It Is Impossible For A 100 ppm Increase In
Atmospheric CO2 Concentration To Cause Global
Anyone who tries to understand how a 100 ppm increase in atmospheric CO2 concentration
causes climate change will soon find out that the entire global warming argument is nothing
more than empirical speculation. No cause and effect linking CO2 and global warming has
ever been demonstrated because none exists. In order to understand how the Earth’s climate
works it is necessary to go back to first principles and look carefully at the surface energy
transfer. The air-ocean and the air-land interfaces behave very differently, so they have to be
considered separately. This article summarizes the work performed by Dr. Roy Clark, the
founder of Ventura Photonics, on the underlying energy transfer physics that sets the Earth's
surface temperature. However, first we must define the energy transfer question that we seek
Ask the Right Question
Over the last 50 years or so, the atmospheric CO2 concentration has increased by about ~70
ppm and over the last 200 years this increases further to ~100 ppm. Under ideal ‘clear sky’
conditions, these increases in CO2 concentration have produced an increase in the downward
atmospheric LWIR flux at the surface of 1.2 and 1.7 Watts per square meter. At the same
time, over the last six solar cycles, the sunspot index has been running about 70% above
normal compared to the index average from 1650. This has produced an average increase in
the solar constant at the top of the atmosphere of ~0.3 Watts per square meter in a flux of
1365 Watts per square meter. The changes in the downward LWIR flux may be determined
from radiative transfer calculations using the HITRAN database [Rothmann et al, 2004]. The
calculated changes in the Kirchoff Exchange Energies are identical to the empirical 'Radiative
Forcing Constants' used by Hansen et al . The change in solar flux is derived from the
SOHO Satellite VIRGO Radiometer data and the sunspot record from 1650. These data are
shown in Figures 1 to 3.
The question that needs to be answered is the following:
Starting from the basic Laws of Physics including the First and Second Laws of
Thermodynamics, the Kirchoff Exchange Law and Beer's Law, determine the changes
in surface temperature caused by both a 100 ppm increase in atmospheric CO2
concentration over 200 years and by the changes in the solar constant due to
variations in the sunspot index. Treat the air-ocean and the air-land interfaces
separately. Explain how the changes in surface temperature are then coupled into the
meteorological surface temperature record.
The Air-Ocean Interface
Water is almost transparent to visible radiation and sunlight can penetrate down through
clear ocean waters to depths of ~100 meters [Hale & Querry, 1973]. The light is absorbed
mainly by the rather weak overtones of the water infrared vibrations and converted into heat.
The oceans cool through a combination of evaporation and long wave infrared (LWIR)
emission from the surface [Yu et al, 2008]. The First Law of Thermodynamics (conservation
Any flux difference is converted into a change in ocean temperature. Over most of the LWIR
spectral region, the ocean surface exchanges radiation with the atmosphere. On average,
there is a slight exchange heating of the atmosphere by the ocean. This net heat transfer
depends on the thermal gradient or air -ocean temperature difference as required by the
Second Law of Thermodynamics. LWIR emissive cooling occurs within a relatively small
spectral emission window in the 8 to 12 micron region (~1200 to 800 wavenumbers). The
penetration depth of LWIR radiation into the ocean is less than 100 micron, about the width
of a human hair.
Small increases in LWIR emission from the atmosphere are converted into increases in
ocean surface evaporation that are too small to detect in the wind driven fluctuations
observed in surface evaporation. Between 1977 and 2003, average ocean evaporation
increased by 11 cm per year from 103 to 114 cm per year. This was caused by an increase
in average wind speed of 0.1 meters per second [Yu, 2007]. The uncertainty in the estimate
was 2.7 cm per year which is larger than the upper 'clear sky' limit to the evaporation
produced by a 100 ppm increase in CO2 concentration over 200 years. It is simply
impossible for a 100 ppm increase in atmospheric CO2 concentration to have any effect on
ocean temperatures. Figure 4 illustrates the basic energy transfer processes at the
air-ocean interface. Figure 5 shows the spectral properties of water in the visible and the IR.
