The Trump administration plans to dismantle a $368m deep-sea observation system that has for more than a decade provided crucial data on ocean systems and climate change.

In a notice, the National Science Foundation (NSF) announced that it had “initiated descoping of the Ocean Observatories Initiative” (OOI), a vast ocean observation network comprising more than 900 instruments that collect data on ocean health, including current patterns, climate variability and marine biodiversity.

The notice, issued on 21 May, came just days after Trump fired all members of the independent board that oversees the NSF.

A statement by NSF head of media affairs, Mike England, said the program was not being cancelled entirely and described the plans as a “descope”, or reduction of elements, though it was not clear what data collection capacity would be left.

The NSF’s notice described plans to remove “all in-water infrastructure” from observation sites off the coasts of North Carolina, Oregon, Washington and Alaska, as well as from the Irminger Sea, a marginal sea between Greenland and Iceland.

Some scientists expressed dismay at the plan, while Democratic lawmakers said they would fight it, including Senator Chris Van Hollen of Maryland, who called it a “shortsighted move” that would “end up costing American taxpayers more not less”, the New York Times reported.

Senator Sheldon Whitehouse, a Rhode Island Democrat, said on X: “Fossil fuel is heating our oceans by the zettajoule, so Trump’s corrupt fossil fuel stooges want to turn off the monitors.”

Following the announcement, the OOI’s principal investigator, Jim Edson, said the NSF’s plan involves a phased recovery and infrastructure removal process expected to take place over the next 15 months. “As infrastructure is recovered from each array, the associated real-time data streams and observing capabilities at those locations will come to an end,” Edson said.

The move will bring to an end more than a decade of continuous ocean monitoring after the system first became operational in June 2016.

Describing the network as having “delivered the world’s most advanced continuously operating ocean observing systems”, Edson added: “We are profoundly grateful for the extraordinary efforts of the scientists, engineers, operators, educators, students and partners who made this facility possible and who continue to advance its legacy through the use of its data.”

The dismantling of the OOI marks another step in the Trump administration’s rollback of science and climate initiatives. It also follows Trump’s push to expand deep-sea mining and loosen fishing regulations, a policy that has alarmed ocean scientists and climate experts.

Hilary Palevsky, a professor focusing on marine biogeochemistry and oceanography at Boston College, pointed to the significance of the data that will be lost, particularly given the sophisticated engineering required to deploy and maintain the instruments.

“One of the real powers of this OOI and a lot of the collection of autonomous data is that scientists like me don’t have to have the expertise or the resources to be able to deploy this kind of infrastructure ourselves,” Palevsky said. “Being able to have instruments, both actually out in the atmosphere floating in the surface ocean, as well as surviving through the really deep mixing and waves in the subsurface.”

She said: “Over the more than 10 years that these things have been deployed, they’ve just gotten better and better at it. And so the data return has also gotten better and better over time … the scientific community was really just getting to the point of being able to capitalize on the data that had been collected so far … I’m really disappointed for the continuation of this important data set.”

Palevsky also warned that rebuilding such a network in the future would be difficult, saying: “If we want to put [the instruments] back out again, we need people who know how to do it and the team that knows how to do it is being dismantled along with the infrastructure program itself.

“We’re potentially at risk of having a gap in our ability to regain the expertise to do things that we had sort of just figured out how to pull off.”

For Palevsky and her students, OOI data has helped shed light on biological production in the ocean and its role in carbon sequestration – the process by which carbon dioxide is captured and stored – as well as deep-ocean processes, marine ecosystems and fisheries.

Data from the OOI has also contributed to research on the Atlantic Meridional Overturning Circulation (AMOC), a critical system of Atlantic Ocean currents that studies suggest may be more vulnerable to collapse than previously thought, with potentially severe consequences for the global climate.

“One of the important processes in the AMOC is what we call convection, this really deep mixing of surface waters into the deep ocean that happens in winter, basically driven by the surface ocean getting really cold because the atmosphere gets super cold in winter and big, windy storms blow across the surface ocean,” Palevsky said.

“We have gained some really important insights into both how that happens in the Irminger Sea in particular, and how the drivers of that process vary from year to year from the observations that have been gained at this site,” she added.

For scientists like Palevsky, the consequences of dismantling the OOI extend far beyond ocean researchers, particularly as climate change intensifies extreme weather events around the world.

“As we reduce the amount of data that we have, the observations, as well as the science more generally to understand what’s happening in the climate system, it makes it much harder for us as a society to understand what we’re facing and what we need to do to plan for and adapt to it,” she said.

In a statement to the Guardian, the NSF head of media affairs, Mike England, said the program was not being cancelled entirely: “The NSF is not cancelling the Ocean Observatories Initiative. The decision to descope aligns with NSF’s wider strategy of a nimbler approach to prioritize support for evolving scientific priorities and emerging technologies, as well as smart lifecycle management within its research infrastructure portfolio.”