«DIRECT TESTIMONY OF JAMES E. HANSEN Q. Please state your name and business address. A. My name is James E. Hansen. My business address is 2880 ...»
A. On long (geologic) time scales CO2 is exchanged between the surface reservoirs (atmosphere, ocean, soil and biosphere) and the solid Earth. Two processes take CO2 out of the surface reservoirs: (1) chemical weathering of silicate rocks, which results in the deposition of (calcium and magnesium) carbonates on the ocean floor, and (2) burial of organic matter, some 12 of which eventually forms fossil fuels. Weathering is the more dominant process, accounting for ~80% of carbon removal from surface reservoirs (Berner 2004).
CO2 is returned to the atmosphere principally via subduction of oceanic crustal plates beneath continents. When a continental plate overrides carbonate-rich ocean crust, the subducted ocean crust experiences high temperatures and pressures. Resulting metamorphism of the subducted crust into various rock types releases CO2, which makes its way to the atmosphere via volcanic eruptions or related phenomena such as ‘seltzer’ spring water. This return of CO2 to the atmosphere is called ‘outgassing’.
Outgassing and burial of CO2, via weathering and organic deposits, are not in general balanced at any given time (Edmond and Huh 2003). Depending on the movement of continental plates, the locations of carbonate-rich ocean crust, rates of mountain-building (orogeny), and other factors, at any given time there can be substantial imbalance between outgassing and burial. As a result, atmospheric CO2 changes by large amounts on geologic time scales.
Q. How much do these geologic processes change atmospheric CO2?
A. Rates of outgassing and burial of CO2 are each typically 2-4 x 10**12 mol C/year (Staudigel et al. 1989; Edmond and Huh 2003). An imbalance between outgassing and burial of say 2 x 10**12 mol C/year, if confined entirely to the atmosphere, would correspond to ~0.01 ppm CO2 per year. However, the atmosphere contains only of order 10**(-2), i.e., about 1%, of the total CO2 in the surface carbon reservoirs (atmosphere, ocean, soil, biosphere), so the rate of geologic changes to atmospheric CO2 is only about 0.0001 ppm CO2 per year. This compares to the present human-made atmospheric CO2 increase of ~2 ppm per year. Fossil fuels burned now by humans in one year contain the amount of carbon buried in organic sediments in approximately 100,000 years.
The contribution of geologic processes to atmospheric CO2 change is negligible compared to measured human-made changes today. However, in one million years a geologic imbalance of 0.0001 ppm CO2 per year yields a CO2 change of 100 ppm. Thus geologic changes over tens of millions of years can include huge changes of atmospheric CO2, of the order of 1000 ppm of CO2. As a result, examination of climate changes on the time scale of tens of millions of years has the potential to yield a valuable perspective on how climate changes with atmospheric composition.
Q. What is the most useful geologic era to consider for that purpose?
A. The Cenozoic era, the past 65 million years, is particularly valuable for several reasons. First, we have the most complete and most accurate climate data for the most recent era. Second, climate changes in that era are large enough to include ice-free conditions. Third, we know that atmospheric greenhouse gases were the principal global forcing driving climate change in that era.
Q. How do you know that greenhouse climate forcing was dominant in the Cenozoic?
A. Climate forcings, perturbations of the planet’s energy balance, must arise from either changes in the incoming energy, changes that alter the planetary surface, or changes within the atmosphere.
Let us examine these three in turn.
Solar luminosity is growing on long time scales, at a rate such that the sun was ~0.5% dimmer than today in the early Cenozoic (Sackmann et al. 1993). Because the Earth absorbs 13 about 240 W/m2 of solar energy, the solar climate forcing at the beginning of the Cenozoic was about -1 W/m2 relative to today. This small growth of solar forcing through the Cenozoic era, as we will see, is practically negligible.
Changing size and location of continents can be an important climate forcing, as the albedo of the Earth’s surface depends on whether the surface is land or water and on the angle at which the sun’s rays strike the surface. A quarter of a billion years ago the major continents were clumped together (Figure 9) in the super-continent Pangea centered on the equator (Keller and Pinter 1996). However, by the beginning of the Cenozoic (65 million years before present, 65 My BP, the same as the end of the Cretaceous) the continents were close to their present latitudes. The direct (radiative) climate forcing due to this continental drift is no more than ~ 1 W/m2.
