

Sources

1. NOAA National Ocean Service. (n.d.). Estuaries tutorial: What is an estuary?; The economy and environment; Water circulation; Estuarine habitats; Monitoring estuaries. oceanservice.noaa.gov
2. NOAA Office for Coastal Management. (n.d.). National Estuarine Research Reserves. coast.noaa.gov
3. NERRS Centralized Data Management Office. (n.d.). SWMP parameters. cdmo.baruch.sc.edu
4. NOAA National Centers for Coastal Ocean Science. (2025, July 31). Below average summer 2025 dead zone measured in Gulf. coastalscience.noaa.gov
5. NOAA National Centers for Coastal Ocean Science. (2026, June 16). NOAA forecasts an above-average summer dead zone in Gulf of America. coastalscience.noaa.gov
6. NOAA. (2026, August 5). NOAA and partners find smaller-than-anticipated hypoxic zone in Gulf of America. noaa.gov

A raft of rare animals below a new highway

In 1937, people driving California's newly built Highway 1 along the cliffs of Big Sur looked down and saw something nobody expected: a cluster of dark, floating bodies in the rough water near Bixby Bridge. On 20 June 1938, Life magazine published a photograph of the scene under the headline "The 'extinct' sea otter swims back to life." California officials had quietly known about this small group for years. They had kept it secret to protect it.

The story of those otters, and of the forests they live in, is one of the clearest examples anywhere of a predator shaping a whole ecosystem. It also has a second act, still unfolding, in which the forests of northern California almost vanished in a few years. This lesson follows both.

A forest made of algae

Kelp are large brown algae, not plants. They have no flowers and no true roots: a kelp holds onto rock with a root-like holdfast, and its long stalks and leaf-like blades are buoyed toward the light. NOAA notes that in ideal conditions kelp can grow up to 18 inches a day. Kelp needs cool, nutrient-rich water and light for photosynthesis, so kelp forests grow on shallow rocky coasts and are rarely found deeper than about 15-40 metres. Along the Pacific coast of North America, the giant kelp forests of central and southern California give way, farther north, to forests of bull kelp, which form the canopy along the rugged northern California coast.

Like a forest on land, a kelp forest is a habitat as much as a plant community. NOAA describes kelp forests as harbouring a greater variety of plants and animals than almost any other ocean community: fish shelter among the stalks, invertebrates graze the blades, and seals, sea lions, whales, sea otters and seabirds feed there. Storms and El Niño years can tear the forests apart, and they regrow each spring.

The fur trade and the long silence

Sea otters once lived around the whole rim of the North Pacific, from northern Japan to Baja California. They keep warm with thick fur rather than blubber, and that fur made them a prize. After Vitus Bering's expedition brought pelts back from the far North Pacific in the 1740s, a multinational fur rush followed. By the time the International Fur Seal Treaty of 1911 banned hunting sea otters in international waters, the U.S. Fish and Wildlife Service estimates that only 13 remnant populations survived, each of about 10 to 100 animals. The group off Big Sur, separated from any other otters by about 2,000 miles, was the last of the southern sea otters.

Those survivors slowly spread north and south along the California coast. As they spread, a pattern appeared that only made sense once ecologists studied otters where they had never recovered.

Islands with otters and islands without

In 1970 a young ecologist named James Estes began work in the Aleutian Islands of Alaska, where some islands had sea otter populations again and some still did not. The difference underwater was stark. Where otters were abundant, as at Amchitka Island, the rocks were covered with kelp. Where they were absent, as at nearby Shemya, sea urchins carpeted the bottom and the kelp was gone, leaving what divers call an urchin barren. In 1974 Estes and his colleague J. F. Palmisano argued in Science that the otters themselves were the cause. Otters eat urchins; urchins eat kelp; take away the otters and the urchins multiply until they mow the forest down.

It was a trophic cascade, the effect Lesson 9 described, and the sea otter became one of the classic keystone species. Later work added a twist. Otters do not need to eat every urchin. The Fish and Wildlife Service describes how, where otters hunt, frightened urchins stop roaming the open rock and chewing through the holdfasts of living kelp, and instead hide in crevices, eating pieces of kelp that break off and drift down. Fear alone protects the forest.

The cascade can also start farther out to sea. In the 1990s, Estes and his colleagues watched sea otter numbers crash across western Alaska. They had rarely seen killer whales attack otters before, but now they saw it repeatedly. In a 1998 paper they argued that killer whales, perhaps short of their usual prey, had switched to eating otters, and that the otter decline had let urchins increase and the kelp forests shrink. A change in the open ocean had reached all the way down to the seafloor by way of four links: whales, otters, urchins, kelp.

So what? One predator, by eating or frightening the main grazer, can decide whether a coast is covered in forest or bare rock. Remove the predator, by hunting it or by a new predator hunting it, and the forest can disappear.

The northern California collapse

Northern California has bull kelp but few otters; there, sea stars were among the main predators of urchins. What happened next was described in 2019 by Laura Rogers-Bennett and Cynthia Catton, who had monitored those kelp forests for two decades for the California Department of Fish and Wildlife and the University of California, Davis. They called it a perfect storm, and it came in steps:

Notice what this story is and is not. It is not a single cause. The authors point to several stressors acting together, and they say plainly that key questions remain, such as the relative roles of warming and sea star disease in the urchin explosion. That honesty is part of the science: a collapse with several causes needs several kinds of response.

Solving the sea star mystery

For more than a decade, nobody knew what caused sea star wasting disease, which made sea stars twist and shed their arms and killed billions of them from Mexico to Alaska. Some researchers had suspected a virus. In August 2025, a team from the University of British Columbia, the Hakai Institute and the University of Washington reported in Nature Ecology & Evolution that the cause is a strain of the bacterium Vibrio pectenicida, which they named FHCF-3. When they grew the strain in pure culture and injected it into healthy sunflower sea stars, the stars died within days of showing symptoms. Other Vibrio bacteria thrive in warm water, so the team is now studying how warming affects the disease.

The sunflower sea star has lost more than 90 percent of its population. In 2023 NOAA Fisheries proposed listing it as threatened under the Endangered Species Act, which would make it the first sea star listed. Researchers are now breeding sunflower stars in captivity to test whether they can be returned to the wild.

What helps, and how we know

Restoration has followed the cascade's logic: if urchins are the problem, remove urchins. In Mendocino County, Reef Check worked with the California Department of Fish and Wildlife, the Ocean Protection Council, The Nature Conservancy and local commercial urchin divers at two sites, Noyo Bay from August 2020 and Albion Cove from July 2021. Over two years the divers removed 45,118 pounds of purple urchins. In Noyo Bay, bull kelp grew back over 15 months to about 20 percent of its historical density, while a nearby urchin barren left alone showed no change. That untouched barren is what makes the result convincing: it is the comparison, like Paine's untouched shore in Lesson 9. Twenty percent is a start, not a recovery, and it took two years of sustained diving to get there.

To judge what works across a whole coast, managers need to know where kelp was, where it is now, and how it has changed. That is the job of Kelpwatch.org, launched in 2022 by The Nature Conservancy, Woods Hole Oceanographic Institution, UCLA and UC Santa Barbara. It turns nearly 40 years of Landsat satellite images, starting in 1984, into maps of kelp canopy, season by season, which anyone can animate and download. Using it, Tom Bell and colleagues found that forests responded to and recovered from the 2014-2016 marine heatwaves very differently from place to place: some recovered within a few years, others did not.

Common misconceptions

• "Kelp is a plant, like seaweed trees." Kelp is a brown alga. It anchors with a holdfast, not roots, and takes its nutrients from the water.
• "Urchin barrens are just areas where kelp never grew." Many were kelp forests until urchins, released from their predators, ate them down.
• "The kelp collapse had one cause." Disease, a marine heatwave and an urchin explosion acted together, and researchers are still weighing their roles.
• "Predators protect kelp only by eating urchins." The fear of otters changes urchin behaviour, keeping them hidden and feeding on drift kelp.
• "Sea star wasting was caused by a virus." That was a leading idea for years; the 2025 study identified a bacterium, Vibrio pectenicida.

Putting it together

• Kelp are large brown algae that need cool, nutrient-rich, shallow water; giant kelp can grow up to 18 inches a day, and kelp forests are among the most diverse ocean communities.
• The fur trade reduced sea otters to 13 small remnant groups by 1911; the southern sea otter survived at Big Sur.
• Estes and Palmisano (1974) showed the cascade: otters eat urchins, urchins eat kelp; without otters, urchin barrens replace forests. Killer whales eating otters in the 1990s spread the cascade from the open sea.
• In northern California, sea star wasting (2013), a marine heatwave (2014-2017) and a 60-fold urchin increase removed more than 90 percent of bull kelp along 350 km and closed the red abalone fishery.
• A strain of Vibrio pectenicida was identified as the cause of sea star wasting in 2025; the sunflower sea star, down more than 90 percent, was proposed as threatened in 2023.
• Kelpwatch.org maps kelp from Landsat images since 1984 and shows recovery varies from place to place.

Sources

1. U.S. Fish and Wildlife Service. (2022, July). On second chances: The southern sea otter's return to ecological relevance. fws.gov
2. Estes, J. A., and Palmisano, J. F. (1974). Sea otters: Their role in structuring nearshore communities. Science, 185(4156), 1058-1060. pubs.usgs.gov
3. Estes, J. A., Tinker, M. T., Williams, T. M., and Doak, D. F. (1998). Killer whale predation on sea otters linking oceanic and nearshore ecosystems. Science, 282, 473-476.
4. Rogers-Bennett, L., and Catton, C. A. (2019). Marine heat wave and multiple stressors tip bull kelp forest to sea urchin barrens. Scientific Reports, 9, 15050. Summarized in University of California, Davis. (2019, October 21). California's crashing kelp forest. ScienceDaily. sciencedaily.com
5. University of British Columbia. (2025, August 4). 'Disease detectives' discover cause of sea star wasting disease that wiped out billions of sea stars. news.ubc.ca
6. NOAA Fisheries. (2025). Sunflower sea star [Species page]. fisheries.noaa.gov
7. NOAA National Ocean Service. (2026). What is a kelp forest? oceanservice.noaa.gov
8. NASA Landsat Science. (2022, April 4). World's largest dynamic kelp map launched. science.nasa.gov
9. Reef Check Foundation. (2022, August 29). Restoration of northern California bull kelp forests report. reefcheck.org
10. Bell, T. W., Cavanaugh, K. C., Saccomanno, V. R., and others. (2023). Kelpwatch: A new visualization and analysis tool to explore kelp canopy dynamics reveals variable response to and recovery from marine heatwaves. PLOS ONE, 18(3), e0271477. journals.plos.org

A drill through 4,000 feet of coral

In 1952, at Enewetak Atoll in the Marshall Islands, geologists drilling as part of a U.S. Atomic Energy Commission test program bored down through a ring-shaped coral island. For more than 4,000 feet the drill brought up limestone, most of it the remains of shallow-water reef. Then, at about 4,150 feet, it struck volcanic basalt. The limestone just above the basalt was of Eocene age, tens of millions of years old, and it had formed in shallow water.

Shallow-water limestone four-fifths of a mile down is a puzzle. Here is a natural first explanation, the one many people carry around: coral is a kind of rock, perhaps with some plants growing on it. If that were true, the Enewetak result would make no sense, because rock does not grow upward toward the light. This lesson debugs that belief, step by step, and the corrected picture explains the drill core, the bleaching events that make the news, and what can be done about them.

The belief, and the clues it cannot explain

The belief is understandable. Coral is hard, it never moves, and many corals branch like trees or spread like shrubs. NOAA notes that it is no wonder many people think corals are plants. But three observations do not fit "rock" or "plant":

• Coral colonies grow, and reefs build upward over thousands of years.
• Corals have tiny tentacles that catch food from the water, which plants do not do.
• Corals can turn white, "bleach," and die in a warm summer. Rocks do not die.

What a coral actually is

NOAA's answer is that corals are sessile animals, fixed in one place, and part of an ancient partnership. A coral colony is made of hundreds to thousands of tiny animals called polyps, most no thicker than a nickel, related to sea anemones and jellyfish. Each soft-bodied polyp has a ring of tentacles around its mouth for catching food, and each secretes a hard cup of limestone, calcium carbonate, beneath itself. Polyps grow, die and are replaced, and their skeletons pile up, so a colony can live for a very long time and a reef can grow for thousands of years.

The partner lives inside the polyps' tissues: microscopic algae called zooxanthellae. The algae get shelter and use the coral's waste products for photosynthesis. In return they supply oxygen, remove wastes and pass the coral the products of photosynthesis. NOAA calls these algae the coral's primary food source, and they also give corals their colour.

Because the algae need sunlight, reef-building corals grow in warm, clear, shallow tropical water, and new corals tend to settle on the skeletons of older ones, climbing toward the light. That is the key that unlocks Enewetak.

Key idea: A coral is an animal colony with algae living in its tissues. The algae feed the animal; the animal builds limestone. A reef is a structure built by living things, and it grows.

Darwin's sinking islands

In the 1830s, on the voyage of the Beagle, Charles Darwin noticed three kinds of reef. A fringing reef hugs the shore of an island. A barrier reef stands offshore, separated from the land by a lagoon. An atoll is a ring of reef around a lagoon with no island in the middle at all. Darwin proposed that they are stages of one story: a fringing reef grows around a volcanic island; the island slowly sinks; the corals, which must stay in the light, keep growing upward, so a lagoon opens between reef and land; and when the island disappears beneath the water, only the ring remains. He could not explain why islands would sink. Today we know that oceanic crust subsides as it moves away from the mid-ocean ridge where it formed, as Lesson 2 described.

Darwin himself suggested drilling into an atoll to test the idea. If he was right, a drill should pass through a great thickness of shallow-water reef limestone and then hit the volcano beneath. That is what Enewetak showed in 1952, and the cores were widely hailed as the final proof of his theory. The geologist who led the drilling, Harry Ladd, later added a caution: the atolls did not seem to have passed through every one of the fringing and barrier stages Darwin pictured. The core of the idea held; some of the details did not.

Now do the arithmetic that makes the picture convincing. Webb's oceanography text gives reef growth of about 3-5 metres per 1,000 years. The limestone at Enewetak is about 1,270 m (4,150 feet) thick. At 3-5 m per 1,000 years, that much reef could build up in roughly 250,000 to 420,000 years. But the oldest limestone there is tens of millions of years old, so the island must have sunk far more slowly than corals can grow. Corals easily kept their heads in the light while the volcano sank beneath them over tens of millions of years.

Why reefs matter

NOAA estimates that coral reefs cover less than one percent of the ocean floor but support about 25 percent of all marine life. They shelter fish that coastal communities eat and sell, break waves before they reach the shore, and draw visitors. That is why the next part of the story is alarming.

Bleaching: when the partnership breaks

When corals are stressed by changes in temperature, light or nutrients, they expel the algae from their tissues. Without the coloured algae, the transparent tissue shows the white skeleton beneath, and the coral looks bleached. A bleached coral is not dead. NOAA explains that if the stress is mild and short, corals can recover their algae; if the loss lasts and the stress continues, the coral starves and dies. Heat from a warming ocean is the leading cause, although cold can do it too: in January 2010 unusually cold water bleached corals in the Florida Keys.

NOAA's Coral Reef Watch tracks heat stress from satellites with two numbers. A HotSpot is how far the sea surface temperature on a reef rises above its normal warmest monthly temperature, and HotSpots of 1 °C or more indicate heat stress that can lead to bleaching. Degree heating weeks (DHW) add up that stress over the most recent 12 weeks, in units of °C-weeks.

The levels above 8 were added in December 2023, because the heat of that year went far beyond the old scale. Work two cases. Reef A runs 1 °C above its normal maximum for 4 weeks: 1 x 4 = 4 °C-weeks, a risk of bleaching. Reef B runs 2 °C above for 6 weeks: 2 x 6 = 12 °C-weeks, and many species are likely to die. A difference of one degree, held for a few extra weeks, separates a warning from a catastrophe.

The fourth global bleaching event

On 15 April 2024, NOAA and the International Coral Reef Initiative confirmed that the world was in its fourth global coral bleaching event. By Coral Reef Watch's count, from 1 January 2023 to 30 September 2025, bleaching-level heat stress reached about 84.4 percent of the world's coral reef area, and mass bleaching was documented in at least 83 countries and territories. The previous record, set in the third event of 2014-2017, was 68.2 percent; the first two global events were in 1998 and 2010. In June 2026 Coral Reef Watch reported that the fourth event had likely ended in 2025.

The scientific assessments are blunt about the future. In its 2018 special report on 1.5 °C of warming, the Intergovernmental Panel on Climate Change projected that coral reefs would decline by a further 70-90 percent at 1.5 °C of global warming, with losses of more than 99 percent at 2 °C. NOAA states that the warming behind mass bleaching is primarily due to greenhouse gases from human activities.

What helps? The evidence points to two scales. Globally, only cutting greenhouse gas emissions reduces the heat stress itself. Locally, NOAA lists actions that help reefs survive and recover between heat events: reducing pollution and runoff, which can bleach and smother near-shore corals, and avoiding overfishing, which removes key reef species and damages coral habitat. Local care cannot stop a marine heatwave, but it can decide whether a reef has the strength to come back from one.

Common misconceptions

• "Coral is a rock" or "coral is a plant." Corals are colonies of small animals; the rock is their skeleton, and the plant-like part is algae living inside them.
• "A bleached coral is a dead coral." Bleached corals have lost their algae and are starving. They can recover if the stress ends soon enough.
• "Bleaching is just a change of colour." The colour change means the coral has lost its main food supply.
• "Reefs are thin layers on the seafloor." At Enewetak, reef limestone is more than 4,000 feet thick, built over tens of millions of years.
• "Protecting a reef locally can stop bleaching." Local protection helps recovery, but only lower greenhouse gas emissions reduce the heat stress that causes mass bleaching.

