Auditing marine survey evidence for submerged prehistoric landscapes on the Spanish Gulf of Cádiz shelf
This paper audits marine survey archives on the Spanish Gulf of Cádiz shelf to demonstrate that while existing data supports local palaeolandscape interpretation and exposed-object verification, it lacks the necessary metadata, geometry, and controlled outcomes to definitively rule out the presence of preserved prehistoric contexts across the region.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
The ocean floor is not a blank slate; it is a layered archive of human history. When sea levels were much lower thousands of years ago, during the last ice age, vast stretches of what is now the deep sea were dry land. These exposed areas hosted rivers, wetlands, and coastlines where people lived, hunted, and traveled. As the ice melted and the oceans rose, these landscapes were submerged, burying the evidence of prehistoric life under layers of sand and mud. For decades, scientists have searched these underwater plains, hoping to find the lost settlements of our ancestors. The challenge is not just finding the old land, but proving that the ground has been looked at closely enough to say with certainty that nothing is there. If a map shows a line where a ship passed, it does not mean the seabed was scanned for ancient tools. To know if a prehistoric landscape is truly empty, researchers must verify that the equipment used was capable of seeing deep enough, that the path taken covered the ground thoroughly, and that the data collected was actually checked for signs of human life.
A team of researchers from the University of Seville set out to test exactly how much of the Spanish Gulf of Cádiz shelf has been truly examined for these lost prehistoric worlds. They did not go out to the sea to dig or scan new ground. Instead, they acted as auditors, sifting through a massive collection of existing records from ships, ports, and scientific institutions. They gathered twenty-one distinct groups of data, ranging from sonar maps and seismic lines to core samples and diver reports. Their goal was simple but rigorous: to determine if any of these existing records could prove that the seabed had been searched well enough to rule out the presence of preserved ancient sites. They treated the data not as a simple list of "surveys done," but as a chain of evidence. Each record had to pass a series of strict checks: Was the data clear enough to see the target? Did the ship's path cover the area in a way that would catch a buried site? Was the equipment sensitive enough to find what they were looking for? And finally, did the results explicitly state that no ancient remains were found under conditions where they should have been visible?
The results of this audit were stark. While the region is covered in a dense web of maritime activity and scientific data, none of the twenty-one source groups passed the full chain of requirements. Seventeen of the groups had clear records of what they were, but when the researchers looked at the actual paths the ships took, eight failed to show a pattern that would guarantee a thorough search. Only two groups had data detailed enough to interpret the shape of the seabed, but even these fell short of the specific goal. The primary target was not just any old landform, but a specific type of buried ground from the last ice age that could hold evidence of human occupation. None of the records matched this target perfectly. Most critically, every single group failed the final test: none provided a "negative inference." In plain terms, this means that no existing record allows scientists to say, "We looked here, we had the right tools, and we found nothing, so this area is definitely empty of prehistoric sites."
The study did find that the data is useful for other things. One specific record from a location called El Saladillo successfully imaged a buried river channel from the last ice age, proving that the equipment used there could see deep into the mud. Another set of records in Cádiz Bay helped divers locate modern objects like chains and concrete blocks on the surface. These findings are valuable for understanding the shape of the ancient landscape and for verifying what is visible on the seabed today. However, they do not prove that the wider shelf is free of hidden prehistoric villages. The data shows that the shelf has been mapped for navigation and geology, but it has not been systematically audited for archaeology. The researchers found that the "gaps" in knowledge are not necessarily empty spaces where no one has looked, but rather areas where the existing data is too vague, too incomplete, or too focused on modern objects to tell us about the ancient past.
This distinction is vital for how we manage the ocean floor. If we assume that a map of the seabed means the area has been searched for archaeology, we might mistakenly believe that a site is absent when it is simply hidden in the noise of unverified data. The researchers concluded that the current archive supports local stories about the ancient landscape and helps verify modern debris, but it cannot be used to declare the entire region archaeologically empty. To move forward, the priority is not necessarily to send out new ships immediately, but to recover the missing details from the old data. Scientists need to find the specific settings of the sonar, the exact paths of the ships, and the full logs of what was seen. Only by recovering these details can they determine if the old data is good enough to trust, or if new, targeted investigations are required to finally reveal whether the submerged prehistoric landscapes of the Gulf of Cádiz are truly silent or just waiting to be heard.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.