Storms of My Grandchildren (16 page)

In my Iowa talk I showed a new analysis of ocean heat, by Josh Willis and colleagues, that indicated heat storage at a rate of 0.6 watt in the upper 750 meters of the ocean during 1993–2003. While that result agreed well with the energy imbalance in our climate model simulations, Catia Domingues and colleagues later showed that instrumental biases affected the results of both Levitus’s and Willis’s analyses. Until instrumental issues are resolved and good heat storage data is obtained for the entire ocean, it is not possible to infer the net climate forcing acting on Earth.

The difficulty of determining the precise value of Earth’s energy imbalance based on existing measurements does not diminish its importance. Earth’s energy imbalance is our best indication of where global temperature is headed and of how much global forcings must be altered to stabilize climate. We know for sure that the ocean is warming and that the planet is out of energy balance, averaged over the past decade or two (averaging minimizes the effect of cyclic solar variability and chaotic atmospheric and ocean variability). But, given the imprecision of the measurements, the imbalance is only known to be somewhere in the range of 0.25 to 0.75 watt. This includes about 0.1 watt for changes in heat reservoirs other than the ocean, i.e., heat that goes into warming the air, lakes, and continents and melting sea ice, land ice, and ice shelves.

So the actual planetary energy imbalance is probably less than the energy imbalance that was calculated in the climate simulations we made in 2004 and 2005—which were the climate model results we submitted to the IPCC assessment that was published in 2007. Those climate model results showed the planetary energy imbalance in the first several years of the twenty-first century, averaged over a series of model runs, to be 0.75 watt. What can we learn from the discrepancy, assuming it is not due to observational error or just the variability of the climate system? We know the discrepancy cannot be due to an error in our model’s climate sensitivity, since the paleoclimate data (discussed in chapter 3) confirm the climate sensitivity of 3 degrees Celsius for doubled carbon dioxide. Instead, the discrepancy has implications about the assumed net climate forcing.

Our climate simulations assumed that the amount of atmospheric aerosol has stayed constant since 1990, with the hope that aerosol reductions in the United States and Europe, due to clean air regulations, would tend to offset aerosol increases in developing countries such as China and India. However, clean air rules in the West, and reduced Russian emissions due to economic collapse, occurred largely prior to 1993. Air pollution in developing countries, on the other hand, surely increased during the past two decades, a conclusion supported by global measurements of decreased atmospheric visibility. So an increase of aerosols in the period from 1990 to 2009 is one candidate for explaining the observed planetary energy imbalance.

Greenhouse gases and aerosols are the two largest human-made climate forcings. But there is a third significant climate forcing we need to look at when we use the planet’s energy imbalance to check on the status of the net climate forcing, and when we try to assess likely climate change in the next few years: solar variability. So let’s consider the sun’s role in climate change.

Indeed, there are many people, including scientists, who believe that the sun is the most important factor in climate change, the dominant climate forcing. It is easy to understand their suspicions. Earth gets its warmth from the sun. The sun is variable. Correlations of solar variability and climate change are well known. But what we need is an objective, quantitative comparison of solar and other climate forcings.

Solar irradiance varies by about a tenth of a percent over the average ten-to-twelve-year solar cycle, as shown by figure 11. Earth absorbs 240 watts of sunlight per square meter of its surface. (The absorbed energy is much less than the sun’s irradiance that hits Earth perpendicular to the Earth-sun direction, because the circular cross section of Earth presented to the sun is a quarter of Earth’s surface area, and 30 percent of incident sunlight is reflected to space without being absorbed.) Thus the sun at solar minimum causes a forcing of about −0.2 watt, relative to the forcing at solar maximum, or about half that much relative to the average brightness of the sun.

A solar forcing of −0.2 watt is significant but not a dominant forcing. By contrast, the carbon dioxide forcing today, relative to preindustrial times, is about 1.5 watts. However, much of the carbon dioxide forcing has already been “used up” in causing the warming of the past century. It is more relevant, then, to compare the solar forcing with the planet’s energy imbalance, which we estimated as about 0.5 watt averaged over the past two decades. But before judging the sun’s importance, we should note that there are mechanisms by which the sun’s effect may be magnified or diminished.

