Major ocean database that will guide deep-sea mining has flaws, scientists warn

A company is expected to request authorization in July, for the first time ever, to mine the ocean floor for metals such as cobalt and nickel. At the same time, researchers warn that a crucial database that maps deep-sea biodiversity and that could factor into the decision to approve such a licence contains errors and data gaps.

The International Seabed Authority (ISA), a body associated with the United Nations that oversees deep-sea mining in international waters, currently allows only mining exploration. According to its website, it has approved 17 companies and government entities to study the mining potential of the Clarion–Clipperton Zone (CCZ), a region of the sea floor that spans up to 6 million square kilometres of the central and eastern Pacific Ocean and that holds metal-rich clumps of sediment. Nauru Ocean Resources, a subsidiary of The Metals Company, based in Vancouver, Canada, has been exploring the sea bed, with an eye towards gathering metals needed for electric-vehicle batteries and other electronics. It plans to apply for a commercial mining licence in a month or so. If approved, operations could begin in 2024.

Scientists worry about allowing companies to start mining the sea bed because little is known about deep-sea habitats and biodiversity, so its environmental effects are unpredictable.

The ISA runs a database called DeepData, which is meant to tackle some of these concerns, as well as to enable research projects. The database contains information that the ISA requires contractors to collect during their deep-sea exploration missions. These biological, geochemical and physical data include, for example, the species that they encounter, and the chemicals present in the water.

But the analysis of DeepData, published in the journal Database on 30 March¹, revealed flaws that worry the researchers who conducted the study.

“It strikes me as irresponsible to be relying on the database in its current form” to assess the impact of mining on the sea-floor environment, says Muriel Rabone, a data scientist at the Natural History Museum in London, who led the analysis. Rabone told Nature that the analysis was performed independently of the ISA, but that the agency cooperated to enable data access. It was also consulted on the scope of the study and an early draft of the manuscript.

The ISA protests some of the findings, however, saying that the report is out of date. The researchers downloaded data collected in the CCZ on 12 July 2021 to run their analysis. Since then, the ISA has made “significant improvements” to address quality assurance and control issues with DeepData, it says.

Responding to this criticism, Rabone maintains that the database still contains flaws. Even with its faults, it’s helping to point to thousands of species on the sea floor that had never been seen before — results published just this week. “There is work to do yet,” she says.

Data gaps

Of the 40,518 records that the researchers analysed for the Database study, about one-quarter were duplicates, which could lead to an underestimation of species richness in the deep sea, they say. The scientists think duplicates can arise partially because the database lacks unique codes to identify individual records.

The ISA says that, like any other database, DeepData’s “features and the quality of its data are improving with the years due to technological advances”. It adds that it has identified and corrected duplicate records. Also, it is collaborating with the World Register of Marine Species, which catalogues and classifies marine organisms, and is sharing data with the Ocean Biodiversity Information System — a data hub that has helped to clean up the data and make it more widely available.

Looking at the database today, however, Rabone says that some duplicate data still exist, and that many records still do not have a unique identifier.

The team also found that DeepData contained inconsistent information — for instance, records that catalogued two species under the same name. And a lot of environmental data were missing. When contractors submit their data, they use a form with fields such as species name and fauna class size. The researchers found that 90% of the total data in various fields were missing.

The ISA says it has already updated its forms to address some of these issues and is designing workshops and training for contractors to ensure that data quality and control are improved.

Rabone would like the workshops to be open to the scientific community, which she says can provide feedback on the database. Stefanie Kaiser, a deep-sea ecologist at Senckenberg Research Institute in Frankfurt, Germany, who was not involved with the study, agrees, and says that if the database were improved, it could be useful for researchers, giving them access to all the information collected by the contractors.

But the ISA says workshops are for contractors only, because they provide the data, although it acknowledges that the academic community has assisted contractors with presentations and preparing annual reports.

Despite the disagreements over DeepData, researchers are already learning from the database. Rabone formed an official partnership with the ISA to lead the first census of metazoan biodiversity on the CCZ’s sea floor. The endeavour found more than 5,500 species in the region, of which 92% are new to science, including many worms and arthropods. The findings were published on 25 May in the journal Current Biology².

Source: Ecology - nature.com