TOTAL SOLAR IRRADIANCE AND CLIMATE

1
TOTAL SOLAR IRRADIANCE AND CLIMATE

• Blanca Mendoza
• Instituto de Geofísica UNAM
• México
• ASSE04 Sao Jose dos Campos INPE

2
(No Transcript)
3
FIRST IDEAS

• Antonii Mariae Schyrlei de Rheita of Antwerp
  (1645)
• More sunspots- colder wheater
• William Herschel (1801)
• More sunspots- more light and heat-more wheat
  (lower price)
• The Suns output should be measured
• (Herschel, 1807, in Philosophical Transactions)

4
THE ROLE OF THE SUN IN CLIMATE CHANGE

• The observed global warming of the Earth since
  the beginning of the 20th century has been
  attributed preponderantly, if not uniquely, to
  the increasing concentration of greenhouse gases

5
SUN-PLANETARY BODIES INTERACTION

• PRODUCTS OF SOLAR ACTIVITY THAT IMPACT ON
  PLANETARY BODIES
• Electromagnetic radiation
• Energetic particles
• Solar wind and transient ejecta with a frozen in
  magnetic field.
• REACTION OF PLANETARY BODIES TO SOLAR
  ACTIVITY DEPENDS ON
• Intrinsic magnetic fields
• Ionospheres
• Neutral atmospheres
• RESPONSE OF THE EARTH TO SOLAR VARIABILITY
• Geomagnetic activity
• Variations of the high atmosphere
• Changes of weather, climate and biota

6
SUN-EARTH RELATIONS
7
SUN-EARTH RELATIONS
8
MECHANISM OF CHANGE OF SOLAR ELECTROMAGNETIC
RADIATION ARRIVING AT THE EARTH

• - Planetary orbital parameter variations
  (eccentricity of the orbit, inclination of the
  rotation axis, precession)
• - Changes of the albedo (due to for instance
  to variations of cloudiness or atmospheric
  composition and changes in the distribution of
  land and ocean masses)
• - Intrinsic variations of the solar
  irradiance
• Some of these mechanisms produce
  variations that are evident on time scales of
  thousands or even millions of years, but in
  particular the observed changes of the solar
  irradiance are occurring from minutes to decades,
  the time scales that matter to human beings.

9
THE TOTAL SOLAR IRRADIANCE

• The total solar irradiance (TSI) is the value of
  the integrated solar energy flux over the entire
  spectrum arriving at the top of the terrestrial
  atmosphere at the mean Sun-Earth distance (the
  astronomical unit AU). The TSI at the Earths
  orbit can be calculated knowing the Suns radius,
  the photospheric temperature and the value of the
  AU, the result is approximately 1367 W/m2.
  Satellite observations indicate a value of 1367?4
  W/m2

10

• Before the spacecraft era, changes of the TSI
  were difficult to detect by ground-based
  observatories due to the lack of knowledge of the
  selective absorption of the Earths atmosphere
  and the insufficient radiometric precision the
  solar constant
• Spacecraft measurements of the TSI started with
  NIMBUS-7 launched in November 1978 and have
  been carried out by their successors.

11
Daily average values of the TSI from different
radiometers Hickey-Frieden cavity radiometer
(HF) on NIMBUS 7, Active Cavity Radiometer
(ACRIM I) on the Solar Maximum Mission (SMM),
Earth Radiation Budget Experiment (ERBE) on the
Earth Radiation Budget Satellite (ERBS), ACRIM
II on the Upper Atmosphere Research Satellite
(UARS), Solar Variability (SOVA2) experiment on
the European Retrievable Carrier (EURECA) and
Variability of Irradiance and Gravity Oscillation
Experiment (VIRGO) on the Solar and heliospheric
Observatory (SOHO).
12
Observed TSI variations and sources

• Changes of minutes to hours granulation, meso
  and supergranulation. Fluctuations on the 5
  minute range solar oscillations
• Short term changes of few days to weeks
  dominated by sunspots. The sunspot-related dips
  produce changes of 0.3 in TSI
• Over the solar cycle, variations of ? 0.1 in
  consonance with sunspot activity, mainly due to
  faculae and bright magnetic elements. Faculae can
  enhance the total flux by 0.08.
• Space-based observations exist only for about 20
  years, variations on time scales longer than the
  11-year cycle uncertain

13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
Composite TSI from Fröhlich (2000) using
different time series at different times as
indicated on the figure

• Willson (1997) and Willson and Mordvinov (2003)
  used the Nimbus 7/ERB results to relate the
  non-overlapping ACRIM I and ACRIM II data
  sets.
• Fröhlich and Lean (1998) and Fröhlich (2000)
  used ERBS to relate ACRIM I and ACRIM II.

17
Spectral solar irradiance at the top of the
atmosphere and its estimated variability along
the 11-year solar cycle

• A maximum near 500 nm
• 50 of the TSI at visible and near-infrared
  wavelengths between 400-800 nm, 99 between 300
  to 10 000 nm.
• VARIATIONS
• For the UV wavelength 20 near 140 nm, 8 near
  200 nm and 3 near 250 nm, but most of them
  occur below 500 nm.
• At visible and infrared wavelengths is
  unreliable. At wavelengths larger than 400 nm
  only theoretical estimates. Observations indicate
  a variability of 0.1 along the solar cycle.

