|
Climate Models & Predictions for the Future
"Time
drops in decay, |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Suggested
|
|
||||
In the last two lectures we discussed paleoclimate over very long time scales. The broad, glacial-interglacial variations observed over the 150,000 year record shown in Figure 1d are explained mainly by solar forcing due to the Milankovitch theory.
|
Shorter term changes in climate, such as the Younger Dryas period shown in Figure 1c, was found to be related to how the formation of North Atlantic deep water and the resulting Conveyor Belt circulation has changed over time. However, it remains to be explained how the observed correlation between the greenhouse gases methane and carbon dioxide as measured in the ice cores, and the temperature shifts during the last several glacial periods, are actually related.
The correlation between methane, CO2 and temperature could perhaps be explained by invoking a temperature dependence to the cycling of CO2 and methane through the environment. For example, the main reservoir for CO2 is the ocean. A dependence of the flux from the ocean on temperature, such that the higher the temperature the greater the flux to the atmosphere, could explain the correlation for CO2. In addition, we know that as temperature warms the oceans can hold less dissolved gas, which would be consistent with the idea that the gas would leave the ocean and reside in the atmosphere. A similar temperature dependence for the decomposition of organic matter, which can produce methane in anaerobic conditions (see later lecture on Microbes), could explain the methane correlation.
As mentioned previously, one thing that we do know for sure is that the physics of heat-trapping gases are unavoidable -- that is, if we increase the concentration of these gases in the atmosphere, they will trap more heat and we will warm the atmosphere. These relationships between heat-trapping ability and gas concentration forms the basis for many of the predictions of future global climate, as discussed below.
Before moving on to the future climate, we need a better understanding of Earth's current climate. Climate differs from weather in that it provides a statistical view of seasonal and daily weather events over a long term period. Thus, for example, the passage of a large rain storm in a frontal system over Ann Arbor is weather event, while the average number of such passages for the month of July (averaged over several years) is part of the climate record. So, climate is the long-term average of weather.
Climate records are most often expressed in terms of temperatures, winds, precipitation, and pressures - all parameters that can be measured at multiple sites around the globe. Over the years a large data base of weather event measurements has been obtained, leading to a good description of today's climate.
We find that climate varies widely around the globe - we have deserts and rain forests, ice caps and "death valleys". As for most subjects discussed in this course, there is a taxonomy of sub-disciplines and we can speak of the following:
Figure 2. Precipitation and temperature seasonal averages. |
---|
The many factors that control local climates include: intensity of overhead sun - including its latitudinal variation; the distribution of land and water; ocean currents; prevailing winds; positions of semi-permanent high- and low-pressure areas; mountain barriers; and altitude. The effects of these controls can be seen in global patterns of temperature and precipitation. Examples of local climatic data are shown in Figure 2. These graphs are derived from temperature and precipitation data available from the National Climate Data Center. The climate data for Detroit is listed in Table 1.
Great differences in climate occur from place to place, even within the continental United States which only accounts for about 2% of the Earth's surface. In 1918, a popular climate terminology was developed by Koppen and is called the Koppen System. It is based on annual and monthly average temperature and precipitation measurements, using evidence from vegetation where data is sparse.
Figures 3 uses this system to provide a summary of the types of climates found on today's Earth.
JAN | FEB | MAR | APR | MAY | JUN | JUL | AUG | SEP | OCT | NOV | DEC | ANNUAL | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Temperature (°F) | 22.9 | 25.4 | 35.7 | 47.3 | 58.4 | 67.6 | 72.3 | 70.5 | 63.2 | 51.2 | 40.2 | 28.3 | 48.6 (MEAN) |
Precipitation (in) | 1.76 | 1.74 | 2.55 | 2.95 | 2.92 | 3.61 | 3.18 | 3.43 | 2.89 | 2.10 | 2.67 | 2.82 | 32.62 (TOTAL) |
Data from the National Climate Data Center |
Figure 4. Average annual sea level temperatures throughout the world (degrees F) |
---|
Note how the isotherms tend to bend along coastlines. This is due to the unequal heating of land and water and the tendency of the winds to blow along coastlines, creating the upwelling and downwelling discussed in the last lecture. Also of significance are the ocean currents both on the surface and where the deep-water currents rise to the surface. Examples of ocean currents include the California Current which flows southwards along the Californian coast and the Gulf Stream which flows northwards in the Atlantic.
Figure 5 summarizes the modern global mean precipitation climate record. Notice the high degree of regional variability.
3. Current Trends in ClimateThe global average surface temperature has increased by 0.6 ±0.2°C since the late 19th century (IPCC, 2001). It is very likely that the 1990s was the warmest decade and 1998 the warmest year in the instrumental record since 1861 (see Figure 6). As indicated in Figure 6, most of the increase in global temperature since the late 19th century has occurred in two distinct periods: 1910 to 1945 and since 1976. The rate of increase of temperature for both periods is about 0.15°C/decade. Recent warming has been greater over land compared to oceans; the increase in sea surface temperature over the period 1950 to 1993 is about half that of the mean land-surface air temperature. The high global temperature associated with the 1997 to 1998 El Niño event stands out as an extreme event, even taking into account the recent rate of warming. |
|
|
Modern changes in temperature and carbon dioxide are shown in Figures 8 and 9. It is clear from these figures that rapid changes are underway - at rates far exceeding anything discussed so far.
