(last updated: Sep 11, 2008)
Paleoclimate is the study of the climate of the past. This
enables us to understand present climate change with the
perspective of what
has happened before. This page examines the history of climate
starting from the present time, and moving towards longer time periods
in the more distant past. In general, we are presently in a
relatively cold period. For most of the past it has been
warmer with little or no polar ice, with carbon dioxide
levels also higher than today. From this perspective,
human-caused changes to the climate up until now appear to be small
compared to what the Earth has already experienced. The present rate of change, if
it continues, is much higher than anything experienced in the past.
Time periods: 25 yr
, 150 yr
, 100 Kyr
, 650 Kyr
Global Warming Case Study: The Paleocene Eocene Thermal Maximum
The Eemian Interglacial -
Lessons for Today?
and the Ice Ages
Cooling: The Younger Dryas Event
Projections for the 21st Century
measured over the course of a day, a week, or even a year.
the average of the weather that occurs over
longer. A common mistake is to confuse a short term trend in the
weather with a change in climate. For example, the chart
the annual average temperature for the province of
can vary by a degree or two each year, and there are periods of a
couple of years when the temperature is relatively high or low.
The change in the long term climate is shown by the blue
The climate is gradually getting warmer, but only by 0.06
per decade, which is far smaller than the annual
variation. The average temperature can vary more in a year
than the climate did in an entire
century. But over a long period of time a slow but steady
rate of warming will become significant.
A typical prediction for the impact of global warming is a rise in
temperature of 3° C by the end of the 21st
figure is not only an average over time, it is also an average over the
entire surface of the Earth, including the oceans. Different
regions will experience different amounts of
warming, so some regions will experience more warming than the global
average. The map below (on the left) shows the annual average
the world at the present time, with each colour representing a
difference of 5° C.
A first approximation of the impact of five degrees of warming
would be to increase the temperature of each colour zone by
degrees. Or another way to look at it is your future climate will be
much like the present climate in the next colour to the south. In
reality, a changing
climate affects wind and ocean currents, which means the actual
changes will be more complex.
The middle graph shows there is a temperature gradient of 1° C per 145
km. If 3° C of
warming is expected in the next 100 years, then each temperature band
northward (in the northern hemisphere) by 3 km per year, on average.
The map on the far right shows the average annual insolation, with each
band representing 20 watts per square meter. These bands are roughly
equivalent to the 5° bands on the left image. Insolation
more near the equator than temperature, reflecting the fact that the
main driver of weather is the transfer of heat from the topics toward
Source for first three images: Global
This chart shows the global average temperature for the past quarter
century. On the monthly or yearly scale it is more
of weather rather than climate. Whereas on a local scale
the annual temperature (averaged over the year) can vary by a
degree or two, we can see that on a global scale it only changes by
about 0.2 degrees. Temperature peaks tend to correspond to El
Nino years, except
when there is a major volcaono. The five year average is
beginning to detect climate change, showing a
steady upward trend. This graph and the one below
show that temperatures are increasing linearly (at
C per decade) rather than
exponentially, at least for now.
The chart below (generated from this NASA site)
illustrates the global distribution of warming between
1970 and 2000. A ten year interval around the start and end
dates was used to
smooth out annual variations. Observe that most of the
occurs in the northern hemisphere, especially in the Arctic.
oceans, covering most of the planet, only warmed by 0.35°
C, while at 45° N (where most people live) warmed by
C and the high Arctic warmed by 1.2 degrees.
The following two charts show the trends in global air temperature for
the last 150 years, covering
the period of industrialization. On the temperature chart three
trends are apparent. In the first half of the century
significantly, even though anthropogenic
greenhouse gas levels were relatively low. In the middle
1940 and 1970 temperatures dropped slightly even though greenhouse gas
continued to rise. There has been a steady
temperature increase since 1970, for which the only explanation is
increased greenhouse gases. Overall there has been
C to 0.8°
C of warming. [ref]
The lower chart shows the distribution of the temperature changes for
past century. The vertical axis represents the average temperature for
each latitude. One can see that the warming in the 1930's was not
global, it was a regional warming in the Arctic with global
consequences. The 1940-1970 cooling can be seen as a return to the
normal climate. This pattern contrasts with the warming after 1980,
which is much
more global in scope.
