By George Weber
Toba Volcano 4. Aftermath: Climate and Environment
|Ice samples from drilling cores are used to establish the chemical
composition of ancient atmospheres (especially the content of greenhouse gases)
as well as the temperatures of past climates. In Greenland for more than hundred
thousand years, snow has fallen regularly every year - and not melted again.
This has created a virtually unbroken sequence of layers that can be analysed.
Most famously the data comes from the "North Greenland Ice Core Project" whose
latest core published in 2004 is more than 3 km long and covers the last 123,000
years (ref. North Greenland Ice Core Project Members, 2004, Nature
One of the most important ways to establish the climate fluctuations of the past is the "Oxygen Isotope" method in which the relative abundance of O16 and O18 isotopes in air bubbles trapped in the snow/ice are compared. The abundance depends on the temperature of the cloud from which the snow has fallen long ago.
The notation for this is "delta-O":
Distinctive warm and cold periods called "Oxygen Isotope Stages" (OIS) have been established with the method. Those relevant to our subject are listed here:
OIS 1 - 12,000 years ago until today (warm)
OIS 2 - 45,000 to 12,000 years ago (cold)
OIS 3 - 63,000 to 45,000 years ago (warm)
OIS 4 - 73,000 to 63,000 years ago (cold)
OIS 5 - 130,000 to 74,000 years ago (warm)
OIS 6 - 190,000 to 130,000 years ago (cold)
It is to be expected that a volcanic eruption of VEI8 magnitude would have a major effect on local and world climate, possibly for an extended period of time. What precisely were the climatic changes caused by Toba, how long have they lasted and how much did they affect life on earth? The following charts, adapted from scientific literature, try to give an answer to these questions.
Fig. 4-1. The climatic effects caused by the Toba YTT event show up far more strongly in the northern hemisphere. Though relatively close to the equator, Toba did erupt into the northern hemisphere which, moreover, is more sensitive to such events than the southern half . The northern hemisphere has more land in relation to sea area than the south and this means that the climatic buffering provided by the sea is reduced in the northern half. (Physical world map curtesy of Natural History Museum).
Numbers refer to the relevant illustrations in this chapter 4,e.g. "2" refers to Fig 4-2)
Fig. 4-2. Volcanic eruptions over the past 600 years have produced evidence that volcanoes do indeed cause "volcanic winters". For example, the Pinatubo eruption in the Philippines in 1991 ejected in the region of "only" 4 sq. km of ash which resulted in a brief world-wide cooling of between 0.5 and 0.7oC (ref. Dutton et al. 1992, Huang et al., 2001). Similar effects have been found in other recent eruptions. While volcanic eruptions are not, of course, the sole cause for climatic anomalies, it is notable that during the last 600 years they have been responsible for all the larger anomalies. The famous "Years without Summer" following the Tambora eruption of 1815 is clearly visible in the chart above (although it is not the deepest ). The relationship between the size of an eruption and the following climatic effects is a complex one and depends, among other things, on the geographic location and the season during which the eruption takes place. It is also noticeable that none of the climatic effects lasted more than a few years even following VEI7 events . The situation , however, is quite different with very large eruptions of VEI8 class. (Chart and following table adapted from ref. Briffa K.R. et al., 1998).
