Was climate change the real dinosaur-killer?

By David Arnold

Why we should wake up to the possibility of dinosaur-scale extinctions in the 21st century

What killed off the dinosaurs? As one of the most famous events in geological history, you might expect scientists would long have resolved the answer to this simple question.

If asked, most people would probably tell you that the dinosaurs were killed by a meteorite. But the widely held view, that an extra-terrestrial rocky behemoth slammed into the Earth in Mexico and immediately wiped out almost all life, is vastly oversimplified. In reality it was a lot more complicated than that.

My PhD research focuses on understanding past changes in climate, particularly during the Late Quaternary, which saw the demise of many large animals, such as mammoths and woolly rhinos. But this mass extinction was just one of many that have occurred throughout Earth’s History (see graph below). One of the largest extinctions in history was the famous end-Cretaceous (K-Pg) mass extinction that killed off the (non-avian) dinosaurs 66 million years ago.

Graph showing past extinction events throughout history -  time is plotted along the horizontal axis in millions of years and the % marine species extinction along the vertical axis.

Graph showing past extinction events throughout history – time is plotted along the horizontal axis in millions of years and the % marine species extinction along the vertical axis.
Image by Smith609 (CC BY-SA 3.0).

A new paper published in Nature Communications1 reveals that Antarctic waters experienced two discrete warming events that coincided with two well-documented pulses of extinction around the K-Pg boundary. The authors suggest that this warming, caused by climate change, is the link between two proposed extinction mechanisms: the famous meteorite impact, and dramatically increased volcanism at this time. It brings more evidence to the table that both events combined to change the climate on a massive scale.

Although in 2010, an international collaboration of geologists weighed up all the evidence and agreed that the meteorite impact was the cause of the K-Pg mass extinction event2, there is still debate over the issue. It all comes down to timing. In deep geological time, such as the K-Pg boundary, it can be tricky to assess the timing of events, as there are limited dating methods available. Dating for these deep geological events is on a completely different scale than the resolution I am used to working with, studying events in the Quaternary.

When you are talking about 66 million years ago, there is no statistical way of separating time scales of a few hundred thousand years. It’s virtually impossible to determine whether two events are synchronous when 1,000 years and 100,000 years are essentially the same in deep time.

This becomes a big problem when you have major events occurring around the same time in the geological record, as is seen at the K-Pg boundary. Within just a few hundred thousand years, the Earth was hit by the famous Chicxulub meteorite, as well as experiencing repeated volcanic eruptions in the Deccan Traps province of India.

Artist's impression of the meteorite impact suggested to have caused the mass extinction event at the K-Pg boundary.

Artist’s impression of the meteorite impact suggested to have caused the mass extinction event at the K-Pg boundary.
Public domain image by NASA.

The 2010 group discussed the mechanisms by which the meteorite impact could have caused the mass extinction event. A 10km wide meteorite hitting the Earth would have been devastating, filling the atmosphere with superhot ash, dust and steam, and causing regional wildfires and a megatsunami across the Gulf of Mexico, which would have affected ecosystems for years. The atmospheric effects could have continued for decades, with an initial super greenhouse effect followed by an ‘impact winter’ of longer-term cooling as ash particles blocked out sunlight. The impact of the meteorite could have also induced earthquakes and volcanic eruptions as shockwaves passed through the Earth.

But eruption of the Deccan Traps would have also induced climatic change through massive increases in greenhouse gas emissions, particularly sulphur dioxide. The 2010 committee ruled out volcanism as a cause of the K-Pg mass extinction, but another school of thought, led by Gerta Keller, suggests that we consistently overestimate the environmental effects of the meteorite impact and that more weight should be given to the climatic change caused by volcanoes.

The prevailing view that the meteor impact directly led to the extinction event has also been challenged by evidence published in PNAS3 that non-avian dinosaurs were already in noticeable decline around 50 million years before the Chicxulub impact. Again, distinguishing the causes of extinction is difficult as many geological sites that show the K-Pg boundary have gaps, poor temporal resolution, and a lack of species continuity across the boundary. Fortunately, a site called Seymour Island in Antarctica exists where these complications do not pose the same problems.