Although the mid century cooling appears to be a notable feature of the
temperature chart, the real question is the cause of the warming in the
early part of the century. Reduce that warming, and the apparent
cooling disappears. Greenhouse gas emissions were relatively low, and
counter balanced by increasing aerosol
causes the reflection of incoming sunlight. The warming has been
attributed to increased
but why did temperature rise most in the one region (the Arctic) where
it shines the least? The explanation may be a regional climate
fluctuation largely unrelated to anthropogenic influences.
According to the 2007 IPCC report [AR4
WG1 Ch03 188.8.131.52],
the peak warmth in the early 1940s is likely to have arisen partly from
closely spaced multiple El Niño events, and also due to the warm
phase of the Atlantic Multi-decadal Oscillation (AMO). The
prolonged 1939–1942 El Niño shows up as a warm
interval. In the Atlantic, the warming from the 1920s to about 1940 in
the NH was focussed on higher latitudes, with the SH remaining cool.
The recent strong warming appears to be related in part to the AMO in
addition to a human caused global warming signal. There were
major volcanic eruptions in 1951 and 1963, corresponding to the
downward temperature spikes on the graph.
anomalies of global land-surface air temperature
(°C), 1850 to 2005, relative to the 1961 to 1990 mean for CRUTEM3
updated from Brohan et al. (2006). The smooth curves show decadal
variations. The black curve from CRUTEM3 is compared with those from
NCDC (Smith and Reynolds, 2005; blue), GISS (Hansen et al., 2001; red)
and Lugina et al. (2005; green).
2007 WG1 Ch3
Because warmer air can hold more water vapor, one effect of global
warming is increased precipitation levels. The following
show that is indeed the case for the warming during the 20th century.
The only major region with less rainfall was central Africa. But note
that warmer temperatures also increase evaporation, so the net effect
of increased moisture may be less than it appears on this chart.
The chart below shows temperatures for the last thousand years, from
left to right.
There are no accurate records from instruments for this time
period, so temperatures must be calculated using indirect
evidence. For example, the width of tree rings can tell us
was like while the tree was growing. Several reconstructions are shown
in this graph, which differ from each other but follow
same pattern. After the "Medieval Warm Period" of 1000 years
the world entered the "Little Ice Age", followed by a rise in
temperatures in the 20'th century [ref].
Below it is a chart of the solar cycle
for the past 1,100 years, based on Carbon-14 isotope
measurements. The varying temperatue of the sun is probably
one of the causes
climate change during this time. But note that the Medieveal
period identified on the temperature
chart lags that in the solar chart by a century. We can see
the present period of global warming coincides with a high point in the
solar cycle. There may have been some solar impact on the warming in
the first half of the 20th century, but there has been no significant
change in solar radiation in the second half when the human caused
global warming signal becomes apparent.
||2008 Reconstruction of the last 1000 years based on
multiple proxies, from this RealClimate
||Solar cycle based on Carbon-14 isotope
This graph covers part of the current (Holocene) interglacial
includes all of human civilization. Following the sudden end
the Younger Dryas (a short period of intense cooling), the Arctic
entered several thousand years of
conditions that were warmer
and probably moister than today. This early Holocene warm
appears to have been punctuated by a severe cold and dry phase about
8200 years ago, which lasted for less than a century, as recorded in
the central Greenland ice cores. Eight different estimates
shown, indicating a lot of uncertainty, but agreement on the general
pattern. Recently tempertures have risen rapidly, but are
still within the same range as the recent past. However, it
global temeratures continue to rise at the rate experienced in the past
few decades, temperatures could exceed any experienced during the
The unusual stability of the Earth’s climate during the Holocene
probably is due to the fact that the Earth has been warm enough to keep
ice sheets off North America and Asia, but not warm enough to cause
disintegration of the Greenland or Antarctic ice sheets. [ref]
Records of NH temperature variation during the last 1.3
Overlap of the published multi-decadal time scale uncertainty ranges of
all temperature reconstructions identified in Table 6.1 (except for
RMO..2005 and PS2004), with temperatures within ±1 standard
error (SE) of a reconstruction ‘scoring’ 10%, and regions
within the 5 to 95% range ‘scoring’ 5% (the maximum 100% is
obtained only for temperatures that fall within ±1 SE of all 10
reconstructions). The HadCRUT2v instrumental temperature record is
shown in black. All series have been smoothed with a Gaussian-weighted
filter to remove fluctuations on time scales less than 30 years;
smoothed values are obtained up to both ends of each record by
extending the records with the mean of the adjacent existing values.
