A brief history of climate science
Progress in climate science dates from more than 200 years ago. By the middle of the 19th century, scientists understood that the earth is heated by sunlight and would keep warming up indefinitely unless it had some way of losing energy. They knew that all objects radiate energy and that the earth radiates it in the form of infrared radiation.
Infrared radiation is a form of light but with longer wavelengths than can be seen by the human eye. However, it can be measured by instruments, including infrared glasses that combat soldiers use to “see” in the dark. The hotter the object, the more radiation it emits, and the shorter the wavelength of the emitted radiation. The sun’s surface temperature is about 6,000°C (11,000°F), and it emits mostly visible light, while the earth’s effective emission temperature is closer to -18°C (0°F) and so it emits much less radiation, and at a much longer (infrared) wavelength.
In 1820, the French mathematician and physicist Jean Baptiste Fourier came to understand that warmer surfaces emit more radiation than colder surfaces and calculated how warm the earth’s surface had to be to emit as much radiation as it receives from the sun, so that the temperature of the planet would neither increase nor decrease over time. He found that his estimate was much colder than the observed temperature. He reasoned that the atmosphere must absorb some of the infrared radiation from the earth’s surface and emit some of it back to the surface, thereby warming it. But he did not have enough information about the atmosphere to test this idea.
It was left to the Irish physicist John Tyndall to solve that problem. He used an experimental apparatus of his own design to carefully measure the absorption of infrared radiation as it passed through a long tube filled with various gases. His measurements astonished him and the whole scientific community of the mid-19th century.
Tyndall found that the main constituents of our atmosphere—oxygen and nitrogen, which together constitute about 98% of air—have essentially no effect on the passage of either visible or infrared radiation. But a few gases he tested, notably water vapor, carbon dioxide, and nitrous oxide, strongly absorb infrared radiation, and water vapor also absorbs some visible light. These gases are called “greenhouse gases” because, like the greenhouses we use to grow plants, they trap heat (although the way they do so is very different from the way actual greenhouses work).
Tyndall’s discovery was entirely based on careful laboratory experiments and measurements. The fundamental physics of the absorption and emission of radiation by matter would not be understood theoretically until the development of quantum mechanics in the early 20th century. According to this physics, symmetrical molecules with only two atoms—nitrogen (N2) and oxygen (O2), for example—hardly interact with radiation, but more complex molecules like water vapor (H2O—two atoms of hydrogen and one of oxygen) and carbon dioxide (CO2—one atom of carbon and two of oxygen) can interact much more strongly with radiation.
02a How it works
Why does the absorption and emission of infrared radiation by the atmosphere warm the planet? When the greenhouse gases (and clouds, which also act as greenhouse agents) absorb infrared radiation, they must re-emit radiation, otherwise the temperature of the atmosphere would increase indefinitely. This re-emission occurs in all directions, so that half the radiation is emitted broadly downward and half broadly upward. The downward part (“back-radiation”) is absorbed by the earth’s surface or lower portions of the atmosphere. Thus, in effect, Earth’s surface receives radiant energy from two sources: the Sun, and the back-radiation from the greenhouse gases and clouds in the atmosphere, as illustrated in Figure 1.
Earth’s surface receives almost twice as much radiation from the atmosphere as it does directly from the Sun.
This is partly because the atmosphere radiates 24/7, while the Sun shines only part of the time.
The warmer a surface, the more radiation it emits. Earth’s surface must get warm enough to lose enough heat to balance both sunlight and back-radiation from the atmosphere and clouds. That is the greenhouse effect. It should be remarked here that none of the preceding is remotely controversial among scientists, not even those few who express skepticism about global warming.
02b Our Atmosphere
Not all greenhouse gases are the same. The most important such gas in our climate system, because of its relatively high concentrations, is water vapor, which can vary from almost nothing to as much as 3% of a volume of air. Also, condensed water (cloud) strongly absorbs and re-emits radiation, and reflects sunlight as well. Next to water, carbon dioxide has the largest effect on surface temperature, followed by methane and nitrous oxide, and a handful of other gases whose concentrations are truly minute.
Water is constantly exchanged between the atmosphere and the earth’s surface through evaporation and precipitation. This process is so rapid that, on average, a molecule of water resides in the atmosphere for only about two weeks.