The short version

• Corals are animal colonies of polyps related to anemones and jellyfish; zooxanthellae in their tissues are their main food source; their skeletons build reefs.
• Darwin explained fringing reefs, barrier reefs and atolls as stages around a sinking island. Drilling at Enewetak in 1952 found more than 4,000 feet of reef limestone on volcanic basalt.
• Reefs cover less than 1 percent of the ocean floor but support about 25 percent of marine life.
• Bleaching is the loss of the algae under stress; degree heating weeks measure accumulated heat stress, with bleaching risk at 4 and reef-wide bleaching with mortality likely at 8.
• The fourth global bleaching event (2023-2025) brought bleaching-level heat stress to about 84 percent of reef area. The IPCC projects 70-90 percent further decline at 1.5 °C and more than 99 percent at 2 °C. Cutting emissions addresses the cause; local protection helps recovery.

Sources

1. NOAA National Ocean Service. (2026). Are corals animals or plants? and What is coral bleaching? oceanservice.noaa.gov
2. Ladd, H. S., and Schlanger, S. O. (1960). Drilling operations on Eniwetok Atoll (U.S. Geological Survey Professional Paper 260-Y). pubs.usgs.gov
3. Sponsel, A. (2006). Core drilling at Bikini and Eniwetok atolls, 1947-1952 [Abstract]. Geological Society of America Annual Meeting, Philadelphia. gsa.confex.com
4. Darwin, C. (1842). The structure and distribution of coral reefs. Smith, Elder and Co.
5. Webb, P. (2019). Introduction to oceanography. Roger Williams University Open Publishing; Geosciences LibreTexts edition, CC BY 4.0. Section 4.10 (coral reefs). geo.libretexts.org
6. NOAA Coral Reef Watch. (n.d.). Daily 5km degree heating week product and Bleaching alert area product (version 3.1). coralreefwatch.noaa.gov
7. NOAA Coral Reef Watch. (2026, June 2). Current global bleaching: Status update. coralreefwatch.noaa.gov
8. IPCC. (2018). Summary for policymakers. In Global warming of 1.5 °C: An IPCC special report. Statement B.4.2 on coral reefs.
9. NOAA National Ocean Service. (2026). How does climate change affect coral reefs? oceanservice.noaa.gov

Warm water and white clams

On 17 February 1977, the submersible Alvin sank 2.5 kilometres to the floor of the Galápagos Rift, a spreading ridge northeast of the Galápagos Islands. Aboard were the pilot Jack Donnelly and two scientists, John Corliss and Tjeerd van Andel. They were looking for hot springs that a towed camera had hinted at a few days earlier. Its thermometer had jumped for less than three minutes, and its photographs from those minutes showed hundreds of white clams on fresh lava. When Alvin reached the spot, the crew saw warm, shimmering water rising from cracks in the rock. Around it lived a crowded community of animals that nobody had ever seen, including clams 30 centimetres long.

The deep sea had long been imagined as a cold, dark, nearly empty desert. It is cold and dark. It is not empty. This lesson compares the ocean's layers from top to bottom, then looks at the three problems every deep animal must solve: darkness, scarce food and pressure.

The zones compared

Oceanographers divide the open ocean into layers by depth. NOAA's descriptions, with the pressures worked out as in Lesson 4, give this picture.

Reading the table: three problems

Read across the rows and three trends stand out. Light disappears within the first kilometre. Temperature falls and then levels off a few degrees above freezing. Pressure climbs steadily all the way down. Read down the last column and a fourth trend appears: below the sunlit layer, nothing makes food from sunlight. Almost everything that lives in the deep depends on food made far above it.

So every deep-sea animal must solve three problems: seeing, or being seen, in the dark; finding enough food; and living under pressure. Lesson 4 dealt with pressure. The rest of this lesson takes the first two, and then an exception to the rule that all food comes from the sun.

Remember: With depth, light disappears, the water cools to a few degrees above freezing, and pressure rises without stopping. Below the sunlit layer, almost all food arrives from above.

Light made by life

In the twilight zone, NOAA notes, fish often have large eyes that face upward, to catch the silhouette of prey against the faint light from above, and bioluminescence, light made by living things, begins to appear. Below 1,000 m, light made by animals is the only light there is.

How common is it? In 2017 Séverine Martini and Steven Haddock of the Monterey Bay Aquarium Research Institute went through video from 240 dives by remotely operated vehicles in and around Monterey Canyon, California, from the surface down to about 4,000 m, and counted more than 350,000 animals larger than a centimetre. About 76 percent of the observed animals were bioluminescent, and the share was remarkably similar at every depth. In the deep sea, making light is the rule, not the exception. Animals use it to find food, to hide and startle predators, and to signal to each other.

The greatest migration on Earth

Food is scarcer the deeper you go, and many animals of the twilight zone solve the problem by commuting. NOAA Ocean Exploration describes how billions upon billions of animals, mostly zooplankton such as copepods but also small fish, shrimp and jellies, swim up toward food-rich surface waters every night and sink back to the depths before sunrise. The likely reason: darkness hides them from predators that hunt by sight while they feed. NOAA calls this daily vertical migration the greatest migration on Earth. As Lesson 9 showed, it also carries carbon down, because the migrators eat near the surface and release carbon at depth.

Below the reach of the migrators, animals live on marine snow, the slow rain of dead plankton and fecal pellets from Lesson 9, and on rare windfalls such as the body of a dead whale sinking to the floor.

An exception: food from below

The 1977 discovery broke the rule that all deep-sea food comes from sunlight. NOAA explains that hydrothermal vents form at spreading ridges and other plate boundaries where seawater seeps down through cracks in the crust, is heated by magma below, and comes back out. In 1979, the same kind of exploration found the first "black smokers," chimneys pouring out water dark with dissolved minerals. Webb's oceanography text describes vent water as hot as 350 °C, chimneys up to 20 m tall, and vent communities with giant tube worms over 2 m long, crabs, shrimp, giant mussels and mats of bacteria. NOAA adds that vent water can exceed 700 °F (about 370 °C) yet does not boil, because of the enormous pressure.

What feeds them? The vent water is rich in hydrogen sulfide. Bacteria use the energy released by oxidizing sulfur compounds to make carbohydrates from carbon dioxide and water. That is chemosynthesis: primary production powered by chemical energy instead of light. The bacteria are the base of the food web. Some animals graze on them; others, such as the giant tube worms, carry bacteria inside their bodies and live on what the bacteria make. By 2010, more than 500 active vent fields had been found around the world.

One honest qualification: the bacteria use oxygen dissolved in the seawater to oxidize the sulfide, and that oxygen was made by photosynthesis at the surface (Lesson 4). So vent life is powered by Earth's chemistry, but it still breathes the sun's leftovers.

An open question: mining the deep

Lesson 2 described polymetallic nodules: lumps a few centimetres across, rich in manganese and iron with smaller amounts of copper, nickel and cobalt, lying on the floors of the central oceans and growing just a few millimetres per million years. Those metals are in demand for batteries, and mining companies have explored nodule fields such as the Clarion-Clipperton Zone of the Pacific with a view to collecting them.

In international waters, mining is regulated by the International Seabed Authority, set up under the United Nations Convention on the Law of the Sea. In 2026 its rules for commercial exploitation were still being negotiated as draft regulations. The scientific worry follows from facts you already know. A nodule that took millions of years to grow will not regrow on any human timescale, and the animals that live on and around nodules would lose that habitat. And as Lesson 1 showed, less than a third of the seafloor has even been mapped in detail. Whether, where and how to mine the deep is partly a scientific question: how much harm, and for how long? It is also a question of values, and one you will be old enough to vote on.

Common misconceptions

• "The deep sea is lifeless." It holds bioluminescent animals at every depth and dense communities around vents; it is sparse in food, not empty of life.
• "All life on Earth depends on sunlight." Vent communities run on chemosynthesis, although they still use oxygen produced by photosynthesis at the surface.
• "Vent water must boil at 350 °C." The pressure at those depths keeps it liquid.
• "Glowing animals are rare oddities." About three quarters of the animals observed off California, at all depths, can make light.
• "Deep-sea animals stay deep." Billions of them rise toward the surface every night and sink again by morning.

What to carry forward

• The open ocean has five zones: sunlit (0-200 m), twilight (200-1,000 m), midnight (1,000-4,000 m), abyss (4,000-6,000 m) and hadal trenches (below 6,000 m). Light vanishes, temperature levels off near freezing and pressure keeps rising.
• About 76 percent of animals observed off California, from the surface to about 4,000 m, are bioluminescent (Martini and Haddock, 2017).
• Daily vertical migration, NOAA's "greatest migration on Earth," brings billions of animals up to feed at night.
• Hydrothermal vents, found in 1977, support communities built on chemosynthesis by sulfide-oxidizing bacteria; black smokers were found in 1979.
• Nodules grow a few millimetres per million years; the rules for mining them in international waters were still in draft in 2026.

Sources

1. Oleson, T. (2015, January 4). Benchmarks: February 17, 1977: Hydrothermal vents are discovered. EARTH Magazine, American Geosciences Institute. earthmagazine.org
2. NOAA. (n.d.). Layers of the ocean. JetStream. noaa.gov
3. Martini, S., and Haddock, S. H. D. (2017). Quantification of bioluminescence from the surface to the deep sea demonstrates its predominance as an ecological trait. Scientific Reports, 7, 45750. Summarized in MBARI. (2017, April 10). New study shows that three quarters of deep-sea animals make their own light. mbari.org
4. NOAA Ocean Exploration. (n.d.). What is vertical migration of zooplankton and why does it matter? oceanexplorer.noaa.gov
5. NOAA National Ocean Service. (2026). What is a hydrothermal vent? oceanservice.noaa.gov
6. Webb, P. (2019). Introduction to oceanography. Roger Williams University Open Publishing; Geosciences LibreTexts edition, CC BY 4.0. Sections 4.11 (hydrothermal vents) and 12.4 (manganese nodules). geo.libretexts.org
7. International Seabed Authority. (2026). The Mining Code: Draft exploitation regulations. isa.org.jm
8. NOAA Ocean Exploration. (2025). Benthic Deepwater Animal Identification Guide, version 4. oceanexplorer.noaa.gov

A drink that would kill you

People adrift at sea without fresh water are in danger of severe dehydration if they drink seawater. The OpenStax biology text explains why: seawater is saltier than human body fluids, and the body cannot adapt to drinking it. Yet an Atlantic cod drinks seawater every day of its life, and thrives. The cod has the same problem you would have. Its body fluids are less salty than the sea around it. So how does it survive?

That question, and two others like it, how to breathe water and how not to sink, explain most of the design of a fish. This lesson takes them one at a time, then asks a question that matters for Lesson 21: which fish can survive being fished?

The problem: water leaks out

Water moves across a membrane by osmosis, from the side with fewer dissolved particles to the side with more. Put a cell in very salty water, a hypertonic environment, and it loses water and shrinks; put it in fresh water, a hypotonic environment, and it gains water and swells. A fish's gills are thin, wet surfaces with a huge area, built to let oxygen through. They also let water through.

For a bony fish in the sea, the sea is hypertonic: water leaks out of its body all the time. For a fish in a river, the water is hypotonic: water floods in. Either way the fish must work constantly to stay in balance, and the solutions are opposite.

Three solutions compared

The marine fish's answer, then, is not to stop drinking but to drink and desalinate: special cells in its gills pump salt out against the gradient, which costs energy every moment. Some fish, called euryhaline, can switch between the two strategies when they move between fresh and salt water.

The shark's answer is cleverer. Instead of pumping salt, it makes its blood as concentrated as seawater by holding onto urea, a waste product most animals get rid of. Urea at that concentration would damage proteins, so sharks also keep TMAO in their blood, which protects proteins from the urea. You met TMAO in Lesson 4, protecting deep-sea fish from pressure: the same molecule does two jobs.

The point: Every fish is in a constant contest with osmosis. Marine bony fish drink and pump out salt, freshwater fish flush out water, and sharks match the sea by keeping urea and TMAO in their blood.

Breathing water

The air is about 21 percent oxygen; water holds far less, about 8 milligrams per litre even when well oxygenated, as Lesson 4 showed. A fish must move a lot of water past a large surface to get enough. Its gills are thin, highly branched and folded filaments, and in bony fish a muscular flap, the operculum, keeps water flowing over them. Oxygen diffuses from the water, where it is more concentrated, into the blood, where it is less.

The arrangement of the flow makes a large difference. In a fish gill, blood runs through the gill filaments in the opposite direction to the water flowing over them. This countercurrent exchange is a highly efficient way to take up oxygen. Here is why, as a thought experiment with simple numbers. Suppose water arrives carrying 100 units of oxygen and blood arrives with none.

• If both flowed the same way, blood would pick up oxygen until it matched the water beside it. They would meet in the middle, near 50 units each, and the transfer would stop.
• Flowing in opposite directions, blood that has already picked up oxygen keeps meeting water that is fresher still. All along the gill, the water beside the blood holds a little more oxygen than the blood does, so oxygen keeps moving across, and the blood can leave carrying nearly as much as the incoming water.

Staying afloat

Many bony fish have a swim bladder, a gas-filled organ that grew as a pouch from the gut. By adding or removing gas, exchanged with the blood, the fish adjusts its buoyancy and can hover at a depth without effort. The swim bladder is thought to be the same organ, in evolutionary terms, as the lungs of lungfish and of land animals. Remember Lesson 4, though: gas changes volume with pressure. A swim bladder is a problem for a fish that moves through large changes of depth, and the deepest fishes, such as the snailfish, have none.

Sharks have no swim bladder at all. Instead, OpenStax notes, a shark has a very large liver full of squalene oil, which is lighter than water (a specific gravity of 0.855), so it helps the shark float. It is not enough on its own: a shark has no way to hold itself at neutral buoyancy and must swim to stay up. Some sharks also need to swim to keep water flowing over their gills, while others pump water with muscles in the mouth, so the popular claim that every shark must swim or die is true for some species only.

Sensing the water

Most fishes have a lateral line, a row of sensory structures you can often see as a dark stripe along each side of the body. It detects movement and vibration in the surrounding water, a sense often compared to hearing, so a fish can feel what moves near it even in dark or murky water. Sharks add another sense, electroreception, which Lesson 15 explores.

Who the fishes are

Fishes are the most diverse group of vertebrates, with more than 30,000 species. OpenStax sorts the living ones into three broad groups:

Which fish can withstand fishing?

How fast a fish population can rebuild after heavy fishing depends on its life history: how old the fish are when they first breed, how many young they produce and how long they live. FishBase, a free global database of fishes, summarizes this as a resilience rating, based on how quickly a population could double, and a fishing vulnerability score from 0 to 100. Here are six North Atlantic species as FishBase listed them in 2026:

Read the extremes. A herring that breeds at 2 to 5 years old and spawns tens of thousands of eggs can replace losses within a few years. A spiny dogfish may not breed until it is 10 to 30 years old, carries its young for up to two years, the longest pregnancy known in any shark or ray, and has a handful of pups. Fished hard, such a population can take decades to recover. Big size does not protect a species, either: the bluefin tuna and the white shark both rate as very highly vulnerable. Lesson 15 follows what this has meant for sharks.

Common misconceptions

• "Fish do not drink." Marine bony fish drink seawater all the time and pump out the salt; it is freshwater fish that barely drink.
• "All fish have swim bladders." Sharks and rays have none, and neither do many deep-sea fishes; sharks rely on an oily liver and swimming.
• "Every shark must keep swimming or it dies." Some sharks must swim to breathe, but others pump water over their gills with muscles in the mouth.
• "Fish that produce millions of eggs can never be overfished." Atlantic cod scores high to very high vulnerability in FishBase despite producing very many eggs; Lesson 21 shows what happened to northern cod.

Pulling it together

• Osmosis pulls water out of marine bony fish and into freshwater fish. Marine fish drink seawater and excrete salt through gills and urine; freshwater fish pass dilute urine and take up salt; sharks keep urea and TMAO in their blood and secrete salt from a rectal gland.
• Gills are thin, folded filaments; countercurrent flow, with blood moving against the water, keeps oxygen moving into the blood along the whole gill.
• Bony fish adjust buoyancy with a gas-filled swim bladder; sharks use an oily liver and must swim to stay up.
• The lateral line senses movement and vibration in the water.
• Fish that mature late, live long and have few young, such as many sharks, have very low resilience to fishing.

Sources

1. OpenStax. (2018). Biology 2e, Section 41.1: Osmoregulation and osmotic balance. Rice University, CC BY 4.0. openstax.org
2. OpenStax. (2018). Biology 2e, Section 29.2: Fishes, and Section 39.1: Systems of gas exchange. Rice University, CC BY 4.0. openstax.org
3. Hou, J.-B., Hughes, N. C., Hopkins, M. J., and Shu, D. (2023). Gill function in an early arthropod and the widespread adoption of the countercurrent exchange mechanism. Royal Society Open Science, 10, 230341.
4. Froese, R., and Pauly, D. (Eds.). (2026). FishBase: species summaries for Clupea harengus, Gadus morhua, Thunnus thynnus, Carcharodon carcharias, Squalus acanthias and Somniosus microcephalus, accessed 2 November 2026. fishbase.se

Sixty-five bites and seventy-one percent

Every year the International Shark Attack File, kept at the Florida Museum of Natural History, investigates reports of sharks biting people. In 2025 it confirmed 65 unprovoked bites worldwide, in line with its five-year average of 61, and nine of them were fatal. Four years earlier, in 2021, a team led by Nathan Pacoureau published an estimate in Nature of what has happened to sharks and rays of the open ocean: since 1970, their global abundance had fallen by 71 percent.^[2]

Many people carry a picture of sharks that fits neither number: sharks as plentiful, dangerous man-eaters, so that fewer sharks means a safer sea at no real cost. This lesson debugs that picture, one link at a time, and replaces it with what the evidence shows.

What a shark is

Sharks, rays and skates are cartilaginous fishes. OpenStax lists their features: a skeleton of cartilage rather than bone, skin covered in tooth-like placoid scales (shark teeth probably evolved from rows of these scales), exposed gill slits, no swim bladder, and internal fertilization. There are more than 500 species of rays and skates, flattened relatives of sharks with gill slits on their undersides. Most sharks are carnivores, but the largest, the whale shark, filters plankton.