Map showing the location of Seymour Island within Antarctica.

Map showing the location of Seymour Island within Antarctica.
Map created by David Arnold, outline credit: d-maps.com.

Seymour Island has an exceptionally well-preserved collection of macrofossils of 66-million-year-old plants and animals, through a continuous section straddling the K-Pg boundary. Previous work at the site on preserved bivalve shells attempted to use the palaeothermometer of oxygen isotope ratios to identify climatic changes around this time. But the results were complicated by uncertainty over the ratio of isotopes in the seawater – if we don’t know the composition of the water these organisms lived in, we can’t calibrate our palaeothermometer. Data that seems to indicate temperature change could actually just be showing us local variability in the isotopic composition of the water.

The new study avoided these issues by using a new carbonate clumped isotope method, which allowed them to calculate the temperature of bivalve shells as they formed, independently of the oxygen isotope composition of sea water. They analysed 29 shells from across the K-Pg boundary exposed at Seymour Island, and found evidence for two discrete periods of ocean warming, rather than a single warming event expected if the meteorite alone was responsible for the K-Pg mass extinction.

Their analysis showed a large spike in ocean temperature of about 8°C, which occurred 150,000 years before the K-Pg boundary. There is also another warming of 1-2°C right on the boundary.

Both of these warming events happen in sections of the geological sequence that also showed a marked decline in the number and diversity of species, known as an ‘extinction pulse’. These two extinction pulses each “coincide” with one of the two proposed causes of extinction that we’ve already discussed.

The problem in the past, which this study tries to overcome, has been linking the two proposed causes of climate change (volcanism and the meteorite) to the geological evidence for extinction. It could be said that warming in the Antarctic is evidence of climatic change that reached well up into the high latitudes, painting a picture of a two-pronged assault on ecosystems worldwide. First, volcanic eruptions battered the environment with global warming, then ecosystems barely had time to think (geologically speaking) before the climate was thrown off again by the meteorite impact.

However the data might also be evidence for more localised environmental changes, like melting of continental or sea ice which may have amplified the warming – a mechanism familiar to most of us in the 21st century.

To be blunt about it, scientists are essentially matching peaks on a graph – lining warming events up with disappearing species in the fossil record. The dating methods available to deep time geologists are limited in their precision. However with little evidence to the contrary, at least at Seymour Island, it is a perfectly reasonable hypothesis that the 8°C spike was caused by increased volcanism and the second smaller warming by the meteorite.

Interestingly there is another warming episode visible in the data, which precedes both events, but the rate of warming is was comparatively slow, and the fossil evidence does not show any marked decline in species. One of the main take-home messages here is that it is clearly the rapidity of the warming which has an effect on biodiversity.

Perhaps this earlier, slower warming is related to the longer-term trend in Dinosaur extinctions in the run up to the catastrophic K-Pg events. Perhaps this the double spike warming was isolated to Antarctica. Unfortunately the deep time evidence may never be good enough to know for sure.

What we can say is that if the conclusions from this study are correct then there really was not much chance for the dinosaurs against the one-two punches of climatic change in response to the two events. It’s quite a worrying thought that warmings of 7-8°C or so can do this over the space of just a million years – with predictions of current warming at 3-5°C by 2100 we could well be on the way to a scary level of environmental change, and the next mass extinction could be far from natural.


1. Petersen, S.V. et al. (2016) End-Cretaceous extinction in Antarctica linked to both Deccan volcanism and meteorite impact via climate change, Nature Communications, 7, 12079, 1-9.
2. Schulte, P. et al. (2010) The Chicxulub Asteroid Impact and Mass Extinction at the Cretaceous-Paleogene Boundary, Science, 1214-1218.
3. Sakamoto, M. et al. (2016) Dinosaurs in decline tens of millions of years before their final extinction, Proceedings of the National Academy of Sciences, 113, 18, 5036-5040.

Tagged with:

See more posts by

Stay informed

Click here to subscribe to our RSS newsletter by email.

Find Us

University College London is the administrative lead.

North-West Wing, UCL, Gower Street, London, WC1E 6BT

Follow us on Twitter