The temperature of the air limits how much water vapor it can hold: warmer air can support more vapor, whereas colder air holds less. Because rain and snow remove water from the air, there is often less water vapor in the air than there could be.
The ratio of the actual amount of moisture in the air to its upper limit is what we refer to as relative humidity. Although relative humidity varies greatly, we observe that its long-term average is fairly stable, so to a first approximation, the actual amount of water in the atmosphere changes in tandem with its upper limit, that is, with temperature.
So, if the temperature rises, the amount of water vapor rises with it. But since water vapor is a greenhouse gas, rising water vapor leads to more back-radiation to the surface, which causes yet higher temperatures. We refer to this process as a positive feedback. Water vapor is thought to be the most important positive feedback in the climate system. (It is important here to distinguish between a “feedback” and a “forcing.” When we discuss climate, a “feedback” is a process that strongly reacts to the climate itself, whereas a “forcing,” like changing solar radiation, CO2 or volcanoes, is not controlled by the climate itself, at least not on the time scales of concern in the problem of global warming.)
At the opposite extreme in terms of atmospheric lifetime is carbon dioxide. It is naturally emitted by volcanoes and absorbed by biological and physical processes that eventually incorporate the carbon into carbonate rocks like limestone. On geologic time scales, these carbonate rocks are pushed down into the earth’s mantle at The places in the earth’s mantle where one tectonic plate slides beneath anotherconvergent boundaries, and the carbon is eventually released back into the atmosphere as carbon dioxide through volcanoes or when the rock is once again exposed to air and weathered. This cycle takes many tens to hundreds of millions of years. But CO2 also cycles through the atmosphere, ocean, and land plants on a different time scale, on the order of hundreds—not millions—of years.
02c The impact of increased CO2
Much of the preceding, save for the details of the processes that control atmospheric CO2, was understood by the end of the 19th century. In particular, the Swedish chemist and Nobel laureate Svante Arrhenius understood the effect of greenhouse gases on climate and that CO2 is the most important long-lived greenhouse gas.
He also understood that we were beginning to emit large amounts of CO2 into the atmosphere from industrial processes and was the first to worry that, owing to its long residence time in the atmosphere, we would perceptibly increase its concentration. (Well before Arrhenius’s time, Eunice Foote speculated that past variations in CO2 might have played a role in past variations in climate.) In 1896, Arrhenius published a paper predicting that if we ever managed to double the concentration of CO2, the average surface temperature of the planet would rise between 5 and 6°C (9 and 11°F), a number he revised downward to 4°C (7°F) in a popular book he published in 1908. Arrhenius arrived at these numbers by performing up to 100,000 calculations by hand, and although he made several incorrect assumptions, the resulting errors partially canceled each other. It is truly remarkable that his 4°C (7°F) is within the range of the most recent estimates of 1.5–4.5°C (2.7–8.1°F).
Arrhenius also understood that the radiative effects of CO2 increase nearly logarithmically (rather than linearly) with its concentration, so that increasing CO2 by a factor of 8 would produce about three (rather than four) times more warming than would doubling it.
Arrhenius predicted that increasing CO2 would warm the planet. How did his prediction fare? Figure 4 compares Arrhenius’s prediction based on atmospheric CO2 concentrations with measured global mean surface temperature for the period from 1880 to 2018. The CO2 content of the atmosphere was measured directly beginning in 1958. Before that time (and going back for hundreds of thousands of years) scientists deduced its abundance by measuring CO2 concentration in gas bubbles trapped in ice cores, as we explore in the next section. Over the period of record, the global mean temperature generally follows the logarithm of the concentration of CO2, just as Arrhenius predicted. But you’ll notice in the graph that Earth’s average temperature is jagged; it’s not a smooth rising line like CO2 concentration. The shorter-period deviations mostly reflect the natural, chaotic variability of the climate system (an example of which is A flow of unusually warm surface waters along the Western coast of South America that disrupts normal weather patterns and contributes to extreme weather events like intense storms and droughts. It occurs irregularly every 2-7 years.El Niño), while longer departures are mostly due to other influences on climate, such as volcanoes and human-made An aerosol is a suspension of fine solid particles or liquid droplets in air or another gas. Aerosols can be natural or human-made.aerosols. While we may not be able to account for each little wobble, it is hard to avoid the conclusion from Figure 4 that the data largely vindicate a prediction made more than a century ago, based on simple physics and hand calculations. It stands to reason that more warming will occur if we continue to increase the concentration of CO2 in the atmosphere.
