Gravitational waves, marine fossils and climate change don’t usually appear in the same sentence, but a unique international research collaboration is building bridges between the disciplines. A team of astrophysicists, palaeontologists and mathematicians used machine learning algorithms – originally developed by gravitational wave astrophysicists – to improve the accuracy of a “paleothermometer”, which looks at fossil evidence of past climate change to predict Earth’s future.
Ice cores and tree rings are perhaps more widely recognised examples of paleothermometers. By studying the trapped air bubbles within ice or the oxygen isotope ratio of tree ring cellulose, researchers can reconstruct the composition of the Earth’s atmosphere over millions of years. This then helps us better understand the planet’s cycles.
But this new research– led by palaeontologist Tom Dunkley Jones, from the University of Birmingham, UK – instead studied biomarkers left over from single-celled organisms called archaea, dating as far back as the Cretaceous (145–66 million years ago).
These little critters produce compounds called Glycerol Dialkyl Glycerol Tetraethers (GDGTs). In modern oceans, the abundance of GDGT varies with the local sea temperature, “most likely driven by the need for increased cell membrane stability and rigidity at higher temperatures,” the researchers explain in their paper, which appears in the journal Climate of the Past.
Archaea preserved in ancient marine sediments therefore have the potential to provide a long-term geologic record of the planet’s surface temperatures.
Previous research has combined GDGT concentrations into a single parameter called TEX86, but its accuracy is lacking; when compared to known modern sea-surface temperatures, predictions from TEX86 in recent sediment were several degrees off.
“After several decades of study, the best available models are only able to measure temperature from GDGT concentrations with an accuracy of around 6°C,” says co-researcher Ilya Mandel, an Australian gravitational wave astrophysicist from the ARC Centre of Excellence in Gravitational Wave Discovery (OzGrav).
In collaboration with colleagues from the UK, Mandel tried a different approach to make high-precision measurements of ancient climates.
They turned to modern machine learning tools that were originally used in gravitational wave astronomy, in order to create predictive models of merging compact objects like black holes and neutron stars.
These tools allowed them take into account all GDGT measurements at once, instead of combining them into a simplified factor of TEX86 – resulting in a far more accurate paleothermometer.
The accuracy of the model nearly doubled, from 6°C to 3.6°C.
In their paper, the authors point out that using GDGT abundances to estimate temperatures is still an area of emerging research, limited by available calibration constraints and our current understanding of underlying biophysical models.
Related reading: Trees tell tales of climate change