Answer: Earth’s climate system is gaining more energy than it loses every year; exactly how the extra energy is distributed across the climate system varies over time, resulting in periods of faster and slower warming.

The Facts: Climate scientists have documented striking changes in the Earth’s climate, including melting Arctic sea ice, melting glaciers and ice sheets, and rising air temperatures. These changes are all symptoms of the same root cause: the Earth’s energy budget is out of balance, causing the climate system to gain more energy than it loses, due to an ever-thickening blanket of greenhouse gases in the atmosphere. Current estimates are that the Earth is accumulating energy at a rate of 0.5 to 1.0 W.m^–2 ( Hansen et al., 2013 ; Trenberth and Fasullo, 2013 ). This is the equivalent of four atomic bombs’ worth of heat every second ( Cook and Nuccitelli, 2013 ), and this accumulating energy is responsible for the melting ice and rising air temperatures. As dramatic as these effects are, however, they represent the work of less than 10% of the extra energy accumulating in the climate system. Over 90% of global warming is heating the oceans, according to measurements of temperature changes from the surface down to around 3,000 m, with only about 2% warming the atmosphere ( Levitus et al., 2001 ). The remarkable fact is that, although Page 34  |  Top of Articlethe changes in our planet discussed so far in this book are significant from a human perspective, when considering the climate system as a whole, they’re really no more than a sideshow to the oceans’ main event.

These percentages do not stay constant over time, however. Although the oceans always receive the overwhelming majority of the energy, sometimes a little more goes into the atmosphere and a little less into the oceans; sometimes the opposite happens. These shifts alter the rate at which the oceans or the atmosphere warms up: more energy results in faster warming, less energy in slower warming. Periods of faster and slower warming of the atmosphere are evident in the surface temperature record, shown in Figure 1.7 (see also question 4 ). These periods indicate shifts in where the energy is being distributed.

A crucial point about these shifts in energy distribution is that they are cyclical. The climate system is full of cyclical patterns; in fact, one way to think about climate in general is that it’s the net result of many Page 35  |  Top of Articlecyclical patterns superimposed on each other, rather like a complex piece of music. Climate consists of many repeating patterns, all playing to their own rhythm, and scientists must tease apart the rhythms in order to better understand how climate works.

Figure 1.7 Global average surface temperature reconstructed from weather station thermometers by Berkeley Earth. Annual average temperature is shown in light gray, with an 11-year moving average, which filters out short-term variability, shown in black. Temperature is shown as difference relative to an estimated global average temperature for 1951–1980 of 14.771°C.

Source: Data from Berkeley Earth, available at http://berkeleyearth.lbl.gov/auto/Global/Land_and_Ocean_summary.txt . Accessed March 27, 2015.

One of the best-known climate cycles occurs in the tropical Pacific Ocean, which alternates between two states known as El Niño and La Niña. Winds push the surface waters of the tropical Pacific westward, away from South America, allowing the ocean to accumulate energy from the intense sunshine of the tropics as it goes, building into a pool of warm water over a hundred meters deep by the time it reaches Australia. Much colder water from the deep ocean wells up to the surface off the coast of South America in the east Pacific, to replace the water moving west. This produces a distinctive pattern of sea surface temperatures across the tropical Pacific: very cold to the east, much warmer to the west.

In El Niño conditions, the winds weaken and the warm water slides back across the Pacific from Australia toward South America, covering the cold upwelling water and releasing heat to the atmosphere. In La Niña years, the opposite happens: the wind and ocean currents become exceptionally strong, building a large pool of cold water off South America that reaches into the central Pacific, cooling the atmosphere above it. El Niños therefore tend to produce noticeable upward spikes in global average temperatures, with sharp downturns during La Niñas. The record El Niño of 1998 shows up in Figure 1.7 as one such spike, followed by a sharp drop as the El Niño faded away. (See Herring, 1999 , for a user-friendly account or Trenberth, 1997 , for a more technical discussion of El Niños and La Niñas.)