But what if we are fooling ourselves? Correlation is not causation, and perhaps the correspondence of temperature and CO2 is a coincidence—maybe something else is causing the warming. Or perhaps the rising temperature is causing CO2 concentrations to increase and not the other way around. How accurate is the curve in Figure 4—can we really measure the global mean temperature? Climate is always changing, so what is so special about the last 100 years? Are there other predictions of climate science that are verified or contradicted by observations?
These are all legitimate questions and deserve serious consideration; indeed, we would not be good scientists if we did not constantly ask ourselves such questions.
03a Historic temperatures
Let’s begin with the instrumental record of global average surface temperature. Thermometers were invented in the 17th century, but it was not until the 19th century that people started to make systematic, quantitative measurements around the globe. Naturally, most of these were made from land-based stations, but it was not long before measurements were being taken from ships, including measurements of the temperature of ocean water at and near the surface. (Benjamin Franklin discovered the Gulf Stream by lowering a thermometer into the ocean from a ship.) Sea surface temperature was measured routinely from buckets of water retrieved from the sea, and then, beginning in the 1960s, by taking the temperature of engine intake water. By the late 1960s, these measurements were being augmented by satellite-based measurements of infrared radiation emitted from the sea surface.
In estimating global mean temperature, one must carefully account for the uneven distribution of temperature measurements around the world, changes in the precise location and instruments used to measure temperature, the effects of growing urban areas that create heat islands that are warmer than the surrounding countryside, and myriad other issues that can bias global mean temperature. Different groups around the world have tackled these issues in different ways, and one way to assess the robustness of the temperature record is to compare their different results, as shown in Figure 5. One of these records, the Berkeley Earth estimate, shown in green with transparent uncertainty bounds, was undertaken by a group led by a physicist who was skeptical of the way atmospheric scientists had made their estimates. Even so, the four records agree with each other quite well after about 1900 and especially well after about 1950. The better and better agreement reflects the increasing number and quality of temperature measurements around the planet.
Different scientific groups have tackled measurement issues in different ways, yet their results agree with each other quite well.
We are therefore very confident that these records are accurate.
Theory and models predict that the air over land and at high latitudes should warm faster than that over the oceans, and this is indeed what we observe when measuring air temperature over land and sea. Global warming is neither predicted nor observed to be globally uniform, which can be seen in Video 1, and there are even places where the temperature has dropped over the second half of the 20th century, thanks to changing Ocean circulation, or current, is a continuous, directed movement of sea water generated by a number of forces acting upon the water, some of which include wind, breaking waves, temperature, and salinity.ocean circulation, melting sea ice, and other processes. Some of the fastest warming is in places far removed from cities, like Siberia and northern Canada; in fact, at most 2%–4% of the earth’s total warming can be attributed to urbanization.
So the measurements that underlie Figure 4 are pretty accurate. But how does that record of temperature and CO2 fit with the longer-term climate record? Is it unusual or is it consistent with natural climate variability on 100-year time scales? Since we do not have good global temperature measurements before the 19th century we must turn to the fascinating field of Ancient climates for which systematic measurements were not taken. Using proxies like ice sheets, tree rings, corals, shells, and microfossils, we can reconstruct ancient climate in order to understand natural variation in climate and the evolution of the current climate.paleoclimate, which seeks proxies for climate variables in the geologic record.
03b Prehistoric records
Ice core samples can tell us the temperature of clouds that produced the snow originally.
There are many different proxies for determining historical temperature; all have advantages and drawbacks. Some are physical, like the temperature of water in deep boreholes—water that has been isolated from the surface for a long time and reflects a long history of temperature. Some are biological, like the width and density of tree rings. All of these are local or at best regional metrics; there is no global “paleothermometer.”
