Solar radiation

The sun as a source of energy

gy
The sun is the main source of energy for the climate system, exceeding the
next importat source (geothermal energy) by 4 orders of magnitude!
Sources of energy for the climate systems
solar
l radiation
di ti

1 74 1017 W
1.74

3.2
1.0
5.0
2.0
80
8.0

geothermal energy
anthropogenic energy generation
infrared emission by the full moon
solar radiation reflected by the full moon
radiation
di ti by
b the
th stars
t

1013 W
1013 W
1012 W
1012 W
108 W

The structure of the sun

radiation zone (150000 km)
convection zone (150000 km)
photosphere (~ 500 km)
chromosphere ((~ 2000 km)
Photo by NASA

corona (> 3000 km)

core

radius ro = 6.96 108 m

sunspots
prominence (>
( 3000 km)

The visible region of the sun is called photosphere. Most of the radiation
reaching the earth originates from this layer. Although the sun is a gaseous
body,
y, the photosphere
p
p
is usuallyy referred to as the surface of the sun. The
temperature of this layer varies between 4000 and 8000 K. A temperature of
5800 K is required to explain the observed spectral distribution of the solar
radiation (Planck curve).
curve)
2

The structure of the sun (2)

Sunspots are colder areas on the suns
photosphere.
photosphere

Spectrum of the solar radiation

Data from MODIS (http://edcdaac.usgs.gov/modis/dataproducts.asp)

The solar constant

Most of the radiation reaching the earth stems from the photosphere. The
effective temperature and effective radius of the photosphere are To = 5800 K
and ro = 6.96 108 m, respectively. Assuming that the photosphere emits as a
blackbody, the radiation flux at its outer limit can be computed according to
Stefan Boltzmannss law as:
Stefan-Boltzmann

Fo = To4
where

= 5.67 10-8 W m-2 K-4, Stefan-Boltzmann constant.

Since the propagating solar radiation has a spherical wavefront, at a distance

d from the suns center the flux of solar radiation has decreased to:

4ro2
ro
=
F(d ) = Fo
F

o
4d 2
d

The solar constant (2)

The solar constant So F(a) is defined as the flux of solar radiation at the top
of the atmosphere (TOA) at the mean distance a between the earth and the
sun. The earths orbit being an ellipse, the mean distance is equal to the major
semiaxis, a = 1.496 1011 m, giving So = 1367 W m-2.
Knowledge
l d off the
h astronomical
i l settings
i
allows
ll
to compute the
h solar
l constant
for all other planets, too. Values of a and So for the three terrestrial planets are
listed in the following table. Astronomical parameters for all of the planets are
provided in the table on the next page.

Planet

a [m]

So [W m-2]

Venus
Erde
Mars

1.082 1011
1.496 1011
2 279 1011
2.279

2637
1367
592

Key data for the planets

After: DMK/DPK,, 1984: Formeln und Tafeln. Orell Fssli Verlag,

g, Zrich.
7

Is the solar constant really a constant?

To answer this question let us have a look at measurements of the solar
constant carried out during the last 3 decades with the help of various
satellites. The compilation shown below is due to Quinn and Frhlich (1999).

Quinn and Frhlich (1999)

Is the solar constant really a constant? (2)

Lean and Rind (1998)

Relation between solar constant and solar activity

To explain the 11-years variability of the solar constant it is necessary to
consider the 11
11-years
years cycle of the solar activity.
activity Although one would expect
that the solar constant is lowest during periods of intense activity because the
number of sunspots is highest, the reverse is actually true.

Lean, J. und D. Rind, 1998: Climate forcing by changing solar radiation. J. Climate, 11,
3096-3094

10

Relation between solar constant and solar activity (2)

The reason is that the brightening caused by the enhanced number of faculae
more than compensate for the darkening caused by the enhanced number of
sunspots.

Lean, J. und D. Rind, 1998: Climate forcing by changing solar radiation. J. Climate, 11,
3096-3094

11

The solar constant, 1600 to present

The solar constant varies also on a
multi-decadal time scale. These
variations are of less than ~ 2
pparts in 1400,, that is of less than 2
.

