Content uploaded by Natalya Andreeva Kilifarska
Author content
All content in this area was uploaded by Natalya Andreeva Kilifarska on Sep 02, 2015
Content may be subject to copyright.
ISSN 10693513, Izvestiya, Physics of the Solid Earth, 2015, Vol. 51, No. 5, pp. 768–785. © Pleiades Publishing, Ltd., 2015.
Original Russian Text © N.A. Kilifarska, V.G. Bakhmutov, G.V. Mel’nik, 2015, published in Fizika Zemli, 2015, No. 5, pp. 160–178.
768
1. INTRODUCTION
Among numerous research fields concerning the
phenomenon of the magnetic field of the Earth,
G.N. Petrova paid special attention to the fundamen
tal problems associated with the environmental
impacts of the geomagnetic field, including its effect
on climate. The correspondence between the changes
in the geomagnetic field and the climatic variations is
studied from the paleodata at the level of searching for
the correlations between these phenomena (e.g.,
(Petrova et al., 1992)). However, the problem of the
relationship between the Earth’s magnetic field
(EMF) and climate is rather controversial. This rela
tionship has been explored on the time scales of thou
sands and millions of years, based on the paleodata
and during the period of direct instrumental observa
tions, in relation to the climatic variations and solar
and magnetic activity. In other words, these studies
address the relationship between the climatic trends
and the changes in EMF caused by different
sources—internal, located in the Earth’s core, and
extraterrestrial, located in the ionosphere and mag
netosphere. These sources have a fundamentally dif
ferent nature and different frequency range, and the
information about them is obtained by different meth
ods. This relationship is studied on the timescales of
hundreds, thousands, and millions of years using the
paleodata and during the period of direct instrumental
observations of the variations in climate, solar and
geomagnetic activity, and other geophysical parame
ters.
To date, plenty of data has been gained both testi
fying to the existence of these relationships and
refuting them. This triggered hot debate (e.g.,
(Courtillot et al., 2007; 2008; Bard and Delaygue,
2007)) mainly focused on establishing the reliable
correlations and identifying the probable driver of
these correlations.
In our previous works (Bakhmutov et al., 2011;
2014), we have demonstrated the correlation between
the changes in the parameters of the main geomag
netic field and climate on time scales from decades to
hundreds of years. Using a unified procedure for data
processing, we calculated the total intensity of the geo
magnetic field vector and the fields of the surface air
temperature and pressure for the time interval of
1900–2010 with a step of 10 yr at the nodes of the reg
ular grid with
10
° spacing in latitude and longitude for
the latitudinal band of
40
°
–70
° N. The analysis of the
integral characteristics and the dynamics of the fields
enabled us to identify their regional features and global
trends.
In this paper we describe the results of identifying
the spatial and time relationships between the main
magnetic field of the Earth and the other candidate
impacts on the climatic variations on the time scale of
decades and suggest the mechanism to account for the
causeandeffect relations between these phenomena.
Geomagnetic Field and Climate: Causal Relations
with Some Atmospheric Variables
N. A. Kilifarska
a
, V. G. Bakhmutov
b
, and G. V. Mel’nik
b
a
National Institute of Geophysics, Geodesy, and Geography, Bulgarian Academy of Sciences, Sofia, Bulgaria
email: nkilifarska@geophys.bas.bg
b
Institute of Geophysics, National Academy of Sciences of Ukraine, Kyiv, Ukraine
email: bakhm@igph.kiev.ua
Received April 4, 2015
Abstract
—The relationship between climatic parameters and the Earth’s magnetic field has been reported by
many authors. However, the absence of a feasible mechanism accounting for this relationship has impeded
progress in this research field. Based on the instrumental observations, we reveal the spatiotemporal relation
ship between the key structures in the geomagnetic field, surface air temperature and pressure fields, ozone,
and the specific humidity near the tropopause. As one of the probable explanations of these correlations, we
suggest the following chain of the causal relations: (1) modulation of the intensity and penetration depth of
energetic particles (galactic cosmic rays (GCRs)) in the Earth’s atmosphere by the geomagnetic field; (2) the
distortion of the ozone density near the tropopause under the action of GCRs; (3) the change in temperature
near the tropopause due to the high absorbing capacity of ozone; (4) the adjustment of the extratropical
upper tropospheric static stability and, consequently, specific humidity, to the modified tropopause temper
ature; and (5) the change in the surface air temperature due to the increase/decrease of the water vapor green
house effect.
DOI:
10.1134/S1069351315050067
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
GEOMAGNETIC FIELD AND CLIMATE 769
This mechanism explains the influence of the geomag
netic field on climate through the processes at the
boundary between the upper troposphere and lower
stratosphere. A detailed account of the separate links
in this chain is given in (Kilifarska, 2012a; 2012b). In
this work, we present the chain of the causeandeffect
relations between cosmic rays, EMP, and variations in
the atmospheric ozone and water vapor in the upper
troposphere/lower stratosphere, which result in the
change of the radiation balance of the Earth and,
hence, cause longterm variations in the surface air
temperature.
2. THE PROBABLE MECHANISMS
OF EXTERNAL IMPACTS
ON THE EARTH’S ATMOSPHERE
The detection of the correlation between the differ
ent geophysical parameters (including those associ
ated with the processes in the upper atmosphere and
nearEarth space) and climatic parameters neither
unravels the causal relations nor hints at the probable
drivers of these correlations. A variety of the probable
mechanisms have been suggested to account for the
impacts from the external (cosmic) factors on the
lower and middle atmosphere (e.g., (Loginov, 2008)).
The state of the upper atmosphere is largely governed
by solar electromagnetic radiation, solar corpuscular
fluxes, galactic cosmic rays, and electron precipitation
from the radiation belts. The changes in these factors
alter the atmospheric structure, composition, and
dynamical characteristics at different heights.
The character of the interactions between the
atmosphere and external cosmic factors is controlled
by EMF. Fast magnetospheric processes barely affect
the lower atmosphere. Considering the complexity of
the perturbation transfer from the upper layers to the
lower ones, the existence of certain triggering mecha
nisms associated with the energetic particles absorbed
in the stratosphere is conceivable. However, as far as
the main geomagnetic field and its secular variations
are concerned, their modulating effect on the galactic
cosmic rays is undoubted.
The influence of energetic particles (galactic cos
mic rays (GCRs) and solar cosmic rays (SCRs)) on the
state of the upper atmosphere has been studied in
many publications (e.g., see the review in (Krivolutskii
and Repnev, 2012)). The mechanisms explaining the
effects of the charged particles, which penetrate into
the Earth’s atmosphere at a height below 100 km, are
still unclear despite the long history of their study and
observations.
Energetic particles may cause dissociation, ioniza
tion, and dissociative recombination of the atmo
spheric species. The resulting lowenergy electrons
(~10–100 eV) then produce the bulk of the ionization.
The GCRs penetrate the troposphere; the SCRs may
in some cases reach heights of 20–30 km when the
energy of the particles is above 100 MeV; and the
5MeV electrons penetrate up to 40 km. The energy
injected into the atmosphere by the energetic particles
is far below the contribution from the other energy
sources in the middle atmosphere; at the same time, it
is one of the key sources of ionization and dissociation
below ~80 km where the solar ultra violet (UV) and
Xray radiation are strongly attenuated. The low
energy particles mainly affect the ionization and heat
ing of the thermosphere and deposit their energy in
high latitudes, whereas the highenergy particles pen
etrate the lower atmosphere at any latitudes.
The interaction of energetic particles with the
atmospheric species results in the production of new
compounds which, inter alia, affect minor atmo
spheric components, including the ozone concentra
tion. Energetic particles may cause significant ozone
depletion by lowering the ozone concentration in the
mesosphere and upper stratosphere (Krivolutskii and
Repnev, 2012). The ionmolecular reactions involving
ozone (
О
3
), which take place in the ozonosphere,
account for at most 10% of the total rate of ozone deg
radation at any arbitrary concentration of the particles
(Larin and Talrose, 1977). The indirect influence of
energetic particles on the ozone layerthrough the for
mation of OH (an active source of ozone destruction
resulting from ionmolecule reactions between the
primary ions and ), as well as water molecules,
at night, significantly increases in the case of the ion
ization of the stratosphere by solar protons above the
ozone layer’s peak heights during the solar proton
events (SPEs) (Vinogradov et al., 1980).
Thus, searching for the probable mechanisms of
the charged particles influence on climate through the
impact on cloudiness, aerosols, changes in the circu
lation, albedo, and ozone, is a topical research direc
tion. A recent review of the publications on the subject
(Krivolutskii and Repnev, 2012) shows that studying
the relevant mechanisms is a challenging task. Never
theless, our analysis of different geophysical time
series describing the dynamics of the processes in the
lower stratosphere–upper troposphere (UTLS) and
on the ground surface suggests a new mechanism driv
ing the geomagnetic field control of the climatic vari
ations.
3. DATA AND METHODS
The analysis of the decadal changes in the atmo
spheric pressure and temperature fields during
XX century was previously carried out by us for Janu
ary, when the change in the temperature regime due to
the global temperature rise is highest. In Bakhmutov
et al. (2014), we used the method described in (Mar
tazinova and Ivanova, 2011). The same method was
applied for calculating the total intensity of the geo
magnetic field vector
F
(nT) and its secular variation
from 1900 to 2010 with a time step of 10 yr from the
coefficients of the international geomagnetic refer
ence field (IGRF) model (http://wdc.kugi.kyoto
+
N
2
+
O
2
770
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
KILIFARSKA et al.
u.ac.jp/igrf/index.html). The detailed comparison of
these data is described in (Bakhmutov et al., 2014). All
the calculations were conducted at the nodes of the
geographic grid with 10° spacing in latitude and longi
tude for the latitudinal interval 40°–70° N, the region
with the densest coverage by the observational moni
toring network.