All temperatures represent anomalies (°C) from the 1961 to 1990
mean. From IPCC
2007 WG1, ch 6.
The chart below shows average annual temperatures in Greenland, using data
from an ice core. The temperature varies by more
than three degrees during this period, with the present time at the low
These three maps of Africa below illustrate how changing temperature
affects rainfall and vegetation patterns. The map on the left is
Africa of today. The middle map is Africa at the peak of the
ice age, about 20,000 years ago. The cooler climate resulted in
expansion of the Sahara desert, and the tropical rainforest almost
disappeared. The map on the right is during the Holocene climatic
optimum, which was warmer than today. The Sahara desert is almost
gone, covered by grassland, and the tropical rainforest is much
larger. This illustrates the general principle that a warmer
climate is a
wetter climate. However, there is another factor that influenced
precipitation in Africa 7,500 years ago.
The Earth's orbit is elliptical, causing it to be closer to the Sun for
half of its orbit. Today, that occurs during northern winter,
causing relatively milder winters and cooler summers. But due the
the precession cycle
the Earth was closer to the sun during northern summer 7,500 years ago.
This difference affected the pattern of monsoon winds that bring
subtropical regions such as the Sahara [ref]. Therefore
it may not be
accurate to assume global warming will produce the same rainfall
pattern. See also this information about the changing climate of
Africa during this time period.
The chart below shows the temperature for the last hundred thousand
years, covering the last ice age, as measured from an ice core in
Greenland. During the ice age, the global average temperature
five or six degrees Celsius colder than today. Half of North
America, including all of Canada, was covered by an ice cap up to 3
kilometers thick, as was a large part of Europe and Russia. In
general, ice ages have lasted about one hundreed years, followed by a
warmer interglacial period lasting about ten or twenty thousand years.
We are presently in the middle of an interglacial period.
A series of rapid warm and cold oscillations, called
Dansgaard–Oeschger events, punctuated the last
often taking Greenland and northwestern Europe from a full-glacial
climate to conditions about as warm as at present.
changes of up to 16º C can take place in few decades.
interstadials lasted for varying periods of time, usually a few
centuries to about 2000 years, before equally rapid cooling (associated
with large amounts of glacial ice discharged into the ocean) returned
conditions to their previous state. Heinrich events are
sudden, brief cold events that seem to have occurred very frequently
over the past 115 000 years, during times of decreasing sea surface
temperatures in the form of brief, exceptionally large discharges of
icebergs in the North Atlantic from the Laurentide and European Ice
Sheets. It is believed that the ice caused the North Atlantic current
to slow down or stop, leading to the sudden cooling.
A few thousand years after the apparent recovery from the ice
age (the Bølling-Ållerød warming),
the Arctic was suddenly (in less than 100 years) plunged back into a
new and short-lived cold event known as the Younger Dryas. This
lasted for 1300 years, and ended very abruptly
with central Greenland temperatures increasing by 7
more in a few decades. This was followed by the much more
interglacial period known as the Holocene, which we are presently still
Each cycle begins with a warming event, which causes the ice to
retreat, leading to less reflection of sunlight and further
warming. Ice caps may grow larger because the warming bring
snowfall in winter which may exceed the extra melting in
summer. At some point there is an ice cap collapse, which
amounts of icebergs into the ocean. These directly reflect
sunlight, but more importantly they weaken the North Atlantic current
that brings the warmth. A reverse cycle of increased ice
and lower temperatures then sets in.