Sharks also have formidable senses. OpenStax describes an excellent sense of smell, good vision, a lateral line for vibrations (Lesson 14), and electroreceptors called ampullae of Lorenzini: pores in the snout that detect the weak electric fields produced by all living animals, including prey buried in sand. Sharks may have the most sensitive electroreceptors of any animal. A shark homing in on a hidden flatfish is not seeing it; it is reading the electricity of its heartbeat and muscles.

Link 1: "Sharks are a major danger to people"

The International Shark Attack File's numbers answer this directly. In 2025 there were 65 confirmed unprovoked bites in a world of eight billion people, many millions of whom swim, surf and dive. The United States had the most, 25, with one death. Australia had 21, with five deaths, three of them surfers bitten by white sharks. The Shark Attack File notes that year-to-year numbers fluctuate with ocean conditions and with how many people are in the water.

Compare a hazard from Lesson 6. NOAA estimates that rip currents kill about 100 people a year in the United States alone. In 2025, U.S. shark bites killed one. Shark bites are real, frightening and sometimes fatal, but they are rare, and the risk depends on place, season and activity.

The link fails: sharks are a small hazard to people.

Link 2: "Sharks are plentiful"

Pacoureau and colleagues combined abundance records for the 31 species of oceanic sharks and rays, those living in the open sea rather than near coasts. They found a 71 percent decline in global abundance between 1970 and 2018, driven by an 18-fold increase in relative fishing pressure, and three quarters of those species are threatened with extinction. The cause is fishing. A separate global assessment published the same year by Nicholas Dulvy and colleagues found that 37 percent of all sharks, rays and chimaeras are threatened, with overfishing a threat to every one of the threatened species.^[3]

The link fails: people catch vastly more sharks than sharks bite people.

What matters here: The danger runs mostly the other way. Shark bites on people number in the dozens a year; overfishing has cut oceanic sharks and rays by 71 percent since 1970.

Link 3: "Sharks can bounce back like other fish"

Lesson 14's FishBase table already showed why this fails. Many sharks are built for long, slow lives. A white shark first breeds at about 8 to 12 years and has a handful of pups. A spiny dogfish may not breed until 10 to 30 years old and carries its young for up to two years. The extreme case is the Greenland shark. In 2016 Julius Nielsen and colleagues dated the lenses of the eyes of 28 Greenland sharks, which form before birth, using radiocarbon. The largest, 5 m long, was estimated at 392 years old, give or take 120, and females seemed not to mature until they were about 156 years old. Greenland sharks are the longest-lived vertebrates known.

A population that takes decades to replace one breeding adult cannot absorb heavy fishing. Remove half its adults and it may take decades to recover, even if fishing stops.

The corrected picture, and what helps

Put the three links right and the picture changes: sharks rarely harm people, people have removed most of the open ocean's sharks, and those sharks recover slowly. What has helped?

• Catch limits, enforced over decades. Off the U.S. East Coast, heavy fishing from the late 1970s through the 1980s cut many coastal shark populations by 60 to 99 percent. In 1993 NOAA introduced a shark fishery management plan that limited commercial and recreational landings. When Cassidy Peterson and colleagues analysed six independent research surveys from 1975 to 2014, six of seven coastal shark species were increasing again. The exception, the Gulf of Mexico blacknose shark, may still be held down by bycatch in shrimp trawls.
• Fin rules. Taking only a shark's fins and discarding the body, called finning, wastes most of the animal. The U.S. Shark Fin Sales Elimination Act, in effect since 23 December 2022, prohibits possessing, transporting or selling shark fins in the United States, with narrow exceptions.
• Trade controls. In November 2022, governments meeting under CITES, the treaty that regulates international trade in wildlife, voted to list the whole requiem shark family, 54 species, on Appendix II, along with the remaining hammerhead sharks. Appendix II does not ban trade; it requires exporting countries to issue permits only when trade will not harm the species' survival.

The evidence is strongest for the first measure, because it can be measured in the water: population surveys show what limits achieve. Fin laws and trade listings are newer, and how well they work depends on enforcement, which is still being studied.

Common misconceptions

• "Sharks hunt people." People are not a normal part of any shark's diet, and unprovoked bites number a few dozen a year worldwide among billions of people.
• "There are plenty of sharks." Oceanic sharks and rays fell 71 percent from 1970 to 2018, and three quarters of oceanic species are threatened.
• "Sharks are a kind of mammal, or not really fish." They are cartilaginous fishes, breathing through gills.
• "A big predator can survive anything." Size does not protect a species with few young and late maturity; it makes it more vulnerable.

Where this leaves us

• Sharks and rays are cartilaginous fishes with placoid scales and no swim bladder; their senses include smell, vision, the lateral line and electroreception through the ampullae of Lorenzini.
• In 2025 the International Shark Attack File confirmed 65 unprovoked bites worldwide, nine fatal; rip currents kill far more people in the United States alone.
• Oceanic sharks and rays declined 71 percent from 1970 to 2018 under an 18-fold rise in relative fishing pressure (Pacoureau and colleagues, 2021).
• Late maturity and few young, extreme in the Greenland shark (maturity around 156 years), make shark populations slow to recover.
• Catch limits since 1993 have let six of seven U.S. coastal shark species studied begin to recover; fin sale bans and CITES trade listings add further protection.

Sources

1. Florida Museum of Natural History. (2026). International Shark Attack File: Yearly worldwide shark attack summary, 2025. floridamuseum.ufl.edu
2. Pacoureau, N., Rigby, C. L., Kyne, P. M., and others. (2021). Half a century of global decline in oceanic sharks and rays. Nature, 589, 567-571. cms.int
3. Dulvy, N. K., Pacoureau, N., Rigby, C. L., and others. (2021). Overfishing drives over one-third of all sharks and rays toward a global extinction crisis. Current Biology, 31, 4773-4787.
4. OpenStax. (2018). Biology 2e, Section 29.2: Fishes (Class Chondrichthyes). Rice University, CC BY 4.0. openstax.org
5. Nielsen, J., Hedeholm, R. B., Heinemeier, J., and others. (2016). Eye lens radiocarbon reveals centuries of longevity in the Greenland shark (Somniosus microcephalus). Science, 353(6300), 702-704.
6. Peterson, C. D., Belcher, C. N., Bethea, D. M., Driggers, W. B., Frazier, B. S., and Latour, R. J. (2017). Preliminary recovery of coastal sharks in the south-east United States. Fish and Fisheries, 18(5), 845-859. Summarized by the Virginia Institute of Marine Science. vims.edu
7. NOAA Fisheries. (n.d.). Frequently asked questions: Shark Fin Sales Elimination Act. fisheries.noaa.gov
8. CITES. (2022). Amendments to Appendices I and II adopted at the 19th meeting of the Conference of the Parties, Panama City, 14-25 November 2022 (requiem sharks, Carcharhinidae, and hammerhead sharks, Sphyrnidae).

A day on the beach in 1947

In 1947 someone with a movie camera filmed a beach near Rancho Nuevo, in the Mexican state of Tamaulipas. The film shows Kemp's ridley sea turtles coming ashore to nest, not by ones and twos but in their tens of thousands, on a single day. NOAA Fisheries calls it an amateur video. It is also the best record anyone has of what the species once was.

Within forty years, almost all of them were gone. This lesson follows the Kemp's ridley from that beach to the brink and partway back. Along the way it explains how sea turtles live, how they find their way across oceans, and why the temperature of a nest of sand has become a conservation problem.

The smallest sea turtle

There are seven species of sea turtle: the green, loggerhead, hawksbill, leatherback, flatback, olive ridley and Kemp's ridley. The Kemp's ridley is the smallest. NOAA describes it as living mainly in the Gulf, with young turtles ranging up the Atlantic coast as far as Nova Scotia. Almost all of the world's Kemp's ridleys, about 95 percent, nest in Tamaulipas.

Kemp's ridleys share a remarkable habit with their close relative the olive ridley. Females gather offshore and then come ashore together in huge groups called arribadas, Spanish for "arrival," and unlike other sea turtles, Kemp's ridleys nest in daylight. Scientists still do not know exactly what triggers an arribada; offshore winds, the cycles of the Moon and chemical signals between females have all been suggested. Nesting together may overwhelm predators, which cannot eat every egg.

The habit had a cost. Arribadas concentrate nearly the whole species on a few beaches at once, which made it easy for people to take enormous numbers of eggs. Offshore, adult turtles drowned in shrimp trawls. NOAA reports that nests fell to a record low of 702 in 1985, representing fewer than 250 nesting females.

A long life on a slow clock

Why could the species not simply bounce back once the pressure eased? Kemp's ridleys reach maturity at about 13 years old, according to NOAA, and live at least 30. Sea turtles in general lay many eggs, but few hatchlings survive to adulthood, and every adult lost takes many years to replace. As with the sharks of Lesson 15, a slow life means a slow recovery.

Finding home by magnetism

NOAA notes that adult females usually return to nest in the general area where they hatched, decades earlier. How do they find it after years at sea? In 2015 J. Roger Brothers and Kenneth Lohmann of the University of North Carolina tested an idea Lohmann had proposed. Perhaps hatchlings imprint on the magnetic field of their home beach and later seek it out. Earth's magnetic field drifts slowly, so the magnetic "address" of each stretch of coast changes over the years. If turtles follow the magnetic address, their nesting should drift with it.

The researchers analysed 19 years of loggerhead nesting records from Florida's Atlantic coast. Nesting shifted just as the imprinting idea predicted. Where the magnetic signatures of neighbouring beaches drew closer together, nests crowded together; where they moved apart, nests spread out. The team describes this as strong evidence that turtles navigate at least partly by magnetic signatures. They also note that turtles probably use other cues as well.

Why this matters: A turtle's nesting beach is not just any beach. Females return to the region where they hatched, so losing a nesting beach can mean losing a population, and protecting one protects turtles that may be thousands of kilometres away for decades.

Sex set by sand

In sea turtles, there are no sex chromosomes deciding male or female. The temperature of the sand during incubation decides: warmer nests produce more females, cooler nests more males. That system worked for millions of years of changing climates, but it makes turtles sensitive to how fast beaches are warming now.

In 2018 Michael Jensen of NOAA and colleagues used hormone tests and genetics to identify the sex and birthplace of green turtles feeding on the Great Barrier Reef, home to one of the largest green turtle populations in the world, with an estimated 200,000 females nesting in the northern part. Turtles from the cooler southern beaches were about 65 to 69 percent female. Turtles from the warmer northern beaches were 86.8 percent female among adults, 99.8 percent among subadults and 99.1 percent among juveniles. In other words, the northern beaches had been producing almost no males for about two decades. The authors warned that such a population might not sustain itself, and that managers may need ways to lower incubation temperatures at key nesting beaches.

Comeback, and a stall

Back on the Gulf coast, the response to the Kemp's ridley crisis came on two fronts. On the beaches, nests were protected from collecting and predators, and Mexico and the United States worked together on the main nesting grounds. At sea, NOAA worked with the shrimp industry on turtle excluder devices, or TEDs. A TED is a grid fitted into a shrimp trawl: shrimp pass through into the back of the net, while turtles slide out through an escape opening. TEDs have been required in U.S. shrimp trawls since 1987. Since 1989 the United States has also banned imports of shrimp caught in ways that harm sea turtles, unless the exporting country has comparable protections.

Read the last row carefully. A recovery is not a straight line. NOAA calls the stall unexpected and says it shows why continued protection and monitoring matter. The Kemp's ridley is still listed as endangered.

The light problem

Hatchlings face one more modern hazard. Hatchlings usually emerge at night. The Florida Fish and Wildlife Conservation Commission explains that they have an inborn tendency to crawl toward the brightest direction, which on a natural beach is the open sky over the sea. Artificial lights behind the beach pull them the wrong way, onto roads and into dunes. FWC reports that this disorientation kills thousands of hatchlings each year in Florida alone. The fix is simple: turn off unneeded lights. Where light is needed, keep it low, use long-wavelength amber or red light, and shield it so the bulb cannot be seen from the beach.

Common misconceptions

• "Sea turtles are male or female from the moment the egg is laid." Sand temperature during incubation decides; warmer nests produce more females.
• "Hatchlings follow the Moon to the sea." They crawl toward the brightest direction; on a dark natural beach that is the sky over the water. FWC also notes that it is a myth that hatchlings emerge only around the full moon.
• "Once a species starts recovering, the job is done." Kemp's ridley nests rose about 15 percent a year until 2009, then stopped rising for reasons not yet understood.
• "It is kind to help hatchlings to the water." Sea turtles are protected animals. The guidance for visitors is to keep your distance, keep lights off and leave hatchlings to make their own way.

Looking back

• A 1947 film showed tens of thousands of Kemp's ridleys nesting near Rancho Nuevo in one day; by 1985 there were only 702 nests.
• Kemp's ridleys nest in daytime arribadas, mature at about 13 years and live at least 30, so losses are slow to replace.
• Loggerheads appear to find their natal coast partly by its magnetic signature (Brothers and Lohmann, 2015).
• Sand temperature sets hatchling sex; northern Great Barrier Reef green turtles hatched in recent decades are about 99 percent female (Jensen and colleagues, 2018).
• Nest protection, binational cooperation and turtle excluder devices, required since 1987, brought Kemp's ridley nests up about 15 percent a year until 2009; since 2010 they have fluctuated.
• Artificial light kills thousands of hatchlings a year in Florida; low, long-wavelength, shielded lighting helps.

Sources

1. NOAA Fisheries. (2026). Kemp's ridley turtle [Species page]. fisheries.noaa.gov
2. NOAA Fisheries. (n.d.). Turtle excluder device regulations. fisheries.noaa.gov
3. Brothers, J. R., and Lohmann, K. J. (2015). Evidence for geomagnetic imprinting and magnetic navigation in the natal homing of sea turtles. Current Biology, 25, 392-396. Summarized in University of North Carolina. (2015, January 15). For sea turtles, there's no place like magnetic home. unc.edu
4. Jensen, M. P., Allen, C. D., Eguchi, T., and others. (2018). Environmental warming and feminization of one of the largest sea turtle populations in the world. Current Biology, 28, 154-159. Summarized in NOAA Fisheries West Coast Region. (2018, January 11). Rising temperatures turning major sea turtle population female. ScienceDaily. sciencedaily.com
5. Florida Fish and Wildlife Conservation Commission. (n.d.). Artificial lighting and sea turtle hatchling behavior. myfwc.com
6. NOAA Fisheries. (n.d.). Marine life viewing guidelines. fisheries.noaa.gov

An albatross named Wisdom

In 1956 a scientist put a numbered metal band on the leg of a Laysan albatross nesting on Midway Atoll in the North Pacific. She was already an adult. Seventy years later she was still coming back. In April 2026 the U.S. Fish and Wildlife Service described Wisdom as the world's oldest known wild bird, at an approximate age of 75, and estimated that she had laid 50 to 60 eggs and fledged about 30 chicks in her lifetime.

Wisdom's story holds two facts that shape everything about seabirds: they can live a very long time, and they raise very few young each year. This lesson compares the main kinds of seabird, then uses those two facts to explain why seabirds are in trouble and what has been shown to help.

Four ways to make a living at sea

Seabirds are birds that get their food from the ocean. They split into groups with very different designs. This comparison uses NOAA's guide to the seabirds of Cordell Bank National Marine Sanctuary, off California, and other sources named below.

The tube-nosed birds are the long-haul specialists. NOAA lists wingspans of 6 to 7 feet for black-footed albatrosses and 6.5 to 7.5 feet for Laysan albatrosses, and the Fish and Wildlife Service notes that adult Laysan albatrosses can fly up to 50,000 miles a year. The skuas' piracy has a name, kleptoparasitism: a south polar skua harasses another bird in flight until it gives up its meal, then snatches the food in mid-air.

Drinking the sea

A seabird far from land has no fresh water to drink, and its prey is as salty as the sea. Lesson 14 showed how fish handle salt; seabirds have their own solution. NOAA describes the salt glands of black-footed albatrosses. These specialized glands above the eyes pull excess salt from the blood and make a concentrated salty solution. It runs out through the nostrils and down grooves in the beak, and drips off the tip. A seabird can therefore drink seawater and get rid of the salt, something a human body cannot do.

Reading the table: slow lives

Look down the breeding column. Most seabirds lay one egg a season, a few lay two, and many take several years to start breeding at all. They make up for it by living a long time, as Wisdom shows. That is a fine strategy when adults rarely die. It works badly when something starts killing adults faster than before, because each lost adult takes many seasons to replace. This is the same lesson that Lesson 15 drew for sharks.

In short: Seabirds live long and breed slowly. Anything that raises the death rate of adults, even a little, can drive a population down for decades.

The evidence for decline

In 2015 Michelle Paleczny and colleagues at the University of British Columbia gathered every seabird population count they could find. The populations with enough records to follow made up about 19 percent of the world's seabirds. Together, that monitored portion declined by 69.7 percent between 1950 and 2010. The largest declines were in families of wide-ranging ocean birds, such as albatrosses and petrels. Because the sample was large and broadly representative, the authors suggested the global trend may be similar.

Three threats, and what helps

1. Hooks. Longline fishing boats set lines with thousands of baited hooks. Albatrosses and petrels dive at the bait as the line goes out, get hooked, and drown. The Agreement on the Conservation of Albatrosses and Petrels (ACAP) reviews the evidence. In its 2026 best practice advice for pelagic longlines, it recommends three measures used at once. Weight the branch lines, so baited hooks sink quickly out of the birds' reach. Set lines at night, when most vulnerable seabirds are inactive. And tow bird-scaring lines, streamers that keep birds away from the sinking hooks. Each works alone but has gaps: night setting fails for night-feeding birds and under bright moonlight, and scaring lines protect hooks only as far as the line reaches. Used together, ACAP says, they cover each other's weaknesses.