Figure 6 shows ocean evaporation and the effect of changes in wind speed. An increase of
1.7 Watts per square meter in downward LWIR 'clear sky' radiation translates into an upper
limit increase in evaporation rate of 2.4 cm per year.
Historically, data on subsurface ocean temperatures has been very sparse. This has changed
ocean temperature profiles and other data every 10 days. In general, as the sun warms the
ocean temperature profiles and other data every 10 days. In general, as the sun warms the
ocean during the spring and summer, a stable thermal gradient develops below a daily uniform
mixing layer. During the fall and winter, cooling from the surface extends to lower depths and a
with the Argo Float Program. There are now 3000 floats in the world’s oceans that record
uniform temperature layer forms that reaches to depths of 100 m or more. Figure 7 shows the
2007 annual profile data from two selected Argo floats located in the S. Central Pacific Ocean.
One is near the equator with an average latitude of -1.5 deg. The other has an average
latitude of -20.9 deg. Near the equator, the daily mixing layer does not generally extend below
50 m. Therefore, subsurface ocean layers below 50 m can accumulate solar heat for
extended periods. This is the most likely origin of the ENSO cycles and related phenomena
such as periodic fluctuations in hurricane intensity. These warm subsurface ocean layers can
be transported and recirculated over long distances by ocean currents and gyres without any
interaction with the ocean surface. At higher latitudes, seasonal cooling mixes the ocean
layers uniformly to depths below the 100 m solar heating region.
The effect of small changes in the solar constant on ocean temperatures may be investigated
using a simple ocean solar heating model with a single absorption coefficient and a fixed
surface cooling flux. Such a model reproduces the annual temperature profile shown in
Figure 7b. The cooling flux is adjusted so that the model is stable and repeats the same
temperature changes for each annual cycle. The effect of sunspot induced changes in the
solar constant may then be investigated by scaling the solar flux using the 350 year sunspot
index from 1650 to 2000 with a scale factor of 1 W.m^-2 per 100 change in the annual sunspot
index. Figure 8 shows the result from such a model with an extinction coefficient of 0.075
m^-1, at 30 deg latitude and 90 m depth. The cooling during the Maunder and Dalton Minima
and the Modern Warming period can be clearly seen [Clark 2010a].
Levitus et al,  have estimated that the warming of the oceans from 1955 to 1998 was
caused by an average increase in heat flux into the oceans over this period of 0.2 Watts per
square meter This is consistent with the average increase in the solar constant of 0.3 Watts
per square meter. However, there was also a significant variation in the temperature changes
between ocean basins. For the upper 300 m of the ocean, from 1955 to 2003 the worldwide
average temperature increase was 0.17 C. The N. Atlantic warmed by 0.35(4) C and the N.
Pacific by 0.09(3) C. This is illustrated in Figure 9. These fluctuations are caused by
differences in ocean circulation, mixing and wind speed. For example, in the Atlantic Ocean,
the Southern Equatorial Current is split by the coast of Brazil and part is diverted northwards
towards the Caribbean.
The Air-Land Interface
The ground is heated by the sun during the day and the increase in thermal flux from the
surface heats the air. It is important to distinguish clearly between surface temperature and
meteorological surface air temperature [Clark, 2010b]. The surface temperature needed for
energy transfer analysis is the ground surface temperature. The meteorological surface air
temperature (MSAT) is the air temperature measured in an enclosure placed 1.5 to 2 m above
the ground. There are no obvious or simple relationships between these two temperatures.
