Climate Modeling of the Early
Earth: What Do We Know and
How Well Do We Know It?
James Kasting
Department of Geosciences
Penn State University
Ice age (Late Cenozoic)
Dinosaurs go extinct
Warm
First dinosaurs
Phanerozoic
Time
Ice age
Age of fish
First vascular plants on land
Ice age
First shelly fossils
Geologic time
First shelly fossils (Cambrian explosion)
Snowball Earth ice ages
Warm
Rise of atmospheric O2 (Ice age)
Ice age
Warm (?)
Origin of life
Evidence for 2.9-Ga
glaciation
• One finds diamictites and
dropstones in the 2.9-Ga
Mozaan Group in South Africa,
within the Pongola Basin
• Diamictites are called tillites once
their glacial origin is established
• Diamictites are also found in
the Belingue Greenstone Belt
in Zimbabwe (E. M. Nisbet et al.
Geology, 1993), although they
were not identified as glacial
by the authors
G.M. Young et al., J. Geol. (1998)
2.9-Ga dropstone
• Dropstones are fistsized (or larger) chunks
of rocks found in
laminated marine
sediments that are
thought to have been
deposited by melting
icebergs
• This one was found in
the muddy upper layers
of the diamictites shown
in the previous slide
G.M. Young et al., J. Geol. (1998)
• The geologic data suggest that
the Archean was cool (but not
frozen), at least some of the
time
• By contrast, oxygen isotope
data from cherts (SiO2),
appear to indicate that the
Achean Earth was hot,
7015oC, at 3.3 Ga
• Indeed, data from carbonates
(not shown) suggest that the
Earth remained warm up until
~400 Ma
• These isotopic data have
multiple interpretations, some
of which will come up during
Paul Knauth’s talk
(SMOW)
Oxygen isotope evidence for a hot
early Earth?
Cold
Warm
P. Knauth, Paleo3 219, 53 (2005)
Surface temperature vs. pCO2 and fCH4
3.3 Ga
S/So = 0.77
• Getting surface temperatures of ~70oC at 3.3 Ga would require of the
order of 3 bars of CO2 (10,000 times present), possibly more, as the
greenhouse effect of CH4 was overestimated in these calculations
• pH of rainwater would drop from 5.7 to ~3.7
• Would this be consistent with evidence of paleoweathering?
Kasting and Howard, Phil. Trans. Roy. Soc. B (2006)
• From a theoretical standpoint, it is
curious that the early Earth was warm,
because the Sun is thought to have
been less bright 
Why the Sun gets brighter with time
• H fuses to form He in the
core
• Core becomes denser
• Core contracts and heats
up
• Fusion reactions proceed
faster
• More energy is produced
 more energy needs to
be emitted
Figure redrawn from D.O. Gough,
Solar Phys. (1981)
• Various astronomers have come up with
proposals for altering the standard solar
evolution model, and we will hear about that
at this meeting
• For now, I will assume that the young Sun
was indeed faint
• This has big implications for planetary
climates, as first pointed out by Sagan and
Mullen (1972)…
The faint young Sun problem
Te = effective radiating temperature = [S(1-A)/4]1/4
TS = average surface temperature
Kasting et al., Scientific American (1988)
• Feedbacks are important in the climate
system
• One way to describe these is with
systems diagrams 
Systems Notation
= system component
= positive coupling
= negative coupling
Positive Feedback Loops
(Destabilizing)
Water vapor feedback
Surface
temperature
Atmospheric
H2O
(+)
Greenhouse
effect
The faint young Sun problem
More
H2O
Less
H2O
• The best solution to this problem is higher concentrations
of greenhouse gases in the distant past (but not H2O, which
only makes the problem worse)
Greenhouse gases
• Greenhouse gases are gases that let most of the
incoming visible solar radiation in, but absorb and reradiate much of the outgoing infrared radiation
• Important greenhouse gases on Earth are CO2, H2O,
and CH4
– H2O, though, is always near its condensation temperature;
hence, it acts as a feedback on climate rather than as a
forcing mechanism
• The decrease in solar luminosity in the distant past
could have been offset either by higher CO2, higher
CH4, or both. Let’s consider CO2 first 
The carbonate-silicate cycle
(metamorphism)
• Silicate weathering slows down as the Earth cools
 atmospheric CO2 should build up
• This is probably at least part of the solution to the faint
young Sun problem
Negative feedback loops
(stabilizing)
The carbonate-silicate cycle feedback
Rainfall
Surface
temperature
(−)
Greenhouse
effect
Silicate
weathering
rate
Atmospheric
CO2
CO2 vs. time if no other greenhouse
gases (besides H2O)
J. F. Kasting, Science (1993)
• In the simplest story, atmospheric CO2 levels should have
declined monotonically with time as solar luminosity increased
• Geochemists, however, insist on making the story more
complicated…
Sheldon’s pCO2 estimate from
paleosols
• First published estimate of
Archean CO2 from paleosols
came from Rye et al. (1995)
(probably incorrect)
• Sheldon presented an
alternative analysis of
paleosols based on mass
balance arguments
(efficiency of weathering)
• Sheldon’s calculated pCO2
at 2.2 Ga is also about 0.008
bar, or 28 PAL, with a
quoted uncertainty of a factor
of 3
• New estimate for pCO2 at
2.7 Ga (Driese et al., 2011):
10-50 PAL
Driese et al.,
2011
N. Sheldon, Precambrian Res. (2006)
Rosing et al.: CO2 from BIFs
von Paris et al.
