Category Archives: Blog

Was the exodus of Homo sapiens out of Africa paced by orbital cycles?

by Ignacio Jara

It is now widely accepted that anatomically modern humans (Homo sapiens) emerged in Africa around 200 ka (ka=thousands of years before the present), as evidenced by now-classic hominid sites in eastern Africa and supported by genetic variations among modern populations. Despite the unquestioned African origin of our species, the time when modern humans first exited Africa and the order in which they colonised the remaining continents is hotly debated.

The most recent of these discussions has centred on the timing of the first exodus and its relationship with Quaternary climate variations1. This is clearly a subject that touches a wider and perhaps more complex topic such as the relationship between the evolution and behaviour of humans and their changing environment. This is an appealing subject which bears not only scientific relevance, but also deep cultural implications. For instance, the extent to which climate variability or other environmental drivers has influenced human migrations is a question that stresses the role of geography over culture.

Homo sapiens remains dating back to 125 ka during Marine Isotope Stage 5e (MIS 5e) have been found in modern day Israel. The commonly accepted view is that these remains represent an early human dispersal that failed to spread outside of the Middle East, becoming extinct soon after it stepped out of Africa. The peopling of Europe, Asia and Australasian was supposed to have occurred much later from a single “successful” migration around 70-60 ka1.

However, this view of a single successful migration is being challenged. In 2015, several modern human remains dated to at least 80 ka were found in the Fuyan Cave in southern China2. These findings followed a series of other surprisingly old human fossils in Asia, yet no remains of modern humans older than 50 ka have been found in Europe. This new evidence questioned the classic single “out of Africa” model, suggesting an alternative scenario where H. sapiens expanded eastward into Asia in one or more waves of migration starting well before 70 ka. While it is still unclear what prevented an early migration into Europe.

Back in 2013, Larrasoaña and collaborators combined a series of continental and marine records from northern Africa to propose what is, without question, an exciting idea for any Quaternarist: that orbital-driven Northern Hemisphere insolation played an important role in the migrations of hominids out of Africa3. It has long been noted that the 23-ka precession cycles are clearly imprinted on the Northern Hemisphere summer insolation between 300-60 ka. Every 23,000 years, decreased precession and corresponding high Northern Hemisphere summer insolation are linked to a northward shift of the Intertropical Convergence Zone. These shifts were, in turn, linked to northward expansion of the tropical belt bringing periods of intensified monsoon rainfall over northern Africa and the Middle East. During these humid intervals, the deserted regions of the Northern Africa, the Middle East and the Arabic Peninsula were transformed into savanna-like regions boasting freshwater lakes and extensive grasslands, becoming transient corridors for modern human populations to migrate out of Africa.

Exit passages for African H. sapiens were therefore opened with “orbital” regularity at around 130 ka (MIS 5e), 105 ka (MIS 5c), 82 ka (MIS 5a) and 60 ka (MIS 4-3). Thus it is not surprising that the earliest human fossils out of Africa found in Israel date back to MIS 5e. Furthermore, all the following wet intervals coincide with dated H. sapiens sites in Northern Africa, the Middle East or the Arabian Peninsula. There are several possible routes that the H. sapiens could have taken. The Sinai Peninsula in northern Africa was the most logical passage to exit Africa (Figure 1). However, there is an alternative southern route which connects eastern Africa with southern Asia via the narrow straits of Bab-El Mandeb and Hormuz. This route could have briefly been opened before the sea level rise associated with MIS 5e.


Figure 1 – Homo sapiens emerged in Africa around 200 ka. The older Homo sapiens remains outside of Africa are found in today’s Israel and date back to 125 ka during Marine Isotope Stage 5e. The timing and routes of the first expansion into the other continents is still in debate. For more information see: Stringer, C. (2003).

In an article published just last month, Timmermann and Friedrich (2016) take the idea of precession-controlled wet periods as a pacemaker for human dispersal and test it by running a human dispersal model4. Their new model, forced by environmental variables such as climate variability, sea level changes and the extension of deserts, is able to reproduce a series of independent migration pulses between 110-60 ka in broad alignment with the orbital moist intervals and the archaeological record of Northern Africa (Figure 2). By reproducing the dates for the first arrival time of H. sapiens in places like South China (100-70 ka), New Guinea and Australia (60 ka), and the Americas (14-10 ka), this new simulation provides new weight towards orbital variations as a key driver of early migrations of H. sapiens.


Figure 2 – Modern human Migration waves (HMW, yellow bars) occurred with “orbital” regularity at around 130 ka (MIS 5e), 105 ka (MIS 5c), 82 ka (MIS 5a) and 60 ka (MIS 4-3), when low precession (light blue, note the inverted scale) and resulting high Northern Hemisphere summer insolation resulted in a series of wet episodes in Africa and the Middle east. Atmospheric CO2 concentration curve (light grey) is shown as a reference. Modified from: Timmermann, A., & Friedrich, T. (2016).

Although the early exodus and multiple orbital dispersal model looks promising, several other issues need to be addressed in future investigations. For instance, the model presented by Timmermann and Friedrich indicates that H. sapiens stepped into Europe between 100-80 ka. This time is at odds with the archaeological record, which shows no evidence for modern humans before 45 ka. Interestingly, the authors suggest the possibility that the first H. sapiens populations in Europe where assimilated by the prevalent Neanderthal population before 50 ka. This hypothesis seems plausible considering the information provided by the recently-cracked genome of Neanderthals which gives hints of the incorporation of modern human genes into European Neanderthals DNA at around 100 ka5.

