Unit 3:  Climate Change 
             Background Information
 
    As explained in previous units, global environmental change refers to both human-induced and natural transformations of all of the earth’s systems.  Yet the popular perception of this change is focused largely on climate change, especially global warming.  In this regard, it is important to distinguish between climate and weather.  Essentially, weather refers to the condition of the atmosphere at any particular time and place (e.g., What is the temperature today?  Is it raining or snowing? What is the humidity and air pressure?)  Climate, however, refers to the long-term characteristics of the climatic system for the area (e.g.,  What is the average temperature?  Is the area generally prone to drought or to tropical storms?  What is the annual average rainfall?).

    Climate change is not a new phenomenon.  In fact, the earth has undergone significant variations in climate during its estimated 5 billion year life.  Cold periods known as ice ages put parts of the earth under glaciers for hundreds of thousands of years.  Of particular concern today is a warming of the earth’s climate, occurring more rapidly than any of the previous climate changes that the earth has experienced.  Unlike previous climate changes, however, this global warming has a very clear human component.
 
 
Global Warming and the Greenhouse Effect
 
    The earth’s atmosphere consists primarily of nitrogen and oxygen, and a small amount of trace gases known as greenhouse gases, which include water vapor, carbon dioxide, methane, and nitrous oxide.  These greenhouse gases act much like the panes of glass in a greenhouse, allowing short-wave energy from the sun to pass through them, but trapping the longer-wave heat radiation that is radiated back to the atmosphere from the earth’s surface.  This greenhouse effect is a vital atmospheric process without which the earth would be a cold and lifeless planet (see Figure 6).

 

Figure 6:  The Earth’s Atmosphere and the Greenhouse Effect
 
Source:  Adapted in part from IPCC. 1990. Climate change:  The IPCC scientific assessment. Houghton, Jenkins, and Ephraums, eds. Cambridge, UK:  Cambridge University Press, pp. xiv.
 
 
    Human activity has begun to alter the composition of the earth’s atmosphere.  The by-products from industrialization, such as carbon dioxide from the combustion of fossil fuels like coal and oil, have augmented the amount of greenhouse gases in the atmosphere.  This is known as the enhanced greenhouse effect.  An increase in greenhouse gases traps additional heat energy within the atmosphere and is predicted to result in an increase in the annual average temperatures on earth.

    Climate change resulting from human activity is much more complex than just the rise in global annual mean temperature (the average temperature per year for the entire planet) that is predicted by the end of the next century.  This unit will examine some of the research that is being undertaken in order to gain some understanding of the nature and magnitude of future climatic variability.

    The Intergovernmental Panel on Climate Change (IPCC, an international committee composed of hundreds of leading scientists), has made the following predictions about the future of the earth’s climate (IPCC 1996a):

    Scientists have predicted that the rise in the global mean annual temperature will affect the frequency of extreme weather events such as hurricanes, typhoons, tornadoes, severe storms, and floods.  When the atmosphere gets warmer, more water evaporates from the oceans, and in turn, more rain and snow fall from the atmosphere.  The extra energy released with this evaporation and precipitation cycle increases the power of storms.  Water vapor also contributes to the greenhouse effect and could create what scientists refer to as a positive feedback, meaning that the processes of warming and evaporation will continue to intensify and feed off each other increasing the rate and magnitude of the effect (in this case, temperature rise).

    Climate change will also affect the amount and timing of precipitation.  Warrick and Farmer (1990) note that small variations in mean precipitation could result in large changes in the risk of extreme weather events such as drought and severe tropical storms.  Pierce (1990) hypothesizes that climatic variability will almost certainty affect the timing and intensity of the monsoon seasons upon which millions in Africa and Asia depend for agricultural productivity.

    Thus, it is fairly certain that climate change will have important impacts on the global environment and the world’s population.  It is also clear that some populations may be more vulnerable to the effects of global climate change than others.  For example, in lesser developed parts of the world, increased frequency of droughts could have devastating effects on crop yields and food supply; sea level rise could increase the risk of flooding to settlements in marginal coastal areas.  These same events may not have equivalent impacts on the more developed parts of the world where people may be able to afford preventative or mitigative measures or even to accommodate some amount of environmental change.

    For this reason, it is important to have the best possible information on the likelihood of the intensity and regional nature of climate change in order to prepare for or mitigate some of the effects of such change.  But how do we go about predicting these changes?  Scientists are working on improved models of the operation of the atmosphere that allow more sensitive and robust assessments of different scenarios in which climate change will take place (e.g., assessments that have a better understanding of the relationship between vegetation characteristics and precipitation illustrated in Activity 2.1.)  Two basic areas of modeling research are currently active -- analogue models and general circulation models.
 