|No.||Year||Volcano||Region||VEI class||Dating method(if not historical)
||Season1 - Dec-Feb|
2 - Mar-May
3 - Jun-Aug
4 - Sep-Nov
|2||1452||Kuwae||Vanuatu, SW Pacific||6||?|
|5||1480||St. Helens||USA (Washington)||5||?|
|6||1482||St. Helens||USA (Washington)||5||Dendrochronology||?|
|7||1580||Billy Mitchell||Bougainville, SW Pacific||6||C14||?|
|13||1660||Long Island||New Guinea||6||?|
|20||1800||St. Helens||USA (Washington)||5||Dendrochronology||1|
|23||1853||Chikurachki||Far Eastern Russia (Kuril islands)||5(?)||1|
|24||1854||Sheveluch||Far Eastern Russia (Kamchatka)||5||1|
|25||1883||Krakatoa||Indonesia (between Java and Sumatra)||6||3|
|26||1886||Okataina (Tarawera)||New Zealand||5||3|
|28||1907||Ksudach||Far Eastern Russia (Kamchatka)||5||2|
|29||1912||Novarupta (Katmai)||USA (Alaska)||6||1|
|30||1932||Azul, Cerro (Quizapu)||Chile||5||2|
|31||1956||Bezymianny||Far Eastern Russia (Kamchatka)||5||2|
|32||1980||St. Helens||USA (Washington)||5||2|
In 1815 an eruption of Tambora (no. 21 above) killed 12,000 people through a tidal wave that destroyed many coastal villages. It also caused at least 80,000 people to die through famine in the northern hemisphere. The 40 cu. km of ash that this VEI7 eruption blew into the earth's atmosphere caused "the Years without Summer" in 1816-18, with widespread crop failure and famine. Pre-industrial Europe was then still suffering from the devastation brought on by the Napoleonic wars and that, of course, aggravated the situation. The climate change affected the northern half more than the southern which was then much more sparsely populated than the north. One fact stands out: the Tambora eruption produced "only" 40 cu. km of ash yet it influenced the world climate over several years. Toba blew at least 16 times more ash into the atmosphere as well as a vast amount of sulphuric acid, and that its debris reached much higher levels, staying in the atmosphere much longer. Particles blown into stratosphere could have remained there for centuries, even millennia, blocking and altering the influx of solar energy to the lower atmosphere.
There is good and growing evidence that the YTT event has done just that (ref. Zielinski G.A. et al., 1996, Zielinski, 2000). The surface sea level temperatures in the South China Sea dropped by 1oC for 1000 years (Huang C.Y. et al. 2001) and the average atmospheric surface temperature in the northern hemisphere cooled by 3-5oC for several years (Rose W.I. et al., 1990; Rampino M.R. et al., 2001). The following figure shows that YTT was indeed followed by an ice age of considerable intensity and long duration. Moreover, Greenland ice cores have produced evidence of a rapid and enormous cooling event in the northern hemisphere at the time of the YTT event : the air temperature in Greenland dropped by 16oC within 160 years and then rose again slowly (ref. Lang C. 1999). The high-resolution pollen record from Grand Pile in France also shows the rapid onset of cold, dry steppe conditions at around 70,000 years ago (ref. Woillard G.M., 1978; and Woillard G.M. et al., 1982). It is also notable that the fossil record along the southern Mediterranean shows an abrupt replacement of the Afro-Arabian by a palaeo-arctic biotic community at precisely the boundary between the isotope stage 5 and 4 (ref. Tchernow E., 1992a and 1992b). Included in this exchange of biota, incidentally, was also the replacement of early modern humans (Homo sapiens) by the more cold-adapted Homo neanderthalensis in the Mediterranean area.
Fig 4-3. Indication of climatic conditions based on the the 18O/16O isotope curve from 15 deep sea cores taken from the Indian Ocean west and northeast of Sumatra. The OIS (Oxygen Isotope Stages) are a convenient way to identify the long climate records of the pleistocene. The various methods used to construct ancient climates vary in their results (climate being a very complex set of more or less interdependent values) but they roughly agree on the sequence of cold and warm periods and on their approximate dates. Note the remarkable similarities between Fig. 4-3 (Oxygen isotope, Sumatra) with Fig. 4-4 (pollen analysis, Italy). Both methods are quite different, yet both show cooling to have started sometime before the YTT event. (Chart adapted from Ninkovic D. et al., 1978).
Fig. 4-4. Another indicator of climatic conditions is based on botanical data (expressed as percentages), in this case the number of tree taxonomic groups flourishing around Lake Monticchio in the Basilicata region of southern Italy. (Chart adapted from Stringer Ch. et al., 2001).
Fig. 4-5. A pollen count from the Grand Pile peat bog (near Belfort in eastern France). The similarity of the pollen count curve to the previous Fig. 4-4 is notable. (Chart adapted from Woillard G.M., 1976;, and Woillard G.M. et al., 1981).
Fig. 4-6. The YTT event erupted vast amounts of climate-forcing sulphuric compounds (above all H2SO4, sulphuric acid) into the atmosphere where aerosols formed that increased atmospheric opacity and caused surface cooling. Greenland ice cores on the other side of the world show a spectacular spike of SO42- (the largest of the last 110,000 years!) of sulphuric acid. That the spike is of volcanic origin was demonstrated by measuring its electroconducitivity (ECM). At the same time, no characteristic ash particles could be found in filtered meltwater samples from the acidity peak, suggesting a distant source. All these pointers plus the dating of the event leave little doubt that this is the signature of the Toba YTT event (ref. Rampino M.R. at al., 2000). In Greenland, the sulphuric acid spike was followed by a marked cooling event lasting at least 1,000 years. (Chart adapted from Zielinski G.A.,1996 ).