Solar radiation is absorbed and reflected by the ground. The resulting surface temperature
depends on the absorption coefficient, the thermal conductivity and heat capacity of the
ground, the surface area and angles of incidence, the balance of the upward and downward
LWIR flux and the direct air convection. If the ground is moist, latent heat effects also have to
be included. The MSAT depends on the origin of the bulk air mass of the local weather
system, surface LWIR flux heating, the latent heat flux, air convection and wind speed. The
energy transfer processes at the air-land interface are illustrated in Figure 10. In common with
the air-ocean interface, the First Law of Thermodynamics imposes energy conservation, but
there is no requirement that the local flux be conserved on any time scale. The Second Law of
Thermodynamics requires that the heat transfer follow the thermal gradient. The solar heating
of the ground and the air must both be considered in the energy transfer analysis.
The fundamental assumption that is generally used in large scale climate models is that the
MSAT can be treated as an ‘equilibrium surface temperature’. Small changes in downward
atmospheric LWIR flux are then assumed to produce measurable changes in this ‘temperature’
that may be calculated using perturbation theory and Stefan’s Law. This is totally absurd. The
daily and seasonal surface temperatures have to be calculated dynamically using the time
dependent changes in the total flux and the temperature gradients at the surface. Under full
summer sun conditions, the daily solar heat load reaching the surface can exceed 25 MJ per
square meter. The corresponding maximum surface temperatures can exceed 50 C. Daily
variations in LWIR cooling flux because of changes in cloud cover and humidity may easily
exceed 5 MJ. Over the last 50 years, the total daily increase in clear sky LWIR flux from a 70
ppm increase in atmospheric CO2 concentration has been ~0.1 MJ. It is impossible to detect
any change in daily or seasonal surface temperature caused by such a small change in flux.
Figure 11 shows the daily maximum and minimum MSAT and the 8 day average satellite
surface (skin) temperatures for an Ameriflux monitoring site near Irvine, S. California (UCI
Grasslands). These sites record flux data as well as basic weather station parameters. The
data for 2008 from this station was analyzed in detail and used to develop a basic ground
heating model. This included the latent heat flux, the solar and IR fluxes, the convection and
the subsurface heating to 1 m below ground. The time step was 2 hours. The calculated
maximum and minimum surface temperatures and the 4 th order polynomial fits to the surface
temperature data from Figure 11 are shown in Figure 12. Although the sun reaches its
maximum intensity with the summer solstice, the surface temperatures do not reach their
maxima until later. This clearly shows the role of subsurface heat storage in setting the
The model was then rerun with an increase in downward LWIR flux of 1.7 Watts per square
meters to simulate a 100 ppm increase in atmospheric CO2 concentration. The changes in
calculated minimum and maximum surface temperature are shown in Figure 13. They do not
exceed 0.07 C. The total daily heat flux is coupled into at least the first meter of the surface
and when these effects are included in the simulation, the role of CO2 in setting the surface
temperature is minimal. When short term latent heat and cloud cover fluctuations are
included, no effect from CO2 can possibly be detected. This is one of the many fundamental
and fatal errors in the global warming argument.
Cherry Picking the Pacific Decadal Oscillation
The next question of course is what temperature change are we really measuring? The
meteorological surface air temperature (MSAT), on which the whole global warming argument
is based has nothing to do with the ground surface temperature. It is the air temperature in an
enclosure placed for convenience at eye level 1.5 to 2 m above the ground. Since it is
impossible for CO2 to have caused any change in ground surface temperature, there is even
less reason to expect it to have caused any change in the MSAT. The simplest way to
interpret the historical MSAT record is to assume that the long term average of the minimum
MSAT is an indicator of the atmospheric lapse rate. The maximum MSAT temperature is just
the daily solar warming superimposed on the minimum MSAT. To raise the minimum MSAT
temperature by 1 C, it is necessary to heat the whole air mass up to the troposphere by 1 C.
This requires about 10 MJ of heat for a 1 m x 1m x 10 km air column. Only a change in ocean
surface temperature can couple this quantity of heat into the air mass of a weather system
through a change in surface evaporation. Global warming is nothing more than the
observation of the 30 year increases in ocean surface temperature that ended about 10 years
ago. The same effects can be seen in the previous ‘dust bowl’ cycle from the 1930’s.
