Where does the ocean’s plastic go?

A 2014 study synthesised net tows from a global circumnavigation and from other expeditions to produce a map of floating plastic and a first estimate of how much is there.

Debris concentrates in the five subtropical gyres, as ocean circulation models predict. That study put the load between 7,000 and 35,000 tonnes, for pieces from 0.2 to 100 millimetres in the Atlantic, Pacific and Indian oceans.

It is not the only figure. A 2015 synthesis standardised more than 11,000 net tows and ran three circulation models over them, and got 93,000 to 236,000 tonnes. A 2023 time series from 11,777 stations put the load in 2019 at 1.1 to 4.9 million tonnes, rising steeply since 2005, though its masses rest on one average weight per particle. The surface estimates now span nearly three orders of magnitude.

The other side of the accounting was estimated in 2015, by linking data on solid waste, population and economic status across 192 coastal countries.

The figure for a single year, 2010, came out between about 4.8 and 12.7 million tonnes. Population and the quality of waste management largely decided which countries contributed most, and without better waste management the cumulative amount available to enter the sea was projected to grow tenfold by 2025.

Set against the floating estimates of 2014 and 2015, the gap was not a rounding error. The 2015 synthesis put the standing stock of floating microplastic at around one per cent of a single year of input. Its authors noted that a stock and an annual flux are not the same quantity, and that the input counts plastics that never float, but judged that large debris outside the nets was unlikely to close two orders of magnitude. A 2023 model, with the 2015 synthesis’s lead author among its authors, found that items larger than 25 millimetres hold more than 95 per cent of the buoyant mass.

Plastic at sea breaks into smaller pieces under sunlight and mechanical stress, so the size distribution should keep rising towards the small end. In the 2014 surface samples it does not: abundance peaks at fragments of about two millimetres and falls away below one.

That is a specific and informative absence. Something removes the small fragments from the surface faster than they are produced.

The candidates are that fragments sink once fouled by organisms, that they are eaten by animals and carried downward in faeces or in bodies, that they wash back onto shores, and that they degrade into pieces too small for the nets to catch.

In 2014 none had been shown to account for the bulk. The team that listed them judged a rigorous attribution not yet possible, and the washing ashore of millimetre fragments already trapped in the central gyres unlikely.

Two later lines of work each claim to close the gap. Extrapolating from samples taken across the Atlantic in 2016, one team put just three polymers, in particles of 32 to 651 micrometres, at 11.6 to 21.1 million tonnes in the top 200 metres, and concluded that the input estimates were too low. Extrapolating from one transect, a 2025 study put nanoplastics, smaller than a micrometre, at about 27 million tonnes in the mixed layer of the temperate and subtropical North Atlantic.

From the other end, a 2021 reanalysis argued that river fluxes had been overestimated by two to three orders of magnitude. A published comment called its conclusion fundamentally flawed, and the authors replied defending it. The 2023 model, assimilating surface, beach and deep-ocean data, put the input of buoyant plastic near half a million tonnes a year and concluded there is no missing sink. One account says the input was too high, the other that it was too low.

Which is right changes the hazard. Plastic sealed into sediment, large debris that lingers at sea for years, and nanoplastic spread through the water column do not reach living things in the same way.

Entanglement and ingestion are documented directly. A 2015 review of 340 studies recorded encounters between marine debris and 693 species, and plastic accounted for 92 per cent of the encounters. Effects at the level of populations, and the effects of the smallest particles on physiology, are far less clear.

A 2016 systematic review of the field counted 366 perceived threats from marine debris. Most of the ones that had been tested were demonstrated, but nearly nine in ten of those sat at the level of molecules, cells and tissues rather than whole animals or communities. Distinguishing what is established from what is assumed is unusually hard here, because the subject arrives already attached to a conclusion.