Jump to ContentJump to Main Navigation
The Sun's Influence on Climate$

Joanna D. Haigh and Peter Cargill

Print publication date: 2015

Print ISBN-13: 9780691153834

Published to Princeton Scholarship Online: October 2017

DOI: 10.23943/princeton/9780691153834.001.0001

Show Summary Details
Page of

PRINTED FROM PRINCETON SCHOLARSHIP ONLINE (www.princeton.universitypressscholarship.com). (c) Copyright Princeton University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in PRSO for personal use. Subscriber: null; date: 26 September 2021

Solar Signals in Surface Climate

Solar Signals in Surface Climate

Chapter:
(p.109) 6 Solar Signals in Surface Climate
Source:
The Sun's Influence on Climate
Author(s):

Joanna D. Haigh

Peter Cargill

Publisher:
Princeton University Press
DOI:10.23943/princeton/9780691153834.003.0006

Abstract and Keywords

This chapter looks at how the Sun varies in terms of its emissions of radiation and particles and how these changes might be associated with variations in weather and climate on Earth. Investigations of climate variability and climate change depend crucially on the existence, length, and quality of meteorological records. Ideally, records would consist of long time series of measurements made by well-calibrated instruments densely situated across the globe. For longer periods, and in remote regions, records have to be reconstructed from indirect indicators of climate known as proxy data. The chapter introduces one well-established technique for providing proxy climate data: dendrochronology, or the study of the successive annual growth rings of trees.

Keywords:   Sun, radiation, weather variations, climate, meteorological records, proxy data, dendrochronology

IN CHAPTER 5 WE CONSIDERED HOW THE SUN VARIES IN terms of its emissions of radiation and particles. In this chapter and the next we look at how these changes might be associated with variations in weather and climate on Earth.

Investigations of climate variability and climate change depend crucially on the existence, length, and quality of meteorological records. Ideally, records would consist of long time series of measurements made by well-calibrated instruments densely situated across the globe. In practice, of course, this ideal cannot be met. Measurements with global coverage have been made only since the start of the satellite era, in the late 1970s, and records from meteorological instruments running for more than a couple of centuries are available from only a few locations in Europe. For longer periods, and in remote regions, records have to be reconstructed from indirect indicators of climate known as proxy data. These proxies provide information about weather conditions at a particular location through records of a physical, biological, or chemical response to the contemporaneous temperature or humidity. Some proxy data sets provide information dating to hundreds of thousands of years (p.110) ago, which make them particularly suitable for analyzing long-term variations in climate and their relation to solar activity.

One well-established technique for providing proxy climate data is dendrochronology, or the study of the successive annual growth rings of trees (which may be analyzed living or dead). It has been found that trees from any particular area show the same pattern of broad and narrow rings corresponding to the weather conditions under which they grew each year. Thus samples from old trees can be used to give a time series of these conditions. Felled logs can similarly be used to provide information about ancient times, providing it is possible to date them. This is usually accomplished by matching overlapping patterns of rings from other trees. One complication that arises with the interpretation of tree rings is that the annual growth of the rings depends on a number of meteorological variables integrated over more than a year, so that the dominant factor determining growth varies with location and type of tree. At high latitudes the major controlling factor is likely to be summer temperature, but at lower latitudes humidity may play a greater role.

Much longer records of temperature have been derived from analysis of oxygen or hydrogen isotopes in ice cores, in particular those obtained from Greenland and Antarctica. The ratio of the concentrations of different isotopes in the water molecules is determined by the rate of evaporation of water from tropical oceans as well as the rate of precipitation of snow over the polar ice caps. Both these factors are dependent on temperature, so (p.111) greater proportions of the heavy isotopes are deposited during periods of higher global temperatures. As each year’s accumulation of snow settles, the layers below become compacted, so that at depths corresponding to an age of more than 800 years it becomes difficult to date the layers precisely. Nevertheless, variations on timescales of more than a decade have been extracted dating back more than hundreds of thousands of years.

An individual proxy record does not give a precise measure of a particular climate parameter and needs to be calibrated by comparison with a contemporaneous instrumental record, as available. The choice of proxy data and of calibration period allows for considerable uncertainty, which is further increased by the treatment of spatial variability and seasonal biases. The result is a sometimes wide divergence between derived records. No individual method is foolproof, and recently there has been increased effort to produce multiproxy records. These probably provide more robust series and also give an indication of the uncertainty in the derived paleo record.