In contrast, atmospheric CO2 reached levels of 1000-2000 ppm in the early Cenozoic (Pagani et al. 2005; Royer 2006), compared with values as low as ~180 ppm during recent ice ages. This range of CO2 encompasses about three CO2 doublings and thus a climate forcing more than 10 W/m2. So it is clear that changing greenhouse gases provided the dominant global climate forcing through the Cenozoic era.
We are not neglecting the fact that dynamical changes of ocean and atmospheric currents can affect global mean climate (Rind and Chandler 1991). Climate variations in the Cenozoic are too large to be accounted for by such dynamical hypotheses.
Q. What caused atmospheric CO2 amount to change?
A. At the beginning of the Cenozoic era, 65 My BP, India was just south of the Equator (Figure 9), but moving north rapidly, at about 15 cm/year. The Tethys Ocean, separating Eurasia from India and Africa, was closing rapidly. The Tethys Ocean had long been a depocenter for carbonate sediments. Thus prior to the collision of the Indian and African plates with the Eurasian plate, subduction of carbonate-rich oceanic crust caused outgassing to exceed weathering, and atmospheric CO2 increased.
The Indo-Asian collision at ~50 My BP initiated massive uplift of the Himalayas and the Tibetan Plateau, and subsequently drawdown of atmospheric CO2 by weathering has generally exceeded CO2 outgassing (Raymo and Ruddiman 1992). Although less important, the Alps were formed in the same time frame, as the African continental plate pushed against Eurasia. With the closing of the Tethys Ocean, the major depocenters for carbonate sediments became the Indian and Atlantic oceans, because the major rivers of the world empty into those basins.
For the past 50 million years and continuing today, regions of subduction of carbonate rich ocean crust have been limited. Thus, while the oceans have been a strong sink for carbonate sediments, little carbonate is being subducted and returned to the atmosphere as CO2 (Edmond and Huh 2003). As a result, over the past 50 million years there has been a long-term decline of greenhouse gases and global temperature.
Q. Can you illustrate this long-term cooling trend?
Yes. Figure 10a shows a quantity, δ18O, that provides an indirect measure of global temperature A.
over the Cenozoic era, with a caveat defined below. δ18O defines the amount of the heavy oxygen isotope 18O found in the shells of microscopic animals (foramininfera) that lived in the ocean and were deposited in ocean sediments. By taking ocean cores of the sediments we can sample shells deposited over time far into the past. Figure 10a shows the average result from many ocean cores around the world obtained in deep sea drilling programs (Zachos et al 2001).
14 The proportion of δ18O in the foraminifera shell depends on the ocean water temperature at the time the shell was formed, and thus δ18O provides a proxy measure of temperature.
However, an ice sheet forming on the Earth’s surface has an excess of 16O in its H2O molecules, because 16O evaporates from the ocean more readily than 18O, leaving behind a relative excess of 18 O in the ocean. As long as the Earth was so warm that little ice existed on the planet, as was the case between 65 My BP and 35 My BP, 18O yields a direct measure of temperature, as indicated by the red curve and the temperature scale on the left side of Figure 10a.
The sharp change of δ18O at about 34 My BP was due to rapid glaciation of the Antarctic continent (Lear et al. 2000; Zachos et al. 2001). From 34 My BP to the present, δ18O changes reflect both ice volume and ocean temperature changes. We cannot separate the contributions of these two processes, but both increasing ice volume and decreasing temperature change δ18O in the same sense, so the δ18O curve continues to be a qualitative measure of changing global temperature, chronicling the continuing long-term cooling trend of the planet over the past 50 million years.
The black curve in Figure 10a shows the rapid glacial-interglacial temperature oscillations, which are smoothed out in the mean (red and blue) curves. Figure 10b expands the time scale for the most recent 3.5 million years, so that the glacial-interglacial fluctuations are clearer. Figure 10c further expands the most recent 425,000 years, showing the familiar Pleistocene ice ages punctuated by brief interglacial periods. Note that the period of civilization within the Holocene is invisibly brief with the resolution in Figure 10a. Homo sapiens have been present for about 200,000 years, and the predecessor species, Homo erectus, for about 2 million years, still rather brief on the time scale of Figure 10a.
