The earth has seen a dramatic change in its climate since the Cretaceous. Before the dinosaurs were wiped out, by whatever cause, the avearge climate of the earth was considerably warmer than it is today. Crocodile skeletons have been found in areas of the Canadian arctic that are now have a tundra climate that is inhospitable to all but the hardiest forms of life. The ice caps of the north and south poles did not exist so that the water now contained within them raised sealevels to the extent that most of the central USA was flooded by a vast shallow sea. What caused this massive change that took us from the greenhouse Cretaceous to the icehouse Holocene? In a time when some think that global warming threatens to melt the ice caps once more, this question has more than just academic interest.
The fossil record shows that warm deciduous forests covered much of the land surface and extended to high latitudes. The surface waters around Antarctica may have been as warm as 18oC, with deep waters only a few degrees cooler, whereas today the waters in this area are close to freezing. Deserts appear to have been relatively rare prior to 40 Ma, such that temperature and rainfall seem to have been much more even over the surface of the earth at the start of the Cenozoic.
One of the most quantitative pieces of evidence that allows us to determine both the timing and rate of changes in climate is provided by the oxygen isotope record of benthic foramifera recovered from deep sea sediment cores
| |
| Figure 1: Variation in the oxygen isotope composition of benthic forams from the Atlantic Ocean. Note that the d18O values are relative to PDB in this diagram |
SAQ1 Why do we have a much better record of the temperature of the oceans than we do of the land?
The oxygen isotope record shows that the change from the Cretaceous greenhouse to the Holocene icehouse has not been smooth and gradual, but rather appears to have been characterised by periods of rapid change interspersed with times of relative stability. In addition, examination of the record left by surface dwelling (planktonic) forams tells us that the climatic changes were greatest at high latitudes, with the conditions at the equator 65 Ma being very similar to those of today.
How then can we explain these changes?
This area is very well covered in the by Jim Kennet in his book on Marine Geology. You should look there for a fuller and more detailed explanation of what is presented below.
It is undoubtedly true that the relative position of the continents has played an important role in controlling ocean circulation and, hence, heat transport on the earth's surface.
If you want to see an excellent WWW site that looks at the movement of the earth's plates through geologic time try this link Global Earth History. North American sites are usually busy after 12 o'clock, so it is best to visit in the morning.
At the end of the Cretaceous the continents were distributed more or less as shown in Figure 2.
 |
| Figure 2: Distribution of the continents at the end of the Cretaceous |
There are two major differences between this arrangement and
that we see today.
Firstly, although the Tethys had already contracted significantly there was still open access to the rest of the oceans at both ends of what is know the Mediterranean. North and South America were also separated from one another, hence there was free flow of water around the equator. This equatorial current was considerably warmer than the equatorial currents of today's ocean as the water contained within it spent a significantly longer time along the equator before being fed off into the northern and southern gyres. This also meant that the water within the north and south flowing currents that diverged from the equator towards the poles was also warmer and fed this heat to high latitudes.
The second big difference between the Cretaceous and modern oceans is the lack of a circum-polar current 70 Ma. Today, Antarctica is ringed by a cold surface current that feeds cool water towards the equator and also acts to insulate the South pole from the low latitude warm waters. In the Cretaceous South America was joined with Antarctica and there was only a small (if any) gap between Australia and Antarctica.
The combination of these two factors means that there is a much greater temperature gradient today between the high and low latitudes.
 |
| Figure 3: Distribution of the continents at the end of the Eocene |
By the end of the Eocene a well developed spreading centre
had developed between Australia and Antarctica, such that there was now vigorous
surface (and possibly deep) water flow between the Indian and Pacific Oceans.
Although the gap between South America and Antarctica was still closed this
increase in circulation around Antarctica led to significant cooling throughout
the period from 60-40 Ma. Surface waters in the Southern most Pacific dropped
from about 20oC in the early Eocene to as low as 10oC
by the end of the Eocene. The temperature remained high enough so that there
is no evidence of significant glaciation during this time and overall, the global
climate remained relatively warm and humid.