Paleoclimate Temperature Records

A record of the deuterium ratio, representing temperature, in an ice core retrieved from Vostok in East Antarctica is shown in Figure 6.1. The roughly 100,000-year periodicity of the transitions from glacial to warm epochs is clear and suggests a relationship with the variations in eccentricity of the Earth’s orbit around the Sun (see the discussion in Chapter 4), although this variation

(p.112)

Solar Signals in Surface Climate

Figure 6.1 Records derived from an ice core taken from Vostok, East Antarctica, showing variations in deuterium ratio (representing temperature) and the concentrations of methane and carbon dioxide over at least 400,000 years.

(Stauffer, 2000)36

does not explain the steep transitions from cold to warm. Evidence of very long term temperature variations can also be obtained from ocean sediments. The skeletons of calciferous plankton make up a large proportion of the sediments at the bottom of the deep oceans, and the oxygen isotope ratio within these is determined by the temperature of the upper ocean at the time when the (p.113) living plankton absorbed carbon dioxide. The sediment accumulates slowly, at a rate of perhaps 1 m every 40,000 years, so that changes over periods of less than about 1000 years are not detectable, but ice age cycles every 100,000 years are clearly portrayed.

Figure 6.1 also presents the concentrations of methane and carbon dioxide preserved in the ice core, showing a strong correlation between these and temperature. (Note that neither concentration is nearly as high as the present-day values of around 1893 ppbv [parts per billion by volume; Northern Hemisphere]/1762 ppbv [Southern Hemisphere] CH4 in 2012, and 397 ppmv CO2 in 2013). One theory proposed to account for these variations suggests that when the tilt of the Earth’s axis is small, the summers are cooler at high latitudes, and thus less of the previous winter’s buildup of ice melts, precipitating an ice age. At larger tilts the warming of southern high latitudes caused by the orbital variations is amplified by the release of CO2 from the southern oceans and further amplified through a reduction in albedo resulting from the melting of Northern Hemisphere ice sheets. Such positive feedback mechanisms might explain the sharp increases in temperature seen in the record. (Note that such increases will likely occur in response to any other warming effect, such as increasing concentrations of anthropogenic greenhouse gases.) Thus cycles of glacial/interglacial periods are probably related to changes in insolation owing to variations in orbital geometry.

This mechanism for ice ages does not involve any variations in intrinsic solar activity or output of (p.114) radiation. Across the Holocene (the period of about 11,700 years since the last Ice Age), however, isotope records from lake and marine sediments, glaciers, and stalagmites provide evidence that solar grand maxima/minima have affected climate. For example, in the North Atlantic, icebergs resulting from the flow of Greenland glaciers into the sea carry minerals which uniquely define their source. During colder periods the ice is able to raft farther south, where it melts, depositing the minerals. Ice cores and ocean sediments may also preserve, in isotopes such as 10Be and 14C, information on cosmic ray flux and thus solar activity. Consequently, simultaneous records of climate and solar activity may be retrieved from a single sediment or ice core record. The example in Figure 6.2 shows fluctuations on the 1000-year timescale well correlated between the two records, suggesting a long-term solar influence on climate. These ice-rafting events correlate with climate extremes measured in other parts of the world, such as weak events of the Asian monsoon, as indicated by stalagmite records from Oman. A caveat regarding all these studies is that they rely on the dating, which is complex and not always precise.

Proxy temperature data over the past millennium have been collated from a wide range of sources across the globe, including tree rings, ocean and lake plankton and pollen, coral, ice cores, and glaciers, to provide global (and hemispheric) average surface temperature and precipitation records. A composite of published temperature estimates, produced by using different data/

(p.115)

Solar Signals in Surface Climate

Figure 6.2 Records of 10Be and ice-rafted minerals extracted from ocean sediments in the North Atlantic. (Bond et al., 2001)37

techniques, is presented in Figure 6.3, which also shows the uncertainty range of the estimates.

Most of these records show relatively warm values over the period between about 950 and 1250 (sometimes referred to as the Medieval Climate Anomaly, MCA) and somewhat cooler values during the sixteenth to nineteenth centuries (sometimes referred to as the Little Ice Age, LIA). It has frequently been remarked that the Spörer, Maunder, and Dalton sunspot minima occurred during the LIA, leading to speculation that solar activity might have been the cause of the cooler temperatures. Care needs to be taken in such interpretation, however, as other factors might have contributed. For example, the higher levels of volcanism prevalent during the seventeenth century would also have introduced a cooling tendency owing to reflection

(p.116)

Solar Signals in Surface Climate

Figure 6.3 Variations in Northern Hemisphere surface temperature over the past millennium compiled from 12 estimates derived from proxy data; the shading indicates the overlap among the uncertainty ranges of the individual records. The black curve indicates the instrumental record since 1850. All values are shown relative to the 1961–1990 average (Based on Figure 6.10c of IPCC AR4)38

of the Sun’s radiation to space by a veil of particles injected into the stratosphere.

