Causes of Bond Events: A song of ice and fire

While there is still an ongoing debate about the presence of the 1500-year Bond Cycle, the processes triggering this oscillation remain even more mysterious. A number of hypotheses have been put forward, but all of them lack palaeological evidence. In this post I will give a brief overview of the ideas that scientists have come up with so far...

1. Orbital insolation

It is pretty common knowledge that our planet spins around its axis and orbits around the Sun and that is the exact reason we have things like times of the day and different seasons. If your familiarity with astronomy ends here, the next thing might slightly blow your mind, but... the Earth's position in relation to the Sun has not actually been the same in the past! We can blame it all on a series of complex gravitational interactions between the Sun and the Earth.

Taking it a step further, an absolute genius of a man, Serbian geophysicist and astronomer Milutin Milankovic mathematically theorised that variations in Earth's position relative to the Sun (aka its orbital parameters) have determined climatic patterns on our planet. Firstly, the Earth's orbit is NOT actually circular in shape, but varies from near circular to an ellipse over a period of about 100,000 years with a long cycle of about 400,000 years. The length of the long axis of the ellipse is described by a parameter known as eccentricity. Secondly, the Earth's axis of rotation with respect to the plane of its orbit varies between 21.8° and 24.4° every 41,000 years. This process, known as the change in obliquity, is responsible for the extent of the differences between seasons. Finally, there is precession (i.e. moving around a full circle), which could relate to either the elliptical orbit of the planet or to its axis of rotation. The combination of the two different precessions then produces the classically quoted precessional periodicity of 21,700 years.

Overall, variations in eccentricity, obliquity and precession significantly affect the amount of solar radiation a particular place on the planet will receive and therefore are some of the key reasons for climatic fluctiations.

I know that is a lot of new terms for you already and you probably wish for a little bit more detail on this topic, but sadly we have a lot to cover today. If all of the above sounds pretty confusing to you I suggest you have a look at my colleague's blog dedicated solely to the Milankovic cycles. Alternatively here is the best video on orbital parameters I've ever seen:

So what's the weakness of this hypothesis? Some scientists (e.g. Butikofer, 2007) believe that the periodicities of the orbital forcing are far too long to be the cause for the formation of the Holocene Bond events. The only possible option, in their opinion, is to associate the Bond Cycles with subharmonics of the orbital forcing.

2. Solar irradiance

The amount of radiation emitted by the Sun is also not constant throughout the time but expresses a significant cyclic variability. There is a very obvious gradual increase and more rapid decrease of the number of sunspots over the period of roughly 11 years, however Hathaway (2010) mentions numerous authors who have noted multi-cycle periodicities in the sunspot cycle amplitudes, which I have summarised for you in the table below:

Solar Cycle	Periodicity	Notes
Schwabe cycle	11 years	The most obvious sunspot cycle caused by the differential rotation of the Sun's convection zone
Halle Cycle	22 years	The magnetic field of the Sun reverses during each Schwabe cycle, so the magnetic poles return to the same state after 2 reversals
Gleissberg Cycle	70-100 years	Proposed periodicity of 7 or 8 cycles
Suess/de Vries Cycle	210 years
Halstatt Cycle	~2300 years

Source: Brooke (2012), http://www.youtube.com/watch?v=hhy9KjI-fLk

Given that the amount of radiation emitted by the Sun significantly influences the climate of the planet today it is logical to suggest that variations in the sunspot activity (shown on the graph below) could have been a trigger for the Holocene climatic variability.

Source: Solanski et al. (2005)

Gerard Bond himself looks like a big fan of this hypothesis, as well as several other authors (e.g. Chapman and Shackleton, 2000; Niggeman et al., 2003; Lamy et al., 2006; Allen et al., 2007), who considered that solar forcing was the determining or at least one of the determining factors of their reconstructed multicentury to millennial scale climate fluctuations. Bond et al. (2001) linked their drift-ice records with the production rate of the two nuclides ¹⁴C and ¹⁰Be (hope you remember me talking about them being proxies for solar irradiance) in Greenland ice cores and showed the both isotopic time series match very well with their drift-ice reconstructions.