Lean, J., 2000: Evolution of the

Sun's Spectral Irradiance Since the
Maunder Minimum. Geophysical
Res. Letters, 27, 2425-2428
12

Decadal to centennial variability of the solar output

and climate
Many researchers
M
h hhave postulated
l d a correlation
l i between
b
the
h decadal
d d l
variability of the solar output and various climatic parameters. A good survey
is provided in the books by Hoyt and Schatten (1993) and Burroughs (2003).
A physical explanation for the observed correlations has yet to be found, and
the foundation of such statistical analyses is to be considered tenuous (a
statistical correlation does not necessarily imply a physical linkage!).
linkage!)
There are, however, a few studies that merit consideration:
(1976 1977);
Eddy (1976,
Lean and Rind (1998);
g 1997))
Labitzke and van Loon ((e.g.
Hoyt, D. V., and K. H. Schatten, 1993: The Role of the Sun in Climate Change. Oxford
University
y Press,, 279 pp;
Burroughs, W.J., 2003: Weather Cycles: Real or Imaginary? Second Edition, Cambridge
University Press, 317 pp;
Labitzke, K. and H. van Loon, 1997: The signal of the 11-year sunspot cycle in the upper
troposphere-lower stratosphere. Space Science Reviews, 80, 393-410
13

Solar activity and winter severity

By looking at the historic record of winter severity (through analysis of proxy
data) and the number of sunspot,
sunspot Eddy (1976,
(1976 1977) found that during the so
socalled Little Ice Age the most severe winter were concomitant with periods
characterized by the absence of sunsports.

14

Solar activity and NH temperature

Moreover, Lean and Rind (1998) found that: The correlation of reconstructed
solar irradiance and Northern Hemisphere (NH) surface temperature
anomalies is 0.86 in the preindustrial period from 1610 to 1800, implying a
predominant solar influence. Extending this correlation to the present
suggests that solar forcing may have contributed about half of the observed
0.55 C surface warming since 1900 and one-third of the warming since 1970.
Newer estimates of long term trends in solar irradiance tend to be smaller.

15

Solar activity and NH temperature (2)

Note, however, that starting about 1800 other processes appear to be
significantly correlated with NH temperatures,
temperatures most prominently greenhouse
gases concentrations.
.

After Lean and Rind (1998)

Note, moreover, that since relative variations in the solar constant at the
decadal scale are of less than 2 (p
(p. 8)) a y
yet unknown amplification
p
of the
solar signal must be in place to physically explain the correlation.

16

Solar activity and the dynamics of the upper

troposphere and lower stratosphere
Labitzke
L
bit k andd van Loon
L
(1997) hhave considered
id d a possible
ibl solar
l signal
i l in
i the
th
height of constant-pressure levels in the upper troposphere and lower
stratosphere. For the 30-hPa level the correlation is as high as ~ 0.7 in the
subtropics.
.

Labitzke, K. and H. van Loon, 1997: The signal of the 11-year sunspot cycle in the upper
troposphere-lower stratosphere. Space Science Reviews, 80, 393-410

17

Orbital geometry

After Liou (2002)

18

Key elements of the orbital geometry

The three key elements in the orbital geometry are:
the eccentricity,
eccentricity e;
the obliquity of the earths rotation axis relativ to the ecliptic plane, ;
the longitude of the perihelion, .

After Liou (2002)

The secular variations of the orbital elements can be computed based on

celestial mechanics and are associated with the perturbations that other
planets exert on earths orbit. According to Milankovitch (1941) theory, these
variations are directly or indirectly responsible for the ice-ages.
19

Secular variations of the orbital elements

eccentricity
P of 100 and 413 kyr

obliquity
P of 41 kyr

precession, esin(),
P of 23 and 18 kyr

Data calculated according to

Berger A. and Loutre M.F., 1991: Insolation values for the climate of the last 10 million years.
Quaternary Sciences Review, Vol. 10 No. 4, pp. 297-317, 1991.
20

Solar zenith angle

Essentially the position of the sun with respect to an observer at the surface is
determined by the solar zenith angle o, that is the angle between the vertical
above the observer (the normal) and the sun. The solar zenith angle can be
expressed in terms of the solar declination (the angle subtended by the sun
with respect to the equatorial plane),
plane) the hour angle (< 0 before true solar
noon, = 0 at noon, > 0 after solar noon), and the geographic latitude of the
observer O.
N

o
O

21

Solar zenith angle (2)