The Greenland data has been extended back in time to cover the Eemian
interglacial by substituting date from Antarctica. This is not entirely
valid, as the Milankovitch forcings in the southern hemisphere lag
behind those in the northern hemisphere by 10,000 years. The
chart at the bottom shows seasonal temperature anomolies at the
latitude of 70º North, which are a result of the Milankovitch
It can be seen that periods of relatively warm temperatures
correspond with strong springtime warming.
The charts below are based on detailed information obtained from two
deep ice cores from Antarctica, which includes ice that is up to
years old. Temperatures and carbon dioxide levels can be
from pockets of air trapped in the ice, and a year by year record can
be produced. We can see the ice age cycle of about 100,000
cold followed by about ten thousand years of relatively warm
interglacial, which we are in now (year zero). Note that the present
interglacial period is not quite as warm
as those before it, but is not decending back into the glacial stage as
quickly. Carbon dioxide levels closely track but slightly lag
behind temperatures. Carbon dioxide
is released from the ocean as temperatures rise, so it behaves
as a positive feedback to the orbital forcing that drives the climate.
Observe in the top chart that the amplitude of the cycles is smaller
before 400,000 years ago, conforming the the pattern of intensifying
cycles shown above.
Note that carbon dioxide levels vary from 180 ppm (parts per million)
during ice ages to 300 ppm during interglacial periods. This
ppm increase takes on the order of 10,000 years, while today we have
increased carbon dioxide levels by 100 ppm in 100 years, a rate one
hundred times as fast.
The chart in the bottom corner shows solar insolation amounts caused
at latitude 80º S over the course of a year. Note
temperature peaks on the graph correspond to the warm periods during
the southern spring.
Here is more detail on the final descent into the current ice
age from the warmer and more stable climate of three million years
ago. Average temperature is falling, and climate fluctuation
becoming more intense.
This period covers the evolution of humans from their common
chimpanzees. Before a million years ago there was a 41,000 year cycle
extreme ice age conditions and a shorter interglacial thaw.
then there has been a more intense 100,000 year cycle.
The 41.000 year cycle is still present, but the declining average
temperature began to prevent a return to interglacial conditions. See
the discussion on Milankovitch
Cycles for the significance of these cycle times.
The Pliocene epoch lasted from 6 to 1.8 million years ago, spanning the
right half of the graph above. The table below compares
vegetation today with the mid-Pliocene (about 3 million years ago),
when the global average temperature was 3°
C warmer than today. The CO2
amount ~3 My ago has been estimated as 380-425 ppm (Raymo et al. 1996) [ref].
Some climate models predict that much warming by
the end of the 21st century. Observe that there is no Arctic sea
ice in summer, only the central part of Greenland has an ice cap, and
that desert regions are smaller than today. A warmer climate is
usually a wetter climate, but it is not entirely valid to assume the
present global warming will lead to the same result, at least not right
away. The paleoclimate data is averaged over a long period of time from
in equlibrium, while the near future will be a climate in
transition. There are several reasons while it may take some
equilibrium rainfall levels
- The ocean does not warm as quickly as the land.
levels are largely determined by sea surface temperature, which lags
behind land temperature. A warmer landmass on its own tends
increase the evaporation rate, which could lead to drying.
- Rainfall is also influenced by existing
vegetation. Increased vegetation generates more rainfall, in a
feedback loop. But it takes time for vegetation to establish
- Rainfall in sub-tropical regions changes in a 21,000 year
precession cycle. The data shown below is an average of the wet and dry
cycles. We are presently in the dry portion of the cycle.
The chart below illustrates how in the past 65 million years (the age
of mammals after the
mass extinction which included the dinosaurs), global temperatures have
gradually fallen, and become more unstable. Click here to see how the world
looked at the time. The Arctic Ocean
been permanently frozen only for the past 3 million years.
that the older, warmer temperatures can be estimated because there was
no permanent ice, and the recent termperatures can be calculated
we have geological means to estimate the size of the ice caps.
The section in the middle is more uncertain. The
labeled PETM on the right side of the graph (55 million years ago) is a
period of rapid global warming known as the Paleocene Eocene
Thermal Maximum, described in more detail below.