18
EMPIRICAL MODELS OF TSI VARIATIONS

• TSI variations processes e.g. photospheric
  temperature changes, changes in solar diameter,
  etc changes in the amount and distribution of
  magnetic flux on the solar surface
• Most recent reconstructions of the TSI assume
  that all variations are due to surface magnetic
  changes.
• Time evolution of superficial magnetic features -
  variation of the magnetic solar activity -
  variations in brightness direct and indirect
  indices of solar activity
• The solar activity indices - irradiance
  indices the facular brightening and sunspot
  darkening.

19
Several indices of solar activity
20
A short-term empirical model of TSI variability
based on sunspot and faculae compared with the
measured TSI (Lean et al., 1995)

• The short-term models time scales of the solar
  cycle. Inputs facular brightening and sunspot
  darkening. These models have reproduced ? 80 of
  the TSI irradiance variability of the space
  observation time span.

21
A long-term empirical TSI model using as
long-term component the smoothed sunspot group
number (Lean et al., 1995)

• Long-term TSI changes are speculative.
• Use besides solar cycle inputs, an index of
  long-term variability such as the smoothed
  sunspot group number or the cycle length.

22
THE INFLUENCE OF TSI VARIABILITY ON EARTHS
CLIMATE

• The 1367 W/m2 of TSI are distributed over the
  planet down to 1/4 of this ? 342 W/m2 . The
  Earths albedo is 0.3, then the income
  radiation is 239 W/m2. Upon entering the
  atmosphere solar irradiance at wavelenghts
  shorter than 300 nm are absorbed in the
  stratosphere and above.
• For the last two solar cycles, the portion that
  arrives at the troposphere presents a solar
  cycle change of ?0.1, - 0.24 W/m2.
• This seems too small to have an appreciable
  effect on surface climate.

23

• Secular TSI variation models indicate changes
  from 0.24 to 0.30 (0.5 to 0.75 W/m2), with
  extreme values during the deepest phase of the
  Maunder minimum of 1.23 decrease (2.9 W/m2).
• Then the solar forcing has been most of the time
  small compared with estimates of the
  anthropogenic forcing by greenhouse gases of
  2.4 W/m2

24
Decadal averages of a reconstructed TSI and
North Hemisphere temperature anomalies to the
present
25
MODELS OF THE INFLUENCE OF TSI VARIABILITY ON
EARTHS CLIMATE

• Cusbach and Voss (2000)introduced TSI in a
  general circulation model (GCM)
• - During the last 100 years, the simulations
  have linear increasing trends of 0.17 to 0.19K,
  while the observed one is 0.6, which means that
  TSI is contributing moderately to the observed
  warming.

26

• Shindell et al. (2001) also introduced TSI
  in a GCM
• Examined the climate response between
  17th and 18th centuries. Included a response of
  the complete stratospheric ozone to TSI.
• Global changes of 0.3 to 0.4 C were
  obtained coinciding with temperature
  reconstructions.
• Regional temperature changes as large as
  1 to 2C in the NH winter are obtained.
• The 20th century simulations show that TSI
  together with ozone variations and climate
  feedbacks (for instance aerosols) change the
  temperature by ? 0.19C, almost a third of the
  warming trend.

27

• Before 1970 although reproducing well the
  observed temperature, TSI variations cannot
  account for all the temperature changes, and that
  after 1970 its influence has conspicuously
  descended.
• Then other sources of solar variability and/or
  sources different from solar variability must be
  present.

28
OTHER FORCINGS USED ON CLIMATE MODELS

• Solar UV irradiance absorbed by the stratospheric
  ozone rising the temperature (Haigh,1999
  Shindell, et al., 1999) warming of the lower
  stratosphere -stronger winds- penetration of
  these winds into the troposphere - Hadley
  circulation.
• The model shows the observed 11-year variation in
  the stratosphere but the amplitude of the
  simulated changes is still too small compared to
  the observations.
• Comparisons of a reconstructed UV solar
  irradiance with global temperature along
  1915-1999, indicate a poor correlation of r
  0. 46. The interaction of UV irradiance and
  climate should be indirect (Foukal, 2002).

29

• A good correlation between total cloud cover
  changes and cosmic rays for 1983-1994 (Svensmark
  and Friis-Christensen,1997). Extrapolated to time
  scales of decades and longer.
• Further work seemed to confirm this for low level
  clouds (Marsh and Svensmark, 2000 Pallé and
  Butler, 2000). Serious criticisms have been
  raised concerning the handling of the data (Laut,
  2003).
• A thermodynamic climatic model for 1984-1990
  showed responses of few tenths of degrees in the
  NH temperature using as forcing change in
  total and low cloud cover (Ramírez et al.,2004
  Mendoza et al., 2004).