In order to understand this in more detail, it is convenient to look at some of the long term
weather station data in the Greater Los Angeles area. This region has a transition from Pacific
coast to desert climate zones within 100 km. In addition, there are local mountain ranges with
their own sets of microclimates. The prevailing weather systems generally arrive from the N.
Pacific Ocean, so there is a clear indication of the Pacific Decadal Oscillation (PDO) in the
weather station data. A common weather pattern in the Los Angeles basin involves a large
thermal gradient from the coast to the desert with maximum daily temperatures increasing from
~20 C at the coast to ~35 C or more in the inland low desert regions. A strong sea breeze
develops during the day that pushes the air from the coast inland towards the desert.
However, under certain conditions when high pressure develops over the Great Basin inland,
the wind pattern reverses and strong, dry Santa Ana winds develop. As the air is pushed from
the desert to the ocean it descends about a kilometer in altitude and may also be compressed
further by funneling through the local canyons. The adiabatic compression can increase the
air temperature by 10 C or more. When the coastal region starts to report record high
temperatures it is generally because of strong Santa Ana wind conditions. Any claim that this
is global warming caused by an increase in atmospheric CO2 concentration is simply wrong.
Figure 14 shows the PDO from 1900 onwards. A five year rolling average of the annual PDO
index is shown. A simple straight line fit to the full data set data is also shown. This was
obtained using the linear trendline algorithm in Excel. Over the 1905 to 2009 time frame, the
line is almost flat. However, it should be noted that a wide range of slopes can be obtained if
the fit is made over selected sections of the curve. The PDO index is the temperature
difference between the central and eastern regions of the N. Pacific Ocean. A positive index
indicates higher temperatures in the eastern Pacific, along the N. American coast. Figure 15
shows the minimum MSAT anomaly recorded by the Los Angeles Civic Center (LACC) from
1921. A five year rolling average is plotted and a simple straight line fit to the data is also
shown. The data were downloaded from the Western Region Climate Center website and
used ‘as received’. The plot may be interpreted as the PDO with an urban heat island effect
from Los Angeles and some local climate variations superimposed.
Figure 16 shows the annual minimum temperature anomaly weather station data from the LA
Civic Center (LA CC), Claremont (CMT), Pasadena (PAS) and Santa Monica (SM). The data
are smoothed using a 5 year rolling average. The linear fit trend line to the data and to the
PDO over the same time period as the weather data sets are also shown. All of the data was
used ‘as received’ from the WRCC website. The PDO ‘signature’ can be clearly seen in all of
the temperature anomaly plots. Deviations from the PDO trend indicate other local weather
influences. The El Nino Southern Oscillation (ENSO) may also be a factor. However, in each
case there is a clear increasing temperature trend in the weather data that may be attributed
to the growth of the Los Angeles urban heat island.
Figure 17 shows the weather data for Los Angeles Airport (LAX), Santa Barbara (SB) and
Pierce College (PC) analyzed in the same way as Figure 16. For these stations, the weather
data trend lines are closer to the PDO lines, showing a lower urban heat island effect. The
LAX and SB stations are close to the Pacific Coast and Pierce College is located at the west
end of the San Fernando Valley. The prevailing weather patterns at these stations tend to
shield them from urban heat island effects. [Note: The Santa Monica data is unusual, since
this is also a coastal station. The trend line here is ‘pulled down’ by the lower 1940 to 1950
temperatures. Also, Santa Barbara is north of Los Angeles and is not in the LA Basin.]