Together, Quaternary scientists, archaeologist and geneticists are providing exciting new information about past links between climate variability and ancient migrations. An active role of environmental variability in the distribution of modern humans across the globe has long been dismissed by paleo-anthropologists, ethnographers or linguistics who traditionally favoured cultural explanations over “geographic or climatic determinism” to account for ancient human migrations. Nonetheless, a bio-geographic approach to the peopling of the planet seems to be gaining ground. What it is perhaps even more fascinating to any Quaternarist is that a look at the Northern Hemisphere insolation curve shows that the potential number of humid events in Africa and the Middle East over the last 2 million years is in the order of thousands. Future paleoclimate studies and a more complete archaeological record will have the potential to test the role of climate events in the diaspora of previous hominid species as well as the emergence of our own species.


  1. Mellars, P., Gori, K. C., Carr, M., Soares, P. A., & Richards, M. B. (2013). Genetic and archaeological perspectives on the initial modern human colonization of southern AsiaProceedings of the National Academy of Sciences110 (26), 10699-10704.
  2. Liu, W., Martinón-Torres, M., Cai, Y. J., Xing, S., Tong, H. W., Pei, S. W., & Li, Y. Y. (2015). The earliest unequivocally modern humans in southern ChinaNature.
  3. Larrasoaña, J. C., Roberts, A. P., & Rohling, E. J. (2013). Dynamics of green Sahara periods and their role in hominin evolutionPloS one8 (10), e76514.
  4. Timmermann, A., & Friedrich, T. (2016). Late Pleistocene climate drivers of early human migrationNature538 (7623), 92-95.
  5. Kuhlwilm, M., Gronau, I., Hubisz, M. J., de Filippo, C., Prado-Martinez, J., Kircher, M., & Rosas, A. (2016). Ancient gene flow from early modern humans into Eastern NeanderthalsNature530 (7591), 429-433.

Blog 11: Past, Present and Future of SAM and the impact on the Australian rainfall

By Ignacio A. Jara

The Southern Annular (SAM) mode is one of the most important atmospheric phenomenon affecting temperature and precipitation in the non-tropical Southern Hemisphere. Perhaps the clearest sign of its growing importance for the scientific community was the mention of SAM in several talks in the SHAPE session in the AQUA biennial conference here in Mildura.

A tendency toward positive SAM polarity over the last couple of decades is expressed as an overall poleward contraction and intensification of the cold and rain-carrying Southern Westerly Winds. In regions exposed to westerly activity such as southern Australia, New Zealand and southern South America this dynamic has resulted in a significant reduction in precipitation and an overall increase in temperatures. On the other hand, positive SAM has a mixed influence on Antarctic temperatures. While most of eastern Antarctica has experienced cooling over the positive SAM timeframe, some areas of western Antarctica and most of the Antarctic Peninsula have been warming over the same period. The Antarctic Peninsula seems to be critical in terms of future SAM variability since  it represents the transitional area between the warming continental Southern Hemisphere and the cooling east Antarctica. Thus, the Antarctic Peninsula was the focus of a group of researchers who have published a late Holocene SAM reconstruction now available online in the journal Nature Climate change1.

This new article presents a composite reconstruction using temperature records from the whole domain of SAM influence: South America, the Peninsula and the main Antarctic continent; modelling its evolution over the last millennium. Negative SAM characterizes the first interval of the reconstruction, followed by two positive excursions between 1400 and 1800 AD and during the 20th century respectively. Furthermore, the authors evaluate potential SAM drivers through a series of modelling simulations and suggest that the first positive SAM pulse can be explained by an increase in the solar irradiance; whereas the latest 20th century positive excursion fails to be replicated by any radiative forcing, it is only fully reproduced by the models when greenhouse gas forcing is added.

El Nino Southern Oscillation and SAM

Another natural SAM forcing explored in the article is tropical climate variability. Instrumental climate data indicates that La Niña years correlate with positive SAM, with evidence of warm conditions in the Antarctic Peninsula and New Zealand2. By comparing their new SAM reconstruction with a highly resolved proxy for ENSO variability, the authors found this correlation operated throughout the last millennium. For instance, the dominance of El Niño conditions during the first part of the millennia seem to have acted as a negative driver for SAM; while the first positive SAM pulse coincides with a tendency towards more La Niña conditions. Critically, this long term correlation breaks down in the 20th century when the latest positive trend of SAM parallels an increase in El Niño, suggesting that the tropical Pacific in its El Niño state is currently not muting or attenuating the positive SAM trend (ENSO was covered in the previous blog 10).SAM reconstructionFigure 1: SAM reconstruction for the past millennium relative to the average during 1961-1990 average (dashed black line). Figure from reference 1.

SAM and rainfall changes in Australia

Another interesting new publication has compared Australian precipitation variability with ENSO and SAM over the last few decades3. This new article shows a reduction in winter precipitation in the coastal areas of southern Australia by 10-20% since 1970; while summer precipitation in the dry inland and northern areas has increased from 40 to 50%.  Interestingly, the authors correlated the reduction of winter precipitation with fewer and weaker westerly fronts, a phenomenon largely documented as occurring during positive SAM. On the other hand, enhanced easterly fronts over the north and the central areas have brought more tropical precipitation, especially during summer to northern areas.

Australian rainfall

Figure 2:  Map depicting rainfall changes in Australia in the period 1997-2009 compared with the 20th century average. Precipitation in South eastern Australia have been significantly reduced over the last decades as result of the southward migration of the westerly storms, due to the positive trend in SAM. Figure from reference 5.