At 38 Ma there was a sudden in drop in temperature over a period of about 100,000 years that appears to mark the time when Antarctic bottom water first started to form and fill the deep oceans with cold water. This time also marks the initiation of widespread formation of sea ice around Antarctica, with growth of the ice cap and a consequent drop in sea levels. The end of the Eocene is also marked by dramatic changes in terrestrial temperatures and, hence, flora and faunas. In Alaska mean temperatures dropped by around 12oC and over large areas of the Northern Hemisphere there was a change in the dominant vegetation from diverse broad leaved evergreen forests to low diversity deciduous forests.
SAQ2 Why would a change in ocean circulation cause such as global change in climate rather than in the circulation change had just been restricted to the atmosphere (assuming that this was possible)?
 |
| Figure 4: Distribution of the continents at the beginning of the Miocene |
By the beginning of the Miocene some 20 Ma the Drake Passage
between South America and Antarctica was fully open. This allowed circum polar
circulation around the whole of Antarctica, completely isolating the continent
from the warming meteorologic influences of the lower latitude continental land
masses. Increased deep water production in the southern oceans was joined by
a source in the north as the Iceland-Greenland-Faeroe ridge subsided and allowed
the cold waters of the Arctic to spill south and form North Atlantic Deep Water.
The eastern end of the Mediterranean was finally sealed, so that although the
Panama Isthmus remained open, circum equatorial circulation was ended. These
changes to oceanic circulation led to a further steep drop in temperatures and
major growth of the Antarctic ice sheets by the mid-Miocene.
SAQ3 The increased formation of cold deep water in the Southern Ocean coincides with an increase in the productivity of the seas around Antarctica. Why is this?
On land, these climatic changes led to the development of the tundra and taiga style climates at high latitudes while the higher diversity tropical flora and fauna remained at the equator.
Another major tectonic development that took place during the Miocene was the start of uplift of the Himalayas and the Tibetan plateau. The possible influence of this event on global climate will be considered below.
 |
| Figure 5: Distribution of the continents today |
The major climatic change since the Miocene has been the growth
of the Arctic ice sheets of the North Pole and Greenland and the onset of ice
ages. The best date for this growth is around 2.7-3.2 Ma, which comes from the
date of the change in the oxygen isotope composition of forams and from the
first appearence of ice rafted debris in the sediments of the North Atlantic.
The exact cause of the growth of northern hemisphere ice sheets is not certain,
but the timing is approximately coincident with the closing gateway between
the tropical Atlantic and Pacific Oceans by the uplift of the Panama isthmus.
This provided further shortened the transit of equatorial currents along the
warmest zone on the earth and also led to the creation of a more intense Gulf
Stream as the Atlantic portion of this current was diverted north.
SAQ4 If we divert more warm water into the North Atlantic why would this lead to a growth in ice sheets on Greenland?
Today the only major continental ice sheet in the northern hemisphere is on Greenland, but during glacial maxima these icesheets extended over much of the northern half of North America, Europe and Asia.
As mentioned above one of the outcomes of the movement of the continents was the uplift of the Himalayas and the Tibetan plateau as the result of the collision of India with Asia. Other areas of the earth that are thought to have undergone signficant uplift during the Cenozoic are the Peruvian antiplano in South America and the Colorado plateau in North America. What are the impacts (if any) of these events on the climate of the earth during this time? First I will present the arguments that suggest that they have had a significant effect and then I will briefly look at some of the evidence that may counter these hypotheses.
 |
| Figure 6: Winter effects over a raised plateau |
During the winter the elevated plateau cools causing cold air
to sink and flow out from the plateau. This air is dry and cold, so that it
leads to seasonal droughts in the area surrounding the plateau.
 |
| Figure 7: Summer effects over a raised plateau |
During the summer the plateau surface warms and draw in warm
air from the surrounding area. If those winds are drawn in from the oceans,
particularly warm oceans, they will carry moisture which will precipitate as
it rises up the sides of the plateau.