The trend lines may be used to remove the urban heat island effects from the temperature
anomalies. This is shown in Figure 18. The ‘detrended’ minimum temperature anomaly curves
clearly indicate the effect of the PDO in all of the weather station data. The temperature
increase ‘predicted’ by the hockey stick ‘calibration’ using the Keeling curve data from 1958 is
also shown. This is based on the radiative forcing assumption used in all of the IPCC climate
models that a 1 Watt per square meter increase in downward atmospheric flux from CO2
produces a 2/3 C rise in surface temperature. There is no CO2 ‘signature’ in the greater Los
Angeles weather station data. The hockey stick in this case is nothing more than a ‘cherry
pick’ fit to the Pacific Decadal Oscillation.
One of the most misunderstood and misused terms in the entire global warming argument is
the greenhouse effect [Clark, 2010c]. The IR active gases in the Earth’s atmosphere absorb
and re-emit LWIR radiation. The downward LWIR flux from the atmosphere keeps the Earth’s
average surface temperature surface about 33 K warmer than it would be without this flux.
However, the full explanation of this so called greenhouse effect requires careful consideration
of the detailed energy transfer processes involved. Simple arguments based on ‘equilibrium
averages’ are inadequate at best and often simply wrong. The atmospheric energy flux is
continuously changing and this must be incorporated into the analysis. Atmospheric energy
transfer is of course constrained by the First and Second Laws of Thermodynamics and by the
Kirchoff Exchange Law. It is also important to understand that the thermal gradient in the
troposphere is set by the lapse rate. This may be calculated thermodynamically from the
surface air temperature and humidity, but the uniformity assumptions generally used may not
always be valid. However, in all cases the lapse rate is not changed by 100 ppm variations in
the concentration of the permanent greenhouse gases such as CO2. The molecular collision
rate is also very fast in the troposphere (>10^9 per second), so that the excited state lifetime of
an IR active molecule is very short. This means that the absorbed IR radiation is rapidly
converted to molecular kinetic energy and the IR active molecules are always at the local air
temperature. The atmospheric absorption bands consist of a large number of overlapping
lines due to transitions between specific rotation-vibration states of the IR molecules involved.
The individual lines are quite narrow with line widths of a few tenths of a wavenumber. The line
profiles are Lorentzian and the line widths decrease with altitude as the pressure decreases.
This means that the upward and downward LWIR fluxes are not equivalent [Rothman, 2005].
Any atmospheric energy transfer analysis must explicitly consider these linewidth effects and
any approximations made to simplify the lineshape calculations have to be properly validated
using high resolution results. These linewidth effects invalidate all of the flux equilibrium
assumptions used in radiative forcing calculations.
Figure 19a shows the lapse rate for moist air calculated for various conditions of surface air
temperature and humidity. Figure 19b shows the corresponding concentration profiles for
H2O. The concentration profiles for 200, 380 and 500 ppm of CO2 are also shown. The
concentration of H20 in the troposphere decreases by 3 orders of magnitude with altitude
because of the decrease in vapor pressure with decreasing temperature. The concentration
of CO2 decreases by less than a factor of three. The line width profiles of the H2O spectral
lines therefore decrease much faster that those of CO2. The means that there is less
absorption and re-emission of LWIR radiation within the H2O bands as the altitude increases.
The radiation from the wings of the broader H2O lines in the lower troposphere is not
reabsorbed at higher altitudes and simply escapes into outer space. H2O is therefore much
more effective at cooling the atmosphere than CO2. This is illustrated in Figure 20. The
normal daily and seasonal changes in atmospheric H2O concentration with temperature and
humidity are so large that any changes in the bulk tropospheric cooling rate from a 100 ppm
increase in CO2 concentration are not detectable.
In general, an air parcel in the troposphere emits equal amounts of LWIR flux in the upward
and downward directions. The air parcel will also absorb LWIR radiation from the air layers
above and below. Usually the downward emitted flux and the upward absorbed flux are similar,
whereas the absorbed downward flux from the cooler air layers above will be less than the
upward emitted flux. The net effect is therefore a cooling of the atmosphere. This is illustrated
in Figure 20. Over the oceans, almost all of the LWIR radiation lost to cooling is replaced by
the latent heat flux from the condensation of water that has evaporated from the ocean. Over
land, the most of the LWIR radiation lost to space is replaced by excess thermal energy from
the ground as it is heated by the sun during the day. However, there may also be a significant
latent heat flux if the ground is moist. It is also important to note that the air is re-circulated by
convection and that as the air rises, it is cools at the local lapse rate.