Fig. 4-7. Traces of Toba in the Greenland ice. The blue line shows a record representing sea level changes. The black line shows a 10,000-year smoothed oxygen isotope profile, representing the climate changes. Toba here is clearly not at the beginning of a cold period but about half-way through, apparently accelerating and deepening it. The climatic downturn intensified by the Toba YTT eruption was a world-wide event. (Chart adapted from Andersen K.K. et al. North Greenland Ice Core Project, 2004).
Fig. 4-8. That the cooling caused by Toba also affected the northern Pacific was revealed with GRAPE (Gamma Ray Attenuation Porosity Evaluator). This is a method originally developed for oil prospecting; it measures the attenuation of gamma rays passed through a sample core, giving information on ancient sedimentation and climatic conditions at the time of deposition. The data produced by GRAPE correlates remarkably well with the GRIP data from distant Greenland, indicating that the Toba YTT event affected the climate of the entire northern hemisphere. The two sites for GRIP are at 50o 21.8' N, 167o 36.0 E (Site 882) and 51o 11.9' N, 167o 46.1 E (Site 883), off the east coast of Kamchatka peninsula in Siberia and southwest of Attu island in the Aleut chain, off Alaska. (Chart adapted from ref. Kotilainen A.T. et al., 1995).
|Grey bars indicate stadials before the onset of full glacial conditions.||LM = oxygen isotope stage (OIS)|
|AP - arboreal pollen data||M IS - marine isotope stage|
|DD = Dry density (grain size indicates maximum grain size of minerogenic detritus)||NAP - non-arboreal pollen data|
|GRIP - Greenland ice project||TOC - total organic carbon|
|1-8||Climate proxy data from the Lago Grande de Monticchio in the Basilicata region of southern Italy|
|9||Arboreal pollen data (ref. Allen et al., 1999)|
|10||Oxygen isotope stages|
|11||Oxygen isotope climate proxy data from the Pacific Ocean core V19-29, off the coast of Ecuador (refs. Pisias et al., 1984; Martinson et al., 1987)|
|12||Oxygen isotope climate proxy data from the Greenland ice core (ref. Johnsen et al ., 1992)|
1. InsolationUpper blue curve:
mid-July insolation at 65oN
Lower black curve:
annual mean insolation at 75oS
2. DeuteriumUpper blue curve:
Delta Deuterium values from EPICA dome C
at 75o 06' S, 123o 21' E
Lower red curve:
comparison curve from Wostok station
3. Marine oxygen isotope recordBlue line:
marine oxygen isotope record
Dotted red line: comparison line from another site to show uncertainties
4. DustRed line:
amount of dust recorded in the ice core
The sharp spike following the Toba YTT event with some delay is the only indicator from Antarctica that might possibly have a connection with Toba
Fig. 4-11. A number of climate proxies are shown here for Lake Baikal in Siberia. The lake is not only the deepest but also the oldest existing fresh-water lake on earth. Bottom sedimentation reaches more than 6 km (3.8 statute miles) in places. Dark shading indicates glacial periods. (Chart adapted from Prokopenko A.A. et al., 2002).
Fig 4-12. Sea levels at the Huon Peninsula, at the Pacific-facing eastern coast of Papua New Guinea. Unlike climate, sea levels do not seem to have been influenced much or for long periods by the YTT event. (Chart adopted from Chappell N.J. et al., 1986).
The graphs shown above have been assembled by scientists through the use of widely different methods of measuring/dating past climatic sequences. Most show that around the time of the Toba YTT event an unusual and major climatic event happened. Also: most charts confirm that while the YTT event did not start the climatic change, it intensified and prolonged it. The climatic downturn that Toba erupted into just after it had started, in fact, lasted with interruptions for 63,000 years until the onset of the present warm holocene around 10,000 years ago. As ref. Rampino et al, 1992, p. 52, states:
The detailed record of climate and delta-18O during the (oxygen isotope) stage 5a-4 transition reveals that although sea-level lowering began before the Toba eruption, North Atlantic surface-ocean temperatures remained warm, and ice sheet growth was beginning to slow. It may be significant, therefore, that the Toba eruption apparently coincided with a precipitous decrease in North Atlantic surface temperatures and global sea level. Increased sea ice and snow cover following the eruption may have provided the extra 'kick' that caused the climate system to switch from warm to cold states.