One of the papers I was working on in 2004, “Efficacy of Climate Forcings,” addressed this matter. In that paper, we showed that it is not sufficient to know only the magnitude of a forcing in watts, because some forcing mechanisms have a greater “efficacy” than others. For example, we showed that the effect of solar irradiance forcing is reduced about 10 percent, relative to the standard carbon dioxide forcing, because solar forcing is greatest at low latitudes, where there is little amplifying feedback from ice or snow. But solar forcing is increased about 20 percent by an indirect effect—the large solar variability at ultraviolet wavelengths alters atmospheric ozone. These known effects yield a net efficacy of about 110 percent for solar forcing relative to an equal carbon dioxide forcing—an amplification too small to substantially alter our assessment of the sun’s role in climate change.

So, if solar variability is to be a more significant climate forcing, there must be another, larger, indirect effect of the sun. The favorite among solar aficionados is an almost Rube Goldberg effect of galactic cosmic rays (GCRs) that goes like this: At solar minimum, GCRs penetrate farther into Earth’s atmosphere and increase atmospheric ionization, the ions serve as condensation nuclei for clouds, cloud cover increases, and the clouds reflect sunlight and cause global cooling. Indeed, it is true that the solar cycle has an effect on atmospheric ionization, but the state of the science does not yet allow definitive quantitative evaluation.

Fortunately, we have a way to skirt such difficult theoretical problems. We can go straight to empirical data to evaluate the effect of solar climate forcing, even its indirect effects. All we need are the measurements of solar variability from recent decades and observed global temperature.

But before leaping to the conclusion that there are mechanisms amplifying the solar forcing by a factor of two, we must look a bit more closely at Tung and Camp’s analysis. First, their method of avoiding volcanic influence on the record, removing two years of temperature data following two of the three large eruptions, neglects the residual longer-term effect of large volcanoes—coincidentally, the three large volcanoes in the period of study, Agung, El Chichón, and Pinatubo, all served to enhance cooling during solar minima. Second, Tung and Camp did not use the observed global temperature record of figure 12; instead they used a temperature record generated by a computer model in a process termed “reanalysis.” Reanalysis uses a global atmospheric simulation of the past few decades constrained by inserting available observations, such as tropospheric temperatures measured by satellite. The global surface temperature resulting from the reanalysis is qualitatively similar to that shown in figure 12, but the solar signal is about one-third stronger when measured this way. Tung and Camp suggest that variations of global temperature used in conventional analyses, as in figure 12, which are based on observations at meteorological stations, are more muted than the variations resulting from their reanalysis method, because of the absence of observing stations in places such as the Arctic and the Sahara, the areas with the largest temperature variability. Their supposition is plausible, but it is possible that instead their reanalysis model magnifies the temperature change.

The bottom line is that the only thing we can say with confidence is that the effective climate forcing due to the ten-to-twelve-year solar cycle has an amplitude in the range of 0.2 to 0.4 watt. In wonkish terminology, the 0.2 watt solar forcing has an “efficacy” between 100 and 200 percent.

Now that we have those figures, we can compare the natural solar climate forcing and the human-made carbon dioxide forcing. The annual increase of carbon dioxide today is about 2 ppm, which causes an annual increase of about 0.03 watt, with efficacy identically 100 percent. Thus even if the efficacy of the solar forcing is 200 percent, and if the sun’s brightness remains at the 2009 solar minimum value for a long period, the cooling effect relative to the average solar irradiance would be offset in seven years by a continuing carbon dioxide increase at recent rates. So there is no chance whatsoever that the sun can cause Earth to go into a new Little Ice Age—the numbers above confirm that human-made forcing now overwhelms the natural climate forcing.

Why, then, am I bombarded by demands to repent, to admit that global warming is a hoax? I have received scores of messages claiming that humans are not responsible for climate change and that Earth is headed into much colder conditions. Usually it is claimed that the sun controls climate and that the sun is moving into a period of reduced luminosity. As “proof,” these messages are often accompanied by a graph showing recent global temperature and an assertion that already “half of the global warming of the past century” has been lost. There is remarkable similarity among the messages, and most end by demanding that I resign from the government.