2. Introduced predators. Seabirds nest in dense colonies on islands that, until people arrived, had no ground predators, so the birds have no defences against them. Rats, cats and mice arrived on ships. Midway Atoll holds the world's largest albatross colony, including more than 70 percent of all Laysan albatrosses. In December 2015 volunteers there found that house mice had begun attacking nesting adults. The solution that works is removing the predators completely. South Georgia, in the South Atlantic, was declared free of rats and mice on 18 May 2018. It was the largest island rodent eradication ever attempted, led by the South Georgia Heritage Trust. Native birds such as the South Georgia pipit soon returned to areas the rats had kept them from.

3. Plastic and oil. Albatrosses feed their chicks on what they pick up at sea, and NOAA lists plastic ingestion among the threats to Laysan and black-footed albatrosses; Lesson 22 deals with plastic in detail. Diving birds face oil: NOAA notes that the common murre, which spends so much of its time on and under the water, is often the species most affected by oil spills.

Common misconceptions

• "Seabirds cannot drink seawater." Many do, and get rid of the extra salt through salt glands that drain through the nostrils.
• "Seabirds are doing fine; there are gulls everywhere." Monitored seabird populations fell about 70 percent from 1950 to 2010, with the steepest losses in ocean wanderers such as albatrosses and petrels.
• "A bird that lays one egg a year must be failing." One egg a year is normal for many seabirds; it works because adults live for decades, which is exactly why extra adult deaths are so damaging.
• "Bycatch is an unsolvable cost of fishing." ACAP's combination of weighted lines, night setting and bird-scaring lines is designed to cut seabird bycatch while fishing continues.

Summing up

• Seabirds include tube-nosed birds (albatrosses, petrels, shearwaters), diving auks (murres, puffins), penguins and predatory skuas and jaegers, each built differently for feeding at sea.
• Salt glands above the eyes let seabirds drink seawater and drain the extra salt out through the nostrils.
• Most seabirds lay one or two eggs a season and live long; Wisdom, banded in 1956, was about 75 in 2026.
• The world's monitored seabirds declined 69.7 percent from 1950 to 2010 (Paleczny and colleagues, 2015).
• Main threats: longline hooks, introduced predators on nesting islands, plastic and oil. ACAP recommends weighted lines, night setting and bird-scaring lines together; South Georgia's rodent eradication (2018) shows islands can be restored.

Sources

1. U.S. Fish and Wildlife Service. (2026, April). Midway Atoll National Wildlife Refuge: 'A great monument to the people who passed here'. fws.gov
2. NOAA Cordell Bank National Marine Sanctuary. (n.d.). Tube-nosed seabirds, Diving seabirds and Predatory seabirds. cordellbank.noaa.gov
3. Paleczny, M., Hammill, E., Karpouzi, V., and Pauly, D. (2015). Population trend of the world's monitored seabirds, 1950-2010. PLOS ONE, 10(6), e0129342. journals.plos.org
4. Agreement on the Conservation of Albatrosses and Petrels. (2026). ACAP summary best practice advice for reducing the impact of pelagic longline fisheries on seabirds. acap.aq
5. U.S. Fish and Wildlife Service. (2019, November 27). World's oldest known wild bird returns to Midway Atoll in time for Thanksgiving. fws.gov
6. U.S. Fish and Wildlife Service. (2017, November). Midway's albatross: A new threat puts the world's largest colony at risk. fws.gov
7. Government of South Georgia and the South Sandwich Islands. (n.d.). Eradication projects: Rats and mice. gov.gs

Three hours and forty-two minutes

You breathe every few seconds, all day, without thinking about it. A Cuvier's beaked whale is a mammal with lungs like yours, and it must come to the surface to breathe. Yet it hunts far below, catching squid in total darkness. In 2014 Gregory Schorr and colleagues published the first long-term records from satellite-linked tags on eight of these whales off southern California.^[1] One dive reached 2,992 m, nearly 3 km down. Another lasted 137.5 minutes. Both were new records for any mammal.

The records did not last. A team based at Duke University, led by Nicola Quick, analysed 3,680 dives by 23 tagged Cuvier's beaked whales and published the results in 2020. The median foraging dive lasted 59 minutes, and 5 percent of dives lasted longer than 77.7 minutes. One whale made two dives of 173 and 222 minutes. The longer one is three hours and forty-two minutes on a single breath.

That is the problem for this lesson. How can an air-breathing mammal stay down that long, at depths where the pressure is close to 300 times the pressure at the surface (Lesson 4), and come back unharmed? Before solving it, meet the animals that face it.

Who the marine mammals are

Marine mammals are not one family. The U.S. Marine Mammal Protection Act of 1972 defines them in two ways. Some are built for life in the sea: whales, dolphins and porpoises (the cetaceans), seals, sea lions and walruses (the pinnipeds), manatees (the sirenians) and sea otters. Others simply live mainly in the marine environment, such as the polar bear. Like every mammal, all of them breathe air and feed their young on milk.

Sea lions and seals are easy to confuse. NOAA's rule of thumb: sea lions have small ear flaps, bark loudly and "walk" on land on their large flippers, while seals have no visible ear flaps, are quieter and wriggle along on their bellies. The baleen whales, the giants of the group, eat some of the smallest animals in the sea. A North Atlantic right whale, up to 52 feet (16 m) long, feeds on copepods, tiny crustaceans, by straining huge volumes of water through its baleen plates, which work like a sieve.

Solving the dive, part 1: keep the oxygen where it stays useful

A natural guess is that a deep diver fills enormous lungs with air. It does nearly the opposite. The Natural History Museum in London explains that whales store oxygen in their blood and muscles, not in their lungs as we do. Their blood is packed with haemoglobin, the protein that carries oxygen in red blood cells. Their muscles are packed with myoglobin, a related protein that holds oxygen inside muscle. There is so much that whale muscle is a very dark red, almost black. High School Anatomy & Physiology (hs-anatomy) compares red, myoglobin-rich muscle fibres with pale ones in your own body; a deep-diving whale takes the red kind to an extreme.

Why not the lungs? Because of pressure. Lesson 4 showed that pressure squeezes gas, and by about 200 m down a whale's lungs, like a human's, have been squeezed flat. Whales come to no harm, because their ribs are joined by flexible cartilage, so the chest folds and the lungs collapse without tearing. But once the lungs have collapsed, no more gas passes from them into the blood. Oxygen kept in the lungs would be locked away for the deep part of the dive. Oxygen held in blood and muscle stays available all the way down.

Part 2: spend the oxygen slowly

Storing oxygen is half the answer. The other half is using it slowly. Diving mammals slow their heart rate and stop the blood flow to some parts of the body, temporarily shutting down organs such as the kidneys and liver while they hunt. Scientists call this set of changes the dive response. Beaked whales also save effort through their shape. Their flippers tuck into pockets along the body, giving them a torpedo shape, and they often glide rather than swim.

Exactly how the longest dives are possible is still being worked out. When muscle runs short of oxygen, it can keep working for a while without it, but acid builds up. Quick and colleagues suggest that Cuvier's beaked whales combine a low metabolic rate, a large store of oxygen and a high capacity to buffer that acid. Together, they argue, these let the whales stretch their dives.

Part 3: avoid the bends

Pressure brings a second danger. Before the lungs collapse, the squeezed air forces extra nitrogen to dissolve into the blood and tissues. If the pressure then falls on the way up, that nitrogen can come out as bubbles, like the fizz when you open a bottle of soda. In human divers this is decompression sickness, "the bends", and the bubbles can kill. Scientists do not yet fully understand how whales avoid it. One idea is that their lungs collapse in a way that pushes air away from the alveoli, the tiny air sacs where gases pass into the blood. If so, little nitrogen would dissolve at depth, and collapsing early would protect the whale instead of harming it.

The upshot: A deep-diving whale carries its oxygen in its blood and muscles, not its lungs. It makes that oxygen last by slowing its heart, shutting down organs that can wait and gliding, and its lungs collapse early, which may also protect it from the bends.

Hunting by echo

No sunlight reaches these depths (Lesson 13), so toothed whales hunt by sound. NOAA describes how sperm whales do it. On a deep dive a sperm whale makes clicks and listens for the echoes that bounce off objects. The longer an echo takes to return, the farther away the object is. This is echolocation, the same idea as the ship's echo sounder in Lesson 1. As the whale closes in on prey, the echoes come back sooner, so it clicks faster and faster until the clicks run together into a "buzz" or "creak". Tagged sperm whales in the Gulf of Mexico typically made about 17 attempts to catch prey on each deep dive, counted from their buzzes.

A sperm whale's huge head holds spermaceti, a waxy oil in a sac that helps focus the sound, and NOAA describes its clicks as the loudest animal sound on Earth. Dolphins find prey with high-frequency echolocation, or simply by listening. Sound also carries the social lives of whales. NOAA's recordings show that baleen whales mostly make low sounds, below about 5 kilohertz, while toothed whales mostly make high ones, from about 5 to 150 kilohertz. Both use their calls to keep in contact, to find mates and to compete for food or territory.

Keeping warm

Cold water pulls heat out of a warm body quickly, so most marine mammals are wrapped in blubber, a thick layer of fat under the skin. The sea otter has none. The U.S. Fish and Wildlife Service explains that it relies on fur with more hairs per square inch than any other mammal, and on a metabolism more than double that of a land mammal of the same size. As Lesson 11 told, that fur nearly cost the sea otter its existence. Manatees have a different weakness: they need warm water, and cold snaps are among the natural events that kill them.

Hunted, protected, and some coming back

People hunted whales, seals and sea otters for centuries, for oil, fur and other products. North Atlantic right whales got their name from being the "right" whale to hunt, because they floated when killed. NOAA says whalers had driven them to the brink of extinction by the early 1890s. Sperm whales, whose spermaceti burned in lamps and candles, were a main target of commercial whaling from 1800 to 1987.

Two kinds of protection followed. The International Whaling Commission's moratorium on commercial whaling took effect in the mid-1980s. In the United States, the Marine Mammal Protection Act of 1972 made it illegal to "take" a marine mammal. The law defines that as to harass, hunt, capture or kill one, or to try to, with exceptions such as subsistence hunting by Alaska Natives. Lesson 25 weighs how well such rules have worked. The table compares four populations.

Reading the table: why two are still in trouble

The humpback and the elephant seal came back once the hunting stopped. The right whale and the vaquita show that ending a hunt is not always enough. Both are dying in fishing gear set for other animals, and right whales are also hit by ships.

The North Atlantic right whale lives along the busy east coast of North America, among shipping lanes and fishing gear. NOAA estimates that more than 85 percent of right whales have been entangled in fishing gear at least once. Rope can cut into a whale's body, and an entangled whale can tow gear for months or even years. A healthy female calves about every three years, but females now average one calf every 7 to 10 years, and biologists believe the stress of entanglement is one reason. With only about 70 breeding females, every death matters.

The vaquita is a small porpoise, about 5 feet (1.5 m) long, that lives only in the northern Gulf of California in Mexico: the smallest range of any whale, dolphin or porpoise. NOAA calls it the most endangered marine mammal in the world, and its only known threat is drowning in gillnets. From 1997 to 2008 it fell from about 600 to about 200, mostly in legal nets set for shrimp. Then, from about 2011, illegal fishing for the totoaba, a large fish whose swim bladder sells for high prices in China, drove declines of close to 50 percent a year.

In 2022 the Mexican Navy placed concrete blocks fitted with 3 m metal hooks across the vaquita's core area, designed to discourage gillnets there. The IUCN's cetacean specialists report that the blocks and naval patrols have nearly eliminated gillnets from this sanctuary, and most vaquitas detected in 2025 were inside it. The survey team judged it 67 percent likely that they had seen between 7 and 10 different vaquitas, including at least one calf. None of the 93 sightings since 2017 has looked unhealthy.

That is not a recovery. It does show that the last vaquitas can still breed, and scientists agree that gillnet fishing must end throughout their habitat if they are to survive. The 2025 surveys also found an unprotected area of heavy vaquita use just northwest of the sanctuary, in water that is still heavily fished with gillnets.

A noisy ocean

An animal that finds food and family by sound can be harmed by noise. NOAA notes that underwater noise from ships, sonar and drilling has increased dramatically. Noise can cause temporary or permanent hearing loss and stress, drive animals from places they need, and disrupt feeding, breeding, nursing and calling. Beaked whales seem especially sensitive. Strandings of Cuvier's beaked whales in the Bahamas, the Caribbean, the Canary Islands and the Mediterranean have been associated with active naval sonar, a link scientists still do not fully understand. The Southern Resident killer whales of the Pacific Northwest, which you can listen to in the activity, face boat traffic and noise, contaminants and a scarce supply of their preferred prey, salmon. NOAA's July 2025 census counted only 74 of them.

Watching without harming

Because harassing a marine mammal counts as a "take", getting too close can break the law as well as harm the animal. NOAA's viewing guidelines give these distances:

• Whales: at least 100 yards (about 90 m), the length of a football field. Federal law sets more for some: 500 yards from North Atlantic right whales anywhere in U.S. waters, 100 yards from humpbacks in Hawaii and Alaska, and 200 yards from killer whales in the inland waters of Washington State.
• Dolphins and porpoises: at least 50 yards. Never feed a wild dolphin; it is harmful and illegal.
• Seals and sea lions: at least 50 yards. You are too close if an animal starts to stare, fidget or flee into the water. Keep dogs on a leash.
• Pups: a seal pup alone on a beach usually has a mother nearby. Leave it alone.
• Time and trouble: limit watching to 30 minutes or less, and report sick, injured, entangled or stranded animals to trained responders instead of touching them.

Common misconceptions

• "Whales stay down so long because they fill giant lungs with air." Deep divers store most of their oxygen in blood and muscle; their lungs collapse at depth.
• "Whales and dolphins are a kind of fish." They are mammals: they breathe air through blowholes and feed their calves on milk.
• "Whaling is over, so whales are safe." Commercial whaling no longer threatens most whales, but fishing gear, ships and noise now kill or harm them; the North Atlantic right whale is the clearest case.
• "A seal pup alone on the beach has been abandoned and needs rescuing." Its mother is probably nearby. Keep at least 50 yards away and call trained responders if it looks hurt.

What you now know

• Marine mammals are several separate groups: cetaceans (baleen and toothed whales, dolphins and porpoises), pinnipeds (seals, sea lions and walruses), sirenians (manatees), sea otters and polar bears.
• Cuvier's beaked whales have dived to 2,992 m and stayed down for 222 minutes.
• Deep divers store oxygen in blood (haemoglobin) and muscle (myoglobin), slow the heart and shut down some organs, glide, and let their lungs collapse by about 200 m, which may also protect them from the bends.
• Toothed whales find prey by echolocation, timing the echoes of their clicks; baleen whales sieve small prey through baleen.
• Most marine mammals keep warm with blubber; the sea otter uses the densest fur of any mammal and a high metabolism.
• Humpbacks and elephant seals recovered after hunting stopped. North Atlantic right whales (about 384) and vaquitas (7 to 10 seen in 2025) are still being killed by fishing gear and, for right whales, ships.
• NOAA's distances: at least 100 yards from whales (500 from right whales) and 50 yards from dolphins, seals and sea lions.

Sources

1. Schorr, G. S., Falcone, E. A., Moretti, D. J., and Andrews, R. D. (2014). First long-term behavioral records from Cuvier's beaked whales (Ziphius cavirostris) reveal record-breaking dives. PLOS ONE, 9(3), e92633. journals.plos.org
2. Quick, N. J., Cioffi, W. R., Shearer, J. M., Fahlman, A., and Read, A. J. (2020). Extreme diving in mammals: First estimates of behavioural aerobic dive limits in Cuvier's beaked whales. Journal of Experimental Biology, 223(18), jeb222109. pubmed.ncbi.nlm.nih.gov
3. Natural History Museum, London. (n.d.). Secrets of the deepest diving whales. nhm.ac.uk
4. NOAA Fisheries. (2026, June 12). Diving deep into sperm whale clicks. fisheries.noaa.gov
5. NOAA Fisheries. (n.d.). Sounds in the ocean: Mammals. fisheries.noaa.gov
6. NOAA Fisheries. (n.d.). Species directory pages: Cuvier's beaked whale, sperm whale, humpback whale, killer whale and northern elephant seal.
7. NOAA Fisheries. (n.d.). North Atlantic right whale. fisheries.noaa.gov
8. NOAA Fisheries. (n.d.). Vaquita. fisheries.noaa.gov
9. IUCN SSC Cetacean Specialist Group. (2025, November 16). Joint visual and acoustic survey finds vaquitas surviving and reproducing, mostly in and near the sanctuary. iucn-csg.org
10. Sea Shepherd Conservation Society. (2025, October 28). Results of the 2025 vaquita monitoring effort. seashepherd.org
11. U.S. Fish and Wildlife Service. (2022, July 28). On second chances: The southern sea otter's return to ecological relevance. fws.gov
12. U.S. Fish and Wildlife Service. (n.d.). Species pages: West Indian manatee and polar bear.
13. NOAA National Ocean Service. (n.d.). What's the difference between seals and sea lions? oceanservice.noaa.gov
14. NOAA Fisheries. (n.d.). Ocean noise and Marine life viewing guidelines. fisheries.noaa.gov; fisheries.noaa.gov
15. Marine Mammal Protection Act of 1972, as amended, 16 U.S.C. 1361 et seq. Text published by NOAA Fisheries. fisheries.noaa.gov

A swim in February

On 18 February 2006, members of the Serpentine Swimming Club dived into the lake in London's Hyde Park. Richard Seager, a climate scientist at Columbia University's Lamont-Doherty Earth Observatory, used a photograph of that swim to make a point. At the same latitude on the other side of the Atlantic, he noted, average February temperatures are typically below -10 °C. In January the British Isles are 15 to 20 °C warmer than places at the same latitude in eastern North America.

Ask almost anyone why, and they will name the Gulf Stream, the warm current that Lesson 7 followed out of the Gulf of Mexico and across the Atlantic. The idea goes back at least to 1855. That year the American naval officer Matthew Fontaine Maury wrote in The Physical Geography of the Sea that without the Gulf Stream's heat, the climates of France and England would be like Labrador's: "severe in the extreme, and ice bound". Scientists still argue over how right Maury was. To follow the argument, you first need one property of water.

