Should we fertilize oceans or seed clouds? Nobody knows

The climate clock is turning. To reject it, the whole world trusts "negative emission technologies". These would suck the carbon dioxide out of the atmosphere and lock it up for centuries on the earth, in the sea or under the seabed (see "Marine Geoengineering"). Although these technologies have not yet been developed, they are nonetheless implicit in the assessments of the Intergovernmental Panel on Climate Change (IPCC). Limit warming to 1.5 ° C from pre-industrial levels, no less than 20 billion tonnes (gigatonnes) of CO₂ may need to be removed from the atmosphere each year until 2100¹.

Carbon storage in the oceans seems promising for some. Oceans are vast and there may be fewer political compromises to make than land. For example, fertilizing water with iron would accelerate the growth of phytoplankton and thus absorb CO₂, some of which would sink into the deep ocean in the form of carbon when the organisms died. Another proposal is to spray seawater in the air to help form clouds that reflect sunlight and cool the planet.

Such techniques should be used on a large scale to limit global warming to safe levels. For example, to mop up CO₂ chemically, it would be necessary to sprinkle a billion tons of powder minerals similar to chalk throughout the Pacific Ocean. And these measures may be needed in a decade if emissions cuts fail and policymakers exert increasing pressure to act.

Little is known about the consequences. Little research has been done for a variety of reasons. The controversial nature of geoengineering divides researchers. And some funded research trials were subsequently dropped due to the lack of rules for their execution and conflicts of interest, such as patent applications (see Nature http://doi.org/hw2; 2012). Even the basic tests of the equipment have not been carried out. Most of the preliminary studies have not been published in peer-reviewed academic journals.

This shortage of information hampers the development of a global framework to regulate geoengineering research, despite more than a decade of debate. Researchers and decision-makers need to know which technologies with negative emissions deserve to be explored and which ones will never work or are too damaging to be pursued. The potential benefits and risks of technologies need to be established before leaders or companies in countries decide to implement them prematurely.

As a first step, the United Nations Panel on the Scientific Aspects of Marine Environmental Protection (GESAMP) established a working group in 2016, which we co-chaired, to examine the ecological and potential social impacts of various marine ecosystems. geoengineering approaches. Its mandate was to advise the International Maritime Organization (IMO) and the parties to the 1996 Protocol to the Convention on the Prevention of Marine Pollution by Dumping of Waste and Other Matter, 1972 (London Protocol of 1996). Our report was published in March 2019². However, we struggled to identify the techniques requiring regulation because of the poor evidence available. In the end, all we could do was to classify marine geoengineering research as too inadequate or too incomplete to allow scientific evaluation.

We call here defenders of geo-engineering, from research to trade, to build a body of basic scientific evidence in the next three years. This would allow decision makers to decide which methods to eliminate and which ones have potential. Knowledge and regulation in geoengineering must progress at the same time.

Uneven progress

The GESAMP task force has examined 27 marine geoengineering ideas – from adding reflective foams to the ocean's surface to the burial of carbon beneath the seafloor.². Most research is in its infancy. Some concepts, such as the deposition of crop waste on the sea floor, have made little progress beyond the theoretical experiments. Some have been studied in the laboratory or modeled on a computer. Less than ten pilot studies were conducted in the field.

Critical research gaps abound. For example, the idea of ​​using a fine spray of seawater or other aerosols to thicken or seed clouds over the ocean – similar to the tracks created by ship emissions – has been widely reported in the media. Nobody reported attempting to spray fine droplets of natural seawater (although lab experiments were conducted on manufactured saltwater³). But the seawater is full of tiny organisms and organic matter that could clog a sprayer (as reported in the GESAMP report).² by John Cullen, working group member, oceanographer at Dalhousie University in Halifax, Canada).

Lack of funding is not the main reason for gaps in research. Although few funding programs have been targeted at experiments and modeling in marine geoengineering, many basic tests are inexpensive and can be done in the laboratory – for example, assessing whether impurities in mineral powders are toxic to the environment. marine life. In addition, a range of negative-release technologies, such as improved rock alteration to increase ocean alkalinity, are already being funded through targeted research programs, including in the United Kingdom. Other areas of research, such as modeling, are under way in Germany and a call for research proposals has been launched in Japan. Private funds are being invested in some marine approaches, such as a proposed pilot study on the impacts of iron fertilization on fisheries off Chile. However, this project is down, largely due to the lack of support from scientists (see Nature 545393-394; 2017).