The graph below shows temperatures for last half billion years,
starting on the right with the Cambrian
at the present time on the left. (Unfortunately, different
go in different directions.) This is the time period for
which there has been animal life in the oceans and (most of it) plants
on land, so the ecology is roughly equivalent to today.
climate is constantly changing, and that we have entered one of the
coldest glacial periods.
How we can Calculate Temperatures
for the Deep Past
There are no direct
measurements of temperature in the past, so scientists must reconstruct
from indirect evidence (or "proxies"). In this case the ratio
isotopes of oxygen found in the shells of fossil microscopic
animals (benthic foraminifera) are used [ref].
The element oxygen occurs
isotopes: the common isotope 16O
the heavier rare
(0.1995%). When ocean
escapes more easily than the 18O,
resulting in a higher concentration of 18O. When the water is warmer, the
molecules are moving faster, so the
difference is less. Therefore colder water retains relatively
than warmer water. The
foraminifera incorporate that oxygen into their shells,
accumulate on the ocean floor after they die. We can estimate
water temperature by
ratio of 18O
(referred to as d18O) in these fossil shells.
When water condenses from the atmosphere as rain or snow, the
precipitation has a higher d18O,
because the heavier molecule condenses more easily. Rain that
falls inland is more depleted in 18O
than rain in coastal areas,
because some of it evaporated from the land surface, where it was
already isotopically depleted. This effect is strongest in Antactica. The present ice sheets
thus strongly depleted in 18O
as compared to ocean
water. Bigger ice sheets mean higher d18O
in the ocean. This
conflicts with fact that colder water also has a higher d18O. So when there are ice
we cannot calculate the water temperature from d18O
unless we also
know the volume of
||Click on these symbols to view a
map of the Earth for that period, or its place on the evolutionary
This graph presents several estimates of carbon dioxide levels for the
same time period as above, and show major glaciations that occured.
Generally they correspond with
temperature (see the Greenhouse
For most of this time period carbon dioxide levels have been
far higher that those of today, or
any that are likely to occur in the near future because of
industrialization. However, we cannot detect how rapidly
carbon dioxide levels changed in the past. See the Global Carbon
Cycle for more information on the role of carbon dioxide.
||Atmospheric CO2 and
continental glaciation 400 Ma to present.
Vertical blue bars mark the timing and palaeolatitudinal extent of ice
sheets (after Crowley, 1998). Plotted CO2
records represent five-point
running averages from each of the four major proxies (see Royer, 2006
for details of compilation). Also plotted are the plausible ranges
from the geochemical carbon cycle model GEOCARB III (Berner and
Kothavala, 2001). All data have been adjusted to the Gradstein et al.
(2004) time scale. From IPCC
2007, WG1, Ch 6.
For details on the early Permian glaciation (265 to
305 Mya), see [Science
Sep 07]. See also the Evolution Timeline
for the relationship of carbon dioxide and mass extinctions.
The graph below show that oxygen levels tend to be
high when carbon dioxide levels are low.
||The atmospheric O2
curve is taken from (23).
The upper and lower boundaries are estimates of error in modeling
atmospheric O2 concentration. The numbered
intervals denote important evolutionary events that may be linked to
changes in O2 concentration (see text
2007 Apr 27)
Eocene Thermal Maximum was a period of rapid global warming that
occured 55 million years ago. At the time, the global average
temperature was about 5°
C warmer than today, and there were no polar ice caps. A
amount of carbon dioxide entered the atmosphere in a geologically short
amount of time, and the global average temperature rose by another
C. This warming event lasted about 170,000 years. It
appears that 2000 to 4000
gigatonnes (Gt) of carbon was added to the atmosphere in a period of
10,000 to 20,000 years, during which surface temperatures rose by up to
Estimates of how much warming occured:
Estimates of how much carbon entered the atmosphere:
- Deep ocean warming of about 4–6°
C has been inferred in many cores taken from the sea floor. Estimates
of surface warming from foraminiferal oxygen isotopes are from
C to 8°
C in the high latitudes, 1–4°
C in the subtropics, and 1–2°
C in lower latitudes [NASA
- Sea surface temperature rose by 5°
C in the tropics and as much as 9°
C at high latitudes, whereas bottom-water temperatures increased by 4
- The initial SST rise was rapid, on the order of 1,000
although the full extent of warming was not reached until some 30,000
years later. [Science
- 1500–2000 Gt of carbon over 10–20 kyr
added to the
system. Detailed analysis of the PETM carbon isotope
[Bains et al., 1999] suggests that the initial decrease occurred in
three distinct steps over 20,000 years, each corresponding to an input
of about 600, 500, and 300 Gt of carbon within about 1000 years.