Ohmoto
Rye et al.
Sheldon
(old)
(pCO2  3 PAL)
J. F. Kasting
Nature (2010)
• More recently, Rosing et al. (Nature, 2010) have tried to place even more
stringent constraints on past CO2 using banded iron-formations (BIFs)
• I have argued with these estimates, but I’ll let the authors present them
first, and then we can argue..
Sagan and Mullen, Science (1972)
• Sagan and Mullen liked ammonia (NH3) and methane (CH4)
as Archean greenhouse gases
• As a result of Preston Cloud’s work in the late 1960’s, they
were aware that atmospheric O2 was low on the early Earth
• But Sagan and Mullen hadn’t thought
about the photochemistry of ammonia

Problems with Sagan and
Mullen’s hypothesis
• Ammonia is photochemically unstable with respect to
conversion to N2 and H2 (Kuhn and Atreya, 1979)
• There may be ways to salvage this hypothesis, or part
of it, by shielding ammonia with fractal organic haze
(see talk by Feng Tian)
CH4 as an Archean greenhouse
gas
• CH4 is a better candidate for providing
additional greenhouse warming
• Substrates for methanogenesis should have
been widely available, e.g.:
CO2 + 4 H2  CH4 + 2 H2O
• Methanogens (organisms that produce
methane) are evolutionarily ancient
– We can tell this by looking at their DNA
Feedbacks in the methane
cycle
• Furthermore, there are strong feedbacks in
the methane cycle that would have helped
methane become abundant
• Doubling times for thermophilic methanogens are shorter than for mesophiles
• Thermophiles will therefore tend to
outcompete mesophiles, producing more CH4
and further warming the climate
CH4-climate positive feedback
loop
Surface
temperature
(+)
CH4
production
rate
Greenhouse
effect
• Methanogens grow faster at high temperatures
But,
• If CH4 becomes more abundant than
about 1/10th of the CO2 concentration, it
begins to polymerize 
Titan’s organic haze layer
• The haze is formed
from UV photolysis of
CH4
• It creates an antigreenhouse effect by
absorbing sunlight up in
the stratosphere and reradiating the energy
back to space
• This cools Titan’s
surface
Image from Voyager 2
Possible Archean climate
control loop
CH4
production
(–)
Surface
temperature
Haze
production
(–)
CO2 loss
(weathering)
Atmospheric
CH4/CO2
ratio
CH4/CO2/C2H6 greenhouse with haze
(Old—Rye et al.)
Late Archean Earth?
Driese et al. (2011)
(10-50 PAL, 2.7 Ga)
Water freezes
• The Archean climate system would arguably have stabilized in the regime
where a thin organic haze was present
• We don’t quite make it to 288 K, using pCO2 from Driese et al. (2011)
• Possible additional warming from higher pN2 or from albedo feedbacks?
J. Haqq-Misra et al., Astrobiology (2008)
• In any case, the climate crashed when
pO2 rose at 2.4 Ga, wiping out much of
the methane greenhouse and triggering
glaciations 
Huronian Supergroup (2.2-2.45 Ga)
High O2
Redbeds
Glaciations
Low O2
S. Roscoe, 1969
Detrital
uraninite
and pyrite
Conclusions
• The Sun really was ~30% dimmer during its early
history, at least after the first 200 m.y.
• CO2, CH4, and C2H6 may all have contributed to the
greenhouse effect back when atmospheric O2 levels
were low
– Higher pN2 and lower planetary albedo could also have
helped warm the climate, but these mechanisms (in my
view) are more speculative
• High atmospheric CH4/CO2 ratios can trigger the
formation of organic haze. This has a cooling effect.
– Stability arguments suggest that the Archean climate may
have stabilized when a thin organic haze was present
• The Paleoproterozoic glaciation at ~2.4 Ga may have
been triggered by the rise of O2 and loss of the
methane component of the atmospheric greenhouse