Another problem is that many geoengineering proposals and analyzes are on temporary websites, not in peer-reviewed journals. For example, only half of the websites refer to ideas, plans and documents cited in a detailed 2009 synthesis study of marine geoengineering approaches.⁴ was still in operation when we examined them in 2018. On the other hand, the academic and intergovernmental documents of this period are easy to find.

Again, the reasons are unclear, but may include underfunding, confidentiality concerns about disclosure of method details, and maintenance of proprietary rights over technologies. Some scientists fear that even launching geoengineering research or publishing results will lead to under-researched approaches.⁵.

However, it is essential that the research be thoroughly researched, discussed openly and made readily available, as shown by the most studied geoengineering approach, iron fertilization in the field. ocean. Much of the work is based on ocean biogeochemistry and includes laboratory experiments, pilot studies in the Southern Ocean and modeling across ocean basins. All this activity showed that the method would not work as expected⁶. Fertilization of 1,000 square kilometers of the upper ocean would increase phytoplankton growth but could have alarming side effects. For example, flowing algae could release methane, a much more potent greenhouse gas than CO₂.

This body of research convinced decision-makers to intervene. In 2008, parties to the United Nations Convention on Biological Diversity voluntarily decided to prevent large-scale iron fertilization experiments in the oceans without scientific risk assessment, control and regulation ( see Nature 453704; 2008). In 2013, the London Protocol added legally binding amendments to regulate the fertilization of iron in the oceans, but they have not yet entered into force.

No other method of marine geoengineering has been advanced enough to impose similar regulations. However, techniques such as alkalization of the oceans will likely have large-scale impacts similar to those of iron fertilization.². The models highlight potential problems with other methods: simulations have shown, for example, that pumping cold waters rich in nutrients to the Pacific surface could lower the temperature of the overlying air and help to cool the planet, but at the expense of creating a large oxygen-depleted area that would threaten the fishery⁷.

The path to follow

As climate hazards increase, marine geoengineering needs a body of evidence to guide research and regulation. We suggest that the United Nations Framework Convention on Climate Change or the International Council for Science play a leading role in encouraging four steps.

First, to establish the foundations, researchers must identify and address key knowledge gaps.². Many basic questions can be answered with minimal funding in the laboratory and without new legislation. For example, will mineral powders make sea water less transparent or will they enter the food chain, as tiny pieces of plastic have made? How much will reflective foams on the surface reduce photosynthesis? How long will the moss last and how will the wind affect its spread?

A shared bank of standard ocean models, similar to those used by the IPCC to ensure a consensus between climate models, should be created through initiatives such as the Carbon Dioxide Model Comparison Project. (see go.nature.com/2wevzhj). Linkages should be strengthened between ocean modelers and related research communities such as SOLAS (Ocean and Low Atmosphere Surface Study), which organized a workshop in April on this topic in Sapporo, Japan.

All results must be published in archived journals and repositories. Standards need to be developed to report the details of the methods so that studies can be repeated or compared under different conditions.

Second, the potential advantages and disadvantages of each geoengineering method must be evaluated and ranked. Such a portfolio, updated as the evidence accumulates, should be held by an international organization such as GESAMP. Benefits should include evidence of effective and permanent removal of CO₂ with minimal side effects. Disadvantages may include unforeseen difficulties in the extension of technology and unintended adverse consequences, such as increasing populations of some toxic phytoplankton species through the fertilization of iron by the oceans.⁸.

Third, researchers and policy makers should develop scientific criteria for assessing risks and designing a set of tests that experiments must pass before they are allowed. These should encompass the expected and unintended consequences of the work, as well as the spread of risk, as more ambitious research is conducted – for example, when a pilot study in a laboratory flask progressively moves towards Other enclosed environments (such as a large volume environment). incubator) and then at sea. Policymakers will need to select methods that merit further consideration and those that should be considered unachievable or unacceptably risky.

Fourth, if these tests are successful, research and regulation must be conducted in parallel. The London Protocol is a good starting point for the management of interventions in the oceans. Researchers and policy makers will need to develop an adaptive framework to gather, evaluate and process evidence for all candidate geoengineering approaches, including marine methods. Research governance must be based on a wide range of outcomes. Resources can then be targeted to the most promising areas.

Adaptive governance forms focused on responsible research and innovation⁹ have been applied to other emerging technologies, such as synthetic biology and nanotechnology^ten. A similar approach to geoengineering will allow scientists to propose a subset of scientifically valid approaches that can be examined from a legal, socio-economic and geopolitical perspective over the next few years.