Immediately following the YTT event, sea levels dropped rapidly in the Atlantic and the first significant peak in "ice raft detritus" accumulation occurred the North Atlantic at precisely this time (Ruddiman W.F., 1977; Heinrich H., 1988). Ice raft detritus consists mostly of sand and stones scraped off the land surface by moving glaciers. Melting ice rafts (icebergs) drifting into warmer waters lose their detritus load which then sinks to the sea-bed where it accumulates to form identifiable and dateable strata.
A sea surface temperature drop of 1oC and a atmospheric cooling of 3-5oC may not sound much, but it should be remembered that these are average figures for one entire hemisphere. Average drops in temperature as estimated (let alone the truly staggering drop reported from Greenland of 16oC, ref. C. Lang, 1999) would have a profound and long-lasting effect on earth climate. A temperature drop of 1-3oC can make the difference between moderate and Siberian conditions in temperate places like central Europe or the USA and southern Canada. At higher latitudes, most temperate and subarctic forests would have been killed or severely damaged. Cold-sensitive tropical vegetation would have suffered similarly. The following text is from Rampino and Ambrose, 2000, pp. 75 -78 (dotted lines ... indicate were original text has been left out):
The climatic and environmental impacts of the Toba super-eruption are potentially so much greater than that of recent historical eruptions ( e.g. Hansen at al., 1992; Stothers, 1996) that instrumental records, anecdotal information, and climate-model studies of the effects of these eruptions may not be relevant in scaling up to the unique Toba event (ref. Rampino et al 1988; Rampino et al., 1993a). Various studies on the effects of extremes of atmospheric opacity and climate cooling on the environment and life have been carried out, however, in connection with studies of nuclear winter and the effects of asteroid impacts on Earth (e.g. Harwell, 1984; Greene et al., 1985; Tinus et al., 1990) and some of these may be more relevant to the Toba situation. Two major effects on plant life from high-atmospheric opacity are reduction of light levels and low temperatures. For aerosol optical depths between ca. 1 and ca. 10 the reduction in light levels expected from the Toba eruption would range from dim-Sun conditions (ca. 75% of sunlight transmitted), like those seen after the 1815 Tambora eruption, to that of an overcast day (ca.10% sunlight transmitted). Experiments with young grass plants have shown how net photosynthesis varies with light intensity. For a decrease to 10% of the noon value for a sunny Summer day, photosynthesis was reduced by ca. 85% (ref. van Keulen et al., 1975), and photosynthesis also drops with decreasing temperatures (ref. Redman, 1974).
Resistance of plants to unusual cold conditions varies considerably. Conditions in the tropical zone are most relevant to possible impacts on early human populations in Africa. Tropical forests are very vulnerable to chilling, and Harwell et al (1985) argue that for freezing events in evergreen tropical forests, essentially all above-ground plant tissues would be killed rapidly (see also Taylor et al., 1971; Sweeney et al., 1975). Average surface temperatures in the tropics today range ca. 16oC to 24oC. Nuclear winter scenarios predict prolonged temperature decreases of 3-7oC in equatorial Africa, and short-term temperature decreases of up to 10oC. Many tropical plants are severely damaged by chilling to below 10-15oC for a few days (ref. Leavitt, 1980; Hutchinson et al., 1985; Greene et al., 1985). Harwell (1985) compiled data showing that the LT50 data (temperatures required to kill at least 50% of the plants after exposure to cold for 2 hours or more) for most tropical plants was in the range +5 to -2oC. Seedlings and saplings are most vulnerable. Even more serious is the fact that most tropical forest plants have limited seed banks, and the seeds typically lack a dormant phase. Furthermore, regrowth tends to produce forests of limited diversity, capable of supporting much less biomass (ref. Harwell et al., 1985).