For the subsequent analysis presented in this paper,
we used the data of the ERA40 and ERA Interim
reanalysis (http://apps.ecmwf.int/datasets/); more
specifically data for atmospheric ozone (total O
3
den
sity and its mixing ratio at 70 hPa) and specific humid
ity (at 150 hPa) for the period of 1957–2011. The both
reanalyses are merged at January 1, 2000. For estimat
ing the specific humidity (SpH) and surface air tem
perature
T2m
at 2 m above the ground, we analyzed the
monthly averages, whereas for the ozone profiles we
calculated the monthly values from the 6h reanalysis
data, taken at 12:00 UT.
The following time series were additionally used for
the statistical analysis applying linear and nonlinear
statistical methods:
—total ozone content for the period 1926–2010
(annual averages) for the Arosa station in Switzerland
(http://www.iac.ethz.ch/en/research/chemie/tpeter/
totozon.html); before 2007, and for 2008–2010
(http://www.woudc.org/data_e.html);
—surface air temperature in the Northern Hemi
sphere for the past 100 yr from the CRUTEM3 and
CRUTEM4 database of the Met Office Hadley Center
(http://www.metoffice.gov.uk/climateguide/science/
sciencebehindclimatechange/hadley) and Cli
matic Research Unit of the University of East Anglia
(http://www.cru.uea.ac.uk/);
—equivalent effective stratospheric chlorine
(EESC) (http://acdbext.gsfc.nasa.gov/Data_
services);
—the Wolf numbers (Solar Spot Numbers, SSN)
(http://spidr.ngdc.noaa.gov); and
—concentration of atmospheric carbon dioxide
CO
2
for 1958–2011 (http://cdiac.ornl.gov/trends/
co2/siokeel.html#).
Climax neutron monitor data reflecting the varia
tions in the intensity of galactic cosmic rays (GCL),
http://cr0.izmiran.ru/clmx/main.htm.
Since the neutron monitor data are only available
after the beginning of the 1950s, we extrapolated them
back to 1926 according to (McCracken and Beer,
2007; Mursula et al., 2003) and used their forecast up
to 2019 by the method (Lantos, 2005) using the
expected sunspot numbers (according to the data of
the Marshall Space Flight Center).
The solar activity was factored in our analysis in
terms of the sunspots number, considering the reliabil
ity of this parameter and its correlation to the other
indices, e.g., Mg II (Rozelot et al., 2004). This allowed
us to average these data and compare them to the
longer time series of ozone variations (Kilifarska et al.,
2012a).
The EESC time series were also extrapolated back
to 1926 and forecast up to 2020 at the Atmospheric
Chemistry and Dynamics unit of the Science
Research portal of NASA’s Goddard Space Flight
Center (http://atmospheres.gsfc.nasa.gov/acd/).
The concentration of carbon dioxide in the atmo
sphere was assumed in accordance with the data from
the Mauna Loa observatory (Hawaii) for the period
1958–2011. The time series were exponentially
extrapolated back to 1900 (http://cdiac.ornl.gov/
trends/co2/).
Similarly to the analysis of pressure and tempera
ture fields, we have used for other variables seasonal
data of the winter months. For exploring the subse
quent links in the mechanism of the ozone O
3
influ
ence on the surface air temperature (
T2m
), we used
their winter values since at this time (a) the ozone con
centration is close to the annual maximum, whereas
the humidity (SpH) is close to its minimal values,
which suggests the maximal expected effect of O
3
on
specific humidity and, as will be shown below, on the
corresponding change in the surface air temperature;
(b) the seasonal variations in the GCR ionization
reach their peak values (Forbush, 1960).
Since the bulk of the ozone is in the lower strato
sphere, we used the 30year TOMS (the Total Ozone
Mapping Spectrometer) record (http://ozoneaq.
gsfc.nasa.gov/)), as an additional independent data set
for estimating the total ozone content (TOC) in the
winter time for the period of 1980–2010 (http://
disc.sci.gsfc.nasa.gov/giovanni/overview/index. html).
The statistical correlations between the studied
parameters were sought by calculating the crosscorre
lation coefficients by the standard approach in which
the covariances are normalized to the standard devia
tions of the time series.
The nonlinear analysis of the data was conducted
with the use of the STATISTICA 6.0 program pack
age. The accuracy of the model (i.e., the difference
between the observations and model predictions) was
estimated by the least square criterion of the Leven
berg–Marquardt algorithm. The significance of the
coefficients of the correlation was assessed by Stu
dent’s twotailed
t
test.
Among the other methods we note estimation of
the ionization efficiency of the atmospheric compo
nents’ by secondary electrons and ions (produced by
GCRs), which was calculated using the Maxwell–
Boltzmann distribution. The efficiency of the ion
molecule or ionatom reactions was determined from
the Saha ionization equation (Kilifarska, 2012b).
4. THE RELATIONSHIP BETWEEN THE MAIN
GEOMAGNETIC FIELD AND SOME
ATMOSPHERIC PARAMETERS
We have previously shown (Bakhmutov et al., 2011;
2014) that during 1900–2000 the regions with the
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
GEOMAGNETIC FIELD AND CLIMATE 771
maximal intensity of the geomagnetic field coincide
with the atmospheric centers of action in the Northern
Hemisphere, where the pressure is high and tempera
ture is low. By collating the curves of the total intensity
of the geomagnetic field
F
, temperature
T
, and pres
sure
P
, one can see a fairly good agreement between
their minima and maxima (Bakhmutov et al., 2014,
Fig. 14). The coefficient of correlation between
P
and
T
(
R
FТ
) is =
⎯
0.77, between
F
and
P
(
R
FP
)
= 0.66, and
between
F
and
T
(
R
FT
)
= –0.83. The correlation
between the temperature and pressure fields is quite
natural; however, the match between the peaks in
Т
and
F
is rather surprising. The coefficient of correla
tion between these variables
R
FT
= –0.83 is also unex
1980
150
100500–50–100–150
2000
1960
1980
2000
1960
1980
150
100500–50–100–150
2000
1960
1980
2000
1960
1980
150
100500–50–100–150
2000
1960
1980
2000
1960
1980
150
100500–50–100–150
2000
1960
1980
2000
1960
5
3
0
0
0
0
53
0000
5
3
00
00
57
0
00
0
57
0000
5
7
0
0
0
0
5
7
0
0
0
0
5
3
0
0
0
0
2
6
2
2
7
1
2
6
2
4.2
4
.
2
4
.
2
4
.
2
4
.
2
4
.
2
3
.
3
3
.
3
3
.
3
2
.
8
2
.
8
3
.
4
2
.
8
2
.
8
2.
8
Mean GeoMag Field F, nT
Mean surf.
T
, K
Mean O
3
(70 hPa), ppmv
Mean SpH (150 hPa) · 1e9, kg/kg
Longitude, deg
Fig. 1.
The longitude–time variations of latidudinally averaged (over 40°–70° N lat) of: (top to bottom) the total intensity of the
geomagnetic field
F
, nT; surface air temperature
T2m
, K; ozone O
3
, ppmv at 70 hPa; and specific humidity SpH, mg/kg at
150 hPa for the period from 1957 to 2011.
772
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
KILIFARSKA et al.
pectedly high, compared to the correlation between
the other two interrelated parameters
R
РТ
= –0.77.
Figure 1 illustrates the spatiotemporal distribution
of the total intensity of the geomagnetic field vector,
the surface air temperature data
T2m
from the merged
ERA40 and ERA Interim reanalyses, the average val
ues of the lower stratospheric ozone
O
3
at 70 hPa, and
the specific humidity SpH at 150 hPa. The data corre
spond to the latitudinal range of
40
°
–70
° N and the
second half of XX century. The latitudinally averaged
O
3
and SpH for the UTLS heights evidently have sim
ilar longitudetime distributions (Fig. 1, third and
fourth panels). Despite the strong interdecadal vari
ability of these parameters, the sectors with a higher
concentration of
O
3
and, at the same time, low SpH
are distinguished in North America and East Asia.
These sectors correspond to the regions of high
F
and
low surface air temperature
T
.
This is more clearly illustrated in Fig. 2, where the
longitudinal distributions of
F
and
T2m
averaged over
1957–2011 and over the 40°–70° N latitude (Fig. 2a)
are shown together with the lower stratospheric ozone
O
3
(at 70 hPa), specific humidity SpH (at 150 hPa),
and surface temperature
T2m
(Fig. 2b). The anticorre
lation between
F
and
T2m
and the inphase behavior
of
T2m
and SpH, which is antiphase to O
3
(Fig. 2),
are observed.
Generally, Figs. 1 and 2 show that the spatiotempo
ral variations in the considered parameters of the
upper atmosphere and surface temperature are in
some way concordant with the general pattern of sur
face distribution of the main geomagnetic field. Con
sidering the correlation between the surface air tem
3.5
–180
–140–100 –60 –20 20 60 100 140 180
3.7
3.9
4.1
4.3
4.5
274
270
266
262
258
254
250
4.8E+04
5.0E+04
5.2E+04
5.4E+04
5.6E+04
5.8E+04
6.0E+04
270
266
262
258
254
250
246
F
, nT
T
, K
T
, K
O
3
(70 hPa), ppmv &
SpH(150 hPa), kg/kg
Longitude, deg
F
T
O
3
SpH
(а)
(b)
Fig. 2.
