Between 1000 and 1900 ad, forests in many areas of the northern mid-latitudes retreated under the onslaught of a growing population. The most dramatic episode of deforestation occurred after 1600 in North America, as European settlers arrived
Figure 6.8. Global temperature history of the last 2,000 years from several sources of tree ring data, showing the "Little Ice Age" dip after about 1300 ad.
Source: CDIAC. 0 ?00 400 800 «00 1000 '400 1R0H 1S00 ?000
and began clearing land for fields. By 1850, most of the forest that had covered eastern North America was gone. It is possible that the conversion of forests to fields cooled the earth's climate slightly, due to the higher albedo of the exposed soil of the fields and of the dry yellow plants as grain and hay ripens in summer. A modeling study suggests that the global cooling from change in land use in various regions after 1000 ad would have been 0.25°C overall, but concentrated in the northern hemisphere lands (which would have become 0.41 °C cooler). The effect would have been especially intense in particular regions: North America, the Middle East and SouthEast Asia would have borne the brunt of the cooling. Europe, India and North Africa would also have been affected, to a lesser extent. Models suggest that this cooling effect of deforestation could have been strong enough to bring about a known cool period—the "Little Ice Age''—which lasted several centuries between 1000 and 1900 ad.
The Little Ice Age has been recognized from many different sources of historical and geological information (Figure 6.8*)—for example, glacier limits extending downslope, sediment isotopes, crop-planting distributions and accounts of rivers freezing over (Figure 6.9). Overall, it involved a temperature decrease similar in size to that modeled from deforestation, but it is not clear that the real climate changes corresponded in time to what the models predict. The timing of the Little Ice Age, and whether it was synchronized between different parts of the world, is still something of an uncertainty. It seems to have come on several centuries earlier than the main phase of deforestation in the Americas, although many of the coolest phases of the Little Ice Age occurred between about 1650 and 1800 when deforestation was at its peak. Whereas forest cover changes were progressive and unremitting, the Little Ice Age often reversed itself for a while; even during the coldest centuries there were brief warm phases in amongst the cool phases. This does not quite look like a simple
* See also color section.
influence of forest cover on temperature. Furthermore, the Little Ice Age was already ending during the mid to late 19th century, when forest cover was actually at its minimum in eastern and Midwestern North America and when the cooling influence would be expected to be at its strongest. Thus, the timing of events in the Little Ice Age does not suggest that deforestation was the sole cause of the cooling, although it may have been a significant factor alongside multiple influences on climate that waxed and waned in importance over several centuries. Furthermore, some recent work suggests deforesting eastern North America actually warmed and not cooled the climate, because it decreased evaporation.
Whereas models dealing with tropical deforestation have often forecast changes in precipitation and evaporation following deforestation, the model that forecast the Little Ice Age cooling does not suggest any major effects of mid-latitude forest loss on precipitation, rather than temperature. This is because mid-latitude precipitation is controlled by the broad sweep of wind systems that blow across whole oceans and continents (an example being the wind belt known as the westerlies), not by local convection as in the tropical rainforest regions. So, the effect on rainfall in the mid-latitudes is "blurred out" by the winds, whereas the equatorial tropics that do not receive these broad wind belts are at the mercy of the aridifying influence of deforestation. However, historical deforestation in some other mid-latitude areas may have had stronger effects on precipitation.
6.5.2 Deforestation around the Mediterranean and drying in North Africa
There is a lot of evidence that the climate around the Mediterranean has gotten drier during the last 2,000 years, especially along the northern edge of North Africa. In the Atlas region of Tunisia (which was part of the Roman Empire), Roman era aqueducts lead down from hills where no streams flow now. Contemporary accounts of grain yields show that rainfall must have been much higher at that time, and there are the remains of Roman farming settlements in areas that are now too dry to farm. A model study has suggested that deforestation in the Mediterranean region from Roman times onwards was the key factor in the drying of climate in that region. It seems that the removal of tree cover would have affected the roughness, albedo and evapotranspiration from the land surface, and this may have altered the whole broad-scale circulation of the region. Even though most of the forest removal occurred around the northern edge of the Mediterranean (e.g., in Italy, Spain and southern France), the most severe effects would have been felt farther south in the Atlas Mountain region, and in the Nile Valley of Egypt farther east. It seems that, when the forests were relatively intact around the northern Mediterranean, the upwelling in the atmosphere that they promoted helped to pull in a trace of the north African monsoon, up to the southern edge of the Mediterranean. This gave just enough rain to sustain crop-growing and much denser vegetation in areas that are now on the edges of the desert. The key factor that led to deforestation seems to have been a breakdown in the traditional land ownership structure that had preserved many forest areas in Roman times. In late Roman times, large "corporations" known as latifundia took over, farming the land with less eye to preservation, and when Roman rule finally disintegrated after about 600 ad, the chaos that followed allowed much further deforestation.
The most dramatic swings in climate have occurred on longer timescales. In the last couple of million years known as the Quaternary, the world has been through numerous global "ice ages" (known as glacials) (Chapter 3). At these times, vast ice sheets built up and spread over Canada and northern Europe, New Zealand and the southern Andes, and temperature zones were pushed equatorwards. Nowhere escaped this global cooling: even the tropics were some 5 or 6°C cooler than now. As well as being colder, the glacial world was much drier. Because of the cold and aridity, broad changes in forest cover around the world occurred during the large climate swings of the last two and a half million years (known to geologists as the Quaternary), on the timescale of tens of thousands of years. At this time, before the invention of agriculture, it is thought that human influence on forest cover would have been minor and any large changes must have been due to the influence of climate itself. Many areas that are now forest-covered were completely treeless during the dry, cold glacial maxima, such as the last one that ended 15,000 years ago.