As south eastern Australia -the most populated and industrialised portion of the country- relies significantly on the westerly winds as its main precipitation source, a better knowledge of the future SAM trends will be critical for better estimates of water availability over the next few decades. If the positive trend of SAM continues under future global warming scenarios, river runoff in places such as Mildura (the location of the current AQUA conference) will surely be severely compromised unless that summer tropical precipitation fronts extend further south. However, models of future SAM projections are not completely clear. The slow recovery of the Ozone layer seems to be forcing SAM to a negative phase4. More research to understand the potential future of  SAM under future climate change, along with adaptation and mitigation programs will be critical for the wellbeing of the Australian community.


  1. Abram, N. J., Mulvaney, R., Vimeux, F., Phipps, S. J., Turner, J.,      England, M. H. (2014).  Nature Climate Change 4, 564–569.
  2.  Fogt R.. L.,      Bromwish, D. H., Hines, K. M. (2011). Climate Dynamics 36, 1555–1576
  3. Raut, B., C.      Jakob, and M. Reeder. (2014). Rainfall Changes over Southwestern Australia and      their Relationship to the Southern Annular Mode and ENSO. Journal of Climate.      In press.doi:
  4. D. W. J. Thompson, S. Solomon, P. J. Kushner, M. H. England, K. M.      Grise, D. J. Karoly, Signatures of the Antarctic ozone hole in Southern      Hemisphere surface climate change. Nature Geosci 4, 741-749 (2011).
  5. Post, D. A., Bertrand, T., Chiew, H. S., Hendon, H., H. Nguyen, H., Moran, R. (2014). Decrease in southeastern Australia water availability linked to ongoing Hadley cell expansion. Earth´s future, 2, 231-238

Blog 10: Tree rings and ENSO

By Ignacio Jara

Since the 1980’s a huge amount of time and effort has been put to build a climate-sensitive tree-ring chronology for New Zealand. A noteworthy achievement of this scientific effort was the publication in 2006 of two independent tree-ring chronologies, a 2,300-year record from Silver Pine (Lagarostrobos colensoi) and a 3,700-year chronology from Kauri (Agathis australis) (1). Since 2006, the kauri chronology has been refined and extended up to 4,500 years, becoming one of the longest tree-ring records from the Southern Hemisphere (2). To generate such a long series, Kauri wood samples were collected from more than 60 sites across its whole distribution range in northern New Zealand; including modern trees, archaeological sites and sub-fossil trees preserved in swamp areas (that is a lot of annual tree rings to count!).


Figure 1: The present-day distribution of Kauri forest in northern New Zealand has been severely reduced by centuries of human forestry and farming. In the picture a remnant Kauri forest patch. Photo courtesy of

But perhaps what makes this chronology most scientifically relevant is that the dominant driver behind the present-day Kauri radial growth is El Nino Southern Oscillation (ENSO (3)). Wider tree-rings tend to occur during the relatively dry/cold El Nino years, while narrower rings are observed during the warmer/wetter La Nina years. This ENSO-kauri correlation seems to be strong throughout the entire distribution range of kauri forest, which facilitates a robust ENSO signal even if samples from multiple areas are integrated.

The potential of using the kauri chronology to decipher the waxing and waning of ENSO at timescales longer that the period instrumental record has been fully exploit in a recent publication in the journal Nature Climate Change (3). Focussing in the last 700 years of the chronology, the period with the highest statistical quality, the published ENSO reconstruction clearly depicts the twentieth century as the most “ENSO active” period in the context of the last 400 years. However, a high ENSO activity period is observed between 1,300 and 1,450 AD.

700 yr Kauri

Figure 2: The 700-years kauri-based ENSO reconstruction (in blue) in conjunction with other ENSO reconstruction from the Tropical Pacific. Image taken from Fowler et al 2012 (reference #2).

By comparing this tree-ring reconstruction with other ENSO proxies from the core of ENSO activity in the tropical Pacific, the authors further investigate the climate connection between the tropics and New Zealand (Figure 2). Relatively low variation in the kauri-based record compared with the tropical ENSO reconstructions between 1300-1750 AD suggests a reduced ENSO-New Zealand connection during the so-called Little Ice Age period. Interestingly, the authors point out that a northward shift of the Southern Westerly Winds and the subtropical front – a miniature version of a climate shift previously identified during the last Glaciation- could have resulted in a northward movement of the ENSO-south Pacific connection, pulling it away from northern New Zealand and therefore explaining the differences between these reconstructions observed during the Little Ice Age. Unfortunately, a confident ENSO reconstruction using the whole extend of the Kauri chronology is still impossible as some segments of the 4,500-year tree-ring record do not have enough samples to provide a statistically significant interpretation (2).

ENSO is one of the most influential modes of climate variation in the world. Abrupt annual anomalies in its behaviour can have a large impact on ecosystems and societies around the Pacific. For instance, in western South America -where ENSO was first identified- El Nino years are associated with very wet and warm conditions, as well as with a reduction in the upwelling of nutrient rich waters along the Pacific coast, with devastating effects on the local ecosystem and the local fishing industry. While in Queensland, Australia – the opposite side of the Pacific- El Nino is associated with drier conditions, which are linked to droughts, fires and mass coral bleaching on the Great Barrier Reef. The biggest question then is how ENSO behaviour might be affected by future global warming. Will the ENSO mode of variation be enhanced, will it increase in frequency, or will be unaffected, or reduced? Over the last few decades (pretty much the whole extension of the ENSO instrumental record), there has been an increase in ENSO activity associated with the warming trend observed during that same period. The continuation of this trend in the future is uncertain since coupled atmospheric-ocean models suggest contrasting results. This highlights the importance of developing high-resolution paleoclimate records able to investigate past associations between temperature changes and ENSO activity during the Holocene and beyond.