In addition to affecting the global average temperature, El Niños and La Niñas affect the climate of many individual locations around the world, causing especially wet or dry seasons for the years when they occur. During an El Niño, for example, California experiences floods, and eastern Australia experiences droughts, with these conditions flipped for La Niña years. Scientists describe the El Niño/La Niña cycle as an example of climate variability, because it affects the day-to-day weather of these places for an extended period, up to a year in some cases. The average weather—that is, the climate—is especially wet or dry because of an El Niño or La Niña. However, because they are cyclical, they are different from climate change.

The difference between climate variability and climate change can be seen in Figure 1.7 . Climate change is evident in the clear warming trend over the complete period of record. Climate variability is the up-and-down “wiggles” on top of the trend. Although there are other sources of Page 36  |  Top of Articleclimate variability, El Niños and La Niñas account for many of the wiggles. Climate change is a result of an overall buildup of heat in our planet’s climate system. Climate variability is a result of the heat being moved across different parts of the climate system.

Another source of climate variability is the tendency of El Niños and La Niñas to cluster in groups where one or the other is dominant—strong El Niños paired with weak La Niñas or vice versa. These clusters of warm events (El Niños) and cold events (La Niñas) can last for several decades, so the climate variability they bring about lasts longer than the single-year effects of El Niños and La Niñas. This tendency to switch between clusters of warm and cold events is known as the Pacific Decadal Oscillation (PDO), and is shown in Figure 1.8 . There is also a longer-lasting version, known as the Pacific Multidecadal Oscillation (PMO), and a cousin in Page 37  |  Top of Articlethe Atlantic, the Atlantic Multidecadal Oscillation (AMO). Just as El Niños and La Niñas can affect the climate in the years when they occur, so these longer-lasting phases of warm versus cold water in the Pacific and Atlantic Oceans also seem to influence climate. However, these longer cycles are still manifestations of climate variability, not climate change.

Figure 1.8 Pacific Decadal Oscillation index, calculated on the basis of sea surface temperatures in the North Pacific Ocean. Shown are monthly values (dark and light gray bars), and a roughly 5-year (61-month) running mean (black line), for the years 1900–2013.

Source: Data from the University of Washington’s Joint Institute for the Study of the Atmosphere and Ocean, available at http://research.jisao.washington.edu/pdo/PDO.latest . Accessed March 27, 2015.

A close examination of Figure 1.7 shows longer periods of more rapid warming separated by periods of stable or cooling temperatures, with the short-term “wiggles” of El Niño and La Niña superimposed on top. This gives the warming trend a stair-step pattern similar to an escalator (for an animated version, see Nuccitelli, 2011 ). A substantial body of research now shows that one of the flat steps in Figure 1.7 , from around 1940 to the mid-1970s, was most likely caused mainly by air pollution, which reflects sunlight, offsetting the warming effects of carbon dioxide (see question 29 for more information on this phenomenon; see, e.g., Tett et al., 1999 , Meehl et al., 2004 , or Booth et al., 2012 , for technical discussions). However, both the PMO and AMO were shifting into their cold phases around this time, so swings in sea surface temperatures may also have played a role.

Scientists are more confident that the swing to warm phases in the oceans in the mid- to late 1970s helped boost the warming from carbon dioxide. The swing back to cold phases after about 1998 helped slow the warming down again. As Figure 1.9 shows, global average air temperatures continued to increase after 1998—just not as quickly as they had in preceding decades when the ocean cycles were working with carbon dioxide’s warming effects, instead of against them. Much as the 24-hour cycle of daytime warming and nighttime cooling occurs in the context of seasonal changes, warming as summer approaches and cooling as winter approaches, so El Niños and La Niñas are superimposed on the longer-term ocean cycles—and both are superimposed on the overall shift toward warming. Climate variability is superimposed on climate change.

Because of this, different features of the climate system come into sharp relief at different timescales. A focus on a period of a few decades will show climate variability clearly, but will probably hide the long-term trend, just as listening to only one or two seconds of music makes it almost impossible to hear the tune. It’s there, but the short-term focus obscures it.

This is the root problem with one especially widespread myth about climate change, that global warming stopped in 1998 (or a similar recent year). The focus is on the short term, where the cyclical patterns of climate variability show up clearly, rather than on the long term, where the indications of climate change are apparent. The year 1998 may seem like a long time in the past, but in terms of patterns like the PDO, PMO, or AMO, it’s still not even one typical complete cycle ago. From the Page 38  |  Top of ArticlePage 39  |  Top of Articleperspective of the climate system, 1998 is very recent. This short-term focus makes it difficult to see the long-term trend.