The fiftyyear means of latitudinally averaged (over 40°–70° N lat) (a) total intensity of the geomagnetic field and surface
air temperature, (b) surface air temperature, ozone at 70 hPa, and specific humidity at 150 hPa.
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
GEOMAGNETIC FIELD AND CLIMATE 773
perature and ozone revealed previously (Kilifarska,
2012a), we will attempt to reveal the main factors con
trolling the variability of
T2m
and O
3
.
5. ANALYSIS OF THE FACTORS
CONTROLLING THE LOWER
STRATOSPHERIC OZONE
AND SURFACE AIR TEMPERATURE
It is generally believed that the degradation of the
ozone layer during the 1980–1990s resulted from an
increase in the concentration of ozonedestroying
substances (chlorides and bromides) in the atmo
sphere, primarily in the lower stratosphere, which
were accompanied by the longterm changes in the
lower stratospheric circulation (WMO, 2006 and ref
erences therein). However, the causes of the changes
in the circulation, including the trends in the Arctic
Oscillation/North Atlantic Oscillation (AO/NAO)
indices (Bronnimman et al., 2009) remain unclear.
Consequently at least 50% of the ozone variability in
the lower stratosphere has not yet been reasonably
explained to date.
The impacts of the factors affecting the ozone vari
ability (presented in the WMO reports) were estimated
by the methods of linear statistics. On the other hand,
many authors have demonstrated the nonlinearity of
the time series of the climatic parameters and some
factors of external forcing (Mende and Stellmacher,
2000;
Miks vsky
and Raidl, 2005). We applied the
nonlinear statistical methods for estimating the con
tributions of the key factors controlling the strato
spheric ozone. The calculations were conducted by
o
ˆ
the STATISTICA 6.0 program; the results are pre
sented in the table.
We have analyzed the statistically significant (at
95%) coefficients of nonlinear regression of TOC from
the Arosa station in Switzerland in comparison to
(1) equivalent effective stratospheric chlorine EESC;
(2) Earth's magnetic filed vector (
F
); (3) (GCR) inten
sity; (4) westerly type of circulation (determinations
by the Vangeigeim–Girs circulation indices); and
(5) sunspot numbers SSN. Parallel to this, we have
also explored the surface air temperature and the key
factors responsible for its changes (Kilifarska, 2012a;
2012b).
The results of nonlinear statistical analysis, shown
in the Table, illustrate that the decadal changes in the
GCR flux account for 55% of the TOC’s variability.
The EESC is responsible for 46%, while 38% are asso
ciated with the westerly zonal circulation. The fact
that 41% of the total ozone variability can be attributed
to the action of the geomagnetic field is more than sur
prising.
At present, the largest contribution to the changes
in the air temperature on the Earth’s surface is
assigned to the increased concentration of carbon
dioxide and other anthropogenic greenhouse gases
((IPCC, 2007) and references therein). Our nonlinear
reanalysis of the factors controlling climate variability
reveals at least three factors, besides CO
3
, each of
which accounts for an almost equal fraction (in per
centage) of the observed variability of the surface air
temperature. These factors are the variations in the
GCR flux (71%), the changes in the total intensity
F
of
0.47
0
.
5
3
0
.
4
7
0
.
4
7
0
.
5
3
0
.
5
3
0.47
0.
47
–
2
–
2
–
2
–
2
–2
–
2
–
2
–
3
–
3
–
2
–
2
–
4
–
2
–
3
–
3
–
3
–
3
–0.9 –0.5 –0.2 0.2 0.6 1.0 –5 –4 –3 –2
Corr. coeff. (GeoMagF & GCR) Time lag, years
(а) (b)
Fig. 3.
The spatial distribution of the statistically significant coefficients of correlation between GCR and TOC (TOZ, the con
tours) for the Northern Hemisphere. The shaded maps at the background represent correlation between geomagnetic field and
GCR. Dark gray (green in colour version) zones denote negative correlation; pale gray (orange) ones positive correlation. The
time lag of TOC response to GCR in years are shown on the right. The TOC and GCR data are shown for winter months.
2
774
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
KILIFARSKA et al.
40
150
100500–50–100–150
60
80
40
60
80
40
150
100500–50–100–150
60
80
40
60
80
1
0
2
0
2
0
2
0
1
0
1
0
3
0
9
0
7
0
5
0
3
0
3
6
0
3
6
0
3
2
0
36
0
40
0
40
0
Longitude, deg
TOZ & GCR corr. coef. + SecVarF
TOZ & GCR corr. coef. + TOMS O
3
–0.1 0.1 0.3 0.5 corr. coeff.
1
0
1
0
2
0
(a)
(b)
Fig. 4.
The latitude–longitude map of the statistically significant coefficients of correlation between the total ozone content
(TOMS data) and GCR (Climax data) for 1986–2010. The 30yr means of the (a) secular variation of the geomagnetic field in
nT/yr and (b) TOC in Dobson’s units are shown by contours.
geomagnetic field vector (71%), and the longterm
total ozone variability (77%). These parameters are
not independent: the coefficient of correlation
between the geomagnetic field at the latitude of the
Arosa station and the 22year running average of GCR
flux is 0.77 for the period of 1900–2010; and the coef
ficient of nonlinear regression between the longterm
changes in GCR flux and ozone is 0.74 for 1926–2010
(see the table).
The modulation of GCR flux by the geomagnetic
field is a wellknown fact (e.g., see (Van Allen, 1962;
Shea and Smart, 2004; Usoskin et al., 2005)). How
ever, the statistical correlation of the geomagnetic
field with the surface air temperature and total
ozone is rather unexpected and it needs to be
explained.
6. THE MECHANISM OF GEOMAGNETIC
FIELD’S CONTROL ON CLIMATE
6.1. Geomagnetic Modulation of Galactic Cosmic Rays
and Spatial Distribution of the Total O
3
Density
The primary cosmic rays entering into the atmo
sphere collide with the atomic nuclei of the most
abundant atmospheric gases—nitrogen and oxygen.
This interaction triggers the process of cascade ioniza
tion, which produces almost all the known elementary
particles. At the qualitative level, the penetration of
the cosmic ray particles into the Earth’s magneto
sphere can be described in terms of Stermer’s theory of
the motion of a charged particle in the field of a mag
netic dipole. Based on extensive observations, it has
been found that most of the primary particles are
charged positively and the intensity of the cosmic rays
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
GEOMAGNETIC FIELD AND CLIMATE 775
incoming from the west is higher than from the east
(Rakobol’skaya, 1971).
In (Kuznetsova and Kuznetsov, 2012), it is noted
that the regions of the global magnetic anomalies in
East Siberia, North America, South Pacific, South
Indian Ocean, and Central Atlantics are marked with
a sharp change in the precipitation rate and energy
spectrum of the energetic particles. Markov and Mus
tel (1983) have shown that the precipitation of the
highenergy particles from the Earth’s magnetosphere
into the lower ionosphere is mainly confined to the
regions of magnetic anomalies. In this respect, the
Brazilian magnetic anomaly has been studied most
profoundly. The intensity of the magnetic field in the
center of this anomaly is as low as almost half the
ambient values, and this anomaly is displayed up to the
heights of ~600 km (Pinto et al., 1992). The charged
particles flux density in the region of the Brazilian
anomaly is by a few orders of magnitude higher than
the particle density in the other regions, distant from
the anomaly (Glassmeier et al., 2002).
Although the energy injected into the atmosphere
from GCR is negligible (a billionth of the fraction of
the incoming solar radiation), the GCR–climatic link
is at present a focus of active discussion by many
authors, because the mechanisms behind this link have
reasonable physical–chemical basis (e.g., see the
review (Kirbym, 2007)). The key works concerning
the ionization of the atmosphere by cosmic rays are
reviewed in (Bazilevskaya et al., 2008) summarizing
the direct and indirect measurements of the atmo
spheric ionization and model calculations.
In (Shea and Smart, 2004), it is shown that the
GCR variations due to the changes in the geomagnetic
field (and, therefore, the geomagnetic cutoff hardness)
during the past 400 years are commensurate with the
GCR modulation during the 11year cycle (averaged
over the previous four last solar cycles). According to
(Kovaltsov and Usoskin, 2007), the regional variations
in the GCR intensity caused by the drift of the geo
magnetic pole can be even stronger than the deviations
associated with the changes in solar activity.
In order to verify our hypothesis for the correlation
between the geomagnetic field, GCR, and TOC, we
carried out a lagged correlation analysis of these
parameters all over the Earth for the period of 1957–
2012. The results for the statistically significant (at
95%) coefficients of correlation calculated for each
node of the grid with 10° spacing in latitude and lon
gitude are illustrated in Fig. 3. For the Northern
Hemisphere, the positive coefficients of GCR/TOC
correlation, reflecting the effects of GCR on TOC
(from TOMS data), fairly well map into the regions of
positive correlation between the geomagnetic field and
GCR. Here, it should be taken into account that the
GCR–TOC link is nonlinear (see section 4 above),
which explains the relatively low linear coefficients of
correlation between these quantities.
Figure 4 shows a different projection of the statisti
cally significant coefficients of the GCR/TOC corre
lation, with the superimposed contours of the 30yr
mean secular variation of the geomagnetic field (the
30yr means of the secular variation in the total inten
sity of the geomagnetic field
F
, determined as the dif
240 280 320 360 400 440 TOZ, DU
0.3 0.9 1.5 2.1 2.7 O
3
(70 hPa), ppmv
(а)
(b)
Fig. 5.
The comparison between the calculated ozone
(mixing ratio) produced in the lower stratosphere by auto
catalytic cycle (shown by the contours, ppmv) and (a) the
measured TOC climatology according to TOMS data
(shading, Dobson’s units) for 1980–2010 and (b) 50yr
mean O
3
mixing ratio climatology of ERA40 and ERA
Interim at 70 hPa (shading, ppmv).