Figure 22 shows a plot of the greenhouse effect for H2O and CO2 with a surface temperature
of 325 K and a surface air temperature of 295 K. The relative humidity is 50%. This is
representative of mid day summer sun conditions…Most of the LWIR radiation from the
surface is absorbed within the first 100 m above the ground and all of the surface LWIR flux is
absorbed within the first kilometer. Most of the atmospheric cooling occurs from the wings of
the water lines in the rotational water band below 600 wavenumbers. The data used in this
plot were derived from high resolution radiative transfer calculations (0.01 wavenumber
spectral resolution, 100 m altitude resolution).
Figure 23 shows the flux data from Figure 22 plotted on an enlarged scale. The transition
from warming to cooling at altitudes above 1 km can be clearly seen. H2O is at least three
times more effective at cooling than CO2.
As the surface cools and the surface thermal gradient decreases, less heat is transferred from
the surface and atmospheric cooling starts at lower altitudes. An increase in water vapor
concentration will increase the atmospheric cooling rate. A 100 ppm increase in CO2
concentration has no measurable effect on overall atmospheric energy transfer rates.
Most of the large scale climate models used to predict global warming ignore the energy
transfer physics at the Earth’s surface and use an approach known as radiative forcing. This
was first proposed by Manabe and Wetherald in 1967 to simplify climate change calculations
so that they could be performed using the limited computer capabilities available at that time.
The assumptions that underlie radiative forcing require very careful examination. Radiative
forcing assumes that long term averages of dynamic, non-equilibrium climate variables such as
radiative flux and surface temperatures can be analyzed using perturbation theory as though
they were in thermal equilibrium. Small changes in ‘average equilibrium flux’ through an
artificial boundary such as an ‘average tropopause’ are assumed to influence ‘average
equilibrium surface temperatures’. Although the mathematical derivation is correct and may
even appear elegant, the underlying physical assumptions are incorrect and the results have
no physical meaning. This is clearly demonstrated by Figure 13 above. Regardless of the
radiative forcing assumptions, a 1.7 Watt per square meter increase in downward LWIR flux
cannot heat the surface. The troposphere is an open thermodynamic system so heat and flux
are not conserved. The temperatures in the upper troposphere are near 220 K. The
assumption that small changes in LWIR flux in the upper troposphere or stratosphere can
influence surface temperatures of 288 K requires a flagrant violation of the Second Law of
Thermodynamics. Heat does not flow from a cooler to a warmer body. The calculated
‘equilibrium surface temperature’ produced by radiative forcing calculations is not even a
physically measurable climate variable. It certainly has no relationship to the MSAT.
However, in the mid 1980’s, a slight increase in the ‘average’ meteorological surface air
temperature was found [Jones et al, 1994]. This was immediately linked by empirical
speculation to the increase in anthropogenic CO2 concentration. It was decreed by the order
of the divine hockey stick that a 1 Watt per square meter increase in downward LWIR flux due
to an increase in atmospheric CO2 concentration produced an increase in meteorological
surface air temperature of 2/3 C. This was almost four times larger than the blackbody value.