1.            E. R. Cook, B. M. Buckley, J. G. Palmer, P. Fenwick, M. J. Peterson, G. Boswijk, A. Fowler, Millennia-long tree-ring records from Tasmania and New Zealand: a basis for modelling climate variability and forcing, past, present and future. Journal of Quaternary Science 21, 689-699 (2006)10.1002/jqs.1071).

2.            G. Boswijk, A. M. Fowler, J. G. Palmer, P. Fenwick, A. Hogg, A. Lorrey, J. Wunder, The late Holocene kauri chronology: assessing the potential of a 4500-year record for palaeoclimate reconstruction. Quaternary Science Reviews 90, 128-142 (2014); published online Epub4/15/ (

3.            A. M. Fowler, G. Boswijk, A. M. Lorrey, J. Gergis, M. Pirie, S. P. J. McCloskey, J. G. Palmer, J. Wunder, Multi-centennial tree-ring record of ENSO-related activity in New Zealand. Nature Clim. Change 2, 172-176 (2012); published online Epub03//print (

Blog 9: Rapid weathering and erosion of the NZ Southern Alps

By Ignacio Jara and Helen Bostock

The debate about the role the Southern Alps (New Zealand) uplifting to offset global climate change has experienced a recent renewal with the publication of a couple of new articles. According to the Uplifting Theory, over geological time the tectonic erection of large mountain systems has been associated with lower global temperatures due to the enhanced consumption of atmospheric CO2 during the chemical weathering of terrestrial (silicate) material from the uplifting landscapes.

This year, Larsen et al (2014) reported new catchment erosion and soil production rates from the Southern Alps. Using Beryllium 10 (10Be) -a radioactive isotope produced by cosmic rays bombarding exposed mineral surfaces- the team found the highest measured rates of erosion and soil production (1). These chemical weathering rates (of about 2.5 mm/yr) are an order of magnitude higher than previously measured and higher than the suggested kinetically controlled limit. Moreover, the authors point out that soil weathering increases with erosion rate. The study also challenges the current dogma that high erosional environments are inefficient for soil production due to low residence time of key minerals. Even though denudation rates –the speed at which the landscape surface is eroding- exceed soil production in every catchment studied, the landscape is able to maintain a continuous soil mantle as a result of the dense vegetation cover (the main biotic soil producer in New Zealand due to the lack of native burrowing mammals). This seems to be a key point, as root expansion is an efficient mechanism for converting rock into soil, either by the physical breaking of bedrock or by increasing the concentration of organic acid and sub-surface CO2.

Thus, contrary to previous assumptions, this study demonstrates that rapidly uplifting mountains with dense vegetation cover are active weathering systems. The authors thus suggest that –

“These high weathering rates support the view that mountains play a key role in global-scale chemical weathering and thus have potentially important implications for the global carbon cycle”

Since weathering can only drive global climate trends if the erosion of silicate rocks removes enough CO2 from the atmosphere, the next step is to evaluate the amount of CO2 consumption resulting from the uplift of the Southern Alps.

This was the focus of another recent study in the Southern Alps using Calcium isotopes and Ca/Na ratios from rivers (2). They also show that weathering rates were an order of magnitude higher than the global mean and that the region has some of the highest chemical weathering rates in the world. However, they found that the majority of the weathering in non-glacial catchments was from carbonate rather than silicate rocks (a reaction that does not use atmospheric CO2), and thus the consumption of CO2 was no higher than the global average. Higher silicate weathering was only found in rivers downstream of glaciated regions. When these increased silicate weathering rates are extrapolated to all glaciated montane regions around the world, however, they only account for <1% of the global silicate weathering and global atmospheric CO2 consumption. Therefore they conclude that silicate weathering in uplifting mountain ranges like the Southern Alps has little effect on the global carbon cycle and long-term climate change.

Further work is clearly needed to better assess the weathering rates in tectonically active regions and then assess the role of mountain building and erosion in the global carbon cycle. Interestingly, enhancing weathering rates has recently been assessed as a potential geo-engineering strategy to offset increasing anthropogenic CO2 (3).

  1. Larsen et al., 2014. Rapid Soil Production and Weathering in the Southern Alps, New Zealand. Science 343, 637-640, doi:10.1126/science.1244908.
  2. Moore et al., 2013. Tracking the relationship between mountain uplift, silicate weathering, and long-term CO2 consumption with Ca isotopes: Southern Alps, New Zealand. Chemical Geology, 341, 110-127, doi:10.1016/j.chemgeo.2013.01.005.
  3. Hartmann, et al., 2013. Enhanced chemical weathering as a geoengineering strategy to reduce atmospheric carbon dioxide, supply nutrients, and mitigate ocean acidification. Reviews of geophysics, 51, 113-149.

Blog 8: Global volcanism linked to late-Quaternary deglaciations

By Ignacio A. Jara

It has been widely documented that volcanic aerosols can alter the radiative balance of the atmosphere, producing measurable temperature depressions following large explosive eruptions. Perhaps the most renowned of these cases is the eruption of the mount Tambora in Indonesia, which in 1815 caused a decline of 0.5°C in the Northern Hemisphere in what was known as “the year without a summer” (1).

But, what if this causal relationship is inverted and climate change affects the number of volcanic eruptions?