Figure 1.9 Global average temperature reconstructions from Berkeley Earth for 1980–2014 (top) and 1996–2014 (bottom). As with Figure 1.7, annual average temperature is shown in gray, as the difference relative to an estimated global average temperature for 1951–1980 of 14.771°C. For the top graph, an 11-year moving average, which filters out short-term variability, is shown in black, highlighting the slowdown in the rate of warming in recent years. For the bottom graph, the temperature trend is shown by the straight black line, calculated using linear regression of temperature against time.

Source: Data from Berkeley Earth, available at http://berkeleyearth.lbl.gov/auto/Global/Land_and_Ocean_summary.txt . Accessed March 27, 2015.

The myth is expressed in several different ways. A 2013 report in The Economist, for example, stated, “Over the past 15 years air temperatures at the Earth’s surface have been flat while greenhouse-gas emissions have continued to soar.” Later reports in the same newspaper identified the period since 1998 as a “pause” in global warming (The Economist, 2013). It is true that the rate of warming since 1998 has been slower than in earlier decades, notably from around 1980 to 1998—but as Figure 1.9 shows, it has not been flat. Furthermore, periods of faster warming interspersed with periods of slower warming are to be expected, given climate variability superimposed on the warming trend. The warming slowdown coincides with the shift of the Pacific and Atlantic Oceans into cool phases, as noted earlier. To be fair, The Economist article took pains to point this out.

Some reports, however, erroneously assert that global warming did not just pause, but actually stopped in the late 1990s. The title of an article by Bob Carter (2006) in Britain’s Telegraph states flatly that global warming stopped in 1998. Nigel Lawson (2009) , of the climate-skeptic think-tank the Global Warming Policy Foundation, draws attention to “the absence of any recorded 21st century global warming” (p. 8), while Patrick Michaels (2011) asks, “Why hasn’t the Earth warmed in nearly 15 years?” Christopher Monckton (2014) , writing for the website ClimateDepot.com , claims that there has been no warming since 1996.

How can these claims be made, in light of surface temperature data graphed in Figure 1.9 , showing warming since the late 1990s? Lord Monckton’s claim is the easiest to illuminate, because it works by using only the data set which shows the smallest warming trend since 1996: the RSS satellite measurements of lower tropospheric temperature. Readers might recall from question 5 that there are challenges in interpreting the satellite data, and different teams get slightly different results. Over a short-enough time period, those slight differences are enough to give different trends, such that UAH shows a greater warming trend since 1996 than RSS. Figure 1.10 shows both the RSS and UAH temperatures since 1996, along with trend lines indicating the overall pattern of temperature change over that time period. Lord Monckton can say that there has been no warming since 1996 because he focuses on just one set of measurements while ignoring all the others. Interactive online graphing tools, such as Kevin Cowtan’s, at the University of York in Britain, make it possible for readers to do their own comparisons of temperature trends from different data sets over different time frames Page 40  |  Top of Article(see Further Reading), highlighting the fallacy of relying on just one set of measurements.

Figure 1.10 Lower tropospheric air temperature 1996–2014, inferred from satellite measurements of microwave radiation by two different teams (RSS and UAH). Trendlines are calculated using linear regression of temperature against time. The RSS trend since 1996 is essentially flat, while the UAH trend shows warming.

Other versions of the myth make the more sophisticated argument that no statistically significant warming has taken place since 1998 (or 1996). This claim may be true for some data sets but is misleading. As Figure 1.9 shows, there has been warming since that period. At such short timescales, however, the up-and-down wiggles of climate variability are large relative to the trend, making it impossible to say with confidence that the trend is anything different from the year-to-year variability. This claim is really only restating the obvious: as you zoom in to progressively shorter and shorter timescales, it becomes harder and harder to identify the trend amid the climate variability. Climate change is happening over longer periods than the past 20 years or so. While to most of us, 20 years seems like a long time, from the standpoint of climate, it’s no more than a few notes out of a symphony.