776
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
KILIFARSKA et al.
ference between the values of
F
for the subsequent and
preceding years, are calculated at each node of the
10°
×
10
° grid, Fig. 4a). Also shown are the 30yr mean
TOC according to the TOMS data (Fig. 4b). It is
clearly seen that the maximal values of the coefficient
of correlation between TOC and GCR have fallen in
the regions characterized by the maximal decrease
(about
100
°
W latitude) and increase (West Europe) in
the intensity of the geomagnetic field during the past
30 years. The comparison of these data with the aver
age TOC (Fig. 4b) reveals the increase in the total
columnar ozone in North America and East Asia and
its decrease above the North Atlantic and West
Europe.
There are two conclusions, following from Figs. 3
and 4: (i) the relation between GCR and total ozone
are not uniformly distributed over the globe, what
could be expected having in mind the latitudelongi
tudinal variations in the GCR ionization rates (e.g.,
see (Velinov et al., 2005; Usoskin et al., 2005; Kov
altsov and Usoskin, 2007)). (2) The connection
between GCR, TOC, and the secular variation in the
geomagnetic field is equivocal.
Thus, the geomagnetic field control of the GCR
intensity, on one hand, and the correlations between
TOC and GCR, on the other hand, requires a mecha
nism through which GCR can affect the ozone, pri
1.56
270190
250230210
3.75
5.00
9.40
11.88
16.25
20.00
28.75
35.62
40.00
48.12 1
3
7
20
50
100
200
300
500
700
925
–2E12 0 2E12 4E12 6E12 8E12 1E13 1.2E13
Temperature, K
Approximate height, km
Height, hPa
Ozone density, cm
–3
T
(90s)_70
°
N
T
(70s)_70
°
N
O
3
(90s) 70
°
N, cm
–3
O
3
(70s) 70
°
N, cm
–3
Fig. 6.
Comparison between the decadal means of O
3
(full symbols) and temperature open symbols along the height profile
according to the ERA40 reanalysis data fro 70° N and Greenwich meridian in the 1970s (squares) and 1990s (dots).
850
0.500
0.0500.005
600
400
250
150
70
30
10
5
Saturated specif. humid.
×
10
–6
, kg kg
–1
Pressure, hPa
Measure. SpH (Antarct.)
UM_SpH (Antarct.
T
) UM_SpH (Antarct.
T
+10)
Fig. 7.
Comparison between the measured (solid line with
filled dots) and model profiles of specific humidity. The
dashed line with filled markers depicts the SpH profile cal
culated from the measured temperatures by formulas (5)
and (6). The dashed line with open markers corresponds to
the calculations for the same profile with the temperature
increased by 10 K.
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
GEOMAGNETIC FIELD AND CLIMATE 777
marily in the region where the ozone concentration is
maximal, i.e., the lower stratospheric ozone O
3
.
6.2. Ion Chemistry of the Lower Stratosphere
and Autocatalytic Cycle for O
3
Production
The galactic cosmic rays, with much higher ener
gies than solar protons, efficiently penetrate into the
lower stratospheric layers and the troposphere. The
ion chemistry at these heights has many uncertainties
(e.g., (Brasseur and Solomon, 2005)). Here, the
changes in the GCR influence on the O
3
during the
11yr solar cycle should be taken into account. For
example, at
60
° N latitude in summer, the NO pro
duction by GCR during the maximal and minimal
solar activity differs by 20% (Jackman et al., 1980).
The lowenergy GCRs, SCRs, and the relativistic
electrons initiate the ionmolecule reactions which
result in the production of the ozoneactive com
pounds
HO
x
and NO
x
. In turn, the latter trigger the
cycle of the ozonedestructing reactions (e.g., (Banks
and Kockarts, 1973; Jackman et al., 1980; Brasseur
and Solomon, 2005; Krivolutskii and Repnev, 2009)).
The ion chemistry in the lower stratosphere, however,
differs substantially from that of the upper atmo
spheric layers. (Kilifarska, 2012b; 2012c; 2013).
The spatial distribution of the ozone produced in
the autocatalytic cycle close to the Pfotzer maximum
(the maximum in the ion production rate due to GCR,
located at an altitude of ~15 km) is shown by the con
tours in Fig. 5. For comparison, we also plotted the
30yr mean total ozone content according to the TOMS
data (Fig. 5a, shading) and
O
3
at 70 hPa (Fig. 5b). The
400
0.039
0.030
0.021
0.012
0.003
250
150
70
30
10
5
400
250
150
70
30
10
5
23
4 5 6
Pressure, hPa
Moist adiabat. lapse rate, deg/km
Specific humidity, kg kg
–1
SpH(part
Γ
w
(
T
+ 10))(L)
Part
Γ
w
(
T
– 10 K)(R)
SpH(part
Γ
w
(
T
– 10))(L)
Part
Γ
w
(
T
+ 10 K)(R)
Fig. 8.
The profiles of the moist adiabatic lapse rate (
Γ
w
)
calculated with the decrease (the solid line with filled
squares) or increase (the dashed lines with open squares) in
temperature by 10 K along the Terra Nova Bay height pro
file (Tomasi et al., 2004). Also shown is the calculation of
specific humidity SpH in relation to the increase in tem
perature and
Γ
w
(the solid line with open squares) and the
decrease in these quantities (the dashed lines with filled
squares).
250
7
654
200
150
100
70
50
30
20
10
250
7
654
200
150
100
70
50
30
20
10
Pressure, hPa
Moist adiabatic lapse rate Saturated (moist) adiabatic lapse rate
Γ
w
(
T
)
Γ
w
(
T
+ 15 K)
Γ
w
(SpH – 20%)
Γ
w
(
T
)
Γ
w
(
T
– 15 K)
Γ
w
(SpH + 20%)
Pressure, hPa
(a) (b)
Fig. 9.
The moist adiabatic lapse rate as a function of temperature and humidity: (I) the solid line corresponds to the standard
Γ
w
(calculated from the observed
T
and SpH for the Terra Nova Bay); (II) the dotted line with squares depicts
Γ
w
calculated with the
temperature profile (a) increased by 15 K and (b) decreased by 15 K; (III) the dashed line with triangles corresponds to
Γ
w
cal
culated with the specific humidity (a) decreased by 20% and (b) increased by 20%.
778
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
KILIFARSKA et al.
similarity in the spatial distribution of TOC and ozone
produced by GCR, as well as the comparable mixing
ratios with the reanalyses’ climatology, shows that the
impact of GCR in NH ozone budget is not negligible.
Summarizing the results of this section, we may
conclude that the modulation of GCR by the heliom
agnetic as well as by the geomagnetic field results in
the changes of the lower stratospheric ozone density,
which accounts for the correlations between TOC
and the geomagnetic field detected by the statisti
cal methods.
It is worth to note, however, the hot discussion
ongoing presently on the importance of the ozone
destructive species of NO
x
family (NO, NO
2
) pro
40
150
100500–50–100–150
60
80
20
0
40
60
80
20
0
150100500–50–100–150
60
80
40
150100500–50–100–150
60
80
40
60
80
40
150100500–50–100–150
60
80
40
150100500–50–100–150
60
80
40
60
80
40
2
7
0
2
6
0
2
5
0
28
0
2
9
0
2
9
0
2
8
0
2
7
0
2
6
0
2
5
0
2
5
0
2
60
2
7
0
2
8
0
2
9
0
0
.
1
0
.
3
–
0.
3
–
0.
1
0.
1
–
0
.
1
–
0
.
1
–
0
.
3
0
.
1
0
.
3
0
.
5
–
1
–
1
–
1
–
0
.
3
–
0
.
1
–
0
.
3
–
0
.
3
–
0
.
3
–
0
.
1
–
0
.1
–5
–
9
–
9
–
5
–
5
–
5
–
5
Longitude, deg Longitude, deg
Latitude, deg
Latitude, deg
Latitude, deg Latitude, deg
Lagged corr. (TOZ &
T2m
)
Instatnt. corr. (TOZ &
T2m
)
Time lag
Time lag
Latitude, deg
Latitude, deg
TOMS TOZ &
T2m
240 300 360 420 TOZ, DU
(а)
(b)
Fig. 10.
(a) Latitudelongitude distribution of 30year averaged TOMS ozone (gradual shading) and near surface air T from ERA
reanalyses (contours). Note that sectors with highest O3 density correspond to the colder wintertime regions in the Northern
Hemisphere; (b) The spatial distribution of the statistically significant (at 95%) coefficients of correlation between the winter
means of TOMS ozone and near surface air T. Top: the
T2m
response to total ozone forcing with a delay of less than 5 years (left)
and the time lag in years (right); bottom: temperature delay by more than 5 years. The time lag in years is indicated on the right
bottom.
