Observations of the Sun

Our sun is a typical “second generation,” or G2, star nearly 4.5 billion years old. The sun is composed of 92.1% hydrogen and 7.8% helium gas, as well as 0.1% of such all-important heavy elements as oxygen, carbon, nitrogen, silicon, magnesium, neon, iron, sulfur, and so forth in decreasing amounts (see Appendix 3). The heavy elements are generated from nucleosynthetic processes in stars, novae, and supernovae after the original formation of the Universe. This has led to the popular statement that we are, literally, the “children of the stars” because our bodies are composed of the elements formed inside stars. From astronomical studies of stellar structure, we know that, since its beginnings, the sun’s luminosity has gradually increased by about 30%. This startling conclusion has raised the so-called faint young sun climate problem: if the sun were even a few percent fainter in the past, then Earth could have been covered by ice. In this frozen state, it might not have warmed because the ice would reflect most of the incoming solar radiation back into space. Although volcanic aerosols covering the ice, early oceans moderating the climate, and other theories have been suggested to circumvent the “faint young sun” problem, how Earth escaped the ice catastrophe remains uncertain. How can the sun generate vast amounts of energy for billions of years and still keep shining? Before nuclear physics, scientists believed the sun generated energy by means of slow gravitational collapse. Still, this process would only let the sun shine about 30 million years before its energy was depleted. To shine longer, the sun requires another energy source. We now believe that a chain of nuclear reactions occurs inside the sun, with four hydrogen nuclei fusing into one helium nucleus at the sun’s center. Because the four hydrogen nuclei have more mass than the one helium nucleus, the resulting mass deficit is converted into energy according to Einstein’s famous formula E = mc2. The energy, produced near the sun’s center, creates a central temperature of about 15 million degrees Kelvin (°K).

Solar and Climate Changes

Until now we have considered only 11-year variations in solar activity and climate. The sun also varies on longer time scales. Since these variations seem to parallel a number of climatic changes, the sun may contribute to climatic changes on time scales of decades to centuries. We now examine several solar indices that vary in parallel with Earth’s climate change. There exist plausible arguments that these indices are proxy indicators of the sun’s radiative output, but there is no proof. We now present the strongest correlations we have seen for a sun/climate connection. First, as it is the most widely publicized index, we consider the mean level of solar activity. In 1801 Herschel first proposed a relationship between climate and the level of solar activity. Second, we examine solar cycle lengths, which have been studied sporadically since 1905. Third, we look at two closely related indices—sunspot structure and sunspot decay rates. Fourth, we consider variations in the solar rotation rate. Lastly, we examine some major solar and climatic events of the last thousand years to see if any indications of solar influence are evident on climate. Although we present the solar-induced changes as arising from total-irradiance variations, as discussed earlier spectral-irradiance changes may be the primary driver. When Rudolf Wolf reconstructed solar activity based on historical observations of sunspots, he found an 11-year cycle going back to at least 1700. In 1853 Wolf also claimed that there is an 83-year sunspot cycle. This longer term variation becomes evident simply by smoothing the data, as in Socher’s 1939 example. Wolf’s original discovery of an 83-year cycle was forgotten, but the long cycle was rediscovered by H. H. Turner, W. Schmidt, H. H. Clayton, and probably others. After W. Gleissberg also discovered this 80- to 90-year cycle around 1938, he published so much material on the subject that ever since it has been called the Gleissberg cycle. All these rediscoveries of the same phenomenon indicate that the 80- to 90-year cycle may be real but not strictly periodic. Rather, the cycle may be a “persistency” with an 80- to 90-year period. During this period solar activity is quite powerful but fails to exhibit a single sharp spectral peak.