The equation reads:
cos o = sin
i sin
i + cos cos cos

The hour angle is equal to zero at true solar noon, increasing by

0 26 radians or 15
0.26
15 pro hour.
hour Therefore = -/2
/2 at 6:00 and = /2 at
18:00.
Since o = /2 at sunrise and sunset (except at the Poles), the hour angles
- at sunrise and at sunset (atronomical values) can be found from:
cos = tan tan

provided that -1 cos +1. If cos > +1 we have polar night, and if
cos < -1 polar day with midnight sun.
According to Iqbal (1983) a useful formula to express solar declination as
a function of the running date (day of the year, D [1,365]) is:
2

D 1.3944
= 0.4093 sin
365

[radians]

Iqbal, M. 1983: An Introduction to Solar Radiation. Academic Press, Toronto, 390 pp

22

Solar path

Source:
http://www.oksolar.com/abctech/solar-radiation.htm

23

Solar insolation at TOA

The distribution of solar insolation at the top of the atmosphere is given by:
2

a
S = So cos o
r
where So is the solar constant,, a the mean distance between the sun and the
earth, and r the current distance at a particular day of the year D.
Following Iqbal (1983), the square of the ratio (a/r) can be calculated as:
a
2
=
1
+
0
.
033

cos
D

r
365
2

Note: more accurate formulas for the solar declination and the relative
distance to the sun can be found in Iqbal (1983) or Liou (2002).
The daily insolation is found by integrating the above equation between
sunrise and sunset:

a
SD = So sin
i sin
i + cos cos cos d
r

a
= 2 So [ sin
i sin
i + cos cos sin
i ]
r

24

Distribution of daily insolation at TOA

After Liou (2002).

25

Annual mean insolation at TOA

As for the latitudinal distribution of the annual mean insolation we have:

26

Zenith angle and air mass

When considering absorption and scattering of solar radiation it is necessary
to know the total mass of absorbing or scattering substance.
substance Recall the Beer
BeerBouguer-Lambert law:
s

N (s) = N 0 exp k ds
d N 0 exp( k u )
0

where us is the optical path.

path The distance covered by a beam in the
atmosphere depends on the solar zenith angle:

27

Zenith angle and air mass (2)

To account for the effect of the zenith angle, we make use of the so-called
relative optical air mass defined as (Paltridge and Platt,
Platt 1976):

dz

m =

ds

0
As seen in the above figure, to the extent that the atmosphere can be
considered as a non-refreacting, plane-parallel medium:
1
sec o
cos o
I practice,
In
ti due
d to
t the
th curvature
t
off the
th earths
th atmosphere
t
h andd the
th density
d it
dependence of the refractive index, this equation holds true only for o < 60o.
A more accurate formula is due to Kasten (1966). It reads:
m =

m = sin o + 0.15 ( o + 3.885)

1.253 1

where o = 90 o is the observed solar altitude (in degrees).

degrees)
28

Radiation at ground: direct and indirect irradiance

Atmospheric scatter ensures that the downcoming flux density has a direct as
well as a diffuse component.
component At any solar zenith angle o the total vertical
flux density, the so-called global radiation, is given by (Paltridge and Platt,
1976):
Gl F = Fdir cos o + N diff d

where Fdir
di is the irradiance of the direct beam on a surface perpendicular to
the beam abd Ndiff is the radiance of diffuse radiation and the integral is over
the all solid angles of the upper hemisphere.
In terms of the total optical depth [(R, + oz, + wv, + D,) m] relative to the
extinction of solar radiation by Rayleigh scattering, ozone and water vapor
p
and extinction by
y aerosols, the direct beam can be expressed
p
as:
absorption

Fdir =

exp[ ( R , + oz , + wv , + D , ) m ]d

where S is the spectral radiance at the top of the atmosphere.