3. Volcanic activity

When volcanoes erupt they can emit things like ash and sulphur dioxide into the air, which reflect sunlight away from the earth and therefore cause a cooling effect. This effect is especially marked in the case of large eruptions able to blast sun-blocking particles all the way up to the stratosphere. Therefore a hypothesis has been put forward that abrupt Holocene cooling events could have been caused by major volcanic eruptions.

Nevertheless, Cronin (2010) suggests that while having large impacts on climate, usually individual stratospheric eruptions lower the mean global temperature only by 0.2-0.3° and last only one to three years. He, however, admits that extended periods of more frequent stratospheric eruptions might, in theory, alter the climate over longer timescales. Still, no definitive link between the observed Holocene climate variability and volcanic eruptions is yet available. Nonetheless, Wanner et al. (2008) provides a reconstruction of strong tropical volcanic eruptions over the last 6000 years and it is striking that 12 out of the 19 eruptions observed during that time span occured in the last millennium with a considerable clustering during the Little Ice Age, which corresponds with the most recent Bond event.

4. Ice-sheet dynamics

When an ice sheet is thin the temperature at its base is <0°, but as the ice sheet thickens its basal temperature  increases until it reaches 0° and the basal melting occurs. Basal melting can lubricate the bed, causing the ice to surge, eventually thinning and refreezing (Cronin, 2010). Ice-surging in turn can lead to the discharge of iceberg armadas into the North Atlantic surface layer, lowering the salinity and reducing deep-water formation form the conveyor belt. As THC diminishes, less heat is drawn to high northern latitudes, and this acts as a negative feedback leading to increased ice growth in the Laurentide area and a self-sustaining cycle is created (Cronin, 2010).

This theory extends the explanation put forward by MacAyeal (1993) for the cause of the glacial Dansgaard-Oeschger oscillations into the Holocene and has now also acquired a decent number of supporters (e.g. Seidov and Maslin, 1999; Jennings et al., 2002).

5. Other processes

Other catalysts for Bond Cycles have been proposed such as tidal forcing (Keeling and Whorf, 1997), the influence of cosmic rays and the global electric circuit (Kirkby, 2007) and changes in the Earth's magnetic field (Courtillot et al., 2007). If you want to undertake some further research on this topic, I suggest you have a look at pages 176-183 in Chronin (2010), where he provides a good review for a number of other proposed trigger mechanisms.

Overall, although lack of supporting evidence means there is still no consensus on the cause for Holocene climatic fluctuations, it is most likely that more than one factor was responsible for the rapid cooling events. Wanner and Butikoffer (2008) tried to represent the complex mechanism of a formation of a cold Holocene event through the following diagram:

Nonetheless, they admit that their representation of Bond event trigger mechanisms is still speculative and must be investigated with further analyses of marine and other palaeoclimatic records as well as suitable model runs.

List of references:

