Storms of My Grandchildren (23 page)

It seemed to me that part of the difficulty may be our emphasis on glacial-interglacial climate fluctuations—the periodic waxing and waning of large ice sheets on North America and Eurasia. To be sure, precise data on glacial-interglacial climate change obtained from ice cores—including which hemisphere a change originates in and how one quantity leads or another lags—are invaluable for sorting out its mechanisms and dynamics. But we researchers, as well as the public, might be able to see the forest for the trees better if we look at how the glacial-interglacial climate swings fit into longer-term, larger planetary change.

Such a longer time-scale perspective is provided by ocean cores. Ice sheets also existed on the planet millions of years ago, but they have long since melted, destroying their treasure of information. In contrast, for many millions of years there has been a slow rain of material sinking to the ocean floor, piling up in sediments. The most useful material in the sediments, for a climatologist, is the shells of the microscopic animals called foraminifera, or, for short, forams. The most useful characteristic of forams is their proportion of oxygen isotopes.

Hold on! This is not difficult! You already know that elements have different isotopes, depending on how many neutrons are in the nucleus. Almost 99.8 percent of oxygen is the garden-variety oxygen-16, with eight protons and eight neutrons. But about two tenths of 1 percent of oxygen is heavy oxygen-18, with 10 neutrons.

The great thing about oxygen-18 is that it gives us a thermometer, which we can use to measure Earth’s temperature over hundreds of millions of years. All we have to do is measure the proportion of oxygen-18 in the dead bodies of critters that lived in the past.

Ocean sediment cores have been extracted from many different places all around the world. A core is obtained by pushing a very long hollow pipe into the ocean sediments, capping the pipe, and pulling it out. The sediments extracted were deposited at times extending from today (the top of the core) to millions of years ago, at the bottom of a long core. The specific data set that I used was obtained from analyses on the cores’ forams, the microscopically small shelled critters living near the ocean floor.

Shells of forams are made of calcium carbonate (CaCO3). This tiny critter grows its shell by taking calcium and carbon dioxide from the water and snitching one oxygen atom from a water molecule. Water molecules in the ocean are all bouncing around, banging against each other, at a speed that depends on the temperature of the water. The light water molecules, those with oxygen-16, are moving faster than the heavier ones, and so they get incorporated into the shell more easily. If the water gets warmer, the oxygen-16 gains even more speed relative to oxygen-18 and is incorporated in the shell in even greater proportion. Laboratory experiments show us just how fast the oxygen-16 portion increases (or oxygen-18 decreases) as temperature increases. Bingo—we have a thermometer.

Except for one catch. There is a second factor altering the proportion of oxygen-16 and oxygen-18 in the foram. Because water molecules with oxygen-16 are lighter and moving faster, they are more successful at penetrating the surface tension of the ocean and escaping to the air—in other words, they evaporate faster. If the escaped water molecule condenses out as rain, it goes back to the ocean, so the proportion of light oxygen in the ocean remains unaffected. But if the water molecules become snow and build an ice sheet, that ice sheet will have little oxygen-18. As the ice sheet gets bigger and bigger, the proportion of oxygen-18 remaining in the ocean gets bigger and bigger. So the amount of oxygen-16 and oxygen-18 in a foram shell depends on the size of global ice sheets as well as the temperature of the ocean water. This ambiguity spoils the thermometer, causing consternation among paleoclimate scientists. I made a simple assumption to deal with this ambiguity. Geologic data show that from the beginning of the Cenozoic until 34 million years ago, there were no large ice sheets on Earth, so the foram thermometer works without any correction in the early Cenozoic. From the time just before Antarctica froze over until the recent ice ages, the total change of oxygen-18 was twice as large as it would have been due to only the known temperature change between these two end points. So my simple assumption was that throughout the 34-million-year period the variations of oxygen-18 should always be assigned equally to temperature and ice volume change.

Even the most hardened antiscience zealot, once he understands figure 18, will have to admit that it is one of the most beautiful curves on the planet (I’m referring to scientific curves). It contains an enormous amount of interesting information about Earth’s history. There are remarkable stories in both the broad sweep of climate over the 65 million years and in the rapid climate fluctuations.

