Journal of Atmospheric and Solar-Terrestrial Physics
The possible connection between ionization in the atmosphere by cosmic rays and low level clouds
Introduction
Galactic cosmic rays (GCR) are known to be the principal agents of ionization in the atmosphere above 1 km altitude. This has led to the suggestion that cloud formation may be affected by cosmic rays through an enhanced production of charged aerosols that may grow to become cloud condensation nuclei. As the cosmic ray flux on Earth is strongly modulated by solar activity, in the sense that increased solar activity leads to a reduction in the cosmic ray flux, if cosmic rays do affect cloudiness, they could provide a link through which solar activity affects climate.
The study by Svensmark and Friis-Christensen (1997) suggested that total cloud cover over the mid-latitude oceans was strongly correlated with the cosmic ray flux measured by neutron monitors—at least over the period 1984–1991 covered by the satellite cloud data they studied. However, the method by which they had fitted together different satellite cloud datasets, led to some criticism of their conclusions (Kernthaler et al., 1999).
Subsequently, a new uniformly calibrated series of satellite cloud observations from the International Satellite Cloud Climatology Project (ISCCP), the ISCCP-D2 series, has made it possible to assess the variability of cloud cover over substantial parts of the world and to compare this with the observed cosmic ray flux for a longer time interval. The good correlation between total cloud cover and cosmic ray flux found by Svensmark and Friis-Christensen (1997) for the period 1984–1991 was confirmed by this data, however, after 1991 the correlation disappeared.
From this data, only the low cloud of altitude 1–3 km correlated with cosmic ray flux; mid-level and higher cloud showed no apparent correlation (Pallé and Butler, 2000, Marsh and Svensmark, 2000). It was further shown that, although the degree of variation was small (1% global cloud coverage), if it persisted over decadal time scales, it could lead to significant changes in the global energy budget. Indeed, if the correlation observed between cosmic ray flux and low cloud cover during the 11 years 1983–1994 had been operational since the late 19th century and there were no feedback mechanisms operating or other cloud changes, then the observed variation in solar activity and cosmic ray flux could conceivably account for an appreciable fraction of global warming observed over the past century (Pallé and Butler, 2000).
Little is known about the physical mechanism that may link cosmic rays and clouds but, if one were to exist, it seems likely that it may occur, either through the effect of ionization on cloud formation or ion-mediated nucleation (IMN) which may result in the formation of condensation nuclei and increases in cloud lifetime (Yu, 2002, Turco et al., 1998, Svensmark and Friis-Christensen, 1997) or through the influence of cosmic rays on the global electrical circuit, the electroscavenging effect (Tinsley et al., 2000). For a detailed discussion on both mechanisms see Tinsley and Yu (2003). GCR have been reported to affect mainly low clouds (Pallé and Butler, 2000, Marsh and Svensmark, 2000), however, some authors have suggested that there may be an anti-correlation of high clouds with GCR if el Niño and volcano events are removed (Yu, 2002, Marsden and Lingenfelter, 2003).
There are other mechanisms proposed in the literature by which solar activity could affect Earth's cloudiness. Haigh (1996) proposed that changes in solar ultraviolet irradiance over a solar cycle may affect weather patterns in the troposphere via changes in the ozone layer and the propagation of planetary waves from the troposphere to the stratosphere. Udelhofen and Cess (2001) found a statistically significant 11-year signal in total cloudiness over the United States. Here the cloud cover variations were found to be in phase with solar irradiance rather than GCR. These results suggest a modulation of the atmospheric circulation following the action of solar ultraviolet radiation on ozone (Haigh, 1996, Shindell et al., 1999). As we will see, over the United States, ISCCP low cloud cover data does not show any significant correlation with GCR. More recently, Kristjansson et al. (2002) have used the ISCCP D2 (1983–1999) dataset to assess the correlation between low cloud cover with total solar irradiance (TSI) and GCR. They conclude that TSI correlates better and more consistently with low clouds than does GCR, and they provide a possible physical explanation following the results of Haigh (1996) and those of White et al. (1997) who found a significant solar signal in multi-decadal time series of sea surface temperature (SST). A review of the several mechanisms by which solar activity could influence cloud and climate has been published by Carslaw et al. (2002).
In this paper we concentrate on the hypothesis that GCR are responsible for the coupling of low cloudiness and solar activity through their effects on cloud processes. We have not explored the correlation with TSI. However, it is the belief of the authors that a simple correlation analysis is, in any case, inadequate to unambiguously decide which physical mechanism lies behind the correspondence.