Even for temperate forests, destruction could be very severe (ref. Harwell, 1984; Harwell et al., 1985). In general, the ability of well-adapted trees to withstand low temperatures (cold-hardiness) is much greater than is needed at any single time of the year, but forests can be severely damaged by unusual or sustained low temperatures during certain times of the year. For example, Tinus et al. (1990) estimated the cold-hardiness of Rocky Mountain Douglas fir based on controlled growth experiments in which they derived the temperature decrease that would be necessary to kill at least 50% (LT50) of the fir trees. They found that a simulation of a 10oC decrease in temperature during Winter would have a minimal effect on the cold-hardy and dormant trees, whereas a similar 10oC drop in temperature during the growing season (when cold-hardiness is decreased) leads to a 50% dieback and severe damage to the surviving trees (including damage of new growth and invasion of pathogens, resulting in at least a year's loss of growth).
The situation for deciduous forest trees would be even worse than that for the evergreens, as their entire foliage would be new and therefore lost. For example, Larcher and Bauer (1981) determined that cold limits of photosynthesis of various temperate zone plants range from -1.3oC to 3.9oC, approximately the same range as the tissue-freezing temperatures for these plants. Lacking adequate food reserves, most temperate forest trees would not be able to cold-harden in a timely manner and would die or suffer additional damage during early freezes in the Fall (ref. Tinus et al., 1990).
Harwell (1984) carried out simulations for a mixed conifer and hardwood temperate forest for a longer-term reduction in annual average temperatures of 3oC, 6oC, or 9oC for a period of 5 years, biomass was predicted to fall by 25%, with recovery in 30-40 years, whereas for a 6oC cooling, biomass fell by 80% and returned to 50% of normal after 50 years. For the extreme 9oC cooling, biomass fell by 90%, and recovered only to 33% of normal after 50 years. After recovery, species composition was different. For grassland ecosystems, simulations of a 5-year period of temperature decrease of 3-9oC led to a reduction in net primary production ranging from 9% to 51%, with recovery times of several years,
The large amount of dead wood produced by the dead and damaged trees, exacerbated by drought conditions, might lead to increased forest fires. Tinus and Roddy (1990 ) estimate that the combined forests of the Northern Temperate Zone have a total of ca. 0.2 x 1012 cu. m of standing wet, live biomass.
Burning this biomass would release large amounts of reactive species such as hydrocarbons, organic acids, and nitrogen compounds, into the global atmosphere (ref. Andreae et al. 1988).
The effect of Toba on the oceans is more difficult to estimate. Regionally, the effect on ocean biota of the fallout of ca. 4 g/sq. cm of Toba ash over an area of 5 x 106 sq. km in the Indian Ocean must have been considerable. Deposition rates of N, organic C, and CaCO3 all rise sharply in the first few cm of the Toba ash layer, indicating that the ash fallout swept the water column of most of its particulate organic carbon and calcium carbonate (ref. Gilmour et al, 1990). Smit et al (1991) reported that the isocline in the Eastern Indian Ocean suddenly became shallower at the time of the Toba eruption.
Another possible effect of a dense aerosol cloud is decreased ocean productivity. For example, satellite observations after the 1982 Chichon eruption showed high aerosol concentrations over the Arabian Sea, and these values were associated with low surface productivity (as indicated by phytoplankton concentrations) from May through October of that year (ref. Strong, 1993).
Oxygen-isotope studies of corals following the 1991 Pinatubo eruption provide evidence that aerosol-induced cooling of the southwestern Pacific could lead to significant weakening of the Hadley Cell circulation and rainfall, and might precipitate long-term El Nino-like anomalies with extensive drought in many tropical areas (ref. Gagan et al, 1995). Climate model simulations predict significant drought in tropical areas from weakening of the trade winds/Hadley circulation, and from reduction in the strength of the Summer monsoon (ref. e.g. Pittock et al., 1986 and 1989; Turco et al., 1990). For example, Pittock et al. 1989 presented General Circulation Model results that showed a 50% reduction in convective rainfall in the tropics and monsoonal regions.
Animal and plant life suddenly hit by the devastating climatic aftermath of the Toba YTT event needed to be tough and extremely adaptable to survive. Among the countless species that had to face this challenge following the eruption 73,000 years ago, there was a relatively new arrival: early Homo sapiens.
Our next chapter will deal with the ways that our ancestors are thought to have come through a period of extreme difficulties and quite possibly near-extinction.
[ Go to HOME ]
[ Go toHEAD OF THIS CHAPTER ]
Last changed 1 February 2006