The statistically significant (at a level of 95%) coefficients of nonlinear regression of the total ozone content (TOC) at the
Arosa station, Switzerland, and surface air temperature (
T2m
) for the main influencing factors. The variations described by
each nonlinear model are indicated in percentage of the total ozone variability (top) and temperature (bottom)
Equivalent
efficient strato
spheric chlorine
(EESC)
Total intensity
of the geomagnetic
field vector (
F
)
22yr average
GCR time series
11yr average
indices of zonal
circulation
11yr average solar
spot numbers
(SSN)
Total ozone content
(TOC)
0.68
46%
0.64
41%
0.74
55%
0.62
38%
0.58
34%
CO
2
F
22yr average
GCR time series
11yr average total
ozone (TOC)
11yr average solar
spot numbers
(SSN)
Surface air tempera
ture (
T2m
)
0.86
74%
0.84
71%
0.84
71%
0.88
77%
0.79
62%
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
GEOMAGNETIC FIELD AND CLIMATE 779
duced by GCR, for the lower stratosphere ozone bal
ance. The continuous inflow of energetic protons of
galactic origin causes permanent ionization and pro
duction of NO, mainly in the polar lower stratosphere,
where the primary GCR penetrate deeper into the
atmosphere. Beyond the polar region, the ability of
GCR to produce NO in the lower stratosphere sharply
drops. The other factors (solar protons, emission of
energetic electrons from the magnetosphere, etc.)
sporadically affect the NO concentration in the upper
stratosphere, mesosphere, and thermosphere. The
production of the significant amounts of NO mole
cules and other nitrogen oxides in the upper strato
sphere matters to a certain extent for the photochem
ical reactions leading to the ozone destruction.
Besides, the air masses with the increased
NO
x
content
can be transported by the descending branch of the
Brewer–Dobson circulation (a simple circulation
model postulating the presence of slow flows in the
winter hemisphere, which redistribute the air from the
tropical to the extratropical atmosphere) in the polar
regions of the lower stratosphere. This process may
sharply reduce the ozone content. However, this effect
is insignificant on the time scales of climatic changes
and within the considered latitudinal interval. In this
work we do not intend to consider the entire range of
the questions concerning the effects of GCR, solar
proton events, and precipitation of relativistic elec
trons on the ozone content but only consider the lower
stratospheric ozone and its relationship with GCR.
Thus the multiyear variability of the lower strato
spheric ozone supposedly related to the GCR inten
sity, modulated in turn by the heliomagnetic field and
the main geomagnetic field, is the first link in the
causal chain connecting the geomagnetic field and the
surface air temperature.
6.3. The Impact of Lower Stratospheric
Ozone on Climate
The high sensitivity of the climate to the changes in
the lower stratospheric ozone density was noted by
many authors (Ramanatan and Callis, 1976; Wang
et al., 1980; 1993; de Foster and Shine, 1997; Stuber
et al., 2001; etc.). Nevertheless, the mechanism of this
relationship is not entirely clear to date. In this sec
tion, we briefly summarize the data which are thor
oughly exposed in (Kilifarska, 2012a; 2012b), which
will help us to understand how the lower stratospheric
ozone can affect the surface air temperature.
Ozone is one of the most radiatively active atmo
spheric gases. It refers to the most important minor
constituents of the atmosphere, which are of utmost
importance for the climate. The changes in the ozone
concentration affect the incoming solar and outgoing
(reflected from the Earth) radiation, moreover, the
2.2
1957
1962 1967 1972 1977 1982 1987 1992 1997 2002 2007 2012
2.4
2.6
2.8
3.0
3.2
3.4
3.6
3.8
5.8
5.6
5.4
5.2
5.0
4.8
4.6
4.4
Specif. humid.(150 hPa), 10
6
×
kg/kg
O
3
(50 hPa), ppmv
smt SpH150 (50
°
N)(L)
smt SpH150 (70
°
N)(L)
smt O
3
50hPa (50
°
N)(R)
smt O
3
50hPa (70
°
N)(R)
smt SpH150 (60
°
N)(L)
smt SpH150 (80
°
N)(L)
smt O
3
50hPa (60
°
N)(R)
smt O
3
50hPa (80
°
N)(R)
Fig. 11.
The time series of the zonally averaged winter ozone at 70 hPa and specific humidity at 150 hPa (smoothed by 5 point
running average), taken from merged ERA40 and Interim reanalyses, at four latitudes. Note that the periods of increased O
3
content are accompanied by the low density of water vapor.
780
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
KILIFARSKA et al.
effect on the latter is highly dependent on the on the
height distribution of ozone.
The estimated integral radiative forcing of colum
nar ozone on the surface air temperature is very small
because the degradation of the stratospheric ozone at
the end of XX century should lead to a decrease in the
surface air temperature (Ramanatan et al., 1976; de
Foster and Shine, 1997), whereas the tropospheric
ozone enhancement (due to the anthropogenic emis
sions resulting from the reactions between CO, NO
x
,
nonmethane hydrocarbon compounds,
СН
4
, etc.)
causes the opposite effect in the Earth’s radiation bal
ance (Wang et al., 1993; Gauss et al., 2006). Ulti
mately, the stratospheric and tropospheric forcing
cancels each other to a great extent. These results sug
gest that not columnar ozone, but its variability at
some atmospheric level is of special importance for the
climate system.
Below we describe a newly proposed mechanism
rendering the influence of the lower stratospheric
ozone on the surface air temperature. This mechanism
includes three key links: (1) the control on the tropo
pause temperature; (2) adjustment of the UTLS
humidity to the changes in tropopause temperature
induced by the ozone variations; and (3) the rise or
drop in the air temperature near the ground surface
due to the variations in the water vapor forcing at the
UTLS heights. Let us briefly outline each of these
links.
6.4. The Lower Stratospheric Ozone Controls
of the Tropopause Temperature and UTSL
Static Stability
The correlation between the lower stratospheric
ozone
O
3
and temperature near the tropopause can be
attributed to the ozone’s ability to absorb the solar
ultraviolet (UV) radiation (with the wavelength
<400 nm) as well as the longwave outgoing radiation
reflected from the Earth. The cooling of the lower
stratosphere during the strong degradation in
O
3
in the
1990s was noted by many authors (e.g., (de Forster and
Tourpali, 2001; Seidel and Randel, 2006; Randel
et al., 2009)). This dependence is illustrated in Fig. 6
which shows the decadal values of the O
3
and temper
ature profiles between 1970 and 1990 based on the
ERA40 reanalysis data. Clearly, the ozone layer
depletion after the 1970s is accompanied by the long
cooling of the lower stratospheric layers.
On the other hand, a striking example of the ther
modynamically active admixture, which strongly
affects many atmospheric processes is water. Its con
centration (in particular, the specific humidity
SpH
)
provide a significant contribution to the air density
and stratification of the atmosphere. It affects the
fluxes of the short and longwave radiation and causes
the greenhouse effect.
Another known fact is that the change in the tem
perature of the tropopause alters the density of the
water vapor in the lower stratosphere (Mote et al.,
1996). On the other hand, the model experiments
(Spencer and Braswell, 1997) show that even minor
changes in the amount of the water vapor at these lev
els lead to nonlinear changes in the radiation balance
of the atmosphere. The experimental measurements
(Inamdar et al., 2004) have confirmed that the main
contribution to the greenhouse effect belongs to the
water vapor in the upper troposphere. Thus, the next
link in the chain transferring the lower stratospheric
ozone impact on the surface air temperature the
–
1
5
–
1
2
–
1
2
–
1
2
–
12
–
1
5
–
1
2
–
1
5
–
1
2
8
(a) dT, K + Lagged corr. SpH(150 hPa) &
T2m
(b) Time lag, years
0.20
0.12
0.01
–0.04
–0.12
–0.20
–9
–12
–15
–18
–21
–24
Time
lag
d
T2m
Fig. 12.
The spatial distribution of the statistically signifi
cant (at 95%) coefficients of correlation between specific
humidity at 150 hPa and surface air temperature (con
tours) overdrawn on: (a) the 50yr means of the dynamic
anomalies in the surface air temperature in K according to
the ERA reanalysis data (shading); (b) the time lag of the
temperature response to the changes in specific humidity
(gray shading).
lasting
is
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
GEOMAGNETIC FIELD AND CLIMATE 781
interaction between temperature and humidity in the
UTLS layer.
According to the presentday understanding
(IPCC, 2007), the response of the climatic systems to
the rise in the surface air temperature (initiated by the
increase in the concentrations of
CO
2
and other green
house gases) and to the growth in the amount of water
vapor in the atmosphere not only occurs in the lower
and middle troposphere but also involves the upper
troposphere. For verifying these assumptions, we
compared the height profile of the specific humidity of
the saturated humid air measured in the Terra Nova
Bay in the Antarctic (Tomasi et al., 2004) with the
model profile calculated by the formula
(1)
where is the watervapor saturation pressure and
is the atmospheric pressure in hPa. For calculating
the watervapor saturation pressure (above ice surface)
we used the Goff–Gratch formula recommended by
WMO
(2)
for the temperature interval 173 K <
Т
< 273.15 K;
T
t
=
273.16 K is the temperature of the triple point of
water; and
T
is the measured temperature along the
height profile in the Terra Nova Bay.
For estimation of the sensitivity of formulas (1) and
(2) to temperature variations, we repeated the calcula
tions with the temperature increased by 10 K. The
results of these calculations, together with the mea
surements of the specific humidity are displayed in
Fig. 7. From this figure it follows that assumed tem
peraturepressure dependence of specific humidity fits
the empirical data up to a level of ~200 hPa. Above this
height level, the model SpH values sharply increase
and even at 100 hPa there is a complete disagreement
between the calculated and measured values. From
Fig. 7 it also follows that a rise in the temperature of
the troposphere by 10 K causes a quasilinear increase
in the model SpH values up to ~300 hPa. Above this
level the dependence is nonlinear. Since the radiation
balance of the Earth is determined by the amount of
water vapor at these heights, the problem of the dis
crepancy between the empirical and model SpH data
deserves special attention.
The measurements show (Randel et al., 2001) that
the atmosphere is very dry at the heights of ~100 hPa
in winter in the middle and high latitudes. Here, the
reflected longwave radiation should be highly sensi
tive even to insignificant fluctuations in the amount of
the water vapor (Spencer and Braswell, 1997;
Lindzen, 1990). Consequently, the discrepancy
between measured and modelled humidity at these
levels needs further attention.
