34 million years ago was a bad time to be in Antarctica. Atmospheric CO₂ was declining and once it hit a low threshold, the seasonal snows that had graced Antarctica suddenly started sticking year-round. This triggered rapid feedbacks that drove the growth of the large, permanent ice cap on the continent that we know today. Commonly known as the Eocene-Oligocene (E/O) Boundary, this dramatic event is widely recognized as the moment that Earth shifted from a warm “greenhouse” planet to our current “icehouse world.” In some ways, it is the exact opposite of today’s rapid anthropogenic-induced global warming and, as such, provides an invaluable natural experiment for studying how ecosystems respond to global climate change.  

And respond ecosystems did. As the South Pole plunged into eternal winter, there were clear changes in marine primary productivity, plankton communities, and carbon cycling preserved in the fossils and chemicals of deep-sea sediments from that time, with lots of evidence pointing towards increased primary production and even diversification in ancient whales – a great time to live in the ocean. These shifts following the E/O hint at a close connection between the climate and the base of the food web. However, it is unknown precisely how changes in ocean primary productivity translate up the food web to larger animals, such as fish and whales.  

Against this backdrop, we set out to investigate the impact of the E/O on marine fishes, the most diverse group of vertebrates on the planet and a major player in today’s modern marine ecosystems. We hypothesized that the climate shift during the E/O would be associated with significant changes in fish abundance and diversity, reflective of changes in the underlying plankton communities and environment.  

In order to study the impact that this extreme climate event had on ancient fish, we first had to find the fish. Fish are the most common vertebrates on the planet, but they’re rare in the fossil record. When a fish dies in the open ocean, their remains sink all the way to the sea floor, where their microscopic teeth can be preserved in the sediment. Fossil fish teeth, usually smaller than the width of a human hair (~100 microns) but are made of tougher stuff than their bones, and are therefore found in nearly all marine sediments. They provide an unparalleled glimpse into past fish and marine ecosystems at a very fine time resolution – the perfect tool for studying the relationship between fish and climate change in the past.  

To find fossil teeth of fish that lived millions of years ago, we turned to one of the longest-running international scientific collaboration currently in operation: the International Ocean Discovery Program (IODP). IODP, in its many iterations, has been drilling into the deep-sea floor for over 50 years to recover sediments that preserve rich records of earth history going back nearly 200 million years. These sediment cores are muddy time capsules, capturing a snapshot of ancient earth systems and ecosystems over millions of years. Using a dissection microscope and a super fine paintbrush, we sorted through 717 chunks of deep-sea mud, ooze, and clay from 7 different sites from around the world, ranging in age from 28 to 42 million years old, and picked out all of the fish teeth we could find: 50,623 in total, most no bigger than a human hair, about 100 microns across. These teeth revealed a fantastic, but perplexing, story that was not at all what we expected.    

We had hypothesized that there would be considerable change in the abundance of fish across the E/O boundary due to the significant climate shift. We couldn’t have been more wrong. Although there were major changes in Antarctic ice volume and associated global and ocean temperatures, we found that fish production – a metric integrating fish biomass over time – remained relatively constant across the E/O at all of our study sites. In other words, the fish populations didn’t seem to “see” the climate perturbation. They simply continued along as if it hadn’t happened, with no changes in abundance. 

Abundance is only one way to look at a community, so next we asked if the types of fish shifted during the climate change to replace warm-loving taxa with cold-tolerant ones. Faced with hundreds of images of wild and wacky teeth dredged up from deep sea cores, we now had to figure out a way to sort them. We narrowed in on a subset of teeth that had at least one dimension >100 microns and described them in extreme detail, sorting them into groups based on their unique shape characteristics which allowed us to identify morphotypes—a pseudo-taxonomic ranking that is used by paleontologists to group fossils that likely all came from the same type of organism, but that can’t be formally identified without a full body fossil. We then looked at which morphotypes were present before and after the E/O, and found a constant increase in morphotype diversity throughout the 14 million year study interval. In other words, there were no major changes at the E/O at all—no major extinctions nor sudden occurrence of new species. Indeed, the only major change in morphotype diversity occurred around 40 million years ago in the Antarctic, which is nearly 6 million years before the climate upheaval of the E/O boundary. 

As can happen in research, it turned out we were wrong on all counts: both fish abundance and diversity were stable across this major global climate change event. But, being wrong in science means that there’s more to learn if you keep digging. So, we dug deeper.  

We developed a cross-disciplinary synthesis of molecular phylogenies, ecology, and geological proxies, to compile all of the main parts of the Eocene-Oligocene ecosystem. Surprisingly, we found that while most papers report on increases or decreases in primary productivity at the E/O, in most regions of the ocean the productivity proxies don’t record a major or long-term shift—in line with our fish results. Furthermore, many canonical Antarctic species, including Antarctic krill, baleen-bearing whales, and icefish, didn’t evolve until well after the E/O, suggesting that community restructuring may have occurred well after the time period of our study which was centered immediately around the E/O boundary. 

Where does this leave us? We expected ecosystem change during a period of great climate upheaval and found none. Instead, we found evidence for stability in the face of a major global climate event—a completely unexpected result with particularly positive implications. This leaves us with a tantalizing mystery still to solve: if the marine pelagic ecosystem did not respond to the Antarctic glaciation 34 million years ago, as we previously thought, then when and how did it develop into the modern, productive, and unique ecosystem we see today? The fish have more stories to tell, and we can't wait to dig them up!  

Q&A with Author Michelle Zill  

What was most exciting to you about the scientific process on this paper? 
This was my first experience of working though my own research project. This project inspired me to change plans and pursue a PhD. Finding my passion research throughout the process of this paper was pretty exciting. 

What day-to-day tasks were your favorites and least favorites?
I loved washing and picking the teeth. I felt very accomplished after a day of hard work in the lab that was difficult to recreate after a day of writing. I also loved the moment of graphing a new data set for the first time and seeing what the record showed. Since I processed samples out of order, it was difficult to guess the general trends. So it was always exciting to find a new piece in our E.O. puzzle, even if  our results were “nothing happened”.

What ideas were the most challenging?
During the project, it was rather difficult to see no change at the E.O. boundary. Though challenging, it became an exciting way to examine many of the preconceived notions about fish during this time period, and I believe left us with a more interesting result.

What particularly stuck out to you as you were doing the research?
How awesome fish teeth are! They are awesome microfossils to work with and give you insight into the changing population of this very important vertebrate population.  

Michelle Zill is now a PhD Candidate at UC Riverside.

Read the paper, “No state change in pelagic fish production and biodiversity during the Eocene-Oligocene Transition” by Elizabeth C Sibert, Michelle E Zill, Ella T. Frigyik, and Richard D Norris, published online by Nature Geoscience on March 2, 2020. https://www.nature.com/articles/s41561-020-0540-2  

This "Behind the Paper" blog was written by Elizabeth Sibert (@elizabethsibert), with contributions from Richard Norris, Michelle Zill, and Ella Frigyik. Special thanks to Ali Freibott, (@AliFreibott on twitter) who provided invaluable editorial advice on this blog post.