Solar geoengineering may not halt ocean warming, study says

Spraying aerosols high in the stratosphere could dampen global warming over land, but may not prevent the oceans from heating up, new research says.

The findings suggest that this type of “solar geoengineering” – a set of techniques that aim to tackle global warming by reflecting sunlight back into space – may not necessarily stem sea level rise or prevent damage to the world’s marine ecosystems.

The research indicates that solar geoengineering could carry “major uncertainties and risks”, the lead author tells Carbon Brief.

The study also raises the issue of whether global average temperature “is the best metric to control” when addressing the impacts of climate change, another scientist tells Carbon Brief.

Harnessing a volcano

Solar Geoengingeering options. Graphic by Ros Pearce for Carbon Brief

Scientists have suggested that releasing aerosols into the atmosphere – a technique known as “stratospheric aerosol injection” – could cool the planet in a similar way to a large volcanic eruption.

When a volcano erupts, it sends an ash cloud high into the atmosphere. The sulphur dioxide released in the plume combines with water to form sulfuric acid aerosols, which reflect away incoming sunlight, temporarily cooling the Earth.

Artificially introducing aerosols into the atmosphere – via a plane or a high-altitude balloon – could have a similar cooling effect, researchers say.

The idea has never been tested, but previous research using computer simulations suggests that releasing aerosols could help limit global temperature rise to 1.5C – the aspirational target of the Paris Agreement – and keep rainfall from becoming irregular.

However, an aerosol sunshade would not protect the planet from rising CO2 emissions – which is causing oceans to become more acidic and crops to become less nutritious, among other problems.

The new study, published in Nature Geoscience, identifies, for the first time, another potential downfall of the proposed technique: it may not be able to limit warming in the deep ocean.

This is because aerosol release could cause rainfall to decrease in some regions, which could influence ocean circulation patterns, says lead author Dr John Fasullo, a project scientist at the National Center for Atmospheric Research (NCAR), Boulder. He tells Carbon Brief:

“The explanation involves two key stages. In the first, the aerosols reduce the amount of rainfall that occurs globally. This change in rainfall is not uniform, rather, reductions are larger in some key regions – namely, the northern Atlantic Ocean – and this increases the salinity of those ocean areas, making the water more dense.”

This increase in water density could cause the “Atlantic Meridional Overturning Circulation” (AMOC) to speed up, he says. The AMOC is a perpetual conveyor belt that transports heat from the equator up to the North Atlantic. It is part of a wider network of global ocean circulations patterns that transports heat all around the world. Fasullo says:

“The North Atlantic is a key region of deep-water formation and higher density therefore leads to an acceleration of this circulation. The speed up in deep-water formation buries more heat in the deep ocean that would have otherwise occurred.”

Other worlds

To explore the impact of aerosol release on oceans, the researchers used a set of 20 computer simulations looking at what would happen if aerosols were released between 2020 and 2099. This set of simulations is known as the “geoengineering large ensemble” (GLENS).

The simulations assume that aerosols are released at uniform points across the globe. (Previous research shows doing this would prevent large regional changes in temperature and weather.)

The researchers compared their geoengineering simulations to another set of simulations where no geoengineering takes place and little action is taken to tackle climate change, leading to high emissions in the coming decades (a scenario known as “RCP8.5”).

The chart below shows how global average temperature (full lines, left-hand y-axis) and global rainfall levels (dashed lines, right-hand y-axis) are expected to differ between the high-emissions scenario (RCP8.5) simulations and GLENS.

Changes to global mean temperature (full lines, left-hand y-axis) and global rainfall levels (dashed lines, right-hand y-axis) under a scenario with solar geoengineering (GLENS; blue) and a scenario with no geoengineering and high greenhouse gas emissions (RCP8.5; red). Source: Fasullo et al. (2018)

The chart shows how, in the geoengineered world, the global average temperature is kept constant while global rainfall levels fall. In comparison, temperatures and rainfall levels continue to rise under the high-emissions scenario.

The maps below show expected changes to ocean heat content (OHC) between 2010-30 and 2075-95 in the upper ocean (0-300m; left) and in the full-depth of the ocean (right) under the geoengineering scenario. On the maps, red shows increases in OHC, while blue shows decreases.

Expected changes to ocean heat content (OHC) between 2010-30 and 2075-95 in the upper ocean (0-300m; left) and in the full-depth of the ocean (right) under a solar geoengineering scenario. Red shows increases in OCH, while blue shows decreases. Source: Fasullo et al. (2018)

The charts show how, in the geoengineering scenario, some shallower parts of the ocean are expected to heat up. This includes the Arctic Sea and waters off the western coast of North America.

When the full depth of the ocean is considered, however, many more regions are expected to see OHC increases – with warming likely to be most severe in the Arctic Sea and in the Southern Ocean.

It is worth noting, however, that increases in heat content in the full depth of the ocean were found to be considerably higher under the scenario with high greenhouse gas emissions and no solar geoengineering.

Rising tides

The findings suggest that – contrary to previous research – solar geoengineering could fail to prevent sea level rise.

Sea level rise is chiefly driven by the expansion of seawater as a result of warming and by the melting of land ice at the world’s poles. In some parts, including in Western Antarctica, melting occurs where land ice comes into contact with warm ocean water. Fasullo says:

“Relative to present day, it is extremely likely that sea level will continue to rise under our geoengineering approach, but it is also likely that the rate of this rise will be less than under a business-as-usual climate trajectory [RCP8.5].

“We cannot say with confidence what that rate of rise might be as we don’t have models of the ice sheets fully coupled to our climate simulations, nor have we sufficiently vetted such models to have confidence in their projections at this point.”

The results could also mean that solar geoengineering would fail to prevent marine heatwaves, which pose a major threat to undersea wildlife, and the expected increases in hurricane intensity. (Although no research has looked at these issues, specifically, so it is still too soon to draw conclusions, Fasullo says.)

The study raises the issue of whether global average temperature “is the best metric to control” when addressing the impacts of climate change, says Prof Govindasamy Bala, a physical scientist from the Divecha Centre for Climate Change at the Indian Institute of Science, who was not involved in the study. He tells Carbon Brief:

“However, as this is a single modelling study, it is not clear to me if the results of this paper are robust. A multi-model based approach would be needed to verify the robustness.”

The findings imply that solar geoengineering could have “large potential downsides”, says Fasullo:

“There have been various high-profile policymakers in the US that have framed climate change as an engineering challenge. I think the key takeaway from our work is that geoengineering itself carries major uncertainties and risks.”