Allen, J., A. Long, C. Ottley, D. Pearson, and B. Huntley (2007) 'Holocene Climate Variability in Northernmost Europe', Quaternary Science Reviews, 26, 1432-1453.
Bond, G., B. Kromer, J. Beer, R. Muscheler, M. Evans, W. Showers, S. Hoffmann, R. Lotti-Bond, I. Hajdas and G. Bonani (2001) 'Persistent Solar Influence on North Atlantic Climate during the Holocene', Science, 278, 1257-1266.
Butikofer, J. (2007) 'Millennial Scale Climate Variability during the Last 6000 Years - Tracking Down the Bond Cycles', Geographichesches Institut, Universitat Bern.
Chapman, M. and N. Shackleton (2000) 'Evidence of 550-Year and 1000-Year Cyclicities in North Atlantic Circulation Patterns during the Holocene', The Holocene, 10, 287-291.
Courtillot, V., Y. Gallet, J.-F. le Mouel, F. Fluteau and A. Genevey (2007) ' Are There Connections between the Earth's Magnetic Field and Climate?', Earth and Planetary Science Letters, 253, 328-339.
Cronin, T. (2010) Paleoclimates: Understanding Climate Change Past and Present, Columbia University Press: New York.
Hathaway, D. (2010) 'The Solar Cycle', Living Reviews in Solar Physiscs, 7, 1, 5-56.
Jennings, A., K. Knudsen, M. Hald, C. Hansen and J. Andrews (2002) 'A Mid-Holocene Shift in Arctic Sea Ice Variability on the East Greenland Shelf', The Holocene, 12, 49-58.
Kirkby, J. (2008) 'Cosmic Rays and Climate', Surveys in Geophysics, 28, 333-375.
Lamy, F., H. Arz, G. Bond, A. Bahr and J. Patzold (2006) 'Multicentennial-Scale Hydrological Changes in the Black Sea and Northern Red Sea during the Holocene and the Arctic/North Atlantic Oscillation', Paleoceonography, 21, PA1008.
MacAyeal, D. (1993) 'Binge/Purge Oscillations of the Laurentide Ice Sheet as a Cause of the North Athlantic's Heinrich Events', Paleoceonography, 8, 775-784.
Niggemann, S., A. Mangini, M. Mudelsee, D. Richter and G. Wurth (2003) 'Sub-Milankovitch Climatic Cycles in Holocene Stalagmites from Sauerland, Germany', Earth and Planetary Science Letters, 216, 4, 539-547.
Keeling, C. and T. Whorf (1997) 'Possible Forcing of Global Temperature by the Oceanic Tides', Proceedings of the National Academy of Sciences, 94, 8321-8328.
Seidov, D. and M. Maslin (1999) 'North Atlantic Deep Water Circulation Collapse during the Heinrich Events', Geology, 27, 23-26.
Solanski, S., I. Usoskin, B. Kromer, M. Schussler and J. Beer (2004) 'An Unusually Active Sun during Recent Decades Compared to the Previous 11,000 Years', Nature, 431, 7012, 1084-1087.
Wanner, H., J. Beer, T. J. Crowley, U. Cubasch, J. Flückiger, H. Goose, M. Grosjean, J. Kaplan, M. Küttel, O. Solomina, T. Stocker, P. Tarasov, M. Wagner, and M. Widmann (2008) 'Mid- to Late Holocene Climate Change: An Overview', Quaternary Science Reviews, 27, 19-20, 1791-1828.

Fact or fiction?

Almost every new proposed theory has to go through a lot of heavy criticism before it becomes accepted (think of plate tectonics, poor Wegener!). Due to the fact that the concept of cyclic Holocene climate oscillations is rather new and goes contrary to the view scientific community adopted until very recently there is a great deal of heated debate relating to this issue.

To understand the nature of this debate in more depth we are going to work with the three questions put forward by Butikofer (2007):

1) Where are Bond Cycles postulated in paleoclimate record?
2) Is there really a global imprint of millennial (∼1500 years) scale cycles with shifts from cold to warm (dry to wet, respectively) climate?
3) What are the possible processes causing Bond Cycles during the Holocene?

Feeling that the despite your passion for climatology there is a limited amount of information you can intake on a casual Tuesday evening, I decided to focus only on the first two questions in this post. Next time our discussion will be centred around all the proposed causes for the Bond events and I am sure you are already very excited for that. But for now let's at first look at how it all started...

The first ever research to suggest that Holocene atmospheric circulation above the ice cap was punctuated by a series of millennial-scale shifts was by O'Brien et al. (1995), which focused on the analysis of soluble impurities from Greenland ice cores. They argued that the increases in soluble impurities in the ice cores probably occured at times of lowered atmospheric temperatures. However, the main breakthrough came when, encouraged by the findings of O'Brien and her colleagues, Bond et al. (1997) launched an investigation of deep sea Holcene sediments in the North Athlantic. These sediments were analysed in terms of presence of lithic/petrologic tracers in them, which led Bond et al. (1997) to suggest that ice-rafting also occured during the Holocene and was caused by a series of ocean surface coolings with a cyclic signal centered on 1470 +/- 532 years. Overall, the investigation presents the evidence for "pervasive, at least quasiperiodic, climate cycle occuring independently of the glacial-interglacial climate stage".