The black curve in Figure 6.3 also represents instrumental measurements of surface temperature compiled to produce a hemispheric average dating to 1860. Much of the current concern about global warming stems from the obvious rise over the twentieth century. Other records suggesting that the climate has been changing over the past century include the retreat of mountain glaciers, rise in sea level, thinning of Arctic ice sheets, and an increased frequency of extreme precipitation events. A key concern of contemporary climate science is to attribute cause(s) to this warming, including ascertaining the role of the Sun.

(p.117) Factors Influencing Global Surface Temperature

A variety of techniques can be used to assign contributions of individual factors to observed variations in global mean temperature. Several of these are reviewed in the appendix. The results of one attempt using multiple linear regression to analyze temperature over the period 1889–2006 are presented in Figure 6.4. Here the effects of human activity (due to emission of greenhouse gases and particulates), ENSO (El Niño–Southern Oscillation; see Chapter 3), and volcanic eruptions are assessed alongside solar variability. The top panel shows the raw temperature time series in gray. The four lower panels show the contributions inferred to be associated with each of the forcing factors. Each panel shows the forcing index on the right-hand axis and the associated temperature signal on the left-hand axis. Finally, the black curve in the top panel shows the temperature series reconstructed from the individual components; comparison with the gray curve reveals the goodness of fit. These data suggest that the Sun may have introduced an overall global warming (disregarding 11-year cycle modulation) of approximately 0.07 °C, although most of that warming was before about 1960. Over the whole period the temperature increased by about 1 °C, so the fractional contribution to global warming that can be ascribed to the Sun over the last century is 7%. However, this conclusion depends fundamentally on the assumed temporal variation of the solar forcing, and as discussed

(p.118)

Solar Signals in Surface Climate

Figure 6.4 Global monthly mean temperature record (black) and reconstruction from multiple regression analysis (gray). The regression contributions are shown for (a) ENSO, (b) volcanic, solar, and anthropogenic (greenhouse gases and particulates) influences (at appropriate lags).

(Lean and Rind, 2008)39

(p.119) in Chapter 5, there are several alternatives. The index of solar variability used in Figure 6.4 has a small long-term trend (relative to the 11-year cycle magnitude); the use of other solar indexes might produce a larger signal of temperature increase before midcentury and a better match between observations and regression model in the top panel. Crucially, however, it is not possible to reproduce the global warming of recent decades without including anthropogenic effects, and this conclusion is confirmed by the use of more sophisticated nonlinear statistical techniques.

Another approach uses estimates of radiative forcing (RF; see Chapter 4) to indicate potential global mean temperature change. A 1 W m−2 increase in TSI implies a RF of 0.175 W m−2 and an increase in global mean surface temperature at equilibrium (using a climate sensitivity of 0.8 K W−1 m−2) of about 0.14 K. Thus the implication of the different TSI series presented in Figure 5.4 becomes clear in assessing the role of the Sun in determining the temperature difference between the LIA and the present. That figure shows a range of values for the difference in TSI from the seventeenth century to the present of about 1–3 W m−2 (although other published results show even greater disparity). This range of values implies a solar radiative forcing of 0.17–0.50 W m2 and thus a corresponding solar-induced increase in global average temperature of 0.13–0.4 K since that date, which can be compared with the approximately 1 K increase observed.

TSI reconstructions may also be input to climate models to simulate the role of the Sun in climate history.

(p.120)

Solar Signals in Surface Climate

Figure 6.5 Northern Hemisphere mean surface temperature anomalies from observations and model simulations. Upper panel, thin gray lines: Ensembles of reconstructions derived from proxy indicators; black line: instrumental record since 1850; thick solid gray line: mean of model simulations using all forcings (see text for details) with Steinhilber TSI; thick dashed gray line: as solid gray line but using Shapiro TSI (see Figure 5.4 for TSI time series). Lower panel: Simulated contribution of each forcing factor, as identified in the legend (VOLC: volcanic aerosol; GHG: greenhouse gases; LUSE: land use; AER: other aerosol) (Schurer et al., 2013)40

(Copyright © 2013, Rights Managed by Nature Publishing Group)

