In a time of shared scientific sovereignty

On June 2nd, just before the 3rd United Nations Ocean Conference (UNOC3), a team of scientists including CNRS researchers, launched an appeal to guarantee the sustainability of the world's largest dedicated global ocean observation programme - the international One Argo programme. This project involving nearly 30 different countries particularly relies on real-time temperature and salinity measurements collected by 4000 profiling floats posted across the world's oceans: "The data they collect are essential for weather forecasting and monitoring rising sea levels. The information obtained is also used to forecast cyclones. This programme is certified by the World Meteorological Organisation," explains Hervé Claustre, a CNRS research professor working at the Villefranche-sur-Mer Oceanography Laboratory and co-head of this international programme's biogeochemical component (BGC-Argo). 

One Ocean, One Argo

One Argo is operated for the most part by the National Oceanic and Atmospheric Administration (NOAA), the American agency for ocean and atmospheric observation. Mr Claustre points out that "the United States contributes 50% of the funding, with Europe only contributing 25%". However, NOAA's budget is likely to be significantly reduced in the months to come and the oceanography researcher laments the fact that "this programme has entered a critical phase because what's going on in the United States is destabilising the entire observation network just when it's becoming vital for us to have access to all this data to help inform political decision-making as regards changes in the ocean". The programme's international structure and the lifespan of the floating robots (which remain operational for 5 to 6 years) give the One Argo programme a certain resilience but "we can't yet measure the harmful effects American disengagement could have," opines Hervé Claustre. There is still a substantial level of investment as the oceanography researcher estimating that "on the international scale, nearly €100 million needs to be invested every year to finance new floats and make the data available to the whole of the scientific community". This is why the scientists working on the One Argo programme co-authored an article published in the Frontiers in Marine Science journal on June 2nd 1  which advocates for "a shift from project-based funding to a sustainable, institutionalised model like for meteorological observation systems which have been supported by stable public and private funding for a long time". Hervé Claustre thinks it particularly necessary for France to increase its support for this programme. "Our country has the second largest maritime domain in the world and is very active in the governance of One Argo. Also, it's one of the only countries, along with the United States, that has an industrial stakeholder with expertise in the technology used to produce the floating robots used in the framework of this programme. France can thus contribute to the resilience of One Argo so we've submitted a request for additional funding to the ministry, our supervisory authority". 

Maintaining this programme is considered crucial for basic research but it also benefits many private stakeholders. The data collected is shared openly and contributes to maritime safety, weather and climate forecasting, the sustainable management of marine resources, understanding the resilience of ecosystems, or effectively adapting to extreme events. "We will be collaborating with economists to try to evaluate the overall value our observation systems provide for society. The idea is to highlight the scale of the indirect benefits the One Argo programme generates and which are potentially considerable," explains Hervé Claustre. "We could also imagine a system in which the different companies that use our data to sell services could also contribute to funding the observation systems". 

• 1In an article published in the Frontiers in Marine Science journal, the authors stress that, given the uncertainties surrounding international funding, "sustained and reinforced financial support are urgently needed to enable full implementation of One Argo and give our societies the means to preserve the many services that the ocean provides".

"Open science appears to be a real tool for resilience"

The storage and sharing of data and information remain key issues because dependence on databases hosted in the United States is now the subject of considerable concern among the research community. "Some systems will reach the limits of viability even though they host and distribute crucial data, some of which was acquired in cooperation with France and Europe," points out Jean-François Doussin, the deputy director of CNRS Earth and Space. "Also, several European databases rely on mirror systems installed in the United States," adds Sylvie Rousset, director of the Open Research Data Department (DDOR) at the CNRS. So how can we preserve this valuable information? "Scientific data and their backup are a crucial issue. Policymakers at the European level are well aware of this and are trying to find solutions," says Jean-Stéphane Dhersin, director of the CNRS's representative office in Brussels. Would it be possible, then, to consider moving some of the data previously stored in American databases to Europe? "The French research infrastructure Data Terra already hosts a vast amount of data in the fields of climate science, meteorology, oceanography, atmospheric science and biodiversity. This could perhaps be an excellent candidate for hosting American data threatened with deletion in these research sectors," suggests Jean-François Doussin but this would call for a sixfold increase in Data Terra's storage capacity - from 100 to 600 petabytes. "At CNRS Earth and Space, we estimate that an operation like that would cost about €17 million which is not an extraordinary amount given what's at stake. This is strategic data in both the fight against global warming and for adapting to this change," explains the Institute's deputy director. The Data Terra infrastructure also has another advantage. At the start of 2025, the Institute integrated the European Open Science Cloud (EOSC), the future European portal that provides access for all scientists to data and software. The data that would be repatriated to France could then be openly shared with the entire scientific community, with Sylvie Rousset indeed stressing the fact that this means "the use of open science appears to be a real tool for resilience". The HAL open archive library was developed on the initiative of the CNRS and also enables researchers from all over the globe to submit their manuscripts online, explains the DDOR's director. "If scientists find it difficult to publish in their own country, they can always communicate their discoveries via a platform like this". 

Software Heritage - one of the first building blocks for strategic independence

Moving beyond the issues of publications and data, it is also crucial to also preserve the free software that underpins all our digital systems. "It's essential to preserve data, but if you don't have the software required to process and analyse them or the software that created or collected them, then that's not much use," emphasises Roberto Di Cosmo, professor of computer science at the Institut de Recherche en Informatique Fondamentale (IRIF - CNRS/Université Paris Cité) who has been seconded to the National Institute for Research in Digital Science and Technology (Inria) as the director of the non-profit organisation Software Heritage. This UNESCO-backed initiative aims to collect, preserve and share all the publicly available software source codes throughout the world. "Since we began collecting in 2015, millions of projects have vanished, particularly after the closures of the major platforms that hosted them. Fortunately, we were prepared for this so we managed to save nearly 1.5 million of them," explains the computer science researcher. The underlying principle is simple as Mr Di Cosmo explains. "We collect the publicly available code we discover and anyone can launch the archiving of interesting code to make sure it can be found when needed. The Software Heritage's central node is in Paris but we also have independent copies, or mirrors, in Italy and Greece". Indeed the French President, Emmanuel Macron, highlighted this strategic archiving when he spoke at the 'Choose Europe for Science' event in Paris on May 5th. Roberto Di Cosmo observes that our dependence on American hosting systems in this area is actually immense. "Three-quarters of the code contributed by the French academic research community is hosted by the GitHub platform based in the United States". The computer science researcher considers that, in this context, "Software Heritage, which already possesses 26 billion source files, could be one of the building blocks for a resilient infrastructure that will guarantee the strategic independence of our research". 

In reality, this fight to preserve knowledge is only just beginning, warns Jean-François Doussin and indeed the European Union has promised to increase its investment in scientific research as a response to this situation. On May 5th, the Commission President Ursula von der Leyen affirmed the EU Member States' commitment to reaching "the target of 3% of GDP for investment in research and development by 2030". Alongside this, €500 million will be made available for the 2025-2027 period in the framework of the 'Choose Europe for Science' initiative aimed at attracting scientists from all over the world to the 'Old Continent'. "We possess excellent research infrastructures in Europe, including at the CNRS, and this is a real asset for convincing the best researchers from other countries to come and work here," emphasises Jean-Stéphane Dhersin. Europe and France will have a role to play in keeping global scientific cooperation afloat in the years to come.