29

Direct radiation
The formula for the direct radiation can be simplified by introducing the
transmissivity given (see class on Radiative
transmissivity,
Radiative Transfer
Transfer)) as
T exp( )

which
hi h allows
ll
one tto write:
it

Fdir =

S ( TR , Toz , Twv , TD , ) d
m

By defining the transmittance (transmission function) q as the average

transmissivity of the atmosphere, such that:

Fdir = q

th formula
the
f
l for
f the
th direct
di t radiation
di ti reduces
d
to:
t
2
a

Fdir = S q m = So q m cos o
r
Note that the global average of q for a cloudless atmosphere has been
estimated in ~ 0.7.
30

Distribution of selected fluxes of solar radiation

All of the following picture refer to annual mean fluxes and were taken from
Raschke and Ohmura (2005).
(2005) First we show a map of the net solar radiation at
the top of the atmosphere.

Raschke, E. and A. Ohmura, 2005: Radiation budget of the climate system. In Hantel
(ed.), Landolt-Brnstein, Group V, Vol. 6, Observed Global Climate, Springer, Berlin.

31

Distribution of the selected fluxes of solar radiation (2)

Map of the global radiation at the earths surface.

Raschke, E. and A. Ohmura, 2005: Radiation budget of the climate system. In Hantel
(ed.), Landolt-Brnstein, Group V, Vol. 6, Observed Global Climate, Springer, Berlin.

32

Distribution of the selected fluxes of solar radiation (3)

Map of the mean transmittance.

Raschke, E. and A. Ohmura, 2005: Radiation budget of the climate system. In Hantel
(ed.), Landolt-Brnstein, Group V, Vol. 6, Observed Global Climate, Springer, Berlin.

33

Global radiation and atmospheric transmittance

Contrary to what expressed by Budyko in 1982, many studies carried out
during the last 15 years have demonstrated that global radiation observed at
the earths surface is far from being stable on a decadal scale. From 1950 to
about 1990 many stations around the world displayed a decrease in solar
radiation (the so-called
so called global
global dimming);
dimming ); in many places this has been
followed by a inverse tendency (global brightening) (Wild et al., 2005).
Part of these changes can be attributed to changes in cloudiness. However,
part of the variations are related to changes in the transmittance of the
cloudless atmosphere, which ultimately are related to changes in the aerosol
load.

Wild,, M.,, H. Gilgen,

g , A. Roesch,, A. Ohmura,, C.N. Long,
g, E.G. Dutton,, B. Forgan,
g , A.
Kallis, V. Russak, A. Tsekov, 2005: From Dimming to Brightening: Decadal Changes in
Solar Radiation at Earths Surface. Science, 308, 847-850.

34

Global radiation and atmospheric transmittance (2)

El Chichon

Pinatubo
35

Radiation in the atmosphere*

Definitions
quantity
radiation
radiant
di t flux
fl
radiant energy
irradiance
radiance

definition
electromagnetic energy emitted
transferred or received
( radiant
(or
di t power)) rate
t off transfer
t
f
of radiant energy
quantity of energy
radiant flux per unit area
incident upon a surface
radiant flux pper unit solid angle
g
at a point in a surface per unit
projected area of the surface

symbol

unit

P = dU/dt
U
J

F = dP/dA

W m-2

N = dF/cosd W m-2 sr-1

* Chandrasekhar,, S.,, 1950: Radiative Transfer. Dover,, New York;; Liou,, K.N.,, 2002: An Introduction to
Atmospheric Radiation (2nd Ed.). Academic Press, Amsterdam; Paltridge, G.W. and C.M.R. Platt, 1976:
Radiative Processes in Meteorology and Climatology, Elsevier, Amsterdam.

36

Electromagnetic waves
The electromagnetic spectrum
X rays
gamma rays

10-6

microwaves

ultraviolet

10-4

infrared

10-2
0.38

1
0.74

visible

102

TV and radio waves

104

106

108

violet
purple
blue
green
yellow
orange
red

Frequency and wave length

c
=

where
= wave length [m]
= frequency [s-1]
c = 2.998 108 m s-1, speed of light in vacuum

37

Geometrical considerations

38

Radiance and irradiance

Radiation falling on a surface at an angle from the normal to the surface
gives at the surface an irradiance confined to the solid angle d equal to
N cos d. The total irradiance at the surface is therefore the integral over the
half-sphere of which the surface is the diametral plane:

N cos d

In terms of polar co-ordinates and :