The example presented in Figure 6.5 shows the Northern Hemisphere mean surface temperature over the past 600 years from a GCM forced with greenhouse gases, tropospheric aerosols (sulfate, dust, and soot particles), stratospheric (volcanic) aerosols, changes in land use, as well as two different estimates of TSI. The TSI data set due to Steinhilber contains a 1 W m−2 overall increase (p.121) since the Maunder Minimum, and the one due to Shapiro a 6 W m−2 increase over the same period. In the upper panel of Figure 6.5 the thin gray lines show a range of reconstructions of the temperature record from proxy measurements; the black line is the instrumental record. The climate model predicted values for all forcings with the Steinhilber TSI series align fairly well, within experimental uncertainties, with the observed/proxy record. The Shapiro reconstruction shows greater warming over the latter half of the twentieth century, but it also extends temperatures outside the range of observed values across much of the 600-year period. In the lower panel, model runs with individual forcings are shown by the various lines identified in the figure caption and legend. The TSI contribution to variations in the temperature anomaly is small throughout the period when the Steinhilber data are used. We noted in Chapter 5 that the Shapiro TSI reconstruction, with its large variations, is an outlier relative to most others, and the larger- amplitude temperature deviations modeled with this TSI series are clear in the figure.

From these reconstructions we conclude that while natural factors—particularly volcanism—are likely to have contributed to variations in temperature over the past 600 years, they cannot account for the sharp warming since about 1970. Studies using optimal estimation approaches (see the appendix) to attribute causes to global temperature change since 1750 come to very similar conclusions without any a priori assumptions on the magnitudes of the forcing factors.41

(p.122) Over the most recent 15 years global mean surface temperatures did not rise as fast as they did during previous decades (see the top panel of Fig. 6.4), nor as fast as predicted by most GCMs for that period. The reasons for this “hiatus” in global warming are currently the subject of intensive study, and variations in the Sun have been proposed as a possible reason. Certainly, overall solar activity was low over that period relative to the previous few decades, but observed changes in TSI were not large enough to compensate for the radiative forcing from greenhouse gases. There may be a small solar component in the recent slowing of the warming trend, as well as a contribution from a number of small volcanic eruptions, but, as demonstrated by similar periods during the century-long data set, it is likely that it represents mainly natural internal variability and a redistribution of energy within the climate system.

It has further been suggested that if the Sun is currently in a state of overall declining activity (see Chapter 5), then the concomitant negative radiative forcing might compensate for the global warming likely to take place in response to greenhouse gas increases on longer timescales. Hypothetically, if the Sun were to enter another grand minimum such as the Maunder event within the next 50–100 years this (by the arguments presented in the preceding discussion) would produce a maximum global cooling of 0.1–0.4 K as compared with the anticipated warming due to unregulated carbon dioxide emissions of 3–4.5 K in the same period. An experiment with an AOGCM42 assuming a reduction of 0.25% in TSI for (p.123) a 50-year period confirms this conclusion, showing that such solar behavior would temporarily slow down but not stop global warming. Thus, to rely on such cooling to counter human-produced global warming not only would be a risky strategy, given uncertainties in predicting solar activity, but would also likely produce only a small, temporary compensation.

Regional Effects

Both the statistical and modeling approaches discussed demonstrate that signals of solar variability can be detected in records of the global (or hemispheric) average surface temperature. More detailed analyses suggest that the response is not spatially uniform but that certain parts of the world may experience greater or, indeed, opposite effects. These signals are more difficult to confirm, because the natural “noise” in the time series is not reduced by spatial averaging, but the responses found suggest that a solar signal can be detected in some of the natural modes of variability in the climate system (see Chapter 3).

Solar signals have been identified in meteorological records from many stations across the globe. For example, signals of period around 11 years have been identified from power spectrum analysis of temperature records from stations across the United States, with positive correlations to the east side of the Rockies and negative to the west.43 In Europe the positions of the storm tracks crossing the North Atlantic have been found to (p.124) shift north and south with solar activity.44 An 11-year periodicity has also been identified in cyclones in the tropical Atlantic.45 There are many other examples. The statistical robustness of these studies is not always well demonstrated, and sometimes the signal changes in and out of phase with solar activity (for a very good discussion see the review by Hoyt and Schatten46), making the task of identifying physical causes for the statistical relationships even more challenging.

Optimal estimation techniques (see the appendix) have been used to explicate the processes involved in the statistical analysis. For example, using an energy balance model to generate a distribution over the globe of the response in surface temperature, together with noise estimates from long runs of AOGCMs, Stevens and North47 identified a small (maximum a few hundredths of a degree Celsius over land in summer) solar signal in the data.

Polar Regions

The Holocene proxy temperature records across the Northern Hemisphere show strong regional variations in the solar response, including an NAO-like pattern (North Atlantic Oscillation; see Chapter 2), such that when the Sun was less active, the climate was characterized in an NAO-negative phase more frequently than normal. Thus during the Sun’s Maunder Minimum the surface temperatures were typically cooler in eastern North America and western Europe, and warmer in Greenland and central Asia.48 The cooler temperatures (p.125) are consistent with the public experience in Europe, where the longer period including this time was named the Little Ice Age.