The effect on the ocean:
- Previous estimates put the
released carbon at 2 Gt, but Zachos showed that more than twice that
amount--about 4.5 Gt--entered the atmosphere over a period of 10,000
June 2005]... It
took 100,000 years after the PETM for carbon dioxide levels in the air
and water to return to normal.
- Estimates for CO2
levels at the time range from 1125
to 3000 ppm [Science
Sep 2006] (compared to 380 ppm today), while for CH4
a range of
7–10 ppm has been estimated.
- Estimates for pre-PETM atmospheric CO2
concentrations range from 600 to 2800 parts per million. Starting
from these conditions, an increase of 750 to 26,000 ppm of
would be required to account for an additional 5°C rise in
temperature, which implies an addition of 1500 to 55,000 PgC to the
atmosphere alone. Sustaining this concentration for tens of
thousands of years implies
partial equilibration with the carbonate system in the ocean,
indicating a total release of 5400 to 112,000 PgC, with 3900 to 57,000
PgC of released carbon residing in the
ocean (and with additional carbon supplied by the dissolution of
carbonates). The extraordinary magnitude of these estimates is evident
when compared against the 5000 PgC estimated for conventional fossil
fuel resources available today. [Science
Dec 8 2006]
Effects on Biology:
- A relative sea-level fall (~30
m) immediately preceded the late Paleocene thermal maximum, during
sea-level rose again by ~20 m. This rise may have been eustatically
possibly through a combination of thermal expansion of the oceanic
column and melting of unknown sources of high-altitude or polar ice
in response to global warming. [ref]
- Excessive carbonate undersaturation of the deep ocean would
likely impede calcification by marine organisms and therefore is a
potential contributing factor to the observed mass extinctionof benthic
foraminifera at the time [Science
June 2005]. See the section on Ocean
Was it caused by a massive methane release?
- A mass extinction of benthic foraminafera (due to ocean
acidification) and a global expansion of subtropical dinoflagellates at
the earliest onset of the event.
- An increase in both origination and extinction of
calcareous phytoplankton, but little overall change in biodiversity [Science
Dec 15 2006].
- The appearance of new orders of mammals, including primates.
- A transient dwarfing of mammal species, and the migration
of large mammals from Asia to North America. [Science
Dec 8 2006]
- Only one source of carbon that is isotopically light and
large enough quantities has been pinpointed so far, this is the
reservoir of methane hydrate deposits buried on the
continental shelves of the oceans. [NASA
Today, methane hydrates are
stored along continental margins (i.e. at intermediate water depths,
250 m to several thousand meters water depth), where they are
by water pressure and temperature. Methane hydrates may become unstable
under influence of ocean warming or slope instability. The
estimated present-day reservoir of carbon stored in methane hydrates
is about 10,000 Gt. [ref]
- Methane is roughly 20 times more effective a greenhouse gas
dioxide. However, the residence time for methane in the
atmosphere, at present concentrations, is ten years, after which it is
converted to carbon dioxide. As methane concentrations
the residence time increases.
- This event can be approximated by
assuming 1500 Gt of
gradually added for 10,000 years, which amounts to 0.3 Gt per year.
To put this in perspective, today there is an estimated 5000
of carbon in fossil fuels, of which we are adding 7 Gt per
year. Given methane is 20 times as effective, in terms of
forcing, these values are very roughly equivalent.