First, note that the temperature increased in the early Cenozoic, reaching 13 degrees Celsius (55 degrees Fahrenheit) 50 million years ago. Then, over the last 50 million years, the planet cooled. In the past few million years, the coldest period in the record, glacial-to-interglacial oscillations became larger and larger. These temperatures were “measured” in the deep ocean, by the forams, but they tell us about the surface. The temperature in the deep ocean is the same as the temperature of the high-latitude ocean surface in the winter, because that is the season and place where ocean surface water becomes most dense and sinks to the ocean bottom.

There are many stories in figure 18. There is information in the broad sweep of the curve, but also in the rapid climate oscillations. Let us first consider the broad sweep, the great warming that peaked 50 million years ago, followed by a long cooling trend. What could have caused such a huge change of Earth’s surface temperature? There are three possibilities: changes of the energy coming into the planet, changes on the surface, and changes in the atmosphere.

First, consider the energy coming in. Astronomers know that our sun is a very normal “main sequence” star. That phrase refers to a diagram that shows how a star changes as it ages. Our sun is a relatively young star, about 4.6 billion years old. It is still in the phase of “burning” hydrogen in its core, by nuclear fusion, making helium. In this phase the sun is slowly getting brighter. Over the past 65 million years, the sun’s brightness has increased 0.4 percent. Earth absorbs about 240 watts (per square meter, averaged over the planet) of solar energy, so the solar forcing over the Cenozoic era has been a linear increase of about 1 watt. By itself, that should have caused a slow warming, of the order of 1 degree over 65 million years. But the planet has actually cooled, so the sun is not the biggest contributor to the climate changes in figure 18.

Third, consider the changes in the atmosphere. The amount of atmospheric carbon dioxide during the Cenozoic varied from as little as 170 ppm in recent ice ages to 1,000 to 2,000 ppm in the early Cenozoic. Thus the largest carbon dioxide amount was probably close to three doublings of the smallest amount (170’ 340’ 680’ 1,360). Large carbon dioxide change is usefully expressed as the number of doublings, because the infrared absorption bands (illustrated in figure 5 on page 62) become saturated as carbon dioxide increases. Additional absorption occurs in weak bands and at the edges of strong absorption bands, but it takes more and more carbon dioxide to yield a given increment of climate forcing. The result is that forcing increases by about 4 watts with each doubling.

So carbon dioxide changes in the Cenozoic caused a forcing of about 12 watts—at least ten times greater than the climate forcing due to either the sun or Earth’s surface. It follows that changing carbon dioxide is the immediate cause of the large climate swings over the last 65 million years.

Before we consider the reasons for this carbon dioxide change, it is important to check whether this greenhouse gas climate forcing is the correct order of magnitude to account for the measured change of Earth’s temperature. If the topic of climate sensitivity is too esoteric for your taste, just skip the following three paragraphs. However, if you digest this stuff, it will help you understand the important matters in global climate change that the “professionals” are contemplating now.

Earth’s temperature changed about 14 degrees Celsius between 50 million years ago and the recent ice ages (figure 18). Between 50 and 34 million years ago, the period when there were no large ice sheets on Earth, we expect climate sensitivity to be 3 degrees Celsius for doubled carbon dioxide (the empirical climate sensitivity we inferred earlier from glacial-interglacial climate change). That means a forcing of four thirds (4/3) of a watt is needed to cause a 1-degree Celsius temperature change. Thus the 8-degree temperature change between 50 and 34 million years ago required almost 11 watts of forcing. Between 34 million years ago and the depth of the last ice age, surface reflectivity change due to ice sheets approximately doubled the climate sensitivity (as discussed in chapter 3). Thus the 6-degree temperature change in that period required a greenhouse gas forcing of 4 watts. The greenhouse forcing required for the total temperature change over the Cenozoic is about 15 watts, assuming that climate sensitivity averages 3 degrees Celsius for doubled carbon dioxide in the absence of ice sheets.