Why water is slow to warm and slow to cool

Heat capacity is the energy it takes to warm something by one degree. A University of Hawaii textbook gives the values per kilogram: seawater needs about 3,985 joules to warm by 1 °C, dry air about 1,004 joules, and land typically less than 1,000. So a kilogram of seawater stores about four times as much heat as a kilogram of air. NASA's Earth Observatory quotes the oceanographer Josh Willis on the result: put energy into the ocean and "its temperature changes only very slightly".

Worked example: how much ocean matches the whole atmosphere?

1. Lesson 4's rule says pressure rises by one atmosphere for every 10.06 m of seawater. So the entire atmosphere above a patch of sea weighs about as much as a layer of seawater 10 m deep.
2. Per kilogram, seawater stores 3,985 ÷ 1,004, or about 4.0 times as much heat as air.
3. So a layer of ocean only about 10 ÷ 4 = 2.5 m deep can hold as much heat, for each degree of warming, as all the air above it.
4. The ocean averages nearly 3,700 m deep (Lesson 2). As a store of heat, the air is tiny by comparison.

Maritime and continental climates

Because the sea warms and cools so slowly, it stores summer heat and gives it back in winter. Any place where the wind blows off the ocean gets some of that stored warmth in winter and some of the sea's coolness in summer. This is a maritime climate. Places deep inside a continent, or downwind of one, have continental climates, with cold winters and hot summers. NOAA's climate normals, the 30-year averages for 1991 to 2020, show the pattern in four U.S. cities.

Read the first two rows. Eureka and New York sit at almost exactly the same latitude, and both are on the coast. Yet New York's year swings four times as far. The difference is the wind. In the middle latitudes the prevailing winds blow from west to east, so Eureka's air usually arrives off the Pacific, while New York's usually arrives off the land. Seattle and Fargo, farther north, show the same contrast. Notice the timing too. On the West Coast the warmest month is August, not July, because the sea is slow to warm and reaches its warmest late in the summer.

The dispute: what keeps Europe mild?

In 2002 Seager and six colleagues published a study with a blunt title: "Is the Gulf Stream responsible for Europe's mild winters?" Using observations and climate models, they weighed three ways the lands around the North Atlantic get their winter warmth.

1. Seasonal storage. The ocean soaks up heat in summer and releases it in winter, warming whatever lies downwind. This needs no current at all.
2. The atmosphere's heat transport and its waves. Winds carry heat toward the poles. Air flowing east over the Rocky Mountains is forced to swing south and then north again. This giant bend in the westerlies, a stationary wave, pulls cold air down into eastern North America and pushes warm air up into western Europe.
3. The ocean's heat transport. Currents such as the Gulf Stream carry heat north and release it to the air.

Their answer was that at Europe's latitudes the atmosphere carries several times more heat poleward than the ocean does. About half of the winter temperature difference across the North Atlantic comes from seasonal storage, and about half from the stationary waves. The ocean's heat transport warms both sides of the Atlantic by a modest few degrees. In model runs with no ocean heat transport, the subpolar North Atlantic cooled by as much as 8 °C in places, but land typically by less than 3 °C. South of northern Norway, the difference across the ocean stayed the same. The one place where the current fundamentally changes the climate, Seager argues, is the coast of northern Norway, which would be icebound without the warm Norwegian Current.

Seager also points to the Pacific. Seattle and British Columbia have mild, damp winters. Yet the Kuroshio, the Pacific's version of the Gulf Stream, turns east toward Oregon and California, so almost no ocean heat is carried north at Seattle's latitude. Ocean heat transport therefore cannot explain why the Pacific Northwest is so much milder than places at the same latitude in eastern Asia, such as Vladivostok. Look back at the table: Seattle's mild winter needs only the stored heat of the sea and a westerly wind.

The other side

Other scientists think this account leaves too much out. In a 2015 review, Jaime Palter of McGill University gathered evidence from ancient climate records, models and instrument records. She concluded that the Gulf Stream's heat transport "profoundly influences" the climate of the whole Northern Hemisphere, and so of Europe, over decades and longer, acting through feedbacks among the ocean, the air and ice. When the Gulf Stream's heat transport has varied, she found, European temperature, rainfall and storminess have changed with it.

In 2011 Yohai Kaspi and Tapio Schneider of Caltech added a twist. Their model showed that heating of the air over warm ocean currents can itself set up stationary waves that chill the eastern edges of continents. If they are right, warm currents help make the contrast in two ways, by warming western coasts and by cooling eastern ones. Then the "ocean" and "atmosphere" explanations are not as separate as they first look.

Weighing the two sides

The two sides agree on more than the headlines suggest. Both accept that seasonal storage and the atmosphere matter, and that the Gulf Stream carries a great deal of heat. Much of the disagreement comes from asking different questions. Seager asks what sets the average winter contrast across the Atlantic today. Palter asks how changes in the Gulf Stream's heat transport, over decades to thousands of years, change Europe's climate. When you meet a scientific dispute, check first whether both sides are answering the same question.

A slowing circulation?

The second argument is about the future. The Gulf Stream is the surface part of the Atlantic Meridional Overturning Circulation (AMOC), the loop in which warm water flows north near the surface, cools, sinks in the far North Atlantic and returns south at depth (Lesson 7). Since 2004 the RAPID project has measured the AMOC with an array of moored instruments across the Atlantic at 26° N. RAPID reports that the AMOC carries about 1.2 petawatts of heat north (a petawatt is a million billion watts), about a quarter of the poleward heat carried by ocean and atmosphere together. Its record shows a major downturn in 2010 and 2011 and a sustained decline since then.

What happens next is disputed, so it matters who says what.

• The IPCC's assessment (2021). The AMOC is very likely to weaken over the 21st century under every emissions scenario. Confidence in the decline is high, but confidence in how large it will be is low. There is medium confidence that it will not collapse abruptly before 2100. If it did collapse, that would very likely cause abrupt shifts in regional weather, such as a southward shift of the tropical rain belt, weaker African and Asian monsoons and drying in Europe.
• A warning from the data (2023). Peter and Susanne Ditlevsen of the University of Copenhagen looked for statistical early-warning signals in observations. They estimated a collapse around the middle of this century under current emissions.
• A test across 34 models (2025). Jonathan Baker and colleagues at the UK Met Office and the University of Exeter pushed 34 climate models with extreme greenhouse gas levels and floods of fresh water. The AMOC weakened in every case but never collapsed completely, kept going by winds that pull deep water up to the surface in the Southern Ocean. They concluded that a collapse is unlikely this century.

All three expect the AMOC to weaken. The argument is about how far, how fast, and whether it could pass a tipping point, a threshold beyond which it would not easily recover. Each method has limits. Models may leave out processes, and statistical warning signs depend on how well the record they use stands in for the circulation itself. Seager argued in 2006 that a slowdown would bring a cooling tendency of at most a few degrees, likely outweighed by greenhouse warming.^[2] A full collapse, judged unlikely this century but not ruled out, would be a different matter.

The missing heat

Greenhouse gases trap extra energy in the climate system. You might expect most of it to show up as warmer air. It does not. The IPCC found that ocean warming accounted for 91 percent of the heating of the climate system since 1971. Warming of the land took about 5 percent, melting ice about 3 percent and warming the atmosphere only about 1 percent. The worked example shows why: the ocean can absorb vast amounts of heat with only a small rise in temperature.

Measuring that heat takes a global network. NOAA notes that a fleet of more than 3,000 Argo floats, the robots you met in Lesson 3, now measures ocean temperature around the world. Satellites add a check, because warming water expands and raises the sea surface. For 1993 to 2024, NOAA puts the ocean's heat gain at about 0.66 to 0.74 watts for every square metre of Earth's surface. That sounds small, but Earth's surface is about 510 million square kilometres, so it adds up to more than 300 million million watts, day and night.

That heat matters for the rest of this course. The IPCC found that the expansion of warming seawater explains half of the global sea level rise from 1971 to 2018. Warmer water drives the marine heatwaves of Lesson 23 and the coral bleaching of Lesson 12. And heat already stored in the deep ocean will eventually come back out, committing Earth to some further surface warming.

Bottom line: The ocean stores heat so well that its top 2.5 m match the whole atmosphere. That storage, carried downwind, is a large part of why coasts like Seattle's are mild, and the ocean is where 91 percent of the extra heat from greenhouse gases has gone. How much the Gulf Stream itself warms Europe, and how far the AMOC will slow, are still argued over, with evidence on both sides.

Common misconceptions

• "Without the Gulf Stream, Britain would be as cold as Labrador." In model experiments without ocean heat transport, land around the North Atlantic cooled by typically less than 3 °C, far less than the 15 to 20 °C difference across the ocean.
• "A slowing AMOC means a new ice age for Europe." The IPCC expects weakening this century and judges an abrupt collapse before 2100 unlikely, with medium confidence; scientists disagree about the risk, not about an ice age.
• "Global warming is about the air; the ocean is just along for the ride." The ocean has taken up about 91 percent of the extra heat and the atmosphere about 1 percent.
• "Every coastal city has a mild climate." New York is on the coast but has a continental-style year, because its winds usually blow off the land.

The takeaway

• Per kilogram, seawater stores about four times as much heat as air, so the top 2.5 m of ocean holds as much heat per degree as the whole atmosphere above it.
• Places downwind of an ocean have maritime climates, with mild winters, cool summers and a late warmest month; places inland or downwind of land have continental climates. At 40.8° N, Eureka's year swings 6 °C and New York's 24 °C.
• Seager and colleagues (2002) argue that Europe's mild winters come mostly from seasonal heat storage and atmospheric waves set up by the Rockies, with ocean heat transport adding a few degrees. Palter (2015) and Kaspi and Schneider (2011) argue that ocean heat transport matters more, through feedbacks and by shaping the waves.
• The AMOC carries about 1.2 petawatts north at 26° N and has declined since 2010 to 2011 in the RAPID record. The IPCC expects it to weaken this century and judges an abrupt collapse before 2100 unlikely; newer studies disagree about how unlikely.
• The ocean has taken up about 91 percent of the extra heat in the climate system, measured by Argo floats and satellites.

Sources

1. Seager, R., Battisti, D. S., Yin, J., Gordon, N., Naik, N., Clement, A. C., and Cane, M. A. (2002). Is the Gulf Stream responsible for Europe's mild winters? Quarterly Journal of the Royal Meteorological Society, 128(586), 2563-2586. ocp.ldeo.columbia.edu
2. Seager, R. (2006). The source of Europe's mild climate. American Scientist, 94(4), 334-341. ocp.ldeo.columbia.edu
3. Seager, R. (n.d.). The Gulf Stream, European climate and abrupt change. Lamont-Doherty Earth Observatory. ocp.ldeo.columbia.edu
4. Palter, J. B. (2015). The role of the Gulf Stream in European climate. Annual Review of Marine Science, 7, 113-137. pubmed.ncbi.nlm.nih.gov
5. Kaspi, Y., and Schneider, T. (2011). Winter cold of eastern continental boundaries induced by warm ocean waters. Nature, 471(7340), 621-624. pubmed.ncbi.nlm.nih.gov
6. IPCC. (2021). Summary for policymakers. In Climate change 2021: The physical science basis. Contribution of Working Group I to the Sixth Assessment Report. Statements A.4.2, A.4.3 and C.3.4. ipcc.ch
7. Ditlevsen, P., and Ditlevsen, S. (2023). Warning of a forthcoming collapse of the Atlantic meridional overturning circulation. Nature Communications, 14, 4254. pubmed.ncbi.nlm.nih.gov
8. Baker, J. A., Bell, M. J., Jackson, L. C., Vallis, G. K., Watson, A. J., and Wood, R. A. (2025). Continued Atlantic overturning circulation even under climate extremes. Nature, 638, 987-994. pubmed.ncbi.nlm.nih.gov
9. National Oceanography Centre. (n.d.). RAPID 26°N array. rapid.ac.uk
10. Lindsey, R., and Dahlman, L. (2025, June 26). Climate change: Ocean heat content. NOAA Climate.gov. climate.gov
11. NASA Earth Observatory. (2006, April 24). Earth's big heat bucket. science.nasa.gov
12. Nugent, A. D., Dacou, D., and Russell, S. (n.d.). 3.4: Specific heat. In Atmospheric processes and phenomena. University of Hawaii at Manoa, via LibreTexts (CC BY-SA 4.0). geo.libretexts.org
13. NOAA National Centers for Environmental Information. (2021). U.S. climate normals, 1991-2020: monthly normals for Eureka (USW00024213), New York Central Park (USW00094728), Seattle-Tacoma (USW00024233) and Fargo (USW00014914). ncei.noaa.gov

A warm current at Christmas

Fishermen on the coast of South America noticed long ago that every few years the coastal water turned warm around Christmas. They called the warming El Niño, the boy child, after the holiday. In the early 20th century Sir Gilbert Walker found a seesaw of air pressure across the tropical Pacific, the Southern Oscillation, without realising it was linked to the ocean. In the late 1960s the meteorologist Jacob Bjerknes and others saw that the two were one system, which is why scientists now speak of the El Niño-Southern Oscillation, or ENSO.

As this lesson was written, in November 2026, a strong El Niño was under way. On 8 October 2026 NOAA's Climate Prediction Center reported water more than 4 °C warmer than normal in parts of the eastern equatorial Pacific. It said a strong to very strong El Niño was likely to last through early 2027, with a good chance of the strongest event since records began in 1950. This lesson works through what El Niño is, step by step, and then shows you how NOAA decides when one is happening.

Step 1: normal conditions

NOAA's Pacific Marine Environmental Laboratory describes the usual state of the tropical Pacific like this.

• The trade winds blow from east to west along the equator, from South America toward Asia.
• They pile warm surface water up in the west, so the sea surface is about half a metre higher at Indonesia than at Ecuador, and about 8 °C warmer off Asia than in the eastern Pacific.
• The thermocline, here marked by the depth of 20 °C water, slopes from deep in the west to shallow in the east.
• Off South America, cold, nutrient-rich water wells up from below (Lesson 7), feeding plankton, rich food webs and major fisheries (Lesson 9).
• Warm, moist air rises over the warm water near Asia, bringing clouds and rain, while the eastern Pacific stays relatively dry.

Step 2: what changes in El Niño and La Niña

The thermocline row is the key to the sea life. In the 1982-83 El Niño, PMEL notes, the 17 °C layer at 110° W sank to about 150 m. Upwelling could no longer reach the cold, nutrient-rich water below the thermocline, so it cooled the surface far less and brought up few nutrients. Plankton production fell sharply, and so did everything that fed on it, including the commercial fisheries.

Step 3: the Bjerknes feedback

Why does a small warming grow into an event that lasts a year? NOAA's ENSO scientist Michelle L'Heureux explains that the answer is a positive feedback between the tropical Pacific's surface winds and its ocean temperature, now called the Bjerknes feedback.^[7] Ocean and atmosphere push each other along. For El Niño, the loop runs like this:

1. The central and eastern Pacific warm a little.
2. The temperature difference between the warm west and the cooler east shrinks, so the trade winds weaken.
3. Weaker trade winds pile up less water in the west. Warm water spreads back east, and the thermocline in the east deepens.
4. Upwelling now brings up warmer water, so the east warms further, and the loop goes round again.

La Niña runs the same loop the other way: cooling in the east strengthens the trade winds, which push more warm water west and draw up more cold water in the east. Scientists often cannot say what starts the loop; sometimes the ocean changes first, sometimes the winds.

Worth holding on to: El Niño is not a storm or a single current. It is a coupled change in winds and ocean across the whole tropical Pacific, kept going by a feedback in which each makes the other stronger.

Step 4: measure it the way NOAA does

To say whether an El Niño is happening, NOAA follows a fixed procedure.

1. Pick the region. Measure sea surface temperature in the Niño 3.4 region of the equatorial Pacific, from 5° N to 5° S and from 120° to 170° W.
2. Find the anomaly. Subtract the 1991-2020 average for that time of year. An anomaly of +1.0 means one degree warmer than normal.
3. Make it relative. Since February 2026 NOAA's official index has been the Relative Oceanic Niño Index (RONI). It subtracts the average anomaly of the whole tropics, from 20° N to 20° S, then rescales the result to the same size of swings as the original index. The older Oceanic Niño Index (ONI) skips this step. NOAA changed because the whole tropical ocean is warming, and an index tied to a fixed 30-year average struggles to keep up.
4. Smooth it. Average three months at a time: December-January-February, then January-February-March, and so on. These are overlapping seasons.
5. Apply the threshold. For the historical record, a period counts as El Niño when the index is +0.5 °C or more for at least five overlapping seasons in a row, and as La Niña when it is -0.5 °C or less for five in a row. Anything in between is neutral.
6. Grade the strength. NOAA's strength outlooks use steps of 0.5 °C: weak from 0.5, moderate from 1.0, strong from 1.5 and very strong from 2.0 (the same steps below zero for La Niña).

NOAA adds a caution: the strength of an event does not necessarily match the strength of its effects. A strong El Niño makes its usual effects more likely, but never guarantees them.

Worked example: classifying 2023 to 2026

Here are the peak values from NOAA's two tables, the official RONI and the older ONI, as read on 3 November 2026. NOAA revises the latest values for up to two months, so check the live tables for later updates.

Read the table row by row. In 2023-24 the old index called the event very strong, but the official one calls it moderate. The whole tropics were unusually warm that year, so Niño 3.4 stood out less against its surroundings. In 2024-25 and 2025-26 the opposite happened: compared with a very warm tropical ocean, the Niño 3.4 region was cool enough for the RONI to record two La Niña winters that the old ONI would not have counted.

For 2026, four seasons had reached +0.5 or more by early November, and NOAA had already issued an El Niño advisory. In October it put the chance of a very strong RONI, +2.0 or more, for September-October-November at 100 percent. The lesson is that the answer to "was it an El Niño?" depends on how you measure, so always say which index you used.

Step 5: follow the effects through the food web

Peru's anchoveta. The cold upwelling off Peru supports the Peruvian anchoveta, the largest single-species fishery in the world. FAO reports that it has averaged 4.4 million tonnes a year recently, 75 percent of all Peru's landings. In a strong El Niño the anchoveta lose habitat and food, which hits their spawning and the survival of their young, and the stock can stay low after the event ends. In 2023, FAO reports, Peru's industrial landings fell by 50 percent compared with 2022. Because anchoveta is ground into fishmeal and fish oil for fish farms, the loss reached aquaculture around the world.