Storms

We now examine some attempts to link storm numbers and storm track locations to solar activity. The number of both tropical cyclones and thunderstorms has increased and decreased with time and location as a function of solar activity. In fact, an early correlation between the number of Indian cyclones and solar activity proved so startling it caused an explosion of related research. In the previous century, tropical cyclones were called hurricanes or typhoons. Today tropical cyclones refer only to the weaker tropical storms with sustained winds above 31 miles per hour. Here, tropical cyclones refer to the stronger storms like those in the previous century. Anywhere from 1 to about 30 hurricane-strength storms can form each year. Among other factors, formation of these storms requires oceanic water temperatures above 26 °C (79 °F). William Gray at Colorado State University has successfully predicted the number of Atlantic Ocean hurricanes each year. This number is a function of the equatorial wind direction, the sea-level air pressure in the Caribbean, the strength of the westerly winds near the top of the lower troposphere, the presence or absence of an El Niño current, and, particularly, the amount of rainfall in the Sahel in Africa. Earlier we noted that increased solar activity produces a corresponding increase in rainfall in some regions. Figure 6.4 indicates that increased rainfall in the Sahel is expected, so based on this expectation and Gray’s theory, hurricanes should increase in number. Higher solar activity and a higher solar irradiance can also be expected to increase the tropical ocean temperatures by a few tenths of a degree. These increased water temperatures tend to increase both the number of tropical cyclones and their intensity. Figure 7.1 illustrates the number of Atlantic Ocean hurricanes observed between 1962 and 1994 as a function of the sea-surface temperatures (SST). A sharp gradient exists in the number of storms produced between 23 and 25 °C. In some regions, even a small increase in SST can lead to sharp increases in the number of tropical cyclones. Changes in solar brightness on the 11-year time scale could be expected to cause a corresponding cycle in the number and strength of tropical cyclones.

Temperature

How the bulk of the sun’s energy variations, which arise from the solar-irradiance changes, contribute to terrestrial changes will be the subject of this and the next few chapters. Although it’s a simplification, the hypothesis that only the largest solar activity–related energy contributors to the Earth’s atmosphere need to be considered allows us to ignore such far-flung ideas as the influence of sector boundary crossings, cosmic rays, and other less energetic phenomena. If the Foukal and Lean model of solar irradiance is a close approximation of known solar behavior on the 11-year time scale, what are the climatic consequences of these variations? We can ignore all the other proposed solar cycles ranging from 6 to 7 days to hundreds or thousands of years. Although some of these other proposed solar cycles may be real, we will not indulge in cyclomania here. North and his colleagues in 1983 developed a theoretical energy-balance model with a geographical distribution of land and ocean. Testing the model to see how well it reproduced the observed annual cycle of temperatures revealed satisfactory agreement, so North et al. subjected their model to other cyclic solar forcings. Figure 5.1 shows a 10-year cycle imposed on the Earth. Climatologists were not too surprised by the conclusions, as the solar-irradiance and temperature changes are nearly in phase because Earth’s time constant (about 3–5 years) is less than the imposed cycle time. The response over the land is greater than that over the oceans because the land holds less heat than water does and responds more strongly. The amplitude of the temperature variations is very small, no larger than about 0.11 °C. The maximum response is also centered near Arabia. Not everyone agrees that Earth’s temperature response to solar cycle changes will follow the scenario shown in Figure 5.1. In the last chapter, we described a novel approach suggesting the possibility of a larger response to solar activity than North’s and similar models provide. Kim in 1994 argues that tropical ocean waters, about 30° north or south of the equator, are the center of response to solar variations, (personal communication).

Climate Measurement and Modeling

Having considered the sun and its variations, we now turn to Earth’s climate and climatic variations. We examine the definition of climate and the difficulties in measuring it. Awareness of these complexities is critical for an appreciation of how difficult it is to demonstrate changing climate. Separating trends from random variations is the first step in defining climate change. After reviewing the statistical properties of climate, we deal with theoretical climate models. This background is important for understanding how solar variations might affect climate. The following four chapters review specific sun/climate relationships, and the statistical and physical guidelines developed now will be used to select pertinent studies. As the heat source that drives Earth’s climate, the variable sun is important when studying climate change. With many, if not most, modern popular accounts focusing on how humanity is altering climate, it is important to realize that solar variations may play a significant role in the background natural variability. To understand anthropogenic (human-made) influences on climate change, we must be able to make distinctions among the contributions that arise from naturally occurring climate variability. Natural climate variations include a possible solar-irradiance component. Man-made climatic changes are not well known, and natural climate variations are uncertain too. For example, we do not know whether a man-made doubling of atmospheric carbon dioxide provides a 1.5 or a 4.5 °C increase in mean global temperature. This uncertainty arises, in part, because natural climate variability acts as “noise” to confuse our measures of man-made influences. To obtain accurate results, we must understand and remove these background noise sources. Although these temperature changes seem small, they can have tremendous global impact on the survivability of species and on many different aspects of life. In addition, the uncertainty factor of 3 is highly important because it tells us that the risk in emitting a quantity of carbon dioxide is uncertain by this same factor.