In 1979 Rampino et al., first used proxy data to argue that periods of climate cooling were associated with catastrophic volcanic events such as the latest Taupo eruption in New Zealand or the mega eruption of the Toba volcano in Indonesia (~70,000 cal yrs) (2). The authors boldly proposed that the changes in ice extent and sea level during glaciations resulted in sufficiently large variations in the Earth’s crustal stress to alter volcanic activity over longer time scales.

Since this pioneering study, a considerable number of publications have added local evidence supporting long-term climate variations as a driver for volcanic activity. However, solid evidence pointing to a definitive link between glacial cycles and volcanism at global scales as remained elusive.

An interesting article published in 2012 adds new insights into this theory. They analysed the timing of more than 400 tephra layers identified in marine sediment records around the Pacific “Ring of Fire” over the last 1 Myr (3). The spectral analysis of tephras deposited off South and Central America, Japan, the Philippines and the Southern Pacific islands reveals that periods of increased volcanic eruptions have a recurrence of 41 kyr, the exact periodicity of the Earth’s obliquity. Moreover, phase analysis indicates that peaks in volcanic eruptions lag behind minimum ice volume and maximum sea level by about 4 kyr.

Since obliquity has been recognized as one of the orbital pacemakers of the Pleistocene ice ages, these results indicate a direct link between orbital cycles, glacial/interglacial climate and global volcanism. The authors further suggest that this link could be mediated by surface pressure variations resulted from ice-ocean mass redistribution during periods of abrupt climate change. In this regard, volcanic eruptions along the edge of continental plates are expected to occur at greater frequency during periods of deglaciation, when ice retreat caused crustal pressure to decrease, lowering the compression of the rock overlying the magma chambers.

Undoubtedly these results are preliminary and much more work is required to better understand this phenomena. Even though the duration of the present-day warming trend is several orders of magnitude quicker than climate variations presented here, the interplay between climate change, ice extent and volcanism might be relevant in high-latitude regions with ice caps and active volcanic zones. As the atmosphere heats up at an unprecedented rate, the removal of large ice loads may contribute to the reactivation of volcanic complexes or the emergence of unknown volcanic systems.


1.            K. R. Briffa, P. D. Jones, F. H. Schweingruber, T. J. Osborn, (1998) Influence of volcanic eruptions on Northern Hemisphere summer temperature over the past 600 years. Nature 393, 450-455.
2.            M. R. Rampino, S. Self, R. W. Fairbridge, (1979) Can rapid climatic change cause volcanic eruptions? Science 206, 826-829.
3.            S. Kutterolf, M. Jegen, J. X. Mitrovica, T. Kwasnitschka, A. Freundt, P. J. Huybers, (2012) A detection of Milankovitch frequencies in global volcanic activity. Geology, (doi: 10.1130/g33419.1).

Blog 7: Dust deposition in the Southern Ocean

By Ignacio A. Jara and Helen Bostock

Terrestrial dust is important to the climate system not only because it may alter the solar radiative balance of the earth, but also because it supplies the oceans with key iron (Fe), a limiting micronutrient for phytoplankton productivity in the Southern Ocean. The “Iron Hypothesis” was first proposed by Martin (1990) [1], who suggested that changes in Fe supplied impact on the biological productivity in the Southern Ocean which, in turn, could influence the glacial-interglacial changes in atmospheric CO2.  However over the last 20 years scientists have struggled to find evidence to support this theory.

In the present-day Atlantic Southern Ocean an increase in terrestrial dust influx or volcanic aerosols has been linked to vast biological blooms. An increase in biological activity is associated with a high consumption of nutrients and transfers carbon back to the deep ocean. In 2009 an article published in Nature [2] explored past dust-climate interactions in this part of the Southern Ocean (42°S) by presenting an offshore dust record extending back to 4 million years.

This study of a long Ocean Drilling Program (ODP) core from the South Atlantic provided -with unprecedented detail- evidence for a consistently enhanced terrestrial dust influx during ice ages. It also showed a tight coupling with dust deposition in Antarctic ice cores over the last 0.8 million years, indicating that large areas of the Southern Ocean and Antarctica were affected by the glacial dust plume. According to the authors, the glacial dust was the result of an increased aridity of the eastern Patagonian plains (just upwind from the coring site), which increased the dust availability; as well as stronger and northward shifted Southern Hemisphere Westerly Winds, which enhanced the offshore transport of the Patagonian dust.

Has something similar occurred in larger Pacific sector of the Southern Ocean?

The dust record from a new set of cores collected by the RV Sonne 2010 from the Pacific Southern Ocean has recently been published in Science [3], and shows a similar pattern of variability in dust fluxes for this region over the last 1 million years [3]. The increased glacial dust deposition in the Pacific Southern Ocean is probably the result of enhanced supply from the Australian continent (with potential contributions from New Zealand), as modelling suggests this landmass is the main present-day dust source in the southern Pacific. However, geochemical fingerprinting of the dust will be required to determine the exact source.

Dust model

Figure: Modern terrestrial dust sources across the Southern Ocean based on modelling data (Lamy et al., 2014)

Interestingly, dust influxes during glaciations prior to 0.5 million years seem to have been significantly lower in the Pacific than in the Atlantic (Figure 2), something that may be explained by a less intense glacial desiccation of Australia due to its relatively northern position compared with Patagonia.

So has the increased dust/Fe translated into increased productivity?

There is evidence for some increase in productivity from opal flux, n-alkanes and barium concentrations in some of the cores north of the Polar Front. But there is a large reduction in productivity south of the Polar Front, resulting in an overall decrease in total biogenic opal production during the glacials in the South Pacific [4], and thus unlikely to draw-down the atmospheric CO2.