The fishery's history shows El Niño and fishing working together. FAO's 2005 review records catches of 13.1 and 11.2 million tonnes in 1970 and 1971, far above the recommended ceiling of 8 to 9 million tonnes a year. Then came the 1972-73 collapse. Heavy fishing played a major role, FAO concludes, but the 1972-73 El Niño was also a primary cause, and a failure to cut fishing quickly made the decline worse and longer. The stock recovered by the early 1990s, then was badly depleted again in the strong 1997-98 El Niño. Lesson 21 returns to how fish stocks collapse and come back.

Seabirds. In May 1998, near the end of the 1997-98 El Niño, the penguin researcher Dee Boersma of the University of Washington surveyed the Galápagos Islands. She found dead marine iguanas and sea lions, underfed flightless cormorants and thin penguins, with no young penguins to be seen. The water was too warm to support their food. Boersma reported that more frequent and stronger El Niño events, with fewer cold La Niña years, had cut the Galápagos penguin population by about half since 1970. In bad years the adults do not lay eggs, or abandon their nests to find food for themselves.

Coral reefs. NOAA's Coral Reef Watch counts four global coral bleaching events: 1998, 2010, 2014 to 2017, and the one that began in 2023. Look at the El Niño record and you will find an El Niño in 1997-98, 2009-10, 2015-16 and 2023-24. Warm El Niño water adds to the long-term warming of the ocean (Lessons 12 and 23), and the two together push reefs past their bleaching thresholds.

Not every effect is bad. FAO notes that El Niño shifts skipjack tuna eastward across the Pacific, which helped the tuna catch of the easternmost Pacific Island nations in 2023.

Common misconceptions

• "El Niño is a storm, or a single warm current." It is a change in winds and ocean temperature across the whole tropical Pacific, lasting months, held together by the Bjerknes feedback.
• "El Niño is caused by global warming." ENSO is a natural cycle that fishermen recognised long before modern warming. Warming changes the background, which is why NOAA now measures El Niño relative to the whole tropics.
• "A strong El Niño guarantees floods here and drought there." NOAA warns that the strength of an event does not necessarily match its impacts; it shifts the odds.
• "The anchoveta collapse of 1972 was just bad luck with the weather." FAO concludes that heavy fishing played a major role alongside the 1972-73 El Niño.

Recap

• Normally the trade winds pile warm water in the west, the thermocline is shallow in the east, and cold, rich upwelling feeds Peru's fisheries.
• In El Niño the trade winds weaken, warm water spreads east, the eastern thermocline deepens and upwelling brings up warm, nutrient-poor water. La Niña is the reverse.
• The Bjerknes feedback between winds and ocean temperature lets either event grow.
• NOAA's official index since 2026 is the RONI: the three-month Niño 3.4 anomaly minus the tropical average, rescaled. Five overlapping seasons at +0.5 °C or more make an El Niño; at -0.5 °C or less, a La Niña. Strength steps are 0.5, 1.0, 1.5 and 2.0.
• The index you choose can change the answer: 2023-24 was very strong by the old ONI but moderate by the RONI.
• El Niño cut Peru's industrial landings by half in 2023, helped cause the 1972-73 anchoveta collapse along with heavy fishing, starves Galápagos penguins, and has accompanied every global coral bleaching event.

Sources

1. NOAA Climate Prediction Center. (2026, October 8). ENSO diagnostic discussion (updated monthly; the 8 October 2026 issue, read 3 November 2026). cpc.ncep.noaa.gov
2. NOAA Climate Prediction Center. (n.d.). Relative Oceanic Niño Index (RONI), table through July-September 2026 (updated monthly; read 3 November 2026). cpc.ncep.noaa.gov
3. NOAA Climate Prediction Center. (2026, February). CPC adopts Relative Oceanic Niño Index (RONI) for reliable, responsive monitoring and tracking of ENSO (read 3 November 2026). cpc.ncep.noaa.gov
4. NOAA Climate Prediction Center. (n.d.). Oceanic Niño Index (ONI), ERSST version 6, table through July-September 2026 (updated monthly; read 3 November 2026). cpc.ncep.noaa.gov
5. NOAA Climate Prediction Center. (2026, October). Official NOAA CPC ENSO strength probabilities (the October 2026 issue, read 3 November 2026). cpc.ncep.noaa.gov
6. NOAA Pacific Marine Environmental Laboratory. (n.d.). What is El Niño? pmel.noaa.gov
7. L'Heureux, M. (2014, May 5). What is the El Niño-Southern Oscillation (ENSO) in a nutshell? NOAA Climate.gov. climate.gov
8. L'Heureux, M. (2020, October 23). The rise of El Niño and La Niña. NOAA Climate.gov. climate.gov
9. FAO. (2024). Impacts of El Niño on marine fisheries and aquaculture. In The state of world fisheries and aquaculture 2024. openknowledge.fao.org
10. FAO. (2005). Review of the state of world marine fishery resources (FAO Fisheries Technical Paper 457), Southeast Pacific section. fao.org
11. Stricherz, V. (1998, May 26). Increasing frequency of El Niño takes toll on Northern Hemisphere's only penguins. University of Washington News. washington.edu
12. Boersma, P. D. (1998). Population trends of the Galápagos penguin: Impacts of El Niño and La Niña. The Condor, 100(2), 245-253.
13. NOAA Coral Reef Watch. (n.d.). 4th global coral bleaching event. coralreefwatch.noaa.gov

A fishery of 800,000 tonnes

Off the northeast coast of Newfoundland and southern Labrador lives a population of Atlantic cod known as northern cod. It was once the backbone of one of the world's largest fisheries. In the late 1960s, fleets landed more than 800,000 tonnes of it in a year. In some years, more than half of all the cod in the stock was caught. That was the finding of Tyler Eddy and Matthew Robertson, fisheries scientists at Memorial University of Newfoundland, writing in 2024.

The catches could not last. By the end of the 1980s the stock was shrinking fast, yet the total allowable catch, the legal limit for the year, was still about 240,000 tonnes. In 1992 the population had fallen by about 99 percent from its historical size, and the Canadian government shut the commercial fishery down. That moratorium put more than 30,000 people in Newfoundland and Labrador out of work. In the years that followed, the province's population fell by about 10 percent as young people left to find work elsewhere.

Thirty years of waiting

Many people expected the cod to bounce back within a few years once the big boats stopped. They did not. Northern cod stayed at very low numbers through the 1990s and early 2000s. Only around 2015 did they show signs of modest recovery. By 2023, Eddy and Robertson report, there were an estimated 407 million northern cod, weighing about 340,000 tonnes. That was still only about 12 percent of the stock's size in the 1960s.

On 26 June 2024 Canada's Minister of Fisheries ended the northern cod moratorium. Fisheries and Oceans Canada (DFO) set a Canadian total allowable catch of 18,000 tonnes and noted that the stock had been in the "cautious zone" of its scale since 2016. That catch was about 6 percent of the stock, far below the old rate. In June 2025, DFO said the stock had been stable since 2017 and larger than its 2024 assessment had shown, and it raised the catch to 38,000 tonnes.

Then, on 12 June 2026, DFO announced that for the first time since the 1992 moratorium the science placed northern cod in the "healthy zone". It raised the catch to 59,000 tonnes and called northern cod the largest spawning stock of any cod stock in the world. Not everyone was reassured. The conservation group Oceans North pointed out that the new catch gave the stock at least a 60 percent chance of staying in the healthy zone. That still leaves a large chance of slipping back, so the group had urged a lower limit. It also warned that recreational catches are not fully monitored, and that capelin, the cod's main food, remain well below historical levels.

Recovery is not guaranteed elsewhere either. Just to the south, NOAA reports that all four U.S. stocks of Atlantic cod, in the Gulf of Maine, on Georges Bank and off southern New England, were overfished in their 2024 assessments.

How much can you take? The idea of MSY

Why does a stock collapse under one level of fishing and hold up under another? A fish stock grows by adding young fish and by its fish growing heavier, and it loses fish to natural deaths. The amount it adds beyond replacing those deaths is its surplus. In principle, you can catch the surplus every year and leave the stock the same size.

NOAA defines maximum sustainable yield (MSY) as the largest long-term average catch that can be taken from a stock under current conditions. Overfishing means fishing at a higher rate than the one that produces MSY. Overfished means the stock has become too small to produce MSY. A stock can be overfished without overfishing still going on, if its catch has already been cut, as with northern cod after 1992.

Worked example: a model stock. The numbers below are made up, using a simple textbook model. Imagine a stock that would hold 1,000 tonnes of fish if it were never fished. A small stock grows slowly because there are few parents. A stock near 1,000 tonnes grows slowly too, because food and space run short. In this model the surplus is 0.4 × stock × (1 − stock ÷ 1,000).

The surplus peaks at 100 tonnes a year when the stock is at half its unfished size, so in this model the MSY is 100 tonnes. Now follow what happens if fishers take 120 tonnes a year from a stock of 500 tonnes. The stock adds 100 but loses 120, so it drops to 480. At 480 its surplus is a little smaller, so the next year it drops further, and the fall speeds up. Taking just 20 percent more than MSY sends the stock toward collapse. A catch of 80 tonnes does the opposite: the stock grows until it settles at about 720 tonnes, a bigger and safer stock. Real stocks are harder to read than this model, and their surplus changes with food and temperature. So real managers usually aim below MSY, not at it.

To use MSY, scientists need a stock assessment: an estimate of how big a stock is and how hard it is being fished, built from research surveys and the records of commercial and recreational fishers. An assessment describes the stock's past and present and predicts how it will respond to different catch limits.

The world's fish stocks

In June 2025 the Food and Agriculture Organization of the United Nations (FAO) published its most detailed review yet, covering 2,570 individual stocks. It found that 64.5 percent were fished within biologically sustainable levels and 35.5 percent were overfished. Weighted by how much each stock produces, 77.2 percent of the world's marine landings came from sustainable stocks, because many of the biggest fisheries are among the best managed. FAO also warned that the share of overfished stocks had been rising by about 1 percent a year.

Read the table for a pattern. FAO's own summary is that where management is effective, sustainability rates far exceed the global average. The weaker regions include countries where small-scale fishing is central to food security but where data and institutions are thin.

What helps, and how we know

1. Science-based catch limits and rebuilding plans. In the United States, the Magnuson-Stevens Act requires fisheries to prevent overfishing and to rebuild overfished stocks. Regional councils set annual catch limits based on the best available science and must write rebuilding plans. NOAA's Status of Stocks report for 2025 showed the overfishing list at an all-time low of 15 stocks, with 42 stocks overfished. One more stock was rebuilt that year, making 52 since 2000.

2. The global evidence. In 2020 Ray Hilborn and 22 colleagues gathered the trends of every scientifically assessed stock they could find, about half of the world's reported marine catch. On average, those stocks were increasing and close to their target levels. Regions with less developed management had, on average, three times the fishing rate and half the abundance. Regions with no assessments at all appeared to have little management and stocks in poor shape. The authors disclosed that all of them work in or advise on fisheries management, so it matters that FAO's independent review points the same way.

3. Bycatch rules. Fishing also kills animals nobody meant to catch. Earlier lessons showed fixes that work: turtle excluder devices (Lesson 16), bird-scaring lines and weighted hooks (Lesson 17), and keeping gillnets out of the vaquita's last refuge (Lesson 18).

4. Action against illegal fishing. FAO groups illegal, unreported and unregulated fishing as IUU fishing: fishing without permission or against the rules, catches that are hidden or misreported, and fishing by vessels outside any management system. The FAO Agreement on Port State Measures, in force since 2016, is the first binding international agreement aimed at IUU fishing. It works by stopping vessels caught fishing illegally from using ports to land their catch.

The core of it: A fish stock can take only its surplus. Catch more than that and it shrinks, and once it is small it can take decades to rebuild, as northern cod did. Where catches are set from good assessments and kept below the limit, stocks hold up or come back.

Common misconceptions

• "The ocean is too big to be fished out." Northern cod, once landed at more than 800,000 tonnes a year, fell by about 99 percent.
• "Stop fishing and a stock bounces back in a few years." Northern cod barely grew for two decades after 1992; recovery also depends on food, such as capelin, and on ocean conditions.
• "A collapsed stock is gone for good." Northern cod reached the healthy zone in 2026, and the United States has rebuilt 52 stocks since 2000.
• "Nearly all the world's fish stocks are overfished." FAO's 2025 review puts the figure at 35.5 percent: too many and rising, but not nearly all.

Putting it together

• Northern cod catches topped 800,000 tonnes in the late 1960s; by 1992 the stock had fallen about 99 percent, and the moratorium put more than 30,000 people out of work.
• The stock stayed low for two decades, reopened in 2024 with an 18,000-tonne catch, and in June 2026 was declared healthy for the first time since 1992, with some scientists urging caution.
• MSY is the largest long-term average catch a stock can support. Overfishing is fishing too hard; overfished means the stock is too small. In a simple model, catching just above MSY drives a stock down, and catching a little below it keeps the stock larger and safer.
• FAO's 2025 review: 64.5 percent of 2,570 stocks are sustainable, 35.5 percent overfished, and 77.2 percent of landings come from sustainable stocks, with large gaps between regions.
• What helps: catch limits from good assessments, legally required rebuilding plans, bycatch fixes, and port controls against illegal fishing.

Sources

1. Eddy, T., and Robertson, M. (2024, July 3). The federal government has lifted the moratorium on northern cod fishing after 32 years. The Conversation. theconversation.com
2. Fisheries and Oceans Canada. (2024, June 26). The Government of Canada announces the historic return of the commercial northern cod fishery in Newfoundland and Labrador. canada.ca
3. Fisheries and Oceans Canada. (2025, June). Canada announces a sustainable increase in northern cod TAC. canada.ca
4. Fisheries and Oceans Canada. (2026, June 12). Northern cod quota to increase 55 per cent: Minister Thompson announces 2026 management decisions. canada.ca
5. Oceans North. (2026, June 12). Northern cod quota 2026. oceansnorth.org
6. NOAA Fisheries. (n.d.). Atlantic cod. fisheries.noaa.gov
7. NOAA Fisheries. (2026). Status of stocks 2025. fisheries.noaa.gov
8. NOAA Fisheries. (n.d.). Understanding fisheries management in the United States. fisheries.noaa.gov
9. FAO. (2025, June 11). FAO releases the most detailed global assessment of marine fish stocks to date (on the Review of the state of world marine fishery resources 2025). fao.org
10. Hilborn, R., Amoroso, R. O., Anderson, C. M., and 20 others. (2020). Effective fisheries management instrumental in improving fish stock status. Proceedings of the National Academy of Sciences, 117(4), 2218-2224. pubmed.ncbi.nlm.nih.gov
11. FAO. (n.d.). What is IUU fishing? and Agreement on Port State Measures. fao.org; fao.org

Clean it up, or turn off the tap?

In 2018 a team of scientists estimated that the Great Pacific Garbage Patch, between California and Hawaii, held about 1.8 trillion pieces of plastic. Faced with a number like that, people tend to reach for one of two answers. Some say: build machines and go out and collect it. Others say: there is no point mopping the floor while the tap is still running, so stop plastic reaching the sea in the first place. This lesson sets out what is known about where ocean plastic comes from and what it does, then weighs the evidence for each answer.

How much plastic reaches the water?

Nobody can weigh the plastic entering the sea directly, so scientists estimate it from data on waste, population, rivers and rainfall. Four widely cited estimates are shown below.

Why do the numbers differ so much? They describe different years and different waters: some count only the ocean, others all rivers and lakes too. Each rests on a model with wide uncertainty, which is why most give a range. What they share matters more. All put the flow in millions of tonnes a year, and all trace most of it to waste on land that is never properly collected. Jambeck's team found that population size and the quality of waste management largely decide which countries contribute most. The OECD reported that only 9 percent of the world's plastic waste was recycled in 2019, while 22 percent escaped waste management altogether. Meijer's team found that more than 1,000 rivers carry 80 percent of the plastic that rivers deliver to the ocean, with small urban rivers among the worst.

The OECD adds a sobering point. About 30 million tonnes of plastic already sit in the seas and oceans, and a further 109 million tonnes have built up in rivers. Plastic will keep flowing out of rivers for decades, even if new waste is cut.

What is really in a garbage patch

NOAA's Marine Debris Program explains that garbage patches form inside gyres, the huge rotating current systems of Lesson 7. There are five gyres, and the best known patch lies in the North Pacific Gyre. "Patch" is a misleading name. There is no island of trash. Most pieces are microplastics, smaller than 5 mm, spread from the surface down through the water. NOAA compares it to pepper flakes swirling in a soup, and notes that you can sail through parts of a patch and see almost nothing.

The 2018 survey of the Great Pacific Garbage Patch, led by Laurent Lebreton, used ships and aircraft to calibrate a model. It estimated at least 79,000 tonnes of floating plastic, with a range of 45,000 to 129,000, spread over 1.6 million square kilometres. Microplastics made up 94 percent of the 1.8 trillion pieces but only 8 percent of the mass. More than three-quarters of the mass was in pieces larger than 5 cm, and at least 46 percent was fishing nets. The team also found the patch was accumulating plastic faster than the waters around it.

What plastic does to animals

NOAA notes that marine debris has been documented harming more than 700 species, from plankton and seaweeds to fish, turtles, seabirds and whales.

• Entanglement. Nets and packing bands wrap around animals, make it hard for them to swim, and cut into their bodies. Lost nets go on catching animals for years, which is called ghost fishing.
• Ingestion. A swallowed item can block an animal's stomach, cut it from inside, or make it feel full so that it starves. Sea turtles may mistake balloons and plastic bags for prey (Lesson 16), and albatrosses feed plastic to their chicks (Lesson 17).
• Smaller and smaller. As plastic breaks into smaller pieces, ever smaller animals, down to zooplankton, can eat it.