Variations in Solar Brightness

In the last chapter we saw that sunspots, aurorae, and geomagnetic disturbances vary in an 11-year cycle. So do many other solar features, including faculae and plages, which are bright regions seen in visible and monochromatic light, respectively. If both bright faculae and dark sunspots follow 11-year cycles, does this mean the sun’s total light output varies? Or are these two contrasting features balanced so that the sun’s output of light remains constant? The light output of the sun is often discussed in two different ways: either as the solar luminosity, which is the sun’s omnidirectional radiant output, or as the solar constant, the output seen in the direction of the Earth. In this chapter, we explore the variable solar light output that has been the subject of vigorous discussions. The total solar irradiance or solar constant is defined as the total radiant power passing through a unit area at Earth’s mean orbital distance of 1 astronomical unit. Today the most common units of solar irradiance are watts per square meter (W/m2). Power is defined as energy per unit time, so the solar irradiance can also be expressed in calories per square centimeter per minute. Modern experiments indicate that the sun’s radiant output is about 1367 W/m2, with an uncertainty of about 4 W/m2. About 150 years of effort by many people have been required to establish the value to this accuracy. The sun’s radiant output is not an easy quantity to measure, and we will discuss some of the struggles required to measure it. In the late 1800s, many scientists considered the solar total irradiance or solar irradiance to be constant. Oceanographers Dove and Maury vigorously supported this viewpoint, so the solar irradiance was called the solar constant. For the next century, virtually every paper concerning the sun’s radiant output used the term solar constant. No physical justification for this nomenclature existed, only a philosophical bias. Yet by the 1950s this bias proved so strong and so prevalent that support for individuals who wished to measure variations in the solar constant became almost nonexistent.

Gaia or Athena? The Early Faint-Sun Paradox

Stellar evolution theory predicts large, long-term solar large, long-term solar luminosity (L⊙) changes over the lifetime of the sun. The most certain prediction is a general monotonic increase (neglecting short-period variations) in L⊙ of about 30% over the past 4.7 billion years, an increase that will continue. This prediction is well founded theoretically (based on the conversion of hydrogen into heavier elements) and supported observationally by the famous Hertzsprung-Russell diagram showing stellar evolution. If the solar luminosity increases monotonically with time, one might expect to find evidence of increasing surface temperatures in the Earth’s paleoclimatic record. Instead, isotopic indicators show Earth’s mean surface temperature is now significantly lower than it was 3 billion years ago. In 1975, R. K. Ulrich termed this the “faint young sun” paradox. Simultaneous solar luminosity increase and terrestrial temperature decrease imply additional strong influences on climate evolution. To understand climate evolution (and, by inference, the present climate), we must first determine the nature of these “compensatory mechanisms.” The positively increasing line in Figure 12.1 shows the evolution of solar luminosity (in units of present luminosity, L). Since terrestrial surface temperatures have remained nearly constant during the last 2.3 billion years, this requires a very effective compensatory mechanism. Several theories attempt to explain why the Earth’s surface temperature has remained relatively constant even while the solar luminosity has increased by 30%. Also, various scenarios have been advanced to explain why the Earth remained ice-free even during periods when the sun was much dimmer than it is today. Some of these ideas are: • Since it had fewer continents and more oceans, the early Earth was much darker. This same darker surface absorbed enough additional incoming solar radiation to remain ice-free. • In the past, energy transport from the equator to polar regions was easier because the continents had lower elevations. This enhanced heat transport allowed the Earth to remain relatively warm. • The early atmosphere had more carbon dioxide and methane, creating an enhanced greenhouse effect sufficient to trap the incoming solar radiation and keep the Earth warm. The enormous amount of carbon trapped in limestone suggests that Earth’s former atmosphere contained much more carbon dioxide than it does today.

Cyclomania

In his 1874 book Contributions to Solar Physics, Sir Norman Lockyer writes the following: . . . Surely in meteorology, as in astronomy, the thing to hunt down is a cycle, and if that is not to be found in the temperate zone, then go to frigid zones, or the torrid zones and look for it, and if found, then above all things, and in whatever manner, lay hold of, study it, record it, and see what it means. If there is no cycle, then despair for a time if you will, but yet plant firmly your science on a physical basis, as Dr. Balfour Stewart long ago suggested, before, to the infinite detriment of English science, he left the Meteorological Observatory at Kew; and having got such a basis as this, wait for results. In the absence of these methods, statements of what is happening to a blackened bulb in vacuo, or its companion exposed to the sky, is, for research purposes, work of the tenth order of importance. . . . As Lockyer notes, looking for cycles is certainly an attractive prospect. If found, a cycle will help with predictions, and successful predictions are a central goal of scientific studies. Cycle hunting is also a relatively straightforward procedure, with several well-developed techniques. Cycle hunting has become even easier with the advent of computers. By feeding a stream of data into an algorithm to detect cycles, one is likely to find cycles even in a series of random numbers. Hence the danger that a cycle, if detected, may prove to be only a random fluctuation. Will the cycle persist with more data, or will it simply disappear? For ardent cycle hunters, and the sun/climate field has attracted its fair share, these questions are hardly a deterrent. The lure seems to be finding something of seeming importance with minimal effort. From a practical point of view, a cycle may be considered important only if it can be plotted. If sophisticated analyses are required to detect the cycle, the cycle probably has only secondary importance. While these criteria are not the usual mathematical criteria for significance, they are a practical, down-to-earth guide to what is important.