The vertical distribution of saturated water vapor is
determined by the moist adiabatic lapse rate which,
=−
ws a ws
0.62198 ( ),
s
qppp
ws
p
a
p
(
)
=− − −
+−++
ws
log
( ) 9.09685 1 3.56654lg( )
0.87682(1 ) 0.78614 2.0
tt
t
pTT TT
TT
according to the data of the American Meteorological
Society (http://www.ametsoc.org/) is calculated by
the following formula:
(3)
where
Γ
w
is the moist adiabatic lapse rate K/m
–1
; g is
the Earth’s gravitational acceleration (9.8076 m/s
–2
);
H
v
is
the heat of vaporization of water (
2.501
×
10
6
J/kg)
–1
);
r
v
is the mixing ratio of the mass of water vapor to the
mass of dry air (kg kg
–1
);
R
sd
is the specific gas constant
of dry air (287 J kg
–1
K
–1
);
ε
= 0.622;
T
is the temper
ature of the saturated air, K; and
c
pd
is the specific heat
of dry air at constant pressure, J kg
–1
K
–1
.
Equation (3) describes the relationship between the
moist adiabatic lapse rate, temperature, and humidity
of air. For assessing the implications of the change in
temperature at the level of UTLS, we conducted some
numerical calculations (Kilifarska, 2012b). At the first
stage, we estimated the sensitivity of
Γ
w
to the varia
tions in temperature. For doing this, temperature pro
file from Terra Nova Bay (Antarctica) has been
increased and decreased by 100 K in the altitude inter
val 400–30 hPa, where are expected T variations, ini
tiated by the ozone changes (Fig. 8, right). At the first
stage, the calculations were conducted for a constant
mixing ratio of the mass of water vapor to the mass of
dry air. The calculations show that the cooling of the
UTLS region diminishes the moist adiabatic lapse
rate, whereas the heating boosts it.
At the second stage, we estimated the sensitivity of
the specific humidity SpH to the variations in temper
ature and, correspondingly, moist adiabatic lapse rate
Γ
w
. The calculated profiles are also shown in Fig. 8.
Here we can see the inverse dependence between SpH
and
Γ
w
: the rise in temperature at the level of UTLS
reduces the amount of water vapor (due to the
increased
Γ
w
), which impedes its upward propagation.
In the case of cooling at the level of UTLS, the specific
humidity grows due to the decrease in
Γ
w
and thus
facilitates the upward propagation of the
H
2
O
vapor.
This result is inconsistent with the climatic models
which predict SpH increasing throughout the entire
troposphere as a result of the rise of the Earth’s surface
temperature.
At the third stage, we assessed the dependence of
Γ
w
on the variations in specific humidity SpH. We cal
culated the height profile of the moist adiabatic lapse
rate assuming a 20% increase/decrease in SpH. In
Fig. 9 the profile of the initial (standard)
Γ
w
is shown
against the profiles of
Γ
w
corresponding to the mea
sured temperature profile increased/decreased by
15 K and against the profile of SpH
decreased/increased by 20%. It is clearly seen that the
drop in temperature or growth in SpH causes depres
sion of
Γ
w
(Fig. 9b), whereas the rise in temperature
+
Γ=
ε
+
sd
w
pd
sd
2
2
1
,
Hr
RT
g
Hr
c
RT
vv
vv
increases it
782
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
KILIFARSKA et al.
and decrease in humidity boost
Γ
w
(Fig. 9a). The
allowance for both factors enhances the resulting
changes in
Γ
w
. For example, if we take into account
that the decrease in temperature humidifies the UTSL
and calculate
Γ
w
for
T
– 15 K and SpH + 20%, the
resulting
Γ
w
along the profile will doubled correspond
ing to the changes in temperature alone.
This process renders continuous interaction
between temperature, moist adiabatic lapse rate, and
specific humidity in the region of UTLS. The decrease
in temperature in this region diminishes
Γ
w
, which
enables a greater amount of water vapor propagating
upwards. The increase in specific humidity addition
ally depresses
Γ
w
and, correspondingly, enhances the
effect caused by the cooling of the tropopause.
The described calculations support the hypothesis
that the ozone variability at the level of UTLS, which
causes long cooling (or heating) of the lower strato
spheric layers (Fig. 6), can affect the fluctuations of
humidity at the heights providing the largest contribu
tion to the Earth’s radiation balance, i.e., close to the
tropopause (Figs. 7–9).
6.5. Water Vapor and Greenhouse Effect
It is believed that water vapor in the free atmo
sphere provides the predominant contribution in the
greenhouse effect of the Earth (
IPCC
, 2007; Schmidt
et al., 2010). The postulated increase in the water
vapor concentration (resulting from global warming)
is considered in the climate models as a feedback
effect. In other words, vapor is not a cause of the
greenhouse effect but the response of the climatic sys
tem to the warming resulting from the increase in the
anthropogenic greenhouse gases
СО
2
, СН
4
, N
2
O
, etc.
However, water vapor in the lower troposphere (where
the rise in the Earth’s temperature most strongly
impacts the specific humidity) contributes little to the
radiation balance of the Earth (IPCC, 2007). In fact,
the greenhouse effect is primarily determined by the
humidity close to the tropopause (Spencer and
Braswell, 1997; Inamdar, 2004). Therefore, the
hypothesis that the entire troposphere becomes
warmer as a result of the anthropogenic warming is
neither substantiated theoretically nor is it validated by
the empirical evidence. Moreover, the measurements
testify to a cooling rather than warming which
occurred at the UTLS level in 1979–2007 (e.g., (Ran
del et al., 2009)).
In our conceptual framework (Kilifarska, 2012b),
the contradictions between the basic models and
empirical data are quite solvable since the amount of
water vapor at the UTLS level is determined by the
interaction between the ozone, temperature, and
humidity. This solves the problem of moistening of the
UTLS region under the decrease in its temperature.
7. THE CAUSAL RELATIONS
IN THE VARIATIONS OF SOME
ATMOSPHERIC PARAMETERS
For verifying the suggested mechanism of the GCR
influence (through the modulation by the geomag
netic field) on the lower stratospheric ozone, temper
ature, and humidity and the hypothesized correlation
of these parameters with the surface air temperature,
let us consider the data of the ERA40 and ERA
Interim reanalyses and compare them to the TOMS
data for the total ozone (TOC) for the past 30 years.
These time series cover more than 50 years, and we
have processed them treated them with the coordinate
bindings, in particular in the framework of the above
proposed method (applied on 10°
×
10° grid and lati
tudinal band 40°–70° N). We have shown previously
that the ozone changes in the lower stratosphere and,
correspondingly, the changes in TOC, are statistically
significantly correlated to the geomagnetic field and
GCR intensity. Below we present additional evidence
in favor of the geomagnetic field’s influence on ozone
and climate.
In Section 4 (see the table), the highest nonlinear
correlation coefficients have been found between the
surface air temperature and total ozone. In Fig. 10a,
the spatial distribution of the 30yr mean TOC based
on the TOMS data are compared to the distribution of
the surface air temperature according to the ERA
reanalysis data for winter. Both the maxima in the total
ozone (in North America and East Asia) fairly well
coincide with the minima in the winter surface air
temperature (see also Fig. 2). At the same time, the
tongue of the warm air above the Atlantic maps onto
the minimal ozone values. We recall that the maximal
contribution to the TOC is provided by the lower
stratospheric ozone (Wirth, 1993).
The spatial distribution of the statistically signifi
cant (at 95%) coefficients of correlation between the
two fields (TOC and surface air temperature
T2m
) are
shown in Fig. 10b. The response of
T2m
to the varia
tions in
O
3
is displayed on two time scales, fast (with a
time lag less than 5 years) and by 5 or more years. The
fast response of
T2m
to the spatial distribution of the
ozone maxima has two clearly pronounced zones of
increased temperatures (above the entire North Amer
ica and East Asia, Fig. 10b, top). However, the delayed
response is definitely negative in the polar regions
(Fig. 10b, bottom).
Our numerical experiments, described in
Section 5, show that the ozone layer’s depletion and
the corresponding cooling of the lower stratosphere
should be accompanied by an increase in humidity.
The analysis of the time series of the zonalmean win
ter ozone at 50 hPa and specific humidity at 150 hPa
for the last 50 years (smoothed by running average over
5 points) shows their antiphase variations (Fig. 11a).
Besides the synchronism in the changes of the win
ter means of
O
3
and SpH, we also note the estimated
rises
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
GEOMAGNETIC FIELD AND CLIMATE 783
mean period of ozonehumidity variations is ~18–
20 years. A similar period is also identified in the
changes of the surface air temperature (e.g., (Miya
hara, 2008)).
The longwave radiation reflected from the Earth is
extremely sensitive to weak fluctuations in humidity in
the dry tropical upper troposphere (Spenser and
Braswell, 1997). The changes at the middle and high
latitudes of the Northern Hemisphere indicate that
the heights of 100–150 hPa are driest during the winter
season (Randel et al., 2001). The absolute value and
variability of the total ozone in winter is much higher
at polar region than in the subtropical and tropical lat
itudes (Fig. 10a). Therefore, it can be expected that at
the middle and high latitudes, even small fluctuations
in the specific humidity initiated by the ozone’s vari
ability at the UTLS level can cause nonlinear effects in
the Earth’s radiation balance and correspondingly in
surface temperature.
In order to check for such a possibility we have cal
culated the coefficients of the cross correlation
between SpH at a level of 150 hPa (the driest level in
winter at middle and high latitudes; Randel et al.,
2001) and surface air temperature
T2m
at each grid
node for the winter months of the interval 1957–2011.