Whereas in our earlier study (Pallé and Butler, 2000), using the ISCCP-D2 mean monthly data, we looked at the correlation of low cloud cover and cosmic ray flux over substantial areas of the globe (e.g., N. Atlantic, mid-high latitudes, global, etc.), in this study, with the data extended up to 2001, we are interested to see if the correlation is regional, if there are significant differences in sensitivity in different areas of the world and, in particular, if we can detect any variation with latitude that could be attributed to the differential shielding of cosmic ray particles by the Earth's magnetic field. If it were possible to quantify this, we hoped to be able to determine the coefficients of regression of the low cloud cover and cosmic ray flux over different regions to build up a more detailed picture of the contribution to climate change of cloud cover changes over the globe. Rather than use measured cosmic ray fluxes, which by and large are available only for relatively few points on the globe and usually over land regions only, we have used predicted fluxes from the LUIN Code. These have the added advantage that they can be extrapolated back to previous centuries using carbon and beryllium isotope proxies and used to predict ionization levels at different levels in the atmosphere.
Section snippets
Cloud data
Continuous global datasets of mean fractional cloud coverage and cloud radiative properties have been provided by the ISCCP from the combined observations of several inter-calibrated geosynchronous and polar satellites. Data are collected from a suite of weather satellites operated by several nations and processed by several groups. All ISCCP data products are archived at the ISCCP Central Archive (isccp.giss.nasa.gov).
The first ISCCP-D dataset, which extended from July 1983 to August 1994, is
Statistical methods
In determining the correlation between low clouds and GCR a problem arises. Cloud data have a strong seasonal cycle which is not present in the flux of GCR. To overcome this problem, some previous studies (Pallé and Butler, 2000, Marsh and Svensmark, 2000) have used a 12-month running mean of the cloud data, however this enhances the serial correlation between data points and reduces the number of degrees of freedom. For smoothed data the equivalent number of degrees of freedom is difficult to
The correlation between low clouds and ionization level in the atmosphere, 1983–2001
Fig. 2 shows the global annual averages of GCR-induced ionization in the atmosphere and low cloud amounts for the period July 1983–June 2000 (ionization data is only updated to December 2000). A quick look at the data reveals the good agreement between those two quantities from 1983 to 1994, however, from 1995 to 2000 the correspondence breaks. The correlation coefficient (0.49) over the full period is significant only at the 85% level. There are several possible causes for the break of
The amplitudes of low cloud and ionization level over the solar cycle
If there is a well-defined correlation between the ionization level produced by cosmic rays and the low cloud cover, as has been suggested, then it might be expected that the geographical variation in the amplitude of the ionization level would be similar to that of the amplitude of the low cloud cover. For such a comparison, it is evidently advantageous to use the geographically dependent modeled ionization rather than GCR fluxes from a few isolated stations. We plot the geographical
Calculating the effects on the Earth's temperature
In previous sections, it has been shown how the evidence for a connection between GCR and low clouds is inconclusive; though some aspects of our study are encouraging, others are basically negative. Despite these uncertainties, because of the potential impact on the Earth's radiation budget, we have explored the consequences for climate change, should the correlation between low cloud and ionization be confirmed.
Estimates of the possible effect of a link between cosmic rays and clouds have been
Conclusions
The correlation between annual mean low cloud and the ionization level at 2 km altitude exceeds the 99% significance level over mid-latitude oceans and globally over the period 1983–1994. However, globally, it drops to non-significant values if the full available cloud dataset (1983–2001) is taken into account, although some data adjustment such as detrending can restore the correlation significance to 99.5% or greater. Nonetheless, the correlation is significant over several large areas of the
Acknowledgements
The authors would like to thank the several anonymous referees for their comments and ideas which led to the improvement of this manuscript. Also, we would like to thank Dr Brian Tinsley for comments. The cloud D2 datasets were obtained from the NASA Langley Research Center Atmospheric Sciences Data Center. Research at Big Bear Solar Observatory is supported in part by a grant from NASA (NAG5-11007). Research at Armagh Observatory is grant-aided by the Department of Culture, Arts and Leisure
References (41)
- et al.
Effects of cosmic rays on the earth's environment
Journal of Atmospheric and Solar-Terrestrial Physics
(2000) - et al.
The proposed connection between clouds and cosmic rayscloud behavior during the past 50–120 years
Journal of Atmospheric and Solar-Terrestrial Physics
(2002) - et al.