 
 
Climate Models
 
 
Analogue Models
 
    Broadly, analogue measurement is a process through which changes in one phenomenon, such as temperature, are measured through changes in a related or “analogous” phenomenon, such as expansion and contraction of mercury in a thermometer.  An analogue model extrapolates the dynamics of climate change from actual data that have been collected and measured (i.e., parts per million of atmospheric dust or millimeters of precipitation).  Two examples of this kind of research include:

 
General Circulation Models
 
    General Circulation Models (GCMs) are extremely sophisticated computer models of the earth’s climate system.  GCMs were designed to help scientists learn about the complexities of the interactions among the atmosphere, the oceans, the lithosphere, and the biosphere.  Figure 7 illustrates the complexity of the earth’s climate system.
Figure 7:  The Earth’s Climate System
 
Schematic view of the components of the global climate system, their processes and interactions (small arrows), and some aspects that may change (large arrows).  Source:  Based on IPCC. 1998. An introduction to simple climate models used in the IPCC 2nd assessment, Figure 1. http://www.ipcc.ch/techreps.htm (July 1998). Reproduced by permision of the IPCC.  Image has been redrawn for this publication.
 

    To reproduce exactly such a system in a computer model requires more computer power than is  available; consequently, a number of strategies have been employed to reduce the computer requirement  One strategy is to represent large regions of the earth with average climate data.  This means that one temperature and precipitation value is used to represent several hundred square kilometers.  In so doing,currently much of the spatial variability in temperature and precipitation is obscured.  The second strategy is to make predictions without running an enormous number of calculations for hundreds of years.  This is accomplished by running the computer models to an equilibrium or balanced state.  GCMs are run in the following way:

  1. Run the GCM under normal conditions until it reaches equilibrium.
  2. Change one variable in the model, such as solar output, or the concentration of a greenhouse gas.
  3. Run the GCM again with the one changed variable until it reaches equilibrium.
  4. Compare the results of the second run with the first run to determine what effect the selected variable had on the climate.
    There are, of course, several problems with this technique.  One is that the modeler begins with an initial temperature solution and ends with a different final temperature solution. By simply comparing the two values, the analyst can determine how much of a temperature change is expected over time.  It is impossible, however, to know how the temperature change occurred over time.  It is easiest to assume that the change in temperature was linear, meaning that for every unit of time, the temperature increased an equal amount.  But it is unlikely that a linear response will occur because of the complex feedback mechanisms in the climate system.  Figure 8 illustrates the variety of temperature responses that could occur between two points in time.  The response could be linear, as mentioned above, allowing scientists to estimate a temperature change over a given time period.  The response could also be non-linear.  A non-linear response could mean that temperatures increase rapidly then level off or that they increase slowly at first then rapidly near the end of the period.  Finally, the response could be random, meaning that the pattern is unpredictable and erratic.
 
Figure 8:  GCM Temperature Response Scenarios
 
 
    We are still left with an important question -- what will be the likely temperature scenario?  With increased computer power in the last few years, a different approach to climate modeling has been developed.  A time-dependent or transient climate model is now a favored technique.  Here the computer model increases carbon dioxide concentrations by 1% every year.  The model is then run for at least 70 years, when atmospheric carbon dioxide levels are predicted to have doubled.

    A direct comparison between the initial conditions and the 70th year will not yield proper climate change figures because of climate model drift.  Every GCM simulates the earth differently, and therefore, every model will drift to its own particular climate as the model progresses.  To eliminate this problem, another model is run to 70 years without any change in carbon dioxide levels.  Then the two model results can be compared.

    Numerous transient climate modeling experiments that use different time frames will help scientists predict the likely temperature scenario we will face during the next 100 years.  Of course, these models introduce another concern.  The rate of greenhouse gas emissions, which is not required in equilibrium climate experiments, becomes important in transient experiments.  Thus the human response of reducing or increasing the rate of greenhouse gas emissions is critical in determining what future climate may prevail.

    There are several other concerns with using GCMs. It is important to emphasize that the mathematical equations that constitute the mechanisms of the models represent only approximations of the climate system. Scientists do not fully understand how all components of the system operate, and only the largest and most powerful computers can handle the levels of processing required.

    It is also important to understand that GCMs provide outcomes for the global and continental scale; that is, they predict general conditions for large areas.  These average conditions will not hold at smaller scales of analysis like regions and places because of the factors that influence climate at these scales (e.g., topography).  It is precisely at these smaller scales, however, that data is needed in order to know how to address the effects of climate change at a local level.  A critical need -- and one the science community is working on -- is to “downscale” GCMs.

    Although climate change research is a sophisticated scientific endeavor, both the complexity of the earth climate system and the spatial resolution of GCMs introduce a certain level of uncertainty into the research.  For example, scientists are uncertain of the complete role of oceans in the global climate system.  Oceans serve as giant “sinks” or storage for carbon dioxide.  But the amount of CO2 that they store is still unknown, therefore introducing uncertainty into computer models of the climate system.  In addition, the spatial resolution of GCM outputs makes it difficult to pinpoint locations where the greatest warming will occur.

    Uncertainties have significant ramifications on the human responses to global change (see Unit 4).  What we know and do not know with certainty will affect how and when we respond to global environmental changes.  For several years, scientists and politicians have debated whether global warming was a real possibility.  More recently, the debate has not been whether global warming will occur, but how much of a warming should we expect when global warming occurs.