It must be assumed that human activities (known as Anthropogenic Effects) are dominating the present changes. In fact, a detailed analysis of the Earth's carbon cycle (discussed in an upcoming lecture) indicate unequivocally that these changes in the atmosphere are due to human activities and not to some random variation in time.
The past
century has seen an increase in the global mean temperature of 0.8oC.,
and nearly all of the top 10 warmest years on record have occurred in just the
past decade.
The upward trend may also be a consequence of the increasing levels of
carbon dioxide, methane and CFCs put into the atmosphere through various
anthropogenic processes.
Figure 10 shows multi-year measurements of carbon dioxide abundances in the
atmosphere from
a single location in
In Figure 11 the modern records for atmospheric methane and carbon dioxide are shown in a slightly different form to highlight the seasonal differences and the latitudinal differences. Both gas species show a strong yearly cycle (peaking in the winter months). The cycle is more pronounced in the Northern Hemisphere (toward the top of the graphs) than in the Southern Hemisphere, where the seasonal variations are muted.
The reason for the high carbon dioxide and methane concentrations during winter months is that during the winter land plants are much less active, and the balance of photosynthesis (uptake of CO2) to respiration (production of CO2) leans toward respiration. This CO2 accumulates in the atmosphere and causes the peak in the winter months.
In the summer months the opposite occurs, and the plants are active and they remove CO2 from the atmosphere for photosynthesis. This results in a lower concentration in the atmosphere, and the corresponding valleys in the graph (Figure 11).
The second aspect of these graphs is to notice the big difference in atmospheric concentrations and seasonal variations between the northern and southern hemispheres. Most of the land mass of Earth is in the Northern Hemisphere, and thus most of the CO2 and methane are produced there. In addition, there is a preponderance of human activities in the Northern Hemisphere, which also contributes to the production of these gases.
Overall, most of the recent increases in global temperature are due to greenhouse gas increases in the atmosphere. Some of the variations, however, are a consequence of volcanic eruptions. Such events can emit large quantities of dust into the stratosphere where sunlight can be intercepted for a period of a few years.
Finally, the idea that changes in the number of sunspots can affect climate is hotly debated. Although there were fewer sunspots during the Little Ice Age than there are now, it remains to be shown that the power (heat) generated by the sunspots is really large enough to control temperatures on Earth.
|
|
|
|
The Intergovernmental Panel on
Climate Change has released a series of Technical Summaries in 2001.
Download: |
1.
"Climate Change 2001: The Scientific Basis"
(especially germane to this topic) 2.
"Climate Change 2001: Impacts, Adaptation and
Vulnerability" |
A number of sophisticated global climate models have been developed over the past 20 years for the purpose of predicting future climatic change. The most highly developed models are three-dimensional and time dependent and divide the globe up into a series of interacting boxes. The reservoirs and fluxes of importance are coded into the computer program which then solves the conservation equations of mass, momentum, and energy in order to calculate the evolving state of the atmosphere/hydrosphere system.
In general, models such as these must be validated against observations. This is done sometimes by running the model backwards in time to specify past, known climates. Often the models are run starting with input data for the year 1950, in part because the climate data from that time forward are particularly complete and detailed. As the model runs it outputs its "predicted" climate, which can be then easily checked against what actually happened. If the predictions match the actual observations then the model is considered to correctly simulate all of the main climatic processes and equations, and the model then continues to run into the future. If the predictions for past years don't match what actually happened, then the model is developed further. However useful, these predictive models have to constantly checked against experimental data to insure accurate results (Figure 12).
Climate models are often used to predict the climate of an Earth in which the carbon dioxide concentration has doubled. This is a prospect very likely to occur within the next 50-100 years, given the current increasing rates of anthropogenic CO2.
Below are some results of climate models run under twice the current global carbon dioxide concentration. The model predictions for future climate are based on forward estimates of the rate of fossil fuel consumption, and thus the production of CO2 and its input to the atmosphere.
These estimates are called SRES estimates, which stands for Special Report on Emissions Scenarios which was produced by the IPCC. These scenarios include situations such as "business as usual", where we do nothing to alter our consumption of energy on the planet. This of course results in the most drastic predictions of warming or sea level rise. Other scenarios (B, C, D, etc.) all account for various strengths of efforts that we might make to curb our use of fossil fuels.
Given these future scenarios of carbon dioxide concentration in the atmosphere, the models are run with their normal prescriptions for the multiple interactions among clouds, land, oceans, and so on in the Earth's climate system. Some results follow:
|
|
|
1. The most recent changes in Earth's
paleoclimate
record are likely due to shifts in ocean circulation, and the effects of
greenhouse gas increases. Volcanoes have had only a small effect, and
the sun spot record has yet to show that this variation in heat input has been
important.
2. Climate is the long-term,
statistical average of short-term "weather" events, such as individual storms
or droughts. The current climate sets a benchmark against which to judge
changes in past, and in the future.
3.
In recent times, temperature
changes and greenhouse gas abundances are correlated. Rapid global warming is
underway and models have been developed to predict the effects of these
changes. The models are sophisticated, mathematical representations of
the interaction between the land-ocean-atmosphere system on Earth, and rely on
"scenarios" of future fossil fuel use in order to make predictions for the
future.
4. These Global Climate Models (GCMs) predict a much altered climate on Earth during the next century. There will be substantial warming and sea-level rise, and altered patterns of precipitation with some areas receiving more and others less.
All materials © the Regents of the University of Michigan unless noted otherwise.
Copyright
İRegents of the