There is some evidence that generally higher solar activity during the Medieval Climate Anomaly coincided with warmer temperatures in western Europe and eastern North America, although this result is less robust than the converse found in the Maunder Minimum/LIA. Simulations with coupled atmosphere-ocean GCMs (global climate models; see Chapter 3) do, however, generally show a response in which the strength of the NAO signal correlates positively with solar irradiance.

Analyses of instrumental records of surface temperature and pressure on the 11-year solar cycle timescale show regional variations in the solar signal consistent with those found on longer timescales. Atmospheric blocking events, during which the jet stream is diverted in a quasi-stationary pattern associated with cold winters in western Europe (consistent with a negative NAO scenario), occur more frequently when the Sun is less active.49 Allowing for a few years’ lag between solar forcing and atmospheric response appears to strengthen this relationship.50

A robust solar signal is also seen in the Pacific Ocean, where a large positive anomaly in sea-level pressure in the northeast Pacific and a negative anomaly at lower latitudes indicates that the Aleutian low-pressure region sits farther to the west and the Hawaiian high farther to the north, and pushing the storm tracks further north in response to a more active Sun.51,52

(p.126) In the Southern Hemisphere analyses of surface pressure measurements at mid–high latitudes, as well as model simulations, suggest that the Southern Annular Mode (SAM; see Chapter 3) is more often in a positive phase when the Sun is more active, with colder polar temperatures and stronger circumpolar winds. The picture is complicated, however, by an indication that the state of the Quasi-Biennial Oscillation (QBO; see Chapter 3) plays a role in the SAM response to the Sun.53

The polar response to solar variability is further discussed in Chapter 7, where it is suggested that these surface effects may be part of a response produced throughout the depth of the atmosphere but initiated by solar heating of the stratosphere.

Tropics

The signal in the North Pacific, noted in the previous section, is similar to the behavior associated in midlatitudes with a cold ENSO event (i.e., La Niña). A link to the solar cycle has been found in sea surface temperatures (SSTs) in the eastern tropical Pacific which is expressed as a cool (La Niña–like) anomaly at sunspot maximum, followed a year or two later by a warm anomaly, although the technique used to derive this result—namely, solar peak year compositing—means that there are only 14 data points, so the robustness of this signal has still to be established.

Analyses of tropical circulations are not conclusive, but a picture is emerging of a slight weakening and expansion of the Hadley cells54,55 (see Chapter 2), so the (p.127) descending branches, and dry air, occur at higher latitudes. There are also signs in precipitation indicators of a strengthening of the Walker circulation and strengthening of the Asian monsoon when the Sun is more active.56 Consistent with these findings but on longer timescales, paleodata of precipitation derived from stalagmites in caves in southern Oman show lower monsoon precipitation associated with greater 14C production (i.e., lower solar activity). On the solar cycle timescale, data reveal changes in tropical circulation, in cloudiness, and in the location and strength of regions of precipitation in the Pacific consistent with a cold ENSO-like signal at higher solar activity.

As outlined in Chapter 4, about 50% of the solar energy incident at the top of the atmosphere, averaged over the globe, reaches the Earth’s surface but is not uniformly distributed. Radiation is most intense in the tropics, but most of it reaches the surface in the cloud-free subtropical regions. Over the oceans a large proportion of this radiant energy is used in evaporation. The resulting high-humidity air is carried by the prevailing trade winds into the tropics, where it converges with the stream from the other hemisphere and ascends, producing the deep cloud and heavy precipitation associated with the ITCZ. One mechanism for solar–climate links suggests that changes in the absorption of radiation in the clear-sky regions provide the driver: greater irradiance would result in enhanced evaporation, moisture convergence, and precipitation. These changes would result in stronger Hadley and Walker circulations and stronger (p.128) trade winds, which would create greater upwelling in the eastern tropical Pacific Ocean and colder SSTs, and thus the observed La Niña–like signal described earlier.57 It is not obvious that the amounts of energy involved in TSI variations (of order tenths of one percent) are sufficient to produce the desired effect, but there is some evidence for its taking place in GCM simulations of paleoclimate, and also over the solar cycle, although the timing of the signal relative to the cycle peak remains contentious.58 It is also uncertain whether variations in the temperature profile within the lower atmosphere are consistent with what might be anticipated through this mechanism. Nevertheless, it remains a plausible candidate for a solar influence on regional climate and will benefit from further investigation.

Having considered the effects of solar variability at the Earth’s surface, we now look higher into the atmosphere.