Then we selected the maximal statistically significant
(at a level of 95%) coefficients of correlation and plot
ted them on the map. The results are presented by the
contours in Fig. 12. For comparison in Fig. 12a is
shown the 50year climatology of the surface dynamic
anomalies in temperature. It is easily seen that the
ozonehumidity correlation resembles quite well the
positive surface temperature anomalies surrounding
the northern geomagnetic pole. The time delay of the
temperature response to the changes in the specific
humidity varies between 10 and 24 years. This is in
fairly close agreement with the results of Hallegatte
et al. (2006), who showed that the climate responds
positively to increases in humidity (with increasing
T
)
only on the long time scales. In agreement with their
result, our statistical analysis shows that the response
of the surface temperature to the increased humidity at
150 hPa on the time scales of 9–10 years is indeed neg
ative all over the Northern Hemisphere.
The results shown in Fig. 12 indicate that the sug
gested ozone–water vapor mechanism and its influ
ence on the surface air temperature is efficient, firstly,
in the regions with the high density and strong vari
ability of the lower stratospheric O
3
(the middle and
high latitudes in winter) and, secondly, on the long
time scales (>10 years).
8. CONCLUSIONS
On time scales ranging from decades to hundreds
of years there is a correlation between the changes in
the parameters of the geomagnetic field and climate.
Since the time series of the instrumental observations
of the magnetic and climatic variables are rarely longer
than 100 years, the correlations between them on the
time scales of the past few decades are believed to be
most reliable although unexpected, and in some cases
rather debatable. Our studies of the main magnetic
field and climatic parameters (surface air temperature
and surface pressure) enabled us to track their global
variations during XX century and identify the spa
tiotemporal relationship between them. The analysis
of the time series of ozone and specific humidity in the
upper troposphere–lower stratosphere for the second
half of the past century has demonstrated a fairly good
correspondence between the maxima and minima of
their spatial distribution. The latter coincide quite well
with the extremes of the total intensity of the geomag
netic field vector, temperature and pressure in the
Northern Hemisphere.
We suggest the physical mechanism which explains
how the magnetic field of the Earth can affect the spa
tial distribution and temporal variations of the surface
air temperature. The process starts with geomagnetic
modulation of the intensity and the depth of penetra
tion of energetic particles into the Earth’s atmosphere,
which initiates the ion–molecular reactions affecting
the ozone concentration close to the tropopause. This
height level in the Northern Hemisphere is marked
with the maximal absorption of GCRs, where suitable
conditions favoring autocatalytic production of O
3
in
the Northern Hemisphere, have been found. The vari
ations in the ozone density close to the tropopause
affect the temperature in the UTLS region due to the
high absorption capability of O
3
. The higher the tem
perature in this region the drier this layer becomes
(due to the reduced static stability of the upper tropo
sphere). Vice versa, the cooling of the UTLS region
facilitates the upward propagation of water vapor.
These small fluctuations of humidity in the UTLS
region in winter (north of 40° N) affect the radiation
balance (through the greenhouse effect) and, as a con
sequence, the surface air temperature.
Certainly, this mechanism needs further testing of
its applicability and does not claim to account for all
the possible causal relations and regional features. We
note that the obtained results are valid for winter, when
the upper troposphere is much drier. Nevertheless, the
relationship between the climate and geomagnetic
field is, in our opinion, quite feasible, and its changes
should also be factored into the longterm climatic
models as one of the climate controls.
ACKNOWLEDGMENTS
The work was conducted under the BlackSeaHaz
Net FP7 (PIRSESGA2009246874) project.
REFERENCES
Bakhmutov, V.G., Martazinova, V.F., Ivanova, E.K., and
Mel’nik, G.V., Changes in the main magnetic field and cli
temperature
dynamical anomalies
784
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
KILIFARSKA et al.
mate in the 20th century,
Dopovidi Natsional’noï Akademiï
nauk Ukraïni, Nauki pro Zemlyu
, 2011, no. 7, pp. 90–94.
Bakhmutov, V.G., Martazinova, V.F., Kilifarska, N.A.,
Mel’nik, G.V., and Ivanova, E.K., Linkage between the
changes in climate and geomagnetic field: 1. Spatiotempo
ral structure of the magnetic field of the Earth and climate
in the 20th century,
Geofiz. Zh.
, 2014, no. 1, pp. 81–104.
Banks, P.M. and Kockarts, G.,
Aeronomy
, New York: Aca
demic, 1973.
Bard, E. and Delaygue, G., Comment on “Are there con
nections between the Earth’s magnetic field and climate?”
by Courtillot V., Gallet Y., Le Mouël J.L., Fluteau F., Gen
evey A.,
EPSL
253, 328, 2007,
Earth Planet. Sci. Lett.
, 2008,
vol. 265, nos. 1–2, pp. 302–307.
Bazilevskaya, G.A., Usoskin, I.G., Flückige, E.O., Harri
son, R.G., Desorgher, L., Bütikofer, R., Krainev, M.B.,
Makhmutov, V.S., Stozhkov, V.I., Svirzhevskaya, A.K.,
Svirzhevsky, N.S., and Kovaltsov, G.A., Cosmic ray
induced ion production in the atmosphere,
Space Sci. Rev.
,
2008, vol. 137, pp. 149–173.
Brasseur, G. and Solomon, S.,
Aeronomy of the Middle
Stratosphere: Chemistry and Physics of the Stratosphere and
Mesosphere
, 3rd ed., Dordrecht: Springer, 2005.
Bronnimman, S., Sticher, A., Griesser, T., Fischer, A.M.,
Grant, A., Ewen, T., Zhou, T., Schraner, M., Rozanov, E.,
and Peter, T., Variability of largescale atmospheric circulation
indices for the Northern hemisphere during the past 100 years,
Meteorol. Zeitschr.
, 2009, vol. 18, no. 4, pp. 379–396.
Courtillot, V., Gallet, Y., Le Mouel, J.L., Fluteau, F., and
Genevey, A., Are there connections between Earth’s mag
netic field and climate?,
Earth Planet. Sci. Lett.
, 2007,
vol. 253, pp. 328–339.
Courtillot, V., Gallet, Y., Le Mouel, J.L., Fluteau, F., and
Genevey, A., Response to comment on “Are there connec
tions between Earth’s magnetic field and climate?, Earth
Planet. Sci. Lett., 253, 328–339, 2007” by Bard, E., and
Delaygue, M., Earth Planet. Sci. Lett., in press, 2007,
Earth
Planet. Sci. Lett.
, 2008, vol. 265, pp. 308–311.
de Foster, P.M. and Shine, K., Radiative forcing and tem
perature trends from stratospheric ozone changes,
J. Geo
phys. Res.
, 1997, vol. 102, no. D9, pp. 10841–10855.
de Forster, P.M. and Tourpali, K., Effect of tropopause
height changes on the calculation of ozone trends and their
radiative forcing,
J. Geophys. Res.
, 2001, vol. 106, no. D11,
pp. 12241–12251.
Forbush, S.E., Time variations of cosmic rays, in
Cosmic
Rays, the Sun and Geomagnetism: The Works of Scott E. For
bush
, AGU Special Publication Series, vol. 37, Van Allen, J.A.,
Ed., Washington: American Geophysical Union, 1993,
pp. 323–411.
Gauss, M., Myhre, G., Isaksen, I.S.A., Grewe, V., Pitari, G.,
Wild, O., Collins, W.J., Dentener, F.J., Ellingsen, K.,
Gohar, L.K., Hauglustaine, D.A., Iachetti, D., Lamarque, F.,
Mancini, E., Mickley, L.J., et al., Radiative forcing since
preindustrial times due to ozone change in the troposphere
and the lower stratosphere,
Atmos. Chem. Phys.
, 2006, no. 6,
pp. 575–599.
Glassmeier, K.H., Neuhaus, A., and Vogt, J.,
Space Clima
tology, invited presentation at the Alpach Summer School
,
2002.
Hallegatte, S., Lahellec, A., and Grandpeix, J.Y., An elici
tation of the dynamic nature of water vapor feedback in cli
mate change using a 1D model,
J. Atmos. Sci.
, 2006, vol. 63,
pp. 1878–1894.
Inamdar, A.K., Ramanathan, V., and Loeb, N.G., Satellite
observations of the water vapor greenhouse effect and col
umn longwave cooling rates: relative roles of the continuum
and vibrationrotation to pure rotation bands,
J. Geophys.
Res.
, 2004, vol. 109, D06104. doi 10.1029/2003JD003980
IPCC (Intergovernmental Panel on Climate Change).
Cli
mate Change 2007: The Physical Science Basis
, Solomon, S.
et al., Eds., Cambridge: Cambridge Univ. Press, 2007.
Jackman, C.H., Frederick, J.E., and Stolarski, R.S., Pro
duction of odd nitrogen in the stratosphere and mesos
phere: an intercomparison of source strengths,
J. Geophys.
Res.
, 1980, vol. 85, no. C12, pp. 7495–7505.
Jonson, J.E., Sudnet, J.K., and Tarrason, L., Model calcu
lations of present and future levels of ozone and ozone pre
cursors with a global and regional model,
Atmos. Environ.
,
2001, vol. 35, pp. 525–537.
Kilifarska, N.A., Climate sensitivity to the lower strato
spheric ozone variations,
J. Atmos. Sol.–Terr. Phys.
, 2012a,
vols. 90–91, pp. 9–14.
Kilifarska, N.A., Mechanism of lower stratospheric ozone
influence on climate,
Int. Rev. Phys.
, 2012b, vol. 6, no. 3,
pp. 279–289.