- Seabed methane hydrates form from methane produced at depth
seabed under warm conditions by methanogenic bacteria, rising toward
the sea floor until it reaches the region of temperature (near
freezing) and pressure for the formation of methane hydrate. The
corollary is that methane hydrates in the seabed will always be near
their critical point. As an ice age deepens and sea levels drop, they
will finally reach the point where sea floor pressures are low enough
that methane hydrate can evolve methane. At that point sea bed slopes
can destablize, unloading more hydrates and releasing more methane,
producing a methane spike that turns the ice age around. [RealClimate
But maybe it was not caused by methane:
How can we tell it happened?
- Evidence that the Paleocene-Eocene Thermal Maximum was
with a large clathrate release. They
preferred the explanation that a large igneous province intruded on a
large coal formation to produce CO2
and methane (e.g. Svenson et al.
(2004) Nature 429, pp 542)... the temperature change and pH change
seems to require more CO2
than the isotope spike will give you, if it's
- The Paleocene-Eocene Thermal Maximum was triggered by the
massive flood basalt eruption that occurred when the North Atlantic
ocean opened over the ancestral Iceland hot spot.
The temperature rise was caused by greenhouse gas release during magma
interaction with basin-filling carbon-rich sedimentary rocks proximal
to the embroyonic plate boundary between Greenland and Europe.
- The size of the methane hydrate reservior was much smaller
today, possibly only 2000 Gt. This is not enough to account
the amount of warming that occured. [Science
- Regarding the Paleocene-Eocene event itself, the field
seems to be
coming around to the idea that it had something to do with a greatly
accelerated oxidation of organic carbon stored on land, perhaps
associated with the drying up of interior shallow seaways.
July 2008] claims that sea surface temperatures during the
Eocene were between 35 to 40°
C, much higher than the commonly accepted 30°
C. These temperatures are above the thermal tolerance for tropical
vegetation, where they die because photorespiration dominates over
photosynthesis. The extra carbon in the atmosphere during the PETM may
have come from dead tropical vegetation, and that the terrestrial
carbon pool could have been much higher than the 6,000 Gt of carbon
today. But where is the evidence for tropical deserts during this time?
Although referenced in this paper, [Science
Mar 2006] clearly shows that tropical biodiversity is positively
correlated with temperature, not compatible with tropical
reports on the abundance of hotspots of biodiversity in the Eocene
tropics, showing that these hotspots correlate with the collision of
continents rather than temperature.
This was accompanied by an
exceptionally large change to the global
carbon cycle as indicated by a large drop in the isotopic ratio of
13C to 12C
in the ocean
and on land. The ratio of these carbon isotopes generally changes as a
of biological activity, since carbon in living matter tends to be
preferentially made up of 12C
as opposed to 13C.
Thus an increase in biological activity "uses up" more 12C,
and therefore the ratio of 13C
in the remaining carbon increases. Conversely, a decrease of biological
uptake, leads to a decrease of the isotopic ratio (i.e., it gets
"lighter"). The change at the PETM was so large that it would
decrease in biological activity equivalent to roughly three times the
total present-day terrestrial biosphere. In other words, if all of the
terrestrial carbon today (in forests, animals, soils, etc.) were
converted to carbon dioxide and returned to the global inorganic carbon
pool, the change in the global carbon isotopic ratio would only be a
third as big as that observed during the PETM.
After the PETM: "pCO2 ranged
between 1000 to 1500 parts per million by
volume in the middle to late Eocene, then decreased in
several steps during the Oligocene, and reached modern levels by the
latest Oligocene. The fall in pCO2
likely allowed for a critical expansion of ice
sheets on Antarctica and promoted conditions
that forced the onset of terrestrial C4
See the Paleocene-Eocene Thermal Maximum in context on the Evolutionary Timeline
The climate of the last interglacial (LIG) period, from 129,000
to 118,000 years ago (1,
was slightly warmer than today's and is often
viewed as an analog of the climate expected during the
next few centuries. Recent assessments of the LIG climate have
provided strong evidence that sea level was 4 to 6 m above the
present level, due to partial melting of both Greenland and
the West Antarctic Ice Sheet (WAIS) (3,
At peak interglacial conditions, summer
temperatures were 2° to 5°C warmer than
today in the North Atlantic (5)
over Greenland (6)
and the Arctic (7).