Cloudiness decreases associated with Forbush decreases of galactic cosmic rays
Journal of Atmospheric and Solar-Terrestrial Physics
(1995) - et al.
The global atmospheric circuit, solar activity and climate change
Journal of Atmospheric Solar-Terrestrial Physics
(2000) - et al.
Variations of Cosmic Ray Flux and global cloud coverage. A missing link in solar-climate relationships
Journal of Atmospheric and Solar-Terrestrial Physics
(1997) - Alsmiller, F.S., 1965. A general category of soluble nucleon–meson cascade equations. ORNL,...
- et al.
Cosmic rays, clouds and climate
Science
(2002) Proceedings of the Royal Astrogical Society
(1955)Are cosmic rays influencing oceanic cloud coverage or is it only El Niño?
Climatic Change
(2000)The impact of solar variability on climate
Science
(1996)
Adjusting for sampling density in grid box land and ocean surface temperature time series
Journal of Geophysical Research
Some doubts concerning a link between cosmic ray fluxes and global cloudiness
Geophysics Research Letters
A new look at possible connections between solar activity, clouds and climate
Geophysics Research Letters
Climate forcing by changing solar radiation
Journal of Climate
Long-term drift of the coronal source magnetic flux and the total solar irradiance
Geophysics Research Letters
Solar activity and cloud opacity variationsa modulated cosmic ray ionization model
Journal of Atmospheric Science
Low cloud properties influenced by cosmic rays
Physics Review Letters
Galactic cosmic ray and El Niño-Southern oscillation trends in international satellite cloud climatology project D2 low-cloud properties
Journal of Geophysical Research
Secular variations in the production of cosmogenic isotopes in the Earth's atmosphere
Journal of Geophysical Research
Cited by (57)
The Sun's role in climate
2016, Evidence-Based Climate Science: Data Opposing CO2 Emissions as the Primary Source of Global Warming: Second EditionPossible solar-climate imprint in temperature proxies from the middle and high latitudes of North America
2016, Advances in Space ResearchCitation Excerpt :The results of Ogurtsov et al. (2002a, 2013, 2015) were produced analyzing centennial-scale paleorecords. During the last few decades more and more evidence have accumulated to support the idea that cosmic corpuscular radiation – galactic cosmic rays (GCR) and solar cosmic rays (SCR) – is a potential source of solar-induced climate variability over decadal and longer time scales (Marsh and Svensmark, 2000, 2003; Palle et al., 2004; Tinsley, 2008; Veretenenko and Ogurtsov, 2012, 2013). Marsh and Svensmark (2000, 2003) and Palle et al. (2004) showed a link between GCR and low cloudiness, Tinsley (2008) considered connections between cloudiness and global electric circuit.
Solar control on the cloud liquid water content and integrated water vapor associated with monsoon rainfall over India
2014, Journal of Atmospheric and Solar-Terrestrial PhysicsCitation Excerpt :Bangalore, Ahmedabad, Mumbai, Chennai, Hyderabad and Kolkata show negative correlation between SSN and ISM Rainfall in solar maximum but positive correlation in solar minimum. Sun can influence the formation of clouds by intervening through GCR, which is known to be the principal agents of ionization in the atmosphere above 1 km (Palle et al., 2004) and controls nucleation of water droplets in the atmosphere. Thus any change in the GCR activity may affect the rainfall variability.
The relation between lightning and cosmic rays during ENSO with and without IOD - A statistical study
2014, Atmospheric ResearchCitation Excerpt :Similarly, for our study region Table 3 shows a negative correlation between the LLCF and the CRF for the total period. A possible link between the CRF and the LLCF is as follows (Palle et al., 2004): it seems likely that it may occur, either through the effect of ionization on the cloud formation, the ion-mediated nucleation which may result in the formation of condensation nuclei and increases in cloud lifetime (Yu, 2002; Turco et al., 1998; Svensmark and Friis-Christensen, 1997) or through the influence of cosmic rays on the global electrical circuit i.e. through the electro-scavenging effect (Tinsley et al., 2000). Thus, there are two possibilities which may link CRF and LLCF.
Possible effects of atmospheric teleconnections and solar variability on tropospheric and stratospheric temperatures in the Northern Hemisphere
2014, Journal of Atmospheric and Solar-Terrestrial PhysicsNew evidence of solar variation in temperature proxies from Northern Fennoscandia
2013, Advances in Space Research