Kilifarska, N.A., Ozone as a mediator of galactic cosmic
rays’ influence on climate,
Sun Geosphere
, 2012c, vol. 7,
no. 1, pp. 97–102.
Kilifarska, N.A., An autocatalytic cycle for ozone produc
tion in the lower stratosphere initiated by galactic cosmic
rays,
Comptes rendus de l’Acad’emie bulgare des Sciences
,
2013, vol. 66, no. 2, pp. 243–252.
Kirkby, J., Cosmic rays and climate,
Surv. Geophys.
, 2007,
vol. 28, pp. 333–375.
Kovaltsov, G.A. and Usoskin, I.G., Regional cosmic ray
induced ionization and geomagnetic field changes,
Adv.
Geosci.
, 2007, vol. 13, pp. 31–35.
Krivolutsky, A.A. and Repnev, A.I., Impact of space ener
getic particles on the Earth’s atmosphere (a review),
Geo
magn. Aeron.
, 2012, vol. 52, no. 6, pp. 685–716.
Kuznetsova, N.D. and Kuznetsov, V.V., Implications of
cosmic radiation and secular geomagnetic variations for the
evolution of life,
Vestn. Sev.Vost. Nauch. Tsentra DVO RAN
,
2012, no. 2, pp. 11–18.
Lantos, P., Predictions of galactic cosmic ray intensity
deduced from that of sunspot number,
Sol. Phys.
, 2005,
vol. 229, pp. 373–385.
Larin, I.K. and Tal’roze, V.L., The conditions and probable
intensity of the impact of the charged particles on ozone
depletion in the stratosphere,
Dokl. Akad. Nauk SSSR
,
1977, no. 3, pp. 410–413.
Lindzen, R.S., Some coolness concerning global warming,
Bull. Am. Meteorol. Soc
, 1990, vol. 7, pp. 277–288.
Loginov, V.F.,
Global’nye i regional’nye izmeneniya klimata:
prichiny i sledstviya
(Global and Regional Climate Changes:
Causes and Consequences), Minsk: TetraSistems, 2008.
Markov, M.N. and Mustel’, E.P., Spatiotemporal effects of
solarterrestrial linkage in the troposphere and thermo
sphere,
Astron. Zh.
, 1983, vol. 60, pp. 417–421.
Martazinova, V.F. and Ivanova, E.K., Characteristic fea
tures of the synoptic processes of different probability at the
end of the 20th and beginning of the 21st centuries, in
Glo
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 51
No. 5
2015
GEOMAGNETIC FIELD AND CLIMATE 785
bal’nye i regional’nye izmeneniya klimata
(Global and
Regional Changes in Climate), Shestopalov, V.M.,
Loginov, V.F., Osadchii, V.I., et al., Eds., Kyiv: NikaTsentr,
2011, pp. 86–95.
McCracken, K.G. and Beer, J., Longterm changes in the
cosmic ray intensity at Earth, 14282005,
J. Geophys. Res.
,
2007, vol. 112, A10101. doi 10.1029/2006JA012117
Mende, W. and Stellmacher, R., Solar variability and the
search for corresponding climate signals,
Space Sci. Rev.
,
2000, vol. 94, pp. 295–306.
Miks vsky
, J. and Raidl, A., Testing for nonlinearity in
European climatic time series by the method of surrogate
data,
Theor. Appl. Climatol.
, 2006, vol. 83, pp. 21–33.
Miyahara, H., Yokoyama, Y., and Masuda, K., Possible link
between multidecadal climate cycles and periodic reversals
of solar magnetic field polarity,
Earth Planet. Sci. Lett.
,
2008, vol. 272, pp. 290–295.
Mote, P.W., Rosenlof, K.H., Holton, J.R., Harwood, R.S.,
and Waters, J.W., An atmospheric type recorder: the
imprint of tropopause temperatures on stratospheric water
vapour,
J. Geophys. Res.
, 1996, vol. 101, pp. 3989–4006.
Mursula, K., Usoskin, I.G., and Kovaltsov, G.A., Recon
structing the longterm cosmic ray intensity: linear relations
do not work,
Ann. Geophys.
, 2003, vol. 21, pp. 863–867.
Petrova, G.N., Nechaeva, T.B., and Pospelova, G.A.,
Khar
akter izmeneniya geomagnitnogo polya v proshlom
(The
Character of Changes in the Geomagnetic Field in the
Past), Moscow: Nauka, 1992.
Pinto, O., Jr., Gonzalez, W.D., Pinto, I.R.C., Gonza
lez, I.L.C., and Mendes, O., Jr., The South Atlantic Mag
netic Anomaly: three decades of research,
J. Atmos. Terr.
Phys.
, 1992, vol. 54, pp. 1129–1134.
Rakobol’skaya, I.V.,
Yadernaya fizika
(Nuclear Physics),
Moscow: MGU, 1971.
Ramanatan, V., Callis, L.B., and Boucher, R.E., Sensitivity
of surface temperature and atmospheric temperature to
perturbations in the stratospheric ozone and nitrogen diox
ide,
J. Atmos. Sci.,
1976, vol. 33, pp. 1092–1112.
Randel, W.J., Wu, F., Gettelman, A., Russel, III J.M.,
Zavodny, J.M., and Oltmans, S.J., The seasonal variation of
water vapour in the lower stratosphere observed in Halogen
Occultation Experiment data,
J. Geophys. Res.
, 2001,
vol. 106, pp. 14313–14325.
Randel, W.J., Shine, K.P., Austin, J., Barnett, J., Claud, C.,
Gillett, N.P., Keckhut, P., Langematz, U., Lin, R., Long, C.,
Mears, C., Miller, A., Nash, J., Seidel, D.J., Thomp
son, D.W.J., et al., An update of observed stratospheric tem
perature trends,
J. Geophys. Res.
, 2009, vol. 114, D02107.
doi 10.1029/2008JD010421
Rozelot, J.P., Pireaux, S., Lefebvre, S., and
Ajabshirizadeh, A., The Sun asphericities: astrophysical
relevance. arXiv:astroph/0403382 v3 (April 1, 2004).
Schmidt, G.A., Ruedy, R.A., Miller, R.L., and Lacis, A.A.,
Attribution of the presentday total greenhouse effect,
J. Geophys. Res.
, 2010, vol. 115, D20106.
doi 10.1029/2010JD014287
Seidel, D.J. and Randel, W.J., Variability and trends in
the global tropopause estimated from radiosonde data,
J. Geophys. Res.
, 2006, vol. 111, D21101.
doi 10.1029/2006JD007363
Shea, M.A. and Smart, D.F., Preliminary study of cosmic
rays, geomagnetic field changes and possible climate
changes,
Adv. Space Res.
, 2004, vol. 34, pp. 420–425.
Spencer, R.W. and Braswell, W.D., How dry is the tropical
free troposphere? Implications for global warming theory,
Bull. Am. Meteorol. Soc.
, 1997, vol. 78, no. 6, pp. 1097–
1106.
Stuber, N., Sausen, R., and Ponater, M., Stratosphere
adjusted radiative forcing calculations in a comprehensive
climate model,
Theor. Appl. Climatol.
, 2001, vol. 68,
pp. 125–135.
Tomasi, C., Cacciari, A., Vitale, V., Lupi, A., Lan
conelli, C., Pellegrini, A., and Grigioni, P., Mean vertical
profiles of temperature and absolute humidity from a 12
year radiosounding data set at Terra Nova Bay (Antarctica),
Atmos. Res.
, 2004, vol. 71, pp. 139–169.
Usoskin, I.G., Schussler, M., Solanki, S.K., and Mursula, K.,
Solar activity, cosmic rays, and Earth’s temperature: a mil
lenniumscale comparison,
J. Geophys. Res.
, 2005, vol. 110,
A10102. doi 10.1029/2004JA010946
Van Allen J.A., Dynamics, composition and origin of the
geomagneticallytrapped corpuscular radiation, an Invited
Discourse at the 11th General Assembly of International
Astronomical Union,
Trans. Int. Astron. Union,a
1962,
XIB, pp. 99–136.
Velinov, P.I.Y., Mateev, L., and Kilifarska, N., 3D model
for cosmic ray planetary ionisation in the middle atmo
sphere,
Ann. Geophys.
, 2005, vol. 23, pp. 3043–3046.
Vinogradov, P.S., Larin, I.K., Poroikova, A.I., and
Tal’roze, V.L., To the mechanism of cosmic rays influence
on the ozonosphere of the Earth, in
Sovremennoe sostoyanie
issledovanii ozonosfery v SSSR. Tr. Vsesoyuznogo sovesh
chaniya po ozonu
(The State of the Art in the Studies of the
Ozonosphere in the USSR. Proc. AllRussia Conference on
Ozone), Moscow: Gidrometeoizdat, 1980, pp. 123–130.
Wang, WCh., Pinto, J.P., and Yunk, Y.L., Climatic effect
due to the halogenated compound in the Earth atmosphere,
Atmos. Sci.
, 1980, vol. 37, pp. 333–338.
Wang, WCh., Zhuang, YCh., and Bojkov, R., Climate
implications of observed changes in ozone vertical distribu
tions at middle and high latitudes of the Northern Hemi
sphere,
Geophys. Rev. Lett.
, 1993, vol. 20, no. 15, pp. 1567–
1570.
Wirth, V., Quasistationary planetary waves in total ozone
and their correlation with lower stratospheric temperature,
J. Geophys. Res.
, 1993, vol. 98, pp. 8873–8882.
WMO,
Scientific Assessment of Ozone Depletion
, World
Meteorological Organization, Global Ozone Research and
Monitoring Project, Report No. 50, Geneva, 2006.
Translated by M. Nazarenko
o
ˆ