Gerard Bond and his colleagues are not the only ones who believe in the existence of Holocene quasi-periodic climate fluctuations. Niggemann et al. (2003) also came to the conclusion about the operation of a 1450-years cycle after looking at stalagmites (another type of proxy!) in the Atta cave in Sauerland, Germany, which were analysed in terms of their δ¹⁸O composition. The peaks of δ¹⁸O in the stalagmites indicate sections where calcite formed during periods of lower humidity in the cave, which could be attributed to potentially drier winters; winter precipitation is in turn related to the strenght of the Northern Hemisphere annular mode (NAM) through the intensity of westerlies.

The final argument here are the findings of Willard et al. (2005), who, when interpreting pollen data from Chesapeake Bay in the mid-Atlantic region of the United States, discovered significant decreases in pine pollen every 1400 years. These decreases have been said to represent the decreases in January temperatures in between 0.2 and 2 degrees Celcius. The timing of pine minima is correlated with a series of quasi-periodic cold intervals documented by various proxies in Alaskan, Greenland and North Atlantic cores.

On the other hand, Butikofer (2007) reviewed a number of papers on Holocene climate variability (including the ones mentioned above) and concluded that there is not enough evidence to confirm the existence of globally extensive climate cycles operating with a distinct pacing near 1500 years during the past 6000 years and "the previously claimed similarity between Holocene millennial-scale variability and the dominant 1470-year glacial periodicity appears questionable". This idea was later further developed in Wanner and Butikofer (2008).

It is rather hard to summarise all the arguments for both supporters and opponents of this proposed theory in a single article. Therefore, if you feel like investigating this topic in more depth, I strongly suggest you look at the Table 1 in Wanner and Butikofer (2008), where a good overview of 28 research papers discussing the phenomenon of Bond Cycles is provided.

List of refecences:

Bond, G., W. Showers, M. Cheseby, R. Lotti, P. Almasi, P. deMenocal, P. Priore, H. Cullen, I. Hajdas and G. Bonani (1997) 'A Pervasive Millennial-Scale Cycle in the North Atlantic Holocene and Glacial Climates', Science, 278, 1257-1266.
Butikofer, J. (2007) 'Millennial Scale Climate Variability during the Last 6000 Years - Tracking Down the Bond Cycles', Geographichesches Institut, Universitat Bern.
Niggemann, S., A. Mangini, M. Mudelsee, D. Richter and G. Wurth (2003) 'Sub-Milankovitch Climatic Cycles in Holocene Stalagmites from Sauerland, Germany', Earth and Planetary Science Letters, 216, 4, 539-547.
O'Brien, S., L. Meeker, D. Meese, M. Twickler and S. Whitlow (1995) 'Complexity of Holocene Climate as Reconstructed from a Greenland Ice Core', Science, 270, 5244, 1962-1964.
Wanner, H. and J. Butikofer (2008) 'Holocene Bond Cycles: Real or Imaginary?', Geografie - Sbornik Ceske Geograficke Spolecnosti, 113, 4, 338-350.
Willard, D., C. Berhardt, D. Korejwo and S. Meyers (2005) 'Impact of the Millennial-Scale Holocene Climate Variability on Eastern North American Terrestrial Ecosystems: Pollen-Based Climatic Reconstructions', Global and Planetary Change, 47, 1, 17-35.

Written in the Earth

It has been a while since I promised to tell you more about climate proxies and palaeoevidence for Bond cycles in my last post. I am sure that since then you have been dying for more information on this exciting subject, so without further ado I shall prove you the fact that we are able to transfer ourselves almost a million years back in history by simply using... a drill.

To do that let's go back to the fancy word proxy we have learnt last week.

A proxy, as defined by IPCC (2001), is "a local record that is interpreted using physical or biophysical principles to represent some combination of climate-related variations back in time". What it basically means, is that when a particular material (i.e. sediment, ice or even a living organism) forms, it reflects the physical and chemical characteristics of the environment in which it has been formed - usually through the character of deposition or the rate of growth. By determining the age of the object scientists can later reconstruct the environmental conditions of the time in which it was laid down or grew. Examples of proxies include ice cores, boreholes, corals, lake/ocean sediments, tree rings and sub-fossil pollen and are nicely summarised in the picture below I found on the NOAA website.

Source: NOAA, http://www.ncdc.noaa.gov/paleo/paleo.html.

