In my first year of graduate school, I was stumped by a big question on my final exam in biological oceanography. Maybe I had missed the relevant lecture or an assigned reading, but I could not answer the question: How can the same species of clams, snails, and other marine invertebrates be found along the coastlines of both the eastern United States and Portugal?
After the exam, I found the answer in a classic research paper published in 1968 by Rudy Scheltema, a biologist at Woods Hole Oceanographic Institution. He described how bottom-dwelling invertebrates disperse, showing that ocean currents carry larvae of these tiny mollusks near the water’s surface from one side of the North Atlantic to the other.
I never forgot that lesson, but I could not have guessed that more than 30 years later, 17,500-year-old clamshells would turn up in some North Atlantic seafloor sediment I was studying, and I would apply the lesson in my climate research.
The goal of my work is to understand the oceans’ role in climate change. Shelled organisms, collected in cores of sediment extracted from the deep ocean, provide telltale clues. My work, and that of my colleagues, has shown that past changes in the oceans’ circulation have been closely coupled with climate changes.
The oceans act as a planetary heating and ventilation system, conveying large amounts of warm equatorial waters toward the poles in ocean currents. In the North Atlantic, these waters transfer their heat to the atmosphere and warm the region. In the process, the waters become colder and denser and sink to the abyss. These dense water masses flow back southward and throughout the rest of the ocean in deep currents near the ocean bottom, before they mix again with warmer, less dense, upper ocean water masses and return to the surface in other parts of the world's ocean.
It is well known that this conveyor-like ocean circulation system has “turned off” several times in the past. When that happened, North Atlantic surface water did not sink to the abyss to move the conveyor. Warm equatorial waters were not drawn northward, and the North Atlantic region cooled. Near the seafloor, the deep southward-flowing currents slowed or stopped, and instead, deep currents flowed northward from the South Atlantic and the Southern Ocean around Antarctica. See "The Once and Future Circulation of the Ocean.")
A key way to determine when the conveyor was turned on in the past is to measure amounts of a naturally occurring radioactive isotope in the preserved skeletons of deep-sea corals, or in the shells of clams or, more usually, of singled-celled animals called foraminifera. These animals live on or near the seafloor and incorporate the chemical characteristics of the deep-ocean water that existed when their shells and skeletons formed.
When the conveyor is “on,” water from the surface of the North Atlantic sinks to the bottom. That water is enriched with carbon-14, a naturally occurring, steadily decaying radioactive isotope of carbon that forms in the atmosphere and dissolves into ocean surface waters.
We can date the preserved shells and skeletons of deep-sea animals, and if they contain relatively high concentrations of carbon-14, we can ascertain when the conveyor was transporting waters to the depths—and climatically, when the North Atlantic region was warmer. Shells and skeletons that are relatively depleted of carbon-14 indicate that they formed in waters that hadn’t had recent contact with the atmosphere—either because the conveyor was “off” and not transporting surface waters to the depths, or because they were bathed in deep waters coming from the Southern Ocean, whose carbon-14 content has decayed.
Using this method, Laura Robinson, then a postdoctoral researcher at the California Institute of Technology and now an assistant scientist at WHOI, led a study to measure carbon-14 in deep-sea corals from the North Atlantic. (See "The Coral-Climate Connection.") I provided carbon-14 data from foraminifera collected in seafloor sediment cores.
The study, published in December 2005 in the journal Science, showed that 20,000 to 15,000 years ago as the last ice age was coming to an end, carbon-14-depleted water filled the deep North Atlantic, indicating that the conveyor was "off." Then, as Earth's climate warmed, the conveyer switched “on” for about 2,000 years, until it abruptly shut down again about 12,900 years ago, and Earth’s climate cooled during a 1,400-year-period known as the Younger Dryas.
Most of the data about waters deeper than about 2,500 meters (8,200 feet) come from carbon-14 measurements in foraminifera extracted from seafloor sediments. Unfortunately, we can only make such measurements in places where these shells accumulate abundantly. In one such locale, on the Bermuda Rise in the western North Atlantic, foraminifera shells are rare in the glacial epoch. But I found small shells from a single clam that were dated at 17,500 years old preserved in two sediment cores.
Curious about the clam's identity, I sent pictures to Scottish marine biologist John Allen. He identified the shell as Spinula scheltemai, named for the very same WHOI biologist Rudy Scheltema whose research I had overlooked on my final exam. I was stunned to learn that S. scheltemai is native to the South Atlantic and had never been reported in the western North Atlantic.
So here was yet another line of evidence suggesting that bottom waters in the western North Atlantic 17,500 years ago came up from the south (and that the conveyor was “off”). After all, how else would the clams arrive, but to come up from their native home in the South Atlantic? But this interpretation of the data would only hang together if bottom currents controlled the distribution of S. scheltemai.
Thus, the question that once stumped me on my final exam resurfaced: How are clams' larvae dispersed? If, somehow, the larvae of this bottom-dweller could reach surface waters three miles above, they would be transported by near-surface currents, as Scheltema’s classic 1968 research paper showed for some bottom-dwelling invertebrate species. In that case, S. scheltemai might be found in deep basins throughout the Atlantic. On the other hand, if the larvae are dispersed by deep currents, they might reach the deep western North Atlantic only when the deep flow from the south was exceptionally strong.
It turns out that the S. scheltemai clams produce relatively large, heavy eggs. This is a strong indication that their young are not dispersed by surface currents, but rather stay near the bottom and are transported by deep currents.
Thus, the biological data are consistent with the geochemical carbon-14 data. Both indicated profound circulation changes in the deep western North Atlantic near the end of the last ice age. And I have the satisfaction of applying a lesson I initially missed, but quickly learned, at the beginning of my career.
Funding for Lloyd Keigwin's research came from the Comer Science and Education Foundation, through a contribution to the WHOI Ocean and Climate Change Institute.