Australasian Science: Australia's authority on science since 1938

Climate Records Reveal North–South Divide


It’s little wonder that the two regions have different climate histories: the Southern Hemisphere is a vast oceanic region that is influenced by ocean circulation features such as El Niño, while the Northern Hemisphere is dominated by most of the Earth’s continental masses.

By Joëlle Gergis

The first comprehensive reconstruction of the Southern Hemisphere’s temperature over the past millennium reveals that Northern Hemisphere warm and cool periods were not global. But what about late 20th century warming?

Long before humans were keeping official weather records, the natural world was tattooing the passing of time, year after year, for centuries. Cycles of good and poor seasons influence the size of a tree’s annual growth ring, while river runoff and changes in sea temperatures affect the thickness of coral skeletons growing in tropical waters.

In cooler locations, the chemical signatures of wet and dry years are layered into cave drip deposits as rain seeps through the porous limestone from the ground above. Ice cores taken from polar regions contain bands of accumulated snow that respond to changes in moisture availability as atmospheric winds shift in response to the seasonal expansion and contraction of sea-ice and high altitude glaciers.

In 1998, Prof Michael Mann of Pennsylvania State University published a seminal temperature reconstruction that used these kinds of natural climate records to estimate temperature variations in the Northern Hemisphere since 1400 AD. Extensions of that study appeared in the Intergovernmental Panel on Climate Change’s (IPCC) third assessment report in 2001 along with four other reconstructions supporting the same conclusion. The graph was termed the “hockey stick” to describe the pattern this showed – a relatively flat period of temperature variations until the start of the 20th century corresponding to the hockey stick’s shaft, followed by a sharp increase in temperatures corresponding to the hockey stick’s blade.

The graph became a target for climate change skeptics trying to undermine the strengthening scientific consensus that late 20th century warmth was exceptional. By the time the IPCC’s fifth assessment report was released in 2013 more than two dozen reconstructions, using various statistical methods and combinations of proxy records, had supported the broad consensus of unprecedented late 20th century warming shown in Mann’s original 1998 paper.

But until this year we didn’t have a comprehensive equivalent temperature reconstruction for the Southern Hemisphere. Previous studies had focused on the Northern Hemisphere and only used a sparse network of records from the vast oceanic south to draw tentative conclusions about our region.

Data from our remote part of the world is essential for climate models to realistically reproduce a global picture of the Earth’s climate history, sharpening predictions of our greenhouse future. So in 2009 I teamed up with Dr Raphael Neukom of the University of Bern to collate all of the annually resolved palaeoclimate records available in the Southern Hemisphere to see if we could get a better understanding of the Earth’s temperature history over the past millennium.

We got to work compiling climate data from hundreds of different locations across the Southern Hemisphere. We assembled a range of climate records collected from the natural world to reconstruct pre-industrial climate variability before instrumental weather records became available. Three years later we had the biggest collection of Southern Hemisphere climate records ever compiled. We gathered Antarctic ice cores, ancient tree rings from South America, Tasmania and New Zealand, coral records from the Indian and Pacific regions, and even early colonial documents from the Andes in South America to Sydney.

After publishing an inventory cataloguing the new Southern Hemisphere dataset in 2012, we then used these natural climate records from hundreds of different locations to reconstruct Southern Hemisphere temperatures over the past 1000 years. The aim was to get a clearer climate picture of the globe’s temperature history than ever possible before.

The landmark study, published in Nature Climate Change, involved an international scientific team with expertise in past climate information from tree rings, lake sediments, corals, ice cores and climate modelling. It revealed that temperature variations over the past 1000 years have differed greatly between the two hemispheres. The Southern Hemisphere was not simply an echo of what was going on in the north; we had our own history with different players involved.

Until recently, little was known about the magnitude and timing of temperature fluctuations in the Southern Hemisphere during the so-called “Medieval Climate Anomaly” warm (A.D. 900–1250) or “Little Ice Age” cool (A.D.1400–1700) intervals described from Northern Hemisphere climate reconstructions. Our study showed that there were considerable decade-to-decade regional temperature variations in the Southern Hemisphere that were different to the Northern Hemisphere.

Our study definitively proved that the Medieval Climate Anomaly, as originally identified in some European locations, was simply a regional phenomenon that influenced parts of the Northern Hemisphere rather than being a global event. During this period, temperatures in the Southern Hemisphere were only average, suggesting that the medieval warmth was a regional event and not the globally synchronous climate event that many people had previously assumed.

Surprisingly, only twice over the past 1000 years have both hemispheres simultaneously shown extreme temperatures. One of these occasions was a global cold period in the 17th century, providing evidence for a global cold phase coinciding with the peak of the Northern Hemisphere’s “Little Ice Age” when there are reports of the River Thames freezing over in London.

The other period of synchronous temperature extremes is the post-1970 warming. Our analysis revealed that the only warm period in the past 1000 years that is shared by both hemispheres occurs during a period of rapid and sustained greenhouse gas -influenced temperature increases. The warmest decade of the millennium occurred after 1970 in 99.7% of the 3000 temperature reconstructions we calculated. Our analysis highlights the robustness of our results, and provides further confirmation of Mann’s original conclusion about the unusual nature of late 20th century global warming.

It’s little wonder that the two regions have different climate histories: the Southern Hemisphere is a vast oceanic region that is influenced by ocean circulation features such as El Niño, while the Northern Hemisphere is dominated by most of the Earth’s continental masses. Our study showed that internal climate cycles like El Niño may have played a stronger role in influencing regional climate in the Southern Hemisphere compared with the land-dominated north, where external changes in volcanic and solar variations have had a more direct influence.

Our research now provides an opportunity to refine regional climate model predictions in the Southern Hemisphere for countries like Australia and South America by extending our understanding of natural temperature variations recorded since 1850 back over the past 1000 years.

Understanding the past is crucial for our ability to plan for future climate change. When we have a more complete idea of the range of natural climate variability that lies beyond the limits of instrumental weather records, we are in a better position to anticipate future climate surprises that might await us. Or in the wise words of Winston Churchill, “the further back we look, the further forward we can see”.

Our study is yet another in a long line that provides solid evidence that our planet has warmed substantially. Future life in Australia – and on the planet – hangs in the balance of how we choose to respond to this evidence. Now is the time for brave policy action to decarbonise our economy and stabilise the rapid warming currently underway.

Joëlle Gergis is a palaeoclimatologist at the University of Melbourne’s School of Earth Sciences. Her website is at