The Norwegian-Greenland Sea experienced large variability,
but during the warmest period, the Arctic oceanic front
was located west of its present location (8).
Consequently, the Arctic climate was warm
enough to explain the shrinking of the
Greenland Ice Sheet during the LIG (9).
In the Southern Hemisphere, an 2°C warming occurred over the
Antarctic Plateau during the LIG (10),
but it could not have resulted in any melting
because local air temperature was still
extremely cold (–50°C). In the
Southern Ocean, summer sea surface temperatures
were about 2°C higher than during the
Over New Zealand and Tasmania, the LIG warming
was between 0° and 2°C (13,
Such increases in surface water or air
temperature seem too small to have resulted in
substantial melting of the WAIS (15).
||Maximum observed LIG summer temperature anomalies
relative to present
derived from quantitative [terrestrial (circles) and marine
(triangles)] paleotemperature proxies as part of CAPE Last Interglacial
Rapid Global Cooling - The
Younger Dryas Event
The chart below shows the observed global average temperature changes
for the 20th century, and projected changes (using the most recent
climate models) for the 21st century. Projections depend on
"emissions scenario", meaning how much greenhouse gases actually get
added to the atmosphere in this century. This, of course, is not
known, so several scenarios (as defined
by the IPCC) are used. As for the three shown here:
See also Estimating
Future Sea Level
is simply unrealistically high. It assumes emissions
keep rising at the same rate throughout the entire 21st century, with
no new technology adapted and continuous economic growth. CO2 levels of 800 ppm and
C of warming is predicted for this worst case.
is a reasonable high end scenario, with emissions rising at
the same rate as now until mid-century and then declining slightly
thereafter. It implies the economic growth of the 20th
continues and little is done to control emissions until new
technologies are adopted later in the century. CO2 levels of 650 ppm and
C of warming is predicted for this scenario.
is a realistic but optimistic scenario where emission rates
rise more slowly until mid century, then decline. It suggests
that some action is taken to reduce fossil fuel emissions. CO2 levels of 570 ppm and a
little more than 1°
C of warming is predicted for this scenario.
- The "Alternative Scenario" is based on a strong but
of measures to reduce fossil fuel use proposed by James Hansen.
This keeps global warming below an increase of
The chart below, from the IPCC
2007 Summary for Policy Makers,
shows the surface temperature of the Earth in 30 and 100 years from
now, as determined by a set of climate models, for three emissions
scenarios (described above). The graphs on the left shows the
probability range for the global average temperature in these two time
periods. For example, in the A1B scenario, the most likely
temperature increase in 30 years is a bit less than one degree, and
less than 3 degrees in 100 years.
Projected surface temperature changes for the early and
21st century relative to the period 1980–1999. The central
right panels show the Atmosphere-Ocean General Circulation multi-Model
average projections for the B1 (top), A1B (middle) and A2 (bottom) SRES
scenarios averaged over decades 2020–2029 (center) and
2090–2099 (right). The left panel shows corresponding
uncertainties as the relative probabilities of estimated global average
warming from several different AOGCM and EMICs studies for the same
periods. Some studies present results only for a subset of the SRES
scenarios, or for various model versions. Therefore the difference in
the number of curves, shown in the left-hand panels, is due only to
differences in the availability of results.
The chart below (also from IPCC
2007) shows the predicted change in precipitation levels in
the next one hundred years, in (northern) winter and summer.
There is far more uncertainty here than for temperatures.
As can be seen from the caption, the "stippled" areas mean
that 90% of the models agree which way rain levels will change, not by
how much. The rest of the map has even less agreement. The
general pattern is increased precipitation in the temperate and polar
areas, and decreased rain in sub-tropical regions.
Relative changes in precipitation (in percent) for the
2090–2099, relative to 1980–1999. Values are
averages based on the SRES A1B scenario for December to February (left)
and June to August (right). White areas are where less than 66% of the
models agree in the sign of the change and stippled areas are where
more than 90% of the models agree in the sign of the change.
See the larger context for paleoclimate on the Evolutionary Timeline.
to the Climate Change Main Page