Q. Can you explain the nature of the global climate change illustrated in Figure 10?
A. The long-term cooling from 50 My BP to the present must be due primarily to decreasing greenhouse gases, primarily CO2, which fell from 1000-2000 ppm 50 My BP to 180-280 ppm in recent glacial-interglacial periods. Full glaciation of Antarctica, at about 34 My BP (Lear et al.
2000; Zachos et al. 2001), occurred when CO2 fell to 500 ±150 ppm (Hansen and Sato 2007).
Between 34 and 15 My BP global temperature fluctuated, with Antarctica losing most of its ice at about 27 My BP. Antarctica did not become fully glaciated again until about 15 My BP. Deglaciation of Antarctica was associated with increased atmospheric CO2 (Pagani et al.
2005), perhaps due to the negative feedback caused by reduction of weathering (Lear et al. 2004) as ice and snow covered Antarctica as well as the higher reaches of the Himalayas and the Alps.
Cooling and ice growth resumed at about 15 My BP continuing up to the current Pleistocene ice age. During the past 15 My CO2 was at a low level, about 200-400 ppm (Zachos et al. 2001; Pagani et al. 2005) and its proxy measures are too crude to determine whether it had a long-term trend. Thus it has been suggested that the cooling trend may have been due to a reduction of poleward ocean heat transports, perhaps caused by the closing of the Isthmus of Panama at about 12 My BP or the steady widening of the oceanic passageway between South America and Antarctica.
We suggest that the global cooling trend after 15 My BP may due to continued drawdown of atmospheric CO2 of a degree beneath the detection limit of proxy measures. Little additional drawdown would be needed, because the increasing ice cover on the planet makes climate sensitivity extremely high, and the logarithmic nature of CO2 forcing (see formulae in Hansen et al. 2000) makes a small CO2 change very effective at low CO2 amounts. There are reasons to expect CO2 drawdown in this period: the Andes were rising rapidly in this period (Garzione et al.
15 2006), at a rate of about 1 mm per year (1 km per My). The mass of the Andes increased so much as to slow down the convergence of the Nazca and South American plates by 30% in the past 3.2 My (Iaffaldano et al. 2007). Increased weathering and reduced subduction both contribute to drawdown of atmospheric CO2. Finally, a suggestion that CO2 has been declining over the relevant period is provided by the increase of C4 plants relative to C3 plants that occurred between 8 and 5 My BP (Cerling et al. 1993); C4 plants are more resilient to low atmospheric CO2 levels (C4 and C3 photosynthesis are alternative biochemical pathways for fixing carbon, the C4 path requiring more energy but being more tolerant of low CO2 and drought conditions). However, given the high climate sensitivity with large ice cover, other small forcings could have been responsible for the cooling trend without additional CO2 decline.
In summary, there are many uncertainties about details of climate change during the Cenozoic era. Yet important conclusions emerge, as summarized in Figure 11. The dominant forcing that caused global cooling, from an ice free planet to the present world with large ice sheets on two continents, was a decrease in atmospheric CO2. Human-made rates of change of climate forcings, including CO2, now dwarf the natural rates.
Q. Is this relevant to the question of whether we need to “wrestle” with climate change?
A. Yes, it may help resolve the conundrum sensed by some lay persons based on realization that the natural world has undergone huge climate variations in the past. That is true, but those climate variations produced a different planet. If we follow “business as usual” greenhouse gas emissions, putting back into the air a large fraction of the carbon that was stored in the ground over millions of years, we surely will set in motion large climate changes with dramatic consequences for humans and other species.
Q. Why are climate fluctuations in the past few million years (Figure 10b) so regular?
A. The instigator is the distribution of sunlight on the Earth, which continuously changes by a small amount because of the gravitational pull of other planets, especially Jupiter and Saturn, because they are heavy, and Venus, because it comes close. The most important effect is on the tilt of the Earth’s spin axis relative to the plane of the Earth’s orbit (Figure 12). The tilt varies by about 2° with a regular periodicity of about 41 Ky (41,000 years). When the tilt is larger it exposes both polar regions to increased sunlight at 6-month intervals. The increased heating of the polar regions melts ice in both hemispheres.