Climatic changes very often influence isotopic ratio of the analysed proxy. Do not worry if this word seems familiar, but you can't actually remember what it means... Isotopes are basically atoms of the same element with the same number of protons and electrons and but different number of neutrons, which means various isotopes will have similar charges but different masses. For elements of low atomic numbers these mass differences will allow various physical, chemical and biological processes to change the relative proportions of various isotopes (aka the isotopic ratio), which could later be used to infer about the processes that formed a particular isotopic ratio. For example, Beryllium-10, Calcium-II and Carbon-14 are the isotopes used to calculate the intensity of solar irradiance at an established epoch (Nahle, 2009).

Some other proxies may enable a direct measurement of the chemical tracers of the past: gas bubbles trapped in ice reflect the composition of the Earth's atmosphere at the time of the ice formation or the concentration of aragonite, calcite and magnesium-calcite in fossial shells of molluscs and foraminiera are analysed to determine the environmental temperature (Nahle, 2009).

In order to successfully track down Bond Cycles scientists need to recreate long term records of temperature and greenhouse gases. Ice cores, cylinders of ice drilled out of glaciers and polar ice sheets, are especially favoured by the palaeoclimatologists and are one of the most effective and widely used proxy archives. Both at high altitudes and in polar regions snow falls during an annual cycle and remains there permanently. Eventually, the layers of snow compact due to their own weight and become ice, which preserves the environmental conditions during the snowfall.

The longest ice record has been available from Antarctica due to the fact that it is mostly located over land. Lorius et al. (1985) reviewing the data from the Antarctic ice gives a special mention to the the EPICA (deepest in the world!) and Vostok cores. At the same time, the only major area in the Arctic covered with snow is Greenland. A number of drilling projects are taking place there at present, with GISP2 and NGRIP being particularly famous, although they are nowhere near as deep as the cores from the Antarctic. The location of the main coring sites is shown on the maps below:

Source: USGS National Ice Core Laboratory, http://nicl.usgs.gov/coresite.htm

Ice cores are most commonly analysed in terms of their δ¹⁸O isotopic record, which is used to infer about past temperatures and snow accumulation. Although the water in the oceans contains primarily ¹⁶O, 12% of the oxygen in the oceans has an atomic weight of 18. The lighter ¹⁶O evaporates more easily, while the heavier ¹⁸O condenses more readily as the conditions get colder. The ratio of two isotopes will vary depending on the temperature of evaporation and can be measured very accurately. Over long periods of time this ratio provides a clear indicator of the average temperature of in the region of the coring site. Moreover, water contains hydrogen isotopes ¹H and ²H (also known as deuterium), which are also used as temperature proxies (USGS, 2004). Normally, ice cores from Greenland are analysed for δ¹⁸O and those from Antarctic for δ-deuterium, although GISP2 investigators are trying to perform the δ-deuterium analysis aswell as it "will allow even finer detail about source temperature and condensation history" (GISP2, 2006).

This post was initially meant to be dedicated to the information inferred from the natural records, however I got bit too excited talking about different proxies and wrote slightly more than the intended small introductory paragraph. Therefore, not wishing to overload you with information I decided to split the big topic into two parts and the next part will focus on all the wonderful stories that have been narrated to us by the ice cores.

List of references:

GISP2 (2006) 'Ice Cores that Tell the Past' (WWW), GISP2, (http://www.gisp2.sr.unh.edu/MoreInfo/Ice_Cores_Past.html), 10/11/2012.
Lorius, C., J. Jouzel, C. Ritz, L. Merlivat, N. Barkov, Y. Korotkevich and V. Kotlyakov (1985) 'A 150,000-year Climatic Record from Antarctic Ice', Nature, 316, 591-596.
Nahle, N. (2009) 'Correlation between Total Solar Irradiance and Iron Stained Grains during the Last 420 Years' (WWW), Biology Cabinet (http://www.biocab.org/Hematite_Stained_Grains_and_TSI.html), 8/11/2012.

IPCC (2001) IPCC Third Assessment Report: Climate Change 2001, Cambridge University Press: Cambridge.
US Geological Survey (2004) 'Fundamentals of Stable Isotope Geochemistry' (WWW), USGS (http://wwwrcamnl.wr.usgs.gov/isoig/res/funda.html), 10/11/2012.