Dust and biology

Figure: Lithogenic and biogenic proxies from the subantarctic waters of the South Pacific (Lamy et al., 2014)

Modern iron fertilization experiments have also had mixed results; early experiments such as Ironex and SOIREE produced large blooms visible from space [5], while several subsequent experiments have not witnessed any major changes in phytoplankton concentrations. There is also the question of whether the blooms actually result in organic carbon being transferred to the deep ocean.

It is clear that more work is required to understand the link between dust, biological productivity and CO2 ventilation in the Southern Ocean .


  • Martin, J.H., 1990. Glacial-interglacial CO2 change: The Iron Hypothesis. Paleoceanography, 5, 1-13.
  • Martinez-Garcia, A., et al., 2011. Southern Ocean dust-climate coupling over the past four million years. Nature 476, 312-315.
  • Lamy, F., et al., 2014. Increased dust deposition in the Pacific Southern Ocean during glacial periods. Science 343, 403-407.
  • Bradtmiller L., et al., 2009. Comparing glacial and Holocene opal fluxes in the Pacific sector of the Southern Ocean. Paleoceanography, 24, PA2214, doi:10.1029/2008PA001693
  • Boyd P.W. et al (2007) Iron enrichment experiments 1993-2005: synthesis and future directions. Science 315, 5812, 612-7.

Blog 6:Changes in radiocarbon surface reservoir ages in the SE Pacific

By Helen Bostock

Over the last decade the main theory to explain the changes in the atmospheric concentration of CO2 between glacials and interglacials, has primarily focussed on changes in the circulation of the Southern Ocean controlling the release of CO2 from the deep-ocean reservoir. However, there is still considerable debate about the path and timing of the CO2 release during the deglacial. A recent study on sediment cores from the SE Pacific has shed new light on this debate. Siani et al., (2013) found changes in the surface reservoir radiocarbon (14C) age, determined from the difference in the 14C age of planktic foraminifera compared to tephra ages in cores from the SE Pacific. The study found that periods of increased surface reservoir ages were coeval with the timing of upwelling events in the Southern Ocean and increases in atmospheric CO2 during the deglaciation. The increased upwelling of old, carbon-rich, deep-waters is supported by reductions in the difference between 14C benthic-planktic foram ages and the stable carbon isotopes (d13C), the latter of which are primarily controlled by biological production and respiration in the water column. They see three periods of upwelling initiating with a short pulse at the start of the deglaciation at 18.5 ka, then between 17.5 and 14.5 ka and finally between 12.5 to 11.5 ka.

The authors also recalculated the deep-water reservoir age for intermediate depths from the SE Pacific (De Pol Holz et al., 2010) and show that there is older Antarctic Intermediate Waters during the deglaciation. Thus providing a pathway for this old CO2 signal from the Southern Ocean.

This is also a critical study for showing that the surface 14C reservoir age can vary considerably overtime and if we do not adjust for these changing background reservoir age we cannot accurately compare paleoclimate records from the oceans with the ice and terrestrial records. Changes in the surface 14C reservoir age during the last deglacial have previously been suggested for the New Zealand region by comparing the 14C age of planktic foraminifera with local tephra (Sikes et al., 2000). But recent improved dating of some of the widely deposited tephra’s for example the original increased surface reservoir age around the time of the Waiohau Tephra no longer exists (Lowe et al., 2013). Perhaps it is time for a relook at the surface 14C reservoir ages around New Zealand before confidently comparing with other global records. This is however, trickier in regions where there is no widely deposited tephra layers that can act as chronostratigraphic timelines (e.g. Australia, South Africa).  One critical region where it will be very important is the Southern Ocean.

This work suggests we need to be careful comparing marine records dated using 14C and an assumed constant reservoir age with other paleoclimate records… of the main objectives of the INTIMATE and SHAPE projects.


De Pol Holz et al., 2010. No signature of abyssal carbon in intermediate waters off Chile during the deglaciation. Nature Geoscience,  3 , 192-195.

Lowe et a., 2013 Ages of 24 widespread tephras erupted since 30,000 years ago in New Zealand, with re-evaluation of the timing and palaeoclimatic implications of the Lateglacial cool episode recorded at Kaipo bog. Quaternary Science Reviews,74, 170-194.

Siani et al., 2013 Carbon isotope records reveal precise timing of enhanced Southern Ocean upwelling during the last deglaciation. Nature Communications, doi:10.1038/ncomms3758.

Sikes et al., 2000. Old radiocarbon ages in the southwest Pacific Ocean during the last glacial period and deglaciation. Nature, 405, 555-559.

Blog 5: Precipitation and human occupation changes in arid environments

Written by Ignacio Jara, Victoria University Wellington.

While the abrupt climate transitions of the glacial termination and the establishment of the present-day modes of climate variation during the Holocene seems to be well characterized in proxy records around the Southern Hemisphere, there is still sparse evidence about the impact of those changes on the migration, settlement and cultural development of ancient human populations.

A recent approach to this subject includes to compile the radiocarbon dates from archaeological sites as a proxy for human density. The underlying assumption is that period of increasing human presence should be reflected as a higher number of sites with dates falling within such the time interval. Whereas biases and limitations may be reduced by using larger radiocarbon data set, the main advantage of this method is that it provides a chronology of human population that can be directly compared with climate proxies. Two newly-published articles use this technique to uncover human responses to late Quaternary climate events.

The first of them combines a data set of more than 5,000 radiocarbon dates from all around the Australian continent with geospatial modelling (1). The main goal is to investigate the association between contractions and expansions of aboriginal population and climate instability during the terminal Pleistocene.