Biota

We now consider insect populations, circumpolar mammal populations, seaweed density, agricultural yields, and similar topics. Good reasons exist to link such biological phenomena to solar activity. For one thing, if such meteorological parameters as temperature and precipitation vary with solar activity, life forms sensitive to small changes in these parameters may show dramatic responses. We will examine various claims from the 100 to 200 articles that either provide support for or criticize these types of ideas. The topics generally start at the lower levels of the food chain (i.e., insects) and proceed to the upper levels (i.e., predatory mammals), concluding with agricultural and economic studies. Insect populations are sensitive climate indicators. Paleontologists have used fossilized insects (see, for example, Coope, 1977) to show that very rapid changes in climate can occur in only a few years. Certain species of insects can tolerate only narrow ranges of temperature or precipitation. If meteorological variables alter that range, a new species of insect will replace the old. Insects occupy one of the lower rungs of the food chain, so fluctuations in their numbers may cause corresponding fluctuations in such predators as birds or spiders. Therefore, correlating insect populations with solar activity is a worthwhile venture. In his doctoral treatise, “Über die Beziehungen der Sonnenfleckenperiode zu meteorologischen Erscheinungen” published in 1877, F. G. Hahn argues that locusts will probably appear in temperate regions only during unusually hot and dry years. Hahn shows that European locusts appear preferentially between the years of sunspot minimums up to the next sunspot maximum, an average of about 4 years. For the 7 years from the sunspot maximum to the next sunspot minimum, locusts are scarcer. Since sunspot minimums produced relatively warm temperatures for the years 1800–1862, this suggests that the sun influences European locust populations. E. D. Archibald, who in his later years was a very ardent advocate of sun/climate relationships, extended Hahn’s findings. In a letter to Nature in 1878, Archibald showed that locusts appeared in Europe in 1613, 1690, and 1748–1749. According to Wolf, these dates occur 1 to 3 years after a sunspot minimum, which is consistent with Hahn’s findings.

Rainfall

This chapter examines rainfall and associated phenomena and their possible relationship to solar activity. Rainfall can be measured directly using rain gauges or estimated by monitoring lake levels and river flows. Satellite and radar rainfall measurements have become increasingly important. Historical documentation on drought, or the absence of rain, also reveals empirical relationships. Both rainfall and evaporation show marked variations with latitude and geography. First, we examine these rainfall-associated variations and estimate how they might change with solar activity. Second, we cover empirical studies of rainfall, lake levels, river flows, and droughts. The sun bathes the Earth’s equator with enormous amounts of surface energy. Much of this absorbed radiant energy evaporates water, causes atmospheric convection, and is later released to space as thermal radiation. Steady-state energy escapes, so tropical temperatures do not rise without limit. Some absorbed energy is transported poleward by winds from the point of absorption. Intense convection near the equator leads to a large updraft known as the intratropical convergence zone (ITCZ), a band of lofty, high-precipitation clouds producing the largest rainfall of any region on Earth. Solar energy in the ITCZ is carried to high elevations where it diverges and moves poleward. It is unable to travel all the way to the poles, so instead creates a large atmospheric circulation cell known as the Hadley cell. The Hadley cell has an upward motion near the equator and downward motions at about 30° north and south latitude. These downflow regions produce clear air with few clouds and create areas of minimum rainfall called deserts. These regions of upflow and downflow are connected by poleward flows in the upper atmosphere and equatorward flows in the lower atmosphere, forming a complete circulation pattern. Outside the Hadley cell are temperate and polar regions. The temperate regions have more rainfall than the deserts, while the cold polar regions have even less precipitation. Figure 6.1 shows the three regions with relative maximum rainfall. The mean evaporation has a much simpler latitudinal variation that tends to follow the surface temperature. Figure 6.1 shows this variation as a parabolicshaped dotted line.