/2

d N cos sin d

For isotropic radiation, that is when N is independent of direction, the integral

reduces to:
2

F N

/ 2

d cos sin d
0

= N

39

Blackbody radiation: Plancks

Planck s law
In 1901 Planck was able to derive an analytical expression for the
(unpolarized) radiant energy emitted by an enclosure in thermal equilibrium
at an absolute temperature T (a blackbody) per unit volume per unit
wavelength interval, the so-called Planck function:
hc
2 5
1
B (T ) 2 h c exp
kT
where
c
h
k

=
=
=
=

[ W m 2 m 1 sr 1 ]

2.998 108 m s-1, speed

p
of light
g in vacuum
6.626 10-34 J s, Planck constant
1.381 10-23 J K-1, Boltzmann constant
absolute
b l temperature [K]

Note that B has units of a spectral radiance. Spectral quantities are related to
their
h i totall counterparts through:
h
h
N

dN
d
40

Spectrum of blackbody radiation

The Planck function implies that the spectrum of blackbody radiation has a
strong dependence on absolute temperature
temperature, as seen by contrasting the spectra
for the solar (5780 K ~ temperature of the photosphere) and terrestrial
(255 K) emission.

41

Wienss and Stefan

Wien
Stefan-Boltzmanns
Boltzmann s laws
Given a spectrum of blackbody radiation, the wavelength max were the
maximum emission takes place is found by setting dB/d = 0:

max T = 2.898 10 3 K m
This is Wiens displacement law.
The total radiance is found by integrating B over all wavelengths:

B(T ) =

T4
B d =

where
= 5.67 10-8 W m-2 K-4, Stefan-Boltzmann constant.
Si
Since
the
h black-body
bl k b d emission
i i is
i isotropic,
i
i the
h corresponding
di irradiance
i di
is:
i
T4
F(T) =

/ 2

cos sin d = T 4

This is Stefan-Boltzmanns law.

42

Kirchhoffss law
Kirchhoff
Contrary to a blackbody, which absorbs all of the incident radiation, a socalled grey body reflects part of the incident radiation (with a reflectivity R).
)
Hence, in thermal equilibrium with a blackbody a grey body can only emit a
radiant energy e B since (see figure) B = e + R.
The emissivity of a grey body is defined as:
e
1 R A 1
B
where
R = reflectivity
y
A = absorptivity
Hence for a grey body = A. This is Kirchhoffs law.
An average emission coefficient can be defined by requiring that the total
radiant energy emitted by a gray body follows Stefan-Boltzmanns law as:

F = e d = B d = ~ B d T 4
43

Emission coefficient
Most natural surfaces have an emission coefficient ~ 1; for vegetation, for
instance, 0.9 (see Oke, 1987 or Garratt, 1992). Snow, too, behaves in the
longwave range almost like a blackbody, with = 0.99 for fresh snow. Metals,
on the other hand, are poor emitters: 0.03 for aluminium, 0.20 for iron,
= 0.02
0 02 for silver
silver.
The effective emissivity of the atmosphere, a, depends on the concentrations
of the greenhouse gases and the presence of clouds.
Note that the total content u of a constituent is usually expressed in terms of
the total mass per unit cross-sectional area (in units of [kg m-2] but more
typically [g cm-2]),
]) which is the path integral of the density:

dz

44

Absorption spectra
Note, however, that molecular compounds such as H2O and CO2 do not emit
or absorb continuously over the whole spectral range.
range The implications for
the atmospheric absorption (emission) are shown in the following figure.

Absorption spectra for various atmospheric gases between the top of the
atmosphere and the surface.
surface
Peixoto, J.P. and A.H. Oort, 1992: Physics of Climate. American Institute of Physics, New York, 520 pp.

45

The atmospheric window

This differential absorption/emission gives rise to the so-called atmospheric
window.
window

46
Courtesy of Rolf Philippona, MeteoSwiss, Payerne.

Short- and longwave radiation

Short
Solar radiation mainly stems from the suns photosphere, which has an
effective temperature of 5800 K,
K more than an order of magnitude larger than
the mean temperature at the earths surface, 288 K. This contrast gives rise to
a considerable difference in the Planck function.
Solar radiation received at the Earths surface has been weakened by its
interaction with the atmosphere (see Solar Radiation). The spectra of the
incoming
g solar radiation (direct
(
beam)) and of the radiant energy
gy emitted byy
the atmosphere are nevertheless distinct (see figure on the previous page),
with only little overlap at approximately 4 m. This limit is used to
discriminate between short- and longwave radiation
radiation.
In summary:
< 4 m shortwave or solar radiation
> 4 m longwave or terrestrial (incl. atmosphere) radiation.

