Fifty six million years ago there was a super global warming event known as the Paleocene-Eocene Thermal Maximum (PETM), which has been suggested to represent a potential analog for our future climate. There are many theories regarding the cause of this extreme climatic event. One theory involves the release of 4,000 petagrams of methane from ocean sediments into the atmosphere. Another is that the Earth was hit by a large meteorite similar to the one that led to the extinction of non-avian dinosaurs. This particular event caused significant atmospheric warming of between 3 and 6°C, as well as ocean acidification and the extinction of many bottom-dwelling marine organisms. It also led to the evolution of the first true primates (our ancestors) and many modern mammal groups such as the hoofed animals (pygmy horses and elk), tapirs, rodents, bats, owls, elephants and early whales.
Scientists have interpreted this warming event by analyzing biogeochemical proxies and using computer modeling. The geochemical properties of certain fossils can be used to interpret the paleoclimate in the area in which they formed. Plankton fossils, for instance, are used to investigate historical sea-surface temperature variations, paleoproductivity, paleo-oxygenation, past ocean circulation, and much more. However, these organisms live in the photic zone of the water column but when they die their remains sink to the bottom, where they are trapped in sediments and become fossilized. These organisms therefore usually live in one environment for just a few months or years and are then preserved in another environment that is characterized by a totally different water chemistry for millions of years.
The Mg/Ca ratio, for example, is a well established proxy for water temperature. But how accurately can we estimate Mg/Ca ratios from these fossils without taking into consideration variations between the two different environments, within the water column and at the bottom of the ocean? This is a major challenge for this approach as the chemical constituents of the organisms could be altered by the chemistry of the sediments in which they are preserved. Another uncertainty in these fossil records is that they may be heavily biased by the influence of high productivity periods, which can occur either seasonally or at longer intervals related to, for example, Milankovitch cycles. This means that these records may not actually represent the average temperature of the investigated area.
PETM carbon release scenario from fossil record and computer models (source Zeebe, 2009)
The use of paleo-records to assist in climate change predictions is based on James Hutton’s well-known principle of uniformitarianism and the concept that “the present is the key to the past”. Hutton argued that the earth processes that operated in the past were similar in magnitude and intensity to those operating today, whereas these processes are now believed to have actually changed with time.
When comparing the climate forcing behind the PETM and the drivers of present-day climate change it can be argued that the major drivers of the former were natural (in the form of massive volcanism, the dissociation of methane hydrate, and the influence of Milankovitch cycles) whereas current climate change is driven by anthropogenic forces resulting from the burning of fossil fuel, deforestation and pollution.
Richard E. Zeebe from the University of Hawaii at Manoa, together with collaborating climate scientists, tried to simulate the CO2 scenario during the PETM by modeling the greenhouse gas emissions that occurred during this warming event. They were, however, not successful because although when they added 3000 Pg of C in a computer model (to equal the negative carbon isotopic excursion recorded in the PETM event) the temperature increased with the increase in CO2, as soon as CO2 injection was halted the temperature dropped immediately down to zero, with no gradual recovery in temperature such as recorded during the PETM. This clearly shows that using paleoclimate models in prediction future global warming may be neither accurate nor feasible. The differences between past and present with respect to the nature and intensity of the processes operating on Earth, the biodiversity, and the compositional dissimilarities in the geosphere may be so great that past climatic variations cannot be reliably used to predict future climate change.
Many climate scientists involved in paleoclimate investigations understand this problem but tend to avoid mentioning it in order to continue to receive funding for their research. Climate scientists interested in predicting future climate change might therefore be better off concentrating on collecting present-day data and fine-tuning climate data collected over the past two centuries, in order to improve the accuracy of their predictions.
This post is part the series published by students of the East Africa Summer school on “collecting, processing and presenting information in bio-geo-sciences”, introduced by Prof. Martin Trauth in a previous snack.
0
0
0
0
0
0
Celestine Nwojiji