In Australia, cold intervals such as the Last Glacial Maximum (LGM; 23,000-18,000 years ago) have been associated with increased aridity and the expansion of grassland and non-vegetated landscapes (see blog number 3). What it is new from this study is that human population seems to have experienced a significant reduction during the LGM, as indicated by a decrease in the total number of radiocarbon dates of that age. Moreover, archaeological sites show what appears to be an interruption of elaborated cultural behaviour common prior to the LGM such as rock art, ritual burials and coloured ornamentation. These types of cultural expressions are only resumed in the early part of the Holocene Period.

The hyper arid Atacama Desert in northern Chile is also an interesting region to investigate relationship between precipitation and population changes since it is a natural passageway to southern South America, an area where the oldest evidence of human settlement in the continent.

The most intense pulses of occupation in Atacama seem to be coeval  with late Quaternary rainfall events in the Central Andes. For instance, a well preserved archaeological site published this year shows continuous human occupation in the rainless part of the dessert during one of the last of these rainfall pulses between 12,700-9700 year ago (2). Under increasing moisture, the previously plantless landscape would have been scattered by small wetlands and woodlands, becoming oases for migrating populations and a bio-geographic corridor connecting the dessert with the more humid biomes of the south.

But perhaps more tantalizing is the cultural developments associated with increasing precipitation between 7,000-4,000 years ago in Atacama. Using a data set of more than 400 radiocarbon dates, a noteworthy scientific contribution published last year correlates increased rainfall with a notable increment in human density (3). The authors point out that more water availability and larger groups of people led to a period of rapid technological and cultural innovation which resulted in the emergence of the oldest examples of artificial mummification in the world (as early as 7,000 year ago).

Atacama mummy
Increasing water availability could have been one of
the environmental drivers behind the oldest known examples of artificial mummification in the Atacama Desert between 8,000 and 7,000 years ago. Photograph from Marquett et al. 2013.

Although arid environments can restrict human settlement and migration, they also provide extraordinary conditions for the conservation of archaeological sites which might be a great advantage for future studies that address climate-human relationship during the past. What makes the results meaningful is the possibility that natural conditions for human preservation became a distinctive and influential cultural trait upon early human groups. The preservation of the soul with the body after death is a common belief among traditional societies. So the lack of decomposition is not just a relevant issue for Quaternarists.


  1. A. N. Williams, S. Ulm, A. R. Cook, M. C. Langley, M. Collard, Journal of Archaeological Science 40, 4612 (2013).
  2. C. Latorre et al., Quaternary Science Reviews 77, 19 (2013).
  3. P. A. Marquet et al., Proceedings of the National Academy of Sciences 109, 14754 (September 11, 2012, 2012).

Blog 4: Long-term orbital changes and glaciations in the Southern South America

Written by Ignacio Jara, Victoria University Wellington

Although the long-term changes in solar insolation between Northern Hemisphere (NH) and the Southern Hemisphere (SH) are in anti-phase, marine and ice core records from both hemispheres show highly synchronous glacial/interglacial cycles over the last 800 kyr. This disparity between the insolation and the paleo-records is probably the most important caveat concerning Milankovitch’s theory of orbital parameters controlling Earth’s climate. Particularly puzzling for the SH is that glacial/interglacial transitions follow the NH summer insolation, and therefore deglaciations in southern latitudes occur under decreasing local summer insolation1. In order to resolve this apparent conflict it is necessary to develop new detailed glacial chronologies which are able to be compared with other glacial and climate reconstructions from both hemispheres.

Southern South America has the most extensive ice sheets of the Southern Hemisphere outside of Antarctica, and thus it is not surprising this area boasts outstanding geomorphologic evidence of late Quaternary glacial fluctuations. Yet, the timing of these fluctuations has remained more or less equivocal until recently. Two newly-published glacial reconstructions from this part of the world provide new interesting insights into this topic.

Firstly, a detailed geomorphological map of the Torres del Paine area (51°S) has been produced. This is a region with extensive glaciers and massive moraine belts extending up to 35 km from present-day ice margins2. The boulder exposure 10Be dates indicate a major glacial advance culminating at 14,200 cal yr BP, while basal 14C dates from peat sections embedded in the moraine complexes indicate that ice sheet remained extended until 12,500 cal yr BP.

This chronology is supported by a more recent publication of surface exposure and radiocarbon ages from moraine complexes in Tierra de Fuego (54°S), the southern-most tip of South America and a region whose geomorphology was previously poorly mapped and dated3. The authors attribute the lack of early-Holocene moraines as evidence for dry and warm conditions during this period and suggest that the glaciers may have reached near present-day positions as early as 11,200 cal yr BP.

Overall, these two reconstructions show clear evidence for a net glacial retreat between 16,000-11,000 cal yr BP, a period of decreasing SH summer insolation. Likewise, the Tierra del Fuego reconstruction shows the absence of sustained glacial retreat during the Holocene, a period of increasing summer SH insolation. Hence, these studies are consistent with the orbital/climate paradox, suggesting that the SH summer insolation does not control (at least not directly) the ice fluctuations on the Southern American continent.

So what caused this change?

Apart from the NH summer insolation, it has been argued that other orbital parameters such as the SH summer duration and SH spring insolation could be important drivers1. The remarkable detail in the glacial reconstructions presented in these two publications may provide other clues. The ice advances during the Antarctic Cold Reversal (ACR; 14,500-12,900 cal yr BP) are observed at both the Torres del Paine and Tierra del Fuego regions, suggesting a strong Antarctic influence. Moreover, ACR glacial advances have also been recorded in the Southern Alps of New Zealand4, suggesting a zonally synchronous response across the SH, which may point to the Southern Westerly Winds and/or the Southern Ocean as important players.