Side one: clean it up

The Ocean Cleanup, a nonprofit founded to remove ocean plastic, aims to clean up 90 percent of floating ocean plastic by 2040. It tows collection systems through the garbage patches, places barriers in rivers to catch plastic before it reaches the sea, and supports work to reduce lost fishing gear. On its website, read on 3 November 2026, it reported more than 65 million kilograms of trash removed from the environment, and estimated that its river work stops 2 to 5 percent of the world's river-borne plastic.

The case for cleanup rests partly on the 2018 survey: most of the patch's mass is still in large pieces, such as nets, that could be taken out before they break into microplastics that can never be collected.

Side two: stop it at the source

Other researchers have tested how much cleanup can achieve. In 2020 a team led by Sönke Hohn modelled cleanup devices at sea and barriers on rivers. They found the projected impact of devices at sea, even many of them, was very modest. A significant reduction came only from collecting plastic at rivers, or from rivers and sea together. They concluded that only reduced plastic emissions combined with stronger collection could rid the ocean of plastic.

Two studies published in Science in 2020 looked at the whole system. Winnie Lau and colleagues modelled every feasible measure together: making and using less plastic, reuse, better collection and recycling, and safe disposal. By 2040 that cut plastic pollution by 78 percent compared with business as usual, yet 710 million tonnes would still enter the environment along the way. Stephanie Borrelle and colleagues warned that even with the ambitious commitments governments had made, yearly emissions to water could reach 53 million tonnes by 2030.

Worked example: the scale of cleanup. The Ocean Cleanup's 65 million kilograms is 65,000 tonnes, the total removed since it began work. The OECD estimates that 1.7 million tonnes flowed into the ocean in 2019 alone. 65,000 ÷ 1,700,000 ≈ 0.04, so everything removed so far equals about 4 percent of a single year's flow into the ocean.

Weighing the two sides

Look at who wrote what. The 2018 garbage patch survey and the 2021 river study were led by scientists at The Ocean Cleanup, as their published affiliations show. The 2020 scenario study was led by researchers at The Pew Charitable Trusts and the consultancy SYSTEMIQ. Who does a study does not make it wrong, and all these papers passed peer review, so judge them on their data and methods.

Here the evidence lines up. Cleanup at sea cannot keep pace with the inflow. Collecting plastic in rivers and on coasts helps more, and preventing waste on land matters most. Cleanup still has a role for lost nets and other large debris before it breaks apart. The OECD adds that bans on single-use plastics, now found in more than 120 countries, cut littering but do little to reduce overall use. It calls for tools such as making producers pay for collecting their products, deposit-refund schemes and landfill taxes.

So what? Plastic in the sea is mostly a waste management problem on land. Cleanup ships make headlines, but the evidence says the biggest gains come from producing and wasting less plastic and from collecting it before it reaches rivers and the coast.

Common misconceptions

• "The Great Pacific Garbage Patch is a floating island you could walk on." It is mostly small pieces spread through a huge area of water, like pepper in soup.
• "Most ocean plastic is dumped from ships." Most of the flow comes from land through poor waste management, although lost fishing nets make up a large share of the garbage patch's mass.
• "Recycling takes care of plastic waste." Only about 9 percent of plastic waste was recycled in 2019, according to the OECD.
• "Cleanup ships will solve the problem." Modelling finds devices at sea have a very modest effect unless the inflow is cut and plastic is collected in rivers.

What to remember

• Estimates of plastic reaching water range from about 1.7 million tonnes into the ocean (OECD, 2019) to 19 to 23 million tonnes into all waters (Borrelle, 2016), because they count different things; all agree the flow is millions of tonnes a year and comes mostly from poorly managed waste on land.
• Garbage patches are not islands: most pieces are microplastics, but most of the mass in the Great Pacific Garbage Patch is larger debris, at least 46 percent of it fishing nets.
• Plastic harms more than 700 species through entanglement, ghost fishing and ingestion.
• Cleanup at sea removes a small fraction of the inflow. Models find the biggest reductions come from using and wasting less plastic and from collecting it on land and in rivers.
• Read sources critically: check data and methods, and notice who did the work.

Sources

1. Jambeck, J. R., Geyer, R., Wilcox, C., and others. (2015). Plastic waste inputs from land into the ocean. Science, 347(6223), 768-771. pubmed.ncbi.nlm.nih.gov
2. Borrelle, S. B., Ringma, J., Law, K. L., and others. (2020). Predicted growth in plastic waste exceeds efforts to mitigate plastic pollution. Science, 369(6510), 1515-1518. pubmed.ncbi.nlm.nih.gov
3. OECD. (2022, February 22). Plastic pollution is growing relentlessly as waste management and recycling fall short, says OECD (on the Global Plastics Outlook). oecd.org
4. Meijer, L. J. J., van Emmerik, T., van der Ent, R., Schmidt, C., and Lebreton, L. (2021). More than 1000 rivers account for 80% of global riverine plastic emissions into the ocean. Science Advances, 7(18), eaaz5803. pubmed.ncbi.nlm.nih.gov
5. Lebreton, L., Slat, B., Ferrari, F., and others. (2018). Evidence that the Great Pacific Garbage Patch is rapidly accumulating plastic. Scientific Reports, 8, 4666. pubmed.ncbi.nlm.nih.gov
6. Lau, W. W. Y., Shiran, Y., Bailey, R. M., and others. (2020). Evaluating scenarios toward zero plastic pollution. Science, 369(6510), 1455-1461. pubmed.ncbi.nlm.nih.gov
7. Hohn, S., Acevedo-Trejos, E., Abrams, J. F., Fulgencio de Moura, J., Spranz, R., and Merico, A. (2020). The long-term legacy of plastic mass production. Science of the Total Environment, 746, 141115. pubmed.ncbi.nlm.nih.gov
8. NOAA Marine Debris Program. (n.d.). Garbage patches and Why is marine debris a problem? marinedebris.noaa.gov; marinedebris.noaa.gov
9. The Ocean Cleanup. (n.d.). Home page: removal totals and goals (read 3 November 2026). theoceancleanup.com
10. NOAA Marine Debris Program. (n.d.). Marine Debris Monitoring and Assessment Project and MDMAP guide for educators: safety tips and transect selection. marinedebris.noaa.gov; marinedebris.noaa.gov

The bug: "a degree or two cannot matter"

Here is a claim you may hear. The sea changes by several degrees between winter and summer, and fish live everywhere from icy polar water at about -2 °C to tropical reefs near 30 °C. So how could a degree or two of warming matter to sea life? It sounds reasonable. This lesson debugs it the way a scientist would: check the measurements, then follow what warmth actually does to animals and ecosystems, and see where the claim breaks.

Step 1: check the measurements

The IPCC's 2021 assessment found that the surface of the ocean was about 0.88 °C warmer in 2011 to 2020 than in 1850 to 1900. It is virtually certain that the upper 700 m of the ocean has warmed since the 1970s, and extremely likely that human influence is the main driver. As Lesson 19 showed, the ocean has taken up 91 percent of the extra heat in the climate system. In 2026 a team of more than 50 scientists, including NOAA researchers, reported that the heat content of the ocean set another record in 2025. So the warming is real and measured. The question is whether its size, under one degree at the surface, is too small to matter.

Step 2: averages hide extremes

An average tells you little about the worst days. When the whole range of temperatures shifts up a little, the rare hot extremes become much more common. In the ocean these extremes are marine heatwaves: periods when the water is much warmer than usual for that place and time of year, lasting for an extended period. NOAA's California Current Marine Heatwave Tracker spots them in satellite records going back to 1982.

The IPCC found that marine heatwaves have approximately doubled in frequency since the 1980s, and that human influence very likely contributed to most of them since at least 2006. The tracker is busy now. On its page updated 30 October 2026, read on 3 November 2026, NOAA was following two large heatwaves. One had sat off California since May 2025. The other had formed in the Gulf of Alaska region in May 2026. Both had grown and intensified over the previous month.

Step 3: follow one heatwave through the food web

The best studied example is the Blob, the vast marine heatwave that spread across the northeast Pacific from 2014 to 2016. Its water was 2 to 3 standard deviations warmer than average, far outside the normal range. Here is what happened, step by step.

1. Less and poorer food at the bottom. The warm water held fewer nutrients. Phytoplankton declined, and the zooplankton shifted toward species with fewer calories.
2. Hungrier fish. Fish are ectotherms: their bodies run at the temperature of the water, so warm water speeds up their metabolism and their need for food. Small forage fish needed more food just as their food got scarcer and poorer, so they became fewer and thinner. Large fish, also hungrier, ate more of them. John Piatt and colleagues called this squeeze an "ectothermic vise".
3. Starving seabirds. About 62,000 dead or dying common murres washed ashore from California to Alaska between summer 2015 and spring 2016. Most were severely emaciated. Because only a fraction of birds that die at sea wash ashore, the researchers estimated that total deaths approached 1 million. Colonies across the region had 22 complete breeding failures from 2015 to 2017.
4. Toxic algae. In spring 2015 the diatom Pseudo-nitzschia bloomed along the whole West Coast, producing the largest outbreak of the nerve toxin domoic acid ever recorded there. Ryan McCabe and colleagues showed the bloom was set off by the unusually warm water. Razor clam, rock crab and Dungeness crab fisheries closed for long periods.
5. Tangled whales. The cool, food-rich upwelling water was squeezed into a narrow band near shore, and feeding whales followed it. Then the toxin delayed California's 2016 crab season, which pushed crab traps and whales into the same waters at the same time. Jarrod Santora and colleagues linked these changes to record numbers of whale entanglements in crab gear.
6. Vanishing kelp. As Lesson 11 told, the same heatwave helped trigger the collapse of northern California's kelp forests.

Key idea: A degree or two on average brings far bigger extremes. In the sea, a temperature change also travels up the food web: cold-blooded animals need more food as they warm, while their food can shrink.

Step 4: species on the move

Animals that can move do. Malin Pinsky and colleagues compiled coastal surveys around North America from 1968 to 2011, covering 128 million individual animals from 360 kinds of marine life. Species had shifted in latitude and depth. What best explained their moves was not their own traits but climate velocity: the speed and direction in which temperature zones were moving in each place. Species follow their preferred temperatures as those temperatures move, which is why different species in different places have moved in different directions. That can carry a fishery away from the ports and the rules built around it. Animals fixed in place, such as corals, kelp and seabirds tied to nesting colonies, cannot follow.

Step 5: less oxygen

Lesson 4 showed that warm water holds less dissolved oxygen than cold water. Sunke Schmidtko and colleagues analysed oxygen measurements through the whole depth of the ocean and found that its total oxygen content had fallen by more than 2 percent since 1960. They traced the losses in the upper ocean mostly to warming, through lower solubility, and to oxygen used up by living things. A 2018 review led by Denise Breitburg found that oxygen in the open ocean and in coastal waters has been declining for at least half a century.^[9] The main causes are human activities that have raised temperatures and poured nutrients into coastal waters (Lesson 10).

Warming also strengthens the layering of the ocean, called stratification. A warmer, lighter surface layer mixes less with the cold water below, so less oxygen reaches the depths and fewer nutrients come up. The IPCC finds it virtually certain that upper-ocean stratification will keep increasing this century, and has high confidence that deoxygenation will continue.

Where the claim breaks

• Averages hide extremes. A small rise in the average means many more marine heatwaves, which have doubled in frequency since the 1980s.
• Bodies feel it twice. A cold-blooded animal in warmer water needs more food and more oxygen, while warm water holds less oxygen.
• Food webs pass it on. A change at the bottom, in plankton, reached murres, whales, crabs and the people who fish for them.
• Some cannot move. Species that can shift do, but fixed habitats such as reefs and kelp forests cannot.

What helps

The IPCC's projections show that ocean warming will continue through this century, but how far depends on emissions. If emissions are cut steeply, the ocean is likely to warm by 2 to 4 times the change seen from 1971 to 2018 by the year 2100. Under very high emissions, the figure is 4 to 8 times. The rates of stratification, acidification and oxygen loss also depend on future emissions. That is why reducing greenhouse gas emissions is the measure that limits the harm at its source.

Other measures reduce the damage in the meantime. Forecasts and trackers, such as NOAA's heatwave tracker, give fishers and managers early warning. Santora and colleagues showed that fishers, managers and scientists working together could have reduced whale entanglements during the Blob, an example of what they call climate-ready fisheries management.^[6] Because coastal oxygen loss comes partly from nutrient pollution, cutting nutrient runoff tackles the part of the problem that is local. Lesson 25 looks at protection and restoration.

Common misconceptions

• "The ocean is too big to warm." It has absorbed 91 percent of the extra heat, and its upper 700 m has warmed since the 1970s.
• "A marine heatwave is a hot day at the beach." The Blob lasted from 2014 to 2016 and covered water from California to Alaska.
• "Fish will just swim somewhere cooler, so warming is harmless." Many species are shifting, but that moves fisheries away from the people who depend on them, and fixed habitats such as reefs cannot move.
• "Ocean oxygen cannot run short." The ocean's oxygen content has fallen by more than 2 percent since 1960, and the loss is projected to continue.

The short version

• The ocean surface warmed about 0.88 °C from 1850-1900 to 2011-2020, and ocean heat content set another record in 2025.
• Marine heatwaves have roughly doubled in frequency since the 1980s; NOAA was tracking two large ones in the northeast Pacific in late 2026.
• The 2014-2016 Blob cut plankton food, squeezed forage fish in an "ectothermic vise", killed perhaps 1 million murres, set off a record toxic algal bloom that closed crab and clam fisheries, and helped cause record whale entanglements.
• Marine species are shifting in latitude and depth, tracking local climate velocity.
• The ocean has lost more than 2 percent of its oxygen since 1960, from warming and, near coasts, nutrient pollution; stratification and oxygen loss will continue.
• How far warming goes depends on emissions; forecasts, climate-ready fisheries and cutting nutrient runoff reduce the harm.

Sources

1. IPCC. (2021). Summary for policymakers. In Climate change 2021: The physical science basis. Statements A.1.2, A.1.6, A.3.1 and B.5.1. ipcc.ch
2. Pan, Y., Cheng, L., Abraham, J., and others. (2026). Ocean heat content sets another record in 2025. Advances in Atmospheric Sciences. NOAA Institutional Repository record. repository.library.noaa.gov
3. NOAA California Current Integrated Ecosystem Assessment. (2026, October 30). California Current Marine Heatwave Tracker (updated regularly; read 3 November 2026). integratedecosystemassessment.noaa.gov
4. Piatt, J. F., Parrish, J. K., Renner, H. M., and others. (2020). Extreme mortality and reproductive failure of common murres resulting from the northeast Pacific marine heatwave of 2014-2016. PLOS ONE, 15(1), e0226087. journals.plos.org
5. McCabe, R. M., Hickey, B. M., Kudela, R. M., and others. (2016). An unprecedented coastwide toxic algal bloom linked to anomalous ocean conditions. Geophysical Research Letters, 43(19), 10366-10376. pubmed.ncbi.nlm.nih.gov
6. Santora, J. A., Mantua, N. J., Schroeder, I. D., and others. (2020). Habitat compression and ecosystem shifts as potential links between marine heatwave and record whale entanglements. Nature Communications, 11, 536. pubmed.ncbi.nlm.nih.gov
7. Pinsky, M. L., Worm, B., Fogarty, M. J., Sarmiento, J. L., and Levin, S. A. (2013). Marine taxa track local climate velocities. Science, 341(6151), 1239-1242. pubmed.ncbi.nlm.nih.gov
8. Schmidtko, S., Stramma, L., and Visbeck, M. (2017). Decline in global oceanic oxygen content during the past five decades. Nature, 542, 335-339. pubmed.ncbi.nlm.nih.gov
9. Breitburg, D., Levin, L. A., Oschlies, A., and others. (2018). Declining oxygen in the global ocean and coastal waters. Science, 359(6371), eaam7240. pubmed.ncbi.nlm.nih.gov

A number that sounds too small

Lesson 4 noted that the surface ocean has a pH of about 8.1, slightly alkaline. NOAA reports that in the 200-plus years since the industrial revolution began, surface ocean pH has fallen by about 0.1 units. A change of 0.1 sounds trivial. This lesson works through, step by step, what that number means, how it is measured and why oyster farmers on the Pacific coast came to care about it so much.

Step 1: follow the carbon dioxide into the water

NOAA estimates that the ocean absorbs about 30 percent of the carbon dioxide (CO₂) released into the atmosphere. Once it dissolves, a chain of reactions begins.

1. Carbon dioxide dissolves into seawater: CO₂ (air) → CO₂ (dissolved).
2. It combines with water to form carbonic acid, a weak acid: CO₂ + H₂O → H₂CO₃.
3. Carbonic acid breaks apart into a hydrogen ion and a bicarbonate ion: H₂CO₃ → H⁺ + HCO₃⁻. The extra hydrogen ions are what lower the pH.
4. Many of those hydrogen ions then join carbonate ions already in the water, turning them into more bicarbonate: H⁺ + CO₃²⁻ → HCO₃⁻.

Add up the steps and the overall result is CO₂ + H₂O + CO₃²⁻ → 2 HCO₃⁻. Each molecule of carbon dioxide that is absorbed uses up a carbonate ion. That matters because carbonate ions are exactly what shell builders need. NOAA explains that oysters, corals and other calcifying organisms build shells and skeletons by combining calcium with carbonate: Ca²⁺ + CO₃²⁻ → CaCO₃. With fewer carbonate ions available, building and keeping calcium carbonate gets harder, and if the pH falls far enough, shells can begin to dissolve.

Step 2: work out what 0.1 means

pH is defined as the negative logarithm of the hydrogen ion concentration: pH = −log₁₀[H⁺]. Because the scale is logarithmic, each whole unit is a tenfold change. Work it through.

1. At pH 8.2, [H⁺] = 10^−8.2 ≈ 6.3 × 10⁻⁹ moles per litre.
2. At pH 8.1, [H⁺] = 10^−8.1 ≈ 7.9 × 10⁻⁹ moles per litre.
3. Divide: 7.9 ÷ 6.3 ≈ 1.26. Equivalently, 10^0.1 ≈ 1.26.
4. So a fall of exactly 0.1 means about 26 percent more hydrogen ions.