One particularly useful proxy makes use of the fact that ice sheets and seawater contain different “flavors” (or isotopes) of water. Water (H2O) is made of one oxygen atom and two hydrogen atoms. A standard oxygen atom consists of a nucleus with 8 protons and 8 neutrons, surrounded by a cloud of 8 electrons. But some oxygen atoms have 9 or 10 neutrons in their nucleus. These variants are called isotopes. Standard oxygen, with 8 neutrons, called 16O to denote the number of protons and neutrons, is by far the most abundant isotope, followed by 18O with 8 protons and 10 neutrons. A tiny percentage of water contains this heavier oxygen isotope, and it turns out that the ratio of the heavy to the light isotope in water is a very useful metric.
Ocean water has a particular oxygen isotope ratio. But when seawater evaporates, its molecules containing the lighter isotope evaporate slightly faster than the molecules containing the heavier isotope. So, water vapor is “lighter” than seawater, meaning the ratio of heavy to light isotopes is smaller. Likewise, when the evaporated water begins to condense into clouds, molecules made of the heavier isotope condense first, so that as the cloud rains out, the water vapor left behind becomes progressively “lighter,” as does the precipitation that subsequently forms from it. So the farther away the water vapor is from its source, the “lighter” it is. By “farther” we really mean “colder,” since the amount of water vapor in a cloud falls rapidly as the air cools.
Likewise, standard hydrogen atoms in water have one proton and no neutrons, but a few atoms have one neutron, and there are even a few with two neutrons. A hydrogen atom with one neutron is called deuterium, and the ratio of deuterium to normal hydrogen in water can also be used as a paleothermometer.
The isotope ratios in rain and snow reflect the temperature of the cloud in which the rain or snow formed. In places like Greenland and Antarctica, much of the snow that falls accumulates and is progressively compacted by the weight of the snow on top of it, eventually forming ice. The ice is thus progressively older with depth in these ice sheets. Scientists drill down to collect solid cylinders of ice—ice cores—which they can analyze for many properties of the ice, including its isotopes, as a function of depth, or equivalently, age. The isotope ratios give a measure of the temperature of clouds that produced the snow originally. Modern measurements of the isotope ratios of recent snow show that they are highly correlated with surface air temperature, which is in turn correlated with the temperature of clouds above it. Thus we can use the isotope ratios as paleothermometers.
On a 100,000-year time scale, temperature is cyclic.
These cycles are the great ice ages and interglacial periods, and we are in an interglacial period right now.
Figure 7 shows the record of temperature inferred from two ice cores in Antarctica, going back 450,000 years, as well as from the volume of ice on the planet. You might be wondering how we know how much ice there was on Earth 450,000 years ago.
As seawater evaporates, the lighter isotopes evaporate faster, and thus ice sheets, which form from condensed water vapor, have a higher concentration of lighter isotopes than seawater. As ice sheets grow, the heavier isotopes get left behind in the ocean, and so the ratio of heavier to lighter isotopes in seawater steadily increases. Thus the isotopic composition of seawater is a measure of how much land ice there is on the planet. Marine microorganisms incorporate these The ratios of the abundances of different isotopes of elements like carbon, oxygen, and nitrogen. These ratios can be used as a diagnostic test of a sediment sample’s origin or age.isotopic signatures in their shells, and when they die some of them settle to the seafloor, where they get incorporated in sediments. We can analyze these sediment cores to get isotope ratios as a function of depth, and by other means determine the age of the sediments. Thus we can obtain a record of global ice volume with time.
You can see in Figure 7 that the lower the temperature, the higher the volume of ice on the planet, and vice versa. This makes sense! That the two curves—obtained from entirely different sources of data—agree so well testifies to the basic quality of the data underlying each.
It is plainly obvious that on the 100,000-year time scale, temperature is cyclic. These cycles are the great ice ages and interglacial periods, and the right edge of Figure 7 shows that we are in an interglacial period right now. The last ice age ended about 10,000 years ago—a geologic blink of the eye.
The figure also shows that the Antarctic temperature varied about 9°C (16°F) between the warmest and coldest periods. Other proxy estimates, models, and theory indicate that the tropics varied quite a bit less, so that the global mean temperature probably varied by about 5°C (9°F) between peaks and valleys.