47

Radiative transfer
Radiation traversing a medium will be weakened by its interaction with
matter This interaction is called extinction or attenuation,
matter.
attenuation an overall
designation for the processes of absorption and scattering.
We assume that the medium has a density and is characterized by a mass
extinction coefficient of k [m2 kg-1].
ds
N + dN

, k
According to the above figure and to first order:
dN = k N ds

48

Beer-Bouguer-Lambert
Beer
Bouguer Lambert law
If scattering and emission can be neglected:
dN
= N
k ds

With N(s = 0) = N0, the equation can be integrated to yield:

s

N (s) = N 0 exp k ds
0

If k is independent of s, then
s

N (s) = N 0 exp k ds N 0 exp( k u )

0

where the optical path u has been defined as u =

This is Beer
Beerss or Beer
Beer-Bouguer-Lambert
Bouguer Lambert law.
law

ds .
0

49

Optical depth, transmissivity, absorptivity and reflectivity

If k depends on s, then it is more convenient to define the so-called optical
depth and the transmissivity (spectral transmittance) T as:
s

(0, s)

k ds
0

and

T exp( )

N
N0

For a non-scattering medium, the fraction of radiation absorbed by the

medium is:
A 1 T

where A is the absorptivity. Note that all of the above are monochromatic or
spectral quantities.
If scattering takes place,
place a certain portion of the incident radiation can be
reflected back into the incident direction. The ratio of the reflected
(backscattered) to the incident intensity is called monochromatic reflectivity,
R. In this case:
T + A + R = 1
50

Clouds radiative properties

Consider for instance the properties of water clouds as discussed by
Stephens (1978).

After: Stephens, G.L., 1978:

Radiation profiles in extended
water clouds. J. Atmos.
Sciences, 35, 2123-2132 .

51

Scattering
Scattering is the physical process by which a particle (or molecule) in the path
of an electromagnetic wave continuously extracts energy from the incident
wave and re-radiates that energy in all directions. In the atmosphere, the
particles responsible for scattering range from gas molecules (~ 10-4 m) to
large raindrops and hail particles (~ 104 m).
m) We can broadly distinguish the
following categories:
solid aerosols (0.1 to 1 m), irregular shape, variable refractive index;
haze water drops (0.1 to 1 m), spherical, known refractive index;
cloud water drops (1 to 10 m), spherical, known refractive index;
particles ((1 to 100
m),
), irregular
g
shape,
p , known refractive index.
cloud ice p
Based on the size of the scattering particles, we distinguish between
Rayleigh scattering,
scattering particle diameter << wavelength of the incident beam
Mie scattering*, particles diameter ~ wavelength of the incident beam

* Mie theory provide a framework for describing scattering caused by spherical particles.

52

Directional dependence of scattering

Rayleigh scattering is characterized by symmetry between forward and
backward scattering. As the particles become larger, an increasing proportion
of the incident radiation is scattered in the forward direction (Mie scattering).
Sketches of the angular pattern of the scattered intensity from particles of
various sizes are shown here.

After: Liou, K.N., 2002: An Introduction to Atmospheric Radiation (2nd Ed.). Academic
Press, Amsterdam

53

Wavelength dependence of scattering

Rayleigh scattering decreases with increasing wavelength of the incident
beam according to:
N sc
4
N0

where N0 and Nsc are the total intensities of the incident and scattered
radiation. This relation explains why the sky appear blue under
cloudless conditions.
Rayleigh scattering is also responsible for the albedo of the clear-sky
atmosphere It can be shown that the planetary albedo of a purely
atmosphere.
Rayleigh atmosphere is ~ 0.2 (assuming a surface albedo of 0.16),
which is somewhat higher than the observed clear-sky planetary albedo
of 0.17.
0 17
The dependence of Mie scatter on the wavelength of the incident beam
is more complex.
complex As a rule of thumb:
N sc
~ 1.3
N0

54