NOAA describes the change since the industrial revolution as approximately a 30 percent increase in acidity. The exact percentage depends on the exact size of the drop, which is usually rounded to 0.1. Either way, a small-looking number hides a large change in chemistry. Note too what acidification does not mean: at about 8.1 the ocean is still alkaline. "Acidification" describes the direction of change, toward lower pH, not water that has turned into acid.

Remember: pH is logarithmic. A fall of 0.1 means roughly a quarter more hydrogen ions, and every one of those extra ions comes with fewer carbonate ions for shells.

Step 3: read a real record

How do we know the change is real, and not a guess? Since the late 1980s, scientists have sampled the same patch of open ocean near Hawaii, called Station ALOHA, on regular cruises. In 2009 John Dore and colleagues reported nearly 20 years of its pH measurements.^[3] Surface pH fell by 0.0019 units a year, plus or minus 0.0002. That rate is what you would expect if the ocean were simply taking up carbon dioxide in step with the rising level in the air. On top of the trend, pH rose and fell with the seasons, driven by temperature, mixing and photosynthesis. Work with the trend:

• Over 20 years: 20 × 0.0019 = 0.038.
• Time to fall by 0.1 at that rate: 0.1 ÷ 0.0019 ≈ 53 years.

The IPCC's 2021 assessment draws the global picture. It is virtually certain that human emissions of carbon dioxide are the main driver of the current acidification of the surface open ocean. A surface pH as low as that of recent decades is unusual in the last 2 million years (medium confidence).

Step 4: from chemistry to shells

Chemists summarise whether water helps or hinders shell building with the saturation state, written Ω (omega). Above 1, the water favours forming calcium carbonate. Below 1, the water is undersaturated, or corrosive: it tends to dissolve calcium carbonate. Aragonite, the form of calcium carbonate used by corals, pteropods and young oysters, is the more easily dissolved form, so scientists usually report Ω for aragonite.

Lesson 2 described the carbonate compensation depth, below which shells dissolve on the deep seafloor. Richard Feely and colleagues showed in 2004 that the ocean's uptake of carbon dioxide has changed its saturation state for calcium carbonate.^[4] In 2008 they surveyed the continental shelf of western North America from central Canada to northern Mexico. They found water undersaturated in aragonite upwelling onto large parts of the shelf, reaching depths of about 40 to 120 m along most survey lines and all the way to the surface off northern California. Upwelling of this water is natural in the region, they noted, but the ocean's uptake of human-made carbon dioxide has increased the area it affects.

Step 5: the oyster hatcheries

In 2007, according to a 2015 review by Alan Barton and colleagues, West Coast shellfish hatcheries producing Pacific oyster larvae began to suffer unprecedented die-offs. At the Whiskey Creek Shellfish Hatchery in Oregon, which draws its water from the sea, hatchery staff and scientists measured the chemistry of the incoming water. In early summer 2009 its aragonite saturation state swung from below 0.8 to above 3.2, and its pH from below 7.6 to above 8.2. Larval production and growth were linked to the chemistry of the water the larvae were spawned in during their first 48 hours of life: the more carbon dioxide, the worse they did. Corrosive upwelled water, the same water Feely's team had found on the shelf, was killing the larvae.

The industry's response is a model of adaptation. The Pacific Coast Shellfish Growers Association installed instruments in hatcheries and coastal waters, building a monitoring network with university researchers and the U.S. Integrated Ocean Observing System. With real-time measurements of saturation state, hatcheries could see corrosive water coming, and they built commercial-scale systems to treat it. Barton's review reports that these systems have generally proven effective, supplying billions of additional oyster larvae to Pacific Northwest growers.^[6][7] It also warns that significant challenges remain, and that selective breeding, more hatchery capacity, continued monitoring and better understanding of biology will all be needed.

Step 6: separate the cure from the coping

Monitoring and treating water helps a hatchery, but it cannot treat the open ocean. The IPCC finds it virtually certain that ocean acidification will continue in the 21st century, at a rate that depends on future emissions. Acidification of the deep ocean is irreversible on timescales of centuries to millennia (very high confidence). Because about 30 percent of emitted carbon dioxide ends up in the sea, the measure that addresses the cause is reducing carbon dioxide emissions. Local actions such as monitoring, water treatment, selective breeding and managing local water quality buy time for particular places and industries. NOAA has also reported, in 2019, that farming native seaweed at a California oyster hatchery improved the water's quality. These are ways of coping, not a cure.

Common misconceptions

• "The ocean is turning into acid." At about pH 8.1 it is still alkaline; acidification means the pH is falling.
• "A pH drop of 0.1 is too small to matter." The scale is logarithmic: 0.1 means about a quarter more hydrogen ions, and fewer carbonate ions for shells.
• "Acidification is a guess from computer models." It is measured directly: Station ALOHA recorded a fall of 0.0019 pH units a year over nearly two decades.
• "Hatchery fixes have solved the problem." Treating hatchery water saved billions of larvae, but it cannot change the open ocean, where acidification continues.

Pulling it together

• The ocean absorbs about 30 percent of emitted carbon dioxide. Dissolved CO₂ forms carbonic acid, which releases hydrogen ions; those ions lower the pH and turn carbonate into bicarbonate.
• Overall: CO₂ + H₂O + CO₃²⁻ → 2 HCO₃⁻, so shell builders have fewer carbonate ions to work with.
• pH is logarithmic: a fall of 0.1 is about 26 percent more hydrogen ions; NOAA puts the increase in acidity since the industrial revolution at about 30 percent.
• Station ALOHA measured a fall of 0.0019 pH units a year, in step with rising carbon dioxide in the air.
• Saturation state below 1 means corrosive water. Corrosive water now upwells onto the shelf of western North America and killed oyster larvae in hatcheries until monitoring and water treatment restored production.
• Reducing carbon dioxide emissions addresses the cause; monitoring, treatment and breeding help particular places cope.

Sources

1. NOAA. (n.d.). Ocean acidification (education resource collection). noaa.gov
2. IPCC. (2021). Summary for policymakers. In Climate change 2021: The physical science basis. Statements A.1.6, A.2.4 and B.5.1. ipcc.ch
3. Dore, J. E., Lukas, R., Sadler, D. W., Church, M. J., and Karl, D. M. (2009). Physical and biogeochemical modulation of ocean acidification in the central North Pacific. Proceedings of the National Academy of Sciences, 106(30), 12235-12240. pubmed.ncbi.nlm.nih.gov
4. Feely, R. A., Sabine, C. L., Lee, K., and others. (2004). Impact of anthropogenic CO2 on the CaCO3 system in the oceans. Science, 305(5682), 362-366. pubmed.ncbi.nlm.nih.gov
5. Feely, R. A., Sabine, C. L., Hernandez-Ayon, J. M., Ianson, D., and Hales, B. (2008). Evidence for upwelling of corrosive "acidified" water onto the continental shelf. Science, 320(5882), 1490-1492. pubmed.ncbi.nlm.nih.gov
6. Barton, A., Hales, B., Waldbusser, G. G., Langdon, C., and Feely, R. A. (2012). The Pacific oyster, Crassostrea gigas, shows negative correlation to naturally elevated carbon dioxide levels: Implications for near-term ocean acidification effects. Limnology and Oceanography, 57(3), 698-710.
7. Barton, A., Waldbusser, G. G., Feely, R. A., and others. (2015). Impacts of coastal acidification on the Pacific Northwest shellfish industry and adaptation strategies implemented in response. Oceanography, 28(2), 146-159. oceanacidification.noaa.gov
8. NOAA Office of Education. (n.d.). Data in the Classroom: Understanding ocean and coastal acidification, for students. noaa.gov

A reef that came back

Cabo Pulmo lies on the Gulf of California coast near the southern tip of Mexico's Baja California peninsula. In 1995, with strong support from local people, the waters off its reef became a national park closed to fishing. For a few years little seemed to change. In 1999, Octavio Aburto-Oropeza and colleagues found no more fish there than in other parts of the Gulf of California. By 2009, fish biomass in the park had risen from 0.75 to 4.24 tonnes per hectare, an increase of 463 percent. Top predators had increased elevenfold. In other protected areas and open fishing grounds nearby, fish biomass had not changed.

It was the largest increase within a decade measured in any marine reserve. The researchers credited strong community leadership, social cohesion and effective enforcement as well as ecology, and they found real economic benefits from the recovery. Earlier lessons in this module set out the threats one at a time. This last lesson compares the tools that have measurably helped, and asks honestly where each one falls short.

Tool 1: marine protected areas

A marine protected area (MPA) is an area of ocean set aside and managed to conserve nature, but the rules inside vary enormously. Some ban all fishing; some allow almost everything. In 2021 a team led by Kirsten Grorud-Colvert published The MPA Guide, a framework that sorts MPAs by how far they have been put into practice and by their level of protection.^[4] Its strictest levels, fully and highly protected, allow little or no fishing or other damaging use.

What does strict protection achieve? In 2009 Sarah Lester and colleagues combined studies of no-take reserves around the world. On average, biomass, numbers of animals, body size and the number of species were all higher inside reserves. These effects did not come from reserves simply being placed in richer spots, or from fishing being pushed outside them. Even small reserves produced measurable effects. But not every species increased, and results varied a great deal from place to place.

Why the variation? In 2014 Graham Edgar and colleagues surveyed fish in 87 MPAs worldwide. They found that benefits rose steeply as an MPA gained five features.

MPAs with four or five features had twice as many species of large fish per survey, five times the biomass of large fish and fourteen times the shark biomass of fished areas. But 59 percent of the MPAs studied had only one or two features, and they could not be told apart ecologically from fished sites. The authors concluded that targets based on area alone will not make the best use of protection. Cabo Pulmo is a case of what the five features can do.

How much of the ocean is protected?

In December 2022 the world's governments adopted the Kunming-Montreal Global Biodiversity Framework. Its Target 3, often called "30 by 30", aims for at least 30 percent of land, inland waters and coastal and marine areas to be effectively conserved and managed by 2030. Where does the ocean stand? These figures change month by month; they were read on 3 November 2026.

• Protected Planet, the official database run by UNEP-WCMC and IUCN, counted 17,432 MPAs covering 9.71 percent of the ocean.^[5] Adding other effective area-based conservation measures (OECMs), areas that conserve nature without being formally named protected areas, brought the total to 9.99 percent.
• Most of that lies in national waters, which make up 39 percent of the ocean and are 22.79 percent protected. The high seas beyond any nation's control make up 61 percent of the ocean but only 1.45 percent of them are protected.
• MPAtlas, run by the Marine Conservation Institute, applies The MPA Guide to these areas. In its assessment of 30 October 2026, only 3.5 percent of the ocean was fully or highly protected.

Worked example. Check Protected Planet's total from its two parts: 0.39 × 22.79 percent + 0.61 × 1.45 percent ≈ 8.9 + 0.9 = 9.8 percent, close to the reported 9.99 percent. Then compare strict protection with the total: 3.5 ÷ 9.7 ≈ 0.36. Only about a third of the area inside MPAs is fully or highly protected.

The high seas may now change faster. The United Nations agreement on marine biodiversity beyond national jurisdiction, known as the High Seas Treaty or BBNJ Agreement, was adopted on 19 June 2023 and entered into force on 17 January 2026. Among other things, it sets up a way to create protected areas on the high seas. The first meeting of its parties is scheduled for January 2027.

Tool 2: restoration

Protection lets a place recover by itself. Restoration rebuilds what was lost.

• Eelgrass in Virginia. Eelgrass had been missing from Virginia's Atlantic coastal lagoons for decades. Since 1999, Robert Orth and colleagues have scattered more than 70 million eelgrass seeds into those lagoons. By 2020, 3,612 hectares of seagrass had recovered. The meadows support rich animal communities, store substantial carbon and nitrogen, and allowed bay scallops to be restored too.
• Oyster reefs in Chesapeake Bay. NOAA and its partners set a goal of restoring oyster reefs in 10 tributaries of the bay by the end of 2025, and met it, with a bonus eleventh. The 10 tributaries now hold more than 2,400 acres of healthy reef, about 1,900 of them actively restored. In Maryland, 99 percent of restored reefs checked at 6 years old met at least the minimum target for oyster numbers and biomass. In Harris Creek alone, partners planted 2.49 billion young oysters on 343 acres, at a cost of $29.06 million.
• Islands cleared of rats. Holly Jones and colleagues reviewed 251 eradications of invasive mammals on 181 islands. They found that 236 native species benefited, seven were harmed, and four threatened species moved to a lower risk of extinction. South Georgia (Lesson 17) is one example.

Restoration has limits. It is slow and expensive: Harris Creek cost about $85,000 an acre. And it works only once the original cause of loss is under control. Eelgrass will not return to water that is still too murky.

Tool 3: better rules

Earlier lessons showed rules with measured results. After the whaling moratorium and decades of legal protection, 9 of the 14 humpback whale populations recovered enough to come off the U.S. endangered list (Lesson 18). Nest protection and turtle excluder devices, required since 1987, helped bring Kemp's ridley turtles back from the brink (Lesson 16). Science-based catch limits have rebuilt 52 U.S. fish stocks since 2000 (Lesson 21). Nearly eliminating gillnets from the vaquita's sanctuary appears to have halted its steep decline (Lesson 18). Each worked because it was monitored and enforced.

Comparing the tools

The point: No single tool is enough. Strict, enforced protection and good rules let ocean life recover where local pressures are the problem. Restoration can rebuild habitats once those pressures are controlled. Only cutting emissions addresses warming and acidification.

What can a student do?

It is worth being honest about scale: one person's choices are small next to laws, enforcement and emissions. Here are actions open to you, with the evidence behind each.

1. Collect data. Community science feeds real research. NOAA publishes examples of how MDMAP surveys like the one in this module are used in research and prevention projects, and projects such as eBird (Lesson 17) collect sightings from the public. Even far from the sea, you can survey litter on a stream bank that drains toward it.
2. Choose seafood from well-managed fisheries. NOAA's species pages and FAO's figures show that management, not the fish itself, decides sustainability (Lesson 21).
3. Waste less plastic and keep it out of drains and rivers. Collection on land matters far more than cleanup at sea (Lesson 22).
4. Watch wildlife at a distance. NOAA's viewing distances protect animals directly (Lesson 18).
5. Take part in decisions. MPAs, fishing rules and climate policy are set by governments that invite public comment. Rules like these drove the recoveries in this lesson.

Common misconceptions

• "Any marine protected area protects ocean life." In Edgar's study, 59 percent of MPAs had only one or two key features and looked like fished sites.
• "A tenth of the ocean is strictly protected." About 10 percent is in MPAs and OECMs, but only about 3.5 percent is fully or highly protected.
• "Protected areas will save reefs from climate change." No boundary keeps out warm or acidified water; those come from global emissions.
• "Restoration is a quick fix." Virginia's eelgrass took two decades, and one oyster restoration cost about $85,000 an acre.

What to carry forward

• Strict, enforced no-take reserves work: biomass, numbers, size and species all rise on average, and Cabo Pulmo's fish biomass rose 463 percent in a decade.
• MPAs succeed when they are no-take, well enforced, old, large and isolated; most existing MPAs lack most of these features.
• As read on 3 November 2026, about 10 percent of the ocean was in protected areas and OECMs, but only 3.5 percent was fully or highly protected. 30 by 30 aims for 30 percent by 2030, and the High Seas Treaty, in force since January 2026, opens the high seas to protection.
• Restoration has rebuilt eelgrass meadows, oyster reefs and island seabird colonies, but it is slow, costly and needs the original cause fixed.
• Rules that were enforced brought back humpbacks, Kemp's ridleys and 52 U.S. fish stocks. Cutting emissions is the only tool for warming and acidification.
• Your strongest contributions are data, informed choices and taking part in the decisions that set the rules.

Sources

1. Aburto-Oropeza, O., Erisman, B., Galland, G. R., and others. (2011). Large recovery of fish biomass in a no-take marine reserve. PLOS ONE, 6(8), e23601. journals.plos.org
2. Lester, S. E., Halpern, B. S., Grorud-Colvert, K., Lubchenco, J., Ruttenberg, B. I., Gaines, S. D., Airamé, S., and Warner, R. R. (2009). Biological effects within no-take marine reserves: A global synthesis. Marine Ecology Progress Series, 384, 33-46. int-res.com
3. Edgar, G. J., Stuart-Smith, R. D., Willis, T. J., and others. (2014). Global conservation outcomes depend on marine protected areas with five key features. Nature, 506, 216-220. pubmed.ncbi.nlm.nih.gov
4. Grorud-Colvert, K., Sullivan-Stack, J., Roberts, C., and others. (2021). The MPA Guide: A framework to achieve global goals for the ocean. Science, 373(6560), eabf0861. pubmed.ncbi.nlm.nih.gov
5. UNEP-WCMC and IUCN. (n.d.). Protected Planet: Marine protected and conserved areas (statistics read 3 November 2026). protectedplanet.net
6. Marine Conservation Institute. (n.d.). MPAtlas: The Marine Protection Atlas (assessment of 30 October 2026, read 3 November 2026). mpatlas.org
7. Convention on Biological Diversity. (2022). Kunming-Montreal Global Biodiversity Framework, Target 3. cbd.int
8. United Nations. (n.d.). Agreement on marine biological diversity of areas beyond national jurisdiction (BBNJ Agreement) (read 3 November 2026). un.org
9. Orth, R. J., Lefcheck, J. S., McGlathery, K. S., and others. (2020). Restoration of seagrass habitat leads to rapid recovery of coastal ecosystem services. Science Advances, 6(41), eabc6434. pubmed.ncbi.nlm.nih.gov
10. NOAA Fisheries. (n.d.). Oyster reef restoration in the Chesapeake Bay: We're making significant progress. fisheries.noaa.gov
11. Jones, H. P., Holmes, N. D., Butchart, S. H. M., and others. (2016). Invasive mammal eradication on islands results in substantial conservation gains. Proceedings of the National Academy of Sciences, 113(15), 4033-4038. pubmed.ncbi.nlm.nih.gov