Taken together, these recent publications are important contributions to better constrain the timing and forces affecting the SH glacial history over the last 16,000 years. They also highlight the potential of Southern South America for glacial reconstructions, and show the importance of glacial geomorphology studies for late Quaternary global climate reconstructions.

S Am glaciers

Glacial retreat between 16,000-11,000 cal yr BP in Tierra del Fuego occurs under decreasing Southern Hemisphere summer insolation. ACR, YD and LIA denote Antarctic Cold Reversal, Younger Dryas and the Little Ice Age respectively. Pink circles and green squares indicate 10Be dates, while red triangles indicate radiocarbon dates from the different sites from Menounos et al. (2013).


1.            Huybers, P. & Denton, G. Antarctic temperature at orbital timescales controlled by local summer duration. Nature Geosci 1, 787-792 (2008).

2.            García, J.L., et al. Glacier expansion in southern Patagonia throughout the Antarctic cold reversal. Geology 40, 859-862 (2012).

3.            Menounos, B., et al. Latest Pleistocene and Holocene glacier fluctuations in southernmost Tierra del Fuego, Argentina. Quaternary Science Reviews 77, 70-79 (2013).

4.            Putnam, A.E., et al. Glacier advance in southern middle-latitudes during the Antarctic Cold Reversal. Nature Geosci 3, 700-704 (2010).

Blog 3: Environmental transformation of Australia linked to the Late Quaternary demise of the megafauna

Australia provides an outstanding case study to resolve the relationship between Late Quaternary environmental drivers such as climate variability, vegetation changes, wildfires, faunal extinctions and human activities. The interval between 50,000-40,000 years BP is critical for understanding the interplay of some of these factors and how they transformed the Australian landscape. During this period, humans arrived and spread throughout most the continent. At the same time, a diverse range of large browsing mammals, reptiles and birds became extinct; and there is evidence for a marked vegetation change, including more intense and frequent wildfires. Despite this paleo-environmental information, the lack of well dated environmental records has prevented scientists from resolving the relationships between these events.

Tim Flannery was probably the first to propose a causal relationship between the disappearance of great browsing mammals and the increase in fires (1). He suggested the disappearance of the big herbivores was caused by over-hunting, which triggered a massive change in the distribution and structure of plant communities that favoured wildfires, and the extinction of several other smaller animal species. Two recent high-resolution vegetation reconstructions have addressed this hypothesis, providing more detail and support for this theory.

Published last year, a radiocarbon-dated pollen and charcoal profile from northern Australia used changes of the abundance of the fungus spore Sporormiella – a genus of fungi that grows in herbivores dung - as a proxy for large browsing animal activity (2). Between 43,000-38,000 yr BP a succession of environmental changes started with a rapid decline of Sporormiella, followed by an increase in charcoal accumulation and subsequently followed by a decline of rainforest pollen taxa at the expense of grasses and Sclerophyll shrub species. Critically, fire and vegetation changes lagged behind the decline of browsing activity, suggesting that neither of these factors was directly responsible for the faunal extinction. Based on these results, the authors further suggest that the decline in herbivory led to a build up of burnable light fuels and resulted in the increase in wildfires.

A more recent environmental reconstruction from a marine sediment core offshore of southern Australia uses novel proxies for regional vegetation and wildfires over the last 130,000 years (3). Based on biochemical changes in lipids derived from leaf waxes, the regional abundance of C3 and C4 plants is inferred. These two groups of plants have different metabolisms reflecting their preferential distribution over the southern (cold climate with winter precipitation) and northern (warm climate with summer precipitation) parts of the continent respectively. Additionally, paleo-fire activity is inferred from the changes in the accumulation of a biomarker formed during burning and transported offshore by dust and smoke. The record shows how warming periods such as the onset of the present and last interglacial periods are associated with increases in C4 plants, while cooling events such as Last Glacial Maximum are associated with increased C3 plants. However, the most prominent drop in C4 plants between 44,000-42,000 yr BP does not match any climate event. This drastic vegetation transformation is accompanied by high fire activity and occurs right after the interval of disappearance of mega fauna. Similarly, the authors argue for a large-scale ecological transformation caused by the disappearance of large browsers.

These two recent articles provide evidence to support a notable ecosystem rearrangement occurring only after faunal extinction, and not the opposite way around. The implication of this is that hunting was probably the main, if not the only, driver responsible for this extinction process. The disappearance of large herbivores may have promoted the accumulation of fire-prone vegetation, permitting the occurrence and spread of human-lit fires. Further support for a leading role of hunting in the Australian late Quaternary mega fauna extinction come from the fact that other processes of faunal disappearance in the Americas (between 15,000-10,000 yr BP) and New Zealand (750 yr BP) occur coincidently with human colonization of these regions.

1.            T. F. Flannery, Archaeology in Oceania 25, 45 (1990).
2.            S. Rule et al., Science 335, 1483 (March 23, 2012, 2012).
3.            R. A. Lopes dos Santos et al., Nature Geosci 6, 627 (2013).

Australian vegetation

Relative abundance of C4 plant and seasonal precipitation regimes in the Australian Continent. A prominent decreased in C4 plant between 44,000-42,000 yr BP is preceded by the Late Quaternary mega fauna extinction period (49,000-44,000 yr BP). Figure taken from reference Lopes dos Santos et al. (2013).

Some Australian mega fauna browsers such as the Diprotodon (in the picture) may have weighed up to several tonnes.  Image courtesy of the South Australian Museum