These effects have been entered into general climate models (GCMs) to predict the effects of raised plateaus in particular regions of the earth. GCMs are large models that require very sophisticated and powerful computers to generate patterns of temperature, precipitation, etc.
 |
| Figure 8: Predicted climatic effects of the uplift of the Colorado plateau |
The predictions for North America show that the coast of California
should have become drier and this is in accord with fossil evidence that shows
the disappearence of wet loving plant species and the present semi-desert or
desert conditions of much of this area. The mid-west is also predicted to have
become drier and we know that this area is naturally covered by grassland today
and (until the arrival Europeans) was the haunt of large grazing mammals like
bison. This grassy area appeared to have started to replace more wet-loving
forests about 15 Ma. There is abundant evidence that glaciers have swept through
much of the area that is predicted to have been colder. The warm wet winds that
are brought in off of the Gulf of Mexico have kept the southwestern USA hot
and humid in accord with the model predictions.
 |
| Figure 9: Predicted climatic effects of the uplift of the Tibetan plateau |
For Asia the model also correctly predicts the establishment
of colder tundra like climates over much of Siberia. The predicted cooler climates
of western Europe are offset by the effects of the Gulf Stream, which does not
appear to be so well handled by the GCMs. The drier conditions predicted for
the middle east and northern Africa are borne out by conditions in these areas
today. Similarly, the warm wet winds of the monsoon lead to heavy rainfall in
the areas to the south and souteast of the plateau.
Although most scientists would accept that the effects caused by uplift of the Tibetan and Colorado plateaus are in line with predictions, there is less agreement about the timing of the initiation of these conditions and their relationship with global climatic conditions, such as formation of the northern hemisphere ice sheets.
 |
| Figure 10: Timing of the uplift of the Tibetan plateau |
Figure 10 shows some of the scenarios proposed for the uplift
rate of the Tibetan plateau. A similar range of estimates has been made for
other uplifted regions of the world. The difficulty lies in the way in which
uplift rates are estimated and how these are open to different interpretations.
Two lines of evidence come from erosion patterns. The appearence of deeply incised
valleys and increased Pleistocene sedimentation rates have both been used to
infer increased erosion resulting from recent rapid uplift. The other main line
of evidence comes from palaeobotany studies which show a change in the flora
from relatively warm loving species in the Miocene to alpine type plants during
the Pleistocene. Again, this is taken to be consistent with uplift of the areas
concerned.
SAQ5 From what you know of changes in climate during the past 5 Ma, what might be an alternative explanation for these trends?
Briefly, silicate weathering can be approximated by the following equation;
Is there any evidence for this effect? The answer is a resounding, maybe, and is to be found in the Sr isotope record of the oceans.
 |
| Figure 11: The Sr isotope systematics of river waters |
I published a paper sometime which included a survey of the
Sr isotope systematics of river waters. We found that the 87Sr/86Sr
ratios of most river waters fall on a simple mixing line between a source with
low Sr isotope ratios and high Sr concentrations (easily eroded limestones)
and high Sr isotope ratios and low Sr concentrations (more resistant silicate
rocks). The exception to this general trend was seen in the Ganges and Brahmaputra,
which both drain the Himalayas. They have high Sr concentrations and 87Sr/86Sr
ratios that could be interpreted to reflect accelerated weathering of silicate
rocks in this area of high monsoonal rainfall and accelerated uplift.
 |
| Figure 12: Evolution of the Sr isotope ratio of seawater since 2.5Ma |
Raymo coupled this interpretation with the observation that
the 87Sr/86Sr ratio of seawater has increased fairly rapidly
over the last 2.5 Ma as a result of this increased weathering in the Himalayas.
SAQ6What might be an alternative model of the river water and seawater Sr isotope data?
(These WWW pages were created by Martin Palmer)