30
Reconstructed temperature change for the northern
hemisphere and model simulations using
different TSI forcings (Cubasch and Voss, 2000)
31
Simulations that take the greenhouse gases
according to observations but without Suns
influence simulate a trend of 0.43 K for the 20th
century (Paeth et al., 1999) . When taking
greenhouse gases increase and a TSI model, the
trend is of 0.6K, very close to observations
however the aerosol cooling effect, neglected in
the model, should lower the temperature below
the
32
DISCUSSION

• A constant or a variable output of the
  quiet Sun?
• Willson (1997, 2003) in his composite TSI
  found a secular upward trend of 0.05 per decade
  between consecutive solar cycles 22 and 23, while
  Fröhlich and Lean ( 1998) and Fröhlich (2000) in
  their composite series do not see a change.
  However the data is too short to draw any
  conclusion.
• Are TSI changes due to surface solar
  magnetic activity?
• de Toma et al. (1999) pointed out that
  although solar cycle 23 seems magnetically weaker
  than solar cycle 22, TSI space observations
  indicate a similar radiative output for both
  cycles. One possibility is that sunspot and
  facular indices may not adequately represent the
  TSI from cycle to cycle, pointing to the possible
  existence of other sources of solar cycle TSI
  variability.

33
DISCUSSION

• Does TSI always increase in consonance with
  solar activity?
• Observations of the last two cycles
  indicated that faculae and bright magnetic
  network elements dominate over spots at maximum
  times within the solar cycle.
• Foukal (1993) showed that for 1874 - 1976,
  for the Sun at maximum this was true except for
  the highest activity cycle in that time span,
  cycle 19, when sunspots dominated over faculae.
• This result implies that the change of solar
  irradiance in consonance with solar activity may
  be reversed if the Sun becomes much more active
  than today. Furthermore, it has been proposed
  that not only the high activity Sun but also the
  low activity Sun can become dimmer when evolving
  from minimum to maximum.

34
DISCUSSION

• Is there a long-term component of TSI?
• TSI secular forcings of climate consider a
  long-term component. Foukal (2002) TSI
  variations are closely proportional to the
  difference between spot and facular areas, which
  varies from cycle to cycle, then there is little
  reason to expect that TSI tracks any of the
  familiar solar activity indices.
• Lean et al. (2002) TSI comes mainly from
  closed solar magnetic flux regions (total solar
  magnetic flux). The open and total magnetic
  flux variations behave differently the total
  flux does not present a long-term trend as does
  the open flux. Then the TSI should not show a
  long-term change, in agreement with Foukals
  (2002) study and with the composite of TSI by
  Fröhlich and Lean) and Fröhlich), and in
  disagreement with Willson) and Willson and
  Mordvinov.

35
DICUSSION

• Why a long-term trend in TSI?
• Is based on extrapolations to the Sun of
  photometric behavior of Sun-like stars there is
  evidence that the Ca II H and K emission in
  solar-type stars exhibit a much wider scatter
  than the Sun does along a solar cycle. On the Sun
  Ca II brightness is well correlated with magnetic
  features (faculae and the network) (Baliunas and
  Jastrow, 1990).
• Also the variation of cosmogenic isotopes
  (Beer, 2000) suggest that the Sun may present a
  wider range of activity (irradiance changes).
  But as the open and total (mainly closed)
  magnetic flux variations are not the same, we
  cannot expect cosmic rays/ cosmogenic isotopes,
  modulated by the open flux, to reproduce TSI
  which is modulated by the closed flux. Then
  cosmogenic isotopes changes do not imply TSI
  changes.

36
DISCUSSION

• Mechanism of amplification of the TSI
  changes at Earth?
• If long-term changes of TSI are inexistent
  then the solar radiative forcing of climate in
  long-term climate models will be reduced by a
  factor of 3 , and those climate models will be
  overestimating the role of TSI variability. Then
  how to explain the close correlation between
  solar irradiance and temperature?
• Perhaps the answer is in some models that
  indicate that the noise of atmospheric natural
  climate fluctuations amplify a weak solar forcing
  (Rahmstorf and Alley, 200). Then long-term
  climate changes may appear to follow the solar
  cycle because the stochastic response increases
  with cycle amplitude, not because there is an
  actual irradiance change.
• Or indirect interactions of TSI with
  climate a high anticorrelation between TSI and
  low cloud cover has been presented by
  Kristjánsson et al. (2003) for 1983-1999, they
  suggest that TSI variations are amplified by
  interacting with sea surface temperature and
  subsequently with low cloud cover in subtropical
  regions.

37
CONCLUSIONS

• The secular reconstructed TSI variations can
  account for a considerable part of the
  temperature variations of the Earth in the
  pre-industrial era. But even for those times the
  temperature changes are not fully reconstructed
  from TSI. Which means that other sources of
  solar variability as well as internal natural
  causes were contributing to the Earths
  temperature variability.
• During the 20th century TSI produces less than
  half of the observed temperature changes,
  confirming suggestions that for this century
  besides natural causes, man-made activities are
  contributing to the Earths temperature
  variability, particularly the latter.
• Several questions on TSI variations and climate
  challenge well established results.