However, instead of admitting that the assumptions were incorrect, mysterious water vapor
feedback effects were invoked to explain model ‘inaccuracies’. These have now been
disproved by Lindzen and Choi, . The Kirchoff exchange energy then was converted into
an empirical ‘radiative forcing constant’. This ‘calibration factor’ was then applied to other
greenhouse gases such as methane and even to aerosols. The ‘radiative forcing constants’
used in climate simulation models are devoid of physical meaning. This approach is empirical
pseudoscience that belongs to the realm of climate astrology. The results derived from climate
simulations that use the radiative forcing approach may be of limited academic interest in
assessing model performance. However, such results are computational science fiction that
have no relationship to the reality of the Earth’s climate. Radiative forcing by CO2 is, by
definition a self-fulfilling prophesy, since the outcome is pre-ordained with a blatant disregard
of the basic laws of physics. An increase in CO2 concentration must increase surface
temperature. No other outcome is allowed and other possible climate effects are by definition
excluded. Radiative forcing is just the hockey stick feeding back on itself.
It is impossible for the observed 100 ppm increase in atmospheric CO2 concentration to
have caused any kind of climate change.
This follows directly from the application of the basic Laws of Physics to the energy
transfer processes that occur at the Earth’s air-ocean and air-land interfaces.
There is a fundamental need for a worldwide monitoring program to provide quantitative data
on energy transfer at the Earth’s air-ocean and air-land interfaces. Most of the data required
cannot be obtained from satellite observations. Float monitoring programs such as Argo need
to be expanded to include the measurement of the coupling of solar radiation into the ocean
and to perform more detailed studies of near surface ocean currents. A worldwide set of
advanced micrometeorological monitoring stations needs to be established that provide
detailed and continuous surface energy transfer data. These data are needed to support a
new generation of properly validated climate simulation models that incorporate the correct
surface energy transfer physics. In addition to more detailed monitoring of key climate
variables, the astrophysical mechanisms that produce sunspots require further study. The
prediction of long term climate change requires the prediction of long term changes in the
solar constant that are of the order of 0.1 Watts per square meter.
|Dr. Roy Clark
Ventura Photonics, 2/23/2010, All rights reserved
Clark, R., [2010a] Energy and Environment 21(4) 171-200 (2010), 'A Null Hypothesis for CO2'
Clark, R., [2010b] 'What surface temperature is your model really predicting?'
Clark, R., [2010c] 'Gravity rules over the photons in the greenhouse effect'
Hale, G. M. and Querry, M. R., Applied Optics, 12 555-563, (1973), ‘Optical constants of
water in the 200 nm to 200 µm region’.
Hansen, J. et al, (45 authors), J. Geophys. Research, 110 D18104 pp1-45 (2005), ‘Efficacy
of climate forcings’.
Jones, P. D., New, M., Parker, D. E., Martin, S. and Rigor, I. G., Rev. Geophysics, 37(2)
173-199 (1999), ‘Surface air temperature and its changes over the past 150 years’.
Levitus, S., Antonov, J. and Bower, T., Geophysical Research Letters, 32 L02604 1-4,
(2005), ‘Warming of the world ocean 1955-2003’.
Lindzen, R. S. & Y-S. Choi, Geophys Res. Letts 36 L16705 pp 1-6 (2009), ‘On the
determination of climate feedbacks from ERBE data’.
Manabe, S. and Wetherald, R. T., J. Atmos. Sci., 24 241-249 (1967), ‘Thermal equilibrium of
the atmosphere with a given distribution of relative humidity’.
NOAA Sunspot Data
Rothman, L. S. et al, (30 authors), J. Quant. Spectrosc. Rad. Trans., 96 139-205 (2005) ‘The
HITRAN 2004 molecular spectroscopic database’.
Yu, L., J. Climate, 20(21) 5376-5390 (2007), ‘Global variations in oceanic evaporation (1958-
2005): The role of the changing wind speed’.
Yu, L., Jin, X. and Weller R. A., OAFlux Project Technical Report (OA-2008-01) Jan 2008.
‘Multidecade Global Flux Datasets from the Objectively Analyzed Air-sea Fluxes (OAFlux)
Project: Latent and Sensible Heat Fluxes, Ocean Evaporation, and Related Surface
Find Out More - Read the Book:
'The Dynamic Greenhouse Effect and the Climate Averaging Paradox'
Click here for more information