At other times there were warmer climates in the high latitudes, and more forest cover than could exist today. One example of such a warmer phase is the early
Holocene, between around 9,000 and 6,000 years ago. At that time, climates seem to have been several degrees warmer along the Arctic coast of Siberia, and forests occurred hundreds of kilometers farther north into areas that are now treeless tundra (Chapter 3).
Forest was not merely a passive participant in all of these changes in the past. There are intriguing signs that the forest cover itself has amplified many of the climate fluctuations during the Quaternary. However, it is important at the outset to emphasize that forest cover was not the primary control on the largest scale climate changes. The precise pacing of warm and cold periods shows that most of the broad swings in climate that occurred were instead brought about by the shifting orbital parameters of the earth—the Milankovitch cycles—which affect the intensity of sunlight reaching the northern hemisphere in summer (Chapter 5). The shifts in sunlight (by several percent of the total energy influx) occur on the timescale of tens of thousands of years, alternately warming and cooling the earth with their effects. The warmer summers that occur during the parts of these cycles with extra northern sunlight are key to bringing about these globally warmer times. Snow and ice have the highest albedo of any natural surfaces on earth; they can reflect back more than 95% of the solar energy that hits them. This means that they are remarkably good at preserving themselves against the heating sunlight that might melt them, and in doing so they make the whole earth cooler.
How can just a warmer summer in the north bring about a warming of the whole earth, all the year round? Well, if a more complete meltback of snow and ice occurs during a warm spring and summer, it means that more sunlight can be absorbed by the surface, instead of being reflected. This has the effect of amplifying the summer warming, further warming the climate. This can also melt more of the snow and ice. And the more the snow and ice melts, the warmer things get; and the warmer they get, the more melting there is. The large change in the temperature of this most-sensitive region alters the temperature of all the surrounding regions; it is rather like opening or shutting a door to the outside of a house on a cold night—all the rooms in the house will be affected to some extent because they all connect together. Furthermore, each of those regions has its own feedbacks which can pick up and amplify the changes brought about by the high northern latitudes. Many of these feedbacks are likely to involve vegetation in some way. In this way the "signal" from the north propagates all around the world. When the warming signal is operating most strongly, at times when the northern summer sunlight is at its greatest, the world tends to go through its very warmest phases such as the one between 9,000 and 6,000 years ago.
What northern forest cover seems to do, overall, is further amplify the changes that would have occurred anyway due to the feedback between summer temperature and ice. The warming triggered by the presence of the forest is substantial: for instance, it increases the temperature by several degrees C along the northern edge of Siberia. The most important influence of the forest is on albedo. If there is forest in a northerly climate, the tree leaves and branches tend to cover over the snow that fell through the canopy during the winter. This is especially so with the evergreen conifers that often dominate in these climates and do not need time to leaf out in spring, so the snow cover on the ground is never left exposed. Although some snow tends to accumulate on the leaves and branches, most of it tends to fall off the trees and end up under the forest canopy. If snow is covered over by the darker canopy, it cannot reflect sunlight back into space, so it cannot exert its cooling effect on the climate. The air can get warmer because of this. And furthermore, because the air is warmer due to the dark canopy under the trees it can also directly melt back the snow (it is protected against gaining heat by radiation, but not by direct conduction!). The warm air can also move beyond the forests, to open areas such as tundra, and melt back the snow there. That melting of exposed snow further amplifies the warming brought about by the forest cover. In addition, the greater the warming, the farther north the trees can grow, and the cycle continues until eventually it runs out of momentum in the farthest north where the background climate—even with positive feedbacks—is just too cold to sustain tree growth and snow melt.
Here then is the general way that forest cover seems to amplify climate change in the far north during phases of Milankovitch cycles with increased summer sunlight:
More summer sunlight ^ warmer temperatures ^ northern forest expansion
^ lower albedo ^ warmer temperatures
^ forest expansion ^ etc.
The effect of the changing forest cover thus seems to be a positive feedback on climate change. Any background change due to summer sunlight changes is amplified, producing broader swings in climate. The climate models suggest that in the Arctic there was a several degrees increase in temperature resulting from the forest cover and feedbacks that it set in place during the mid-Holocene period about 9,000-6,000 years ago. The feedbacks induced by forest itself also helped to propel the forest hundreds of kilometers farther north at that time.
When the orbital parameters shifted after 6,000 years ago, everything went into reverse. The sustaining summer sunlight that had formerly ensured the warmth now began to decline. Trees did not do so well and the canopy thinned, and snow cover increased. The northern climates cooled and forest retreated, a change once again amplified by the change in forest cover. If humans had left the climate system untouched, in several thousand years' time the earth would be ready to begin its slide into the next major glacial phase, accelerated by this forest-snowcover feedback.
It is likely that this same vegetation-climate feedback has worked in the past to help pull the world into ice ages. Models suggest that an initial cooling event about 115,000 years ago at the end of the last interglacial was greatly amplified by the loss of forest, and its replacement by tundra and snow with a higher albedo. When this extra albedo feedback is included, it turns out to cool summers in the northern lands by a massive 17-18 °C, sending the world plummeting into a major glaciation.
The feedback effect between tree cover and temperature is not only relevant to understanding the distant past and the very long-term future. It might also be very important in the next few centuries. It seems likely that over the next several decades temperatures will continue to increase in the high latitudes due to greenhouse gases placed in the atmosphere by humans. As tree cover responds to the warming, expanding northwards, it is likely to further amplify the temperature increase through the same feedbacks that would have operated during the warm phase several thousand years ago.
Was this article helpful?