Calspace Courses

 Climate Change · Part One

      Climate Change 1 Syllabus

    1.0 - Introduction
    2.0 - The Earth's Natural Greenhouse Effect
    3.0 - The Greenhouse Gases
    4.0 - CO2 Emissions
    5.0 - The Earth's Carbon Reservoirs
    6.0 - Carbon Cycling: Some Examples
    7.0 - Climate and Weather
    8.0 - Global Wind Systems
    9.0 - Clouds, Storms and Climates
    10.0 - Global Ocean Circulation
    11.0 - El Niño and the Southern Oscillation

  12.0 Outlook for the Future
         · 12.1 - Introduction to Climate Change
         · 12.2 - Advances in Computer Modeling
         · 12.3 - Physics versus Fudge Factors

 Climate Change · Part Two
 Introduction to Astronomy
 Life in the Universe

 Glossary: Climate Change
 Glossary: Astronomy
 Glossary: Life in Universe

Advances in Computer Modeling

Examples of two different computer models and their corresponding resolution. The top picture represents North America with a grid size of 480 km, typical of most climate models. The bottom picture is a higher resolution with a grid size of 60 km, typical of weather forecasting models. Note how much topographic detail is lost in the low resolution model; the loss of detail about ocean-atmosphere interaction is of an equal scale. (From: CO2 Science Magazine)
How Are Climate Models Designed?
As we have learned, the activities of humanity have demonstrably changed the climate in the last two decades and have likely been changing it for the past several decades. The most important factor is the addition of greenhouse gases to the atmosphere, mainly from energy production (carbon dioxide) but also from animal farming and waste management (methane). Other factors are deforestation, air pollution, excess fertilization, soil degradation, and drastic interference in the water cycle on land.

How will this great geophysical experiment shape the future? We would like to get some answers to this question before the experiment has run its course. The system is sufficiently complex so that mere guessing can be accorded but little credit. As we have discussed in previous chapters, one needs to build a simulated climate machine, in the form of a computer program, containing the best ingredients of our knowledge, incorporating everything from the principles of radiation balance to the equations of motion. Such a simulated program consists of thousands of instructions of the kind “IF (A) THEN (B),” converting a momentary condition (A) into another condition (B) in one computing step. For example, (A) might describe the temperature of a small area on the ground, while (B) might refer to the radiation received by the atmosphere, the evaporation experienced at this place, or the transfer of heat to a neighboring area. The "forcing" of the model is ultimately caused by the known seasonal changes in insolation (that is, solar radiation). The machine "responds" by producing seasonal cycles of surface temperature, rainfall, snowfall, cloud cover and winds. The more this "output" resembles the real world of changing seasons, the more credence is given to the program as a valid simulation of what is happening in the real world.

Early Computer Models of Climate
The first attempts at climate modeling were done in the 1960’s, making this a very young scientific field. In these early days, the climate system was represented in greatly simplified fashion, for example by the balance of heat radiation entering and exiting the Earth’s atmosphere as a function of latitude. But the complexity of models grew rapidly together with power of the computers that were running them, and as Dr. Richard Somerville of SIO likes to point out, one requirement of a useful model is that their computations should be fast enough to get ahead of the real climate changes! In other words, the complexity of the models is constrained by the computing power; our models need to be designed to give us answers in a reasonable amount of time, a factor that is limited by the speed of the computer. For instance, we could attempt to design a very complex climate model that takes into account the most intricate detail of atmospheric physics. However, if even the world’s fastest computer takes years to come up with a final answer to our computer model, we have defeated the purpose of making a model, which is supposed to give us relatively quick answers.

In a computer model the distance between neighboring points describing conditions on the surface of Earth is referred to as the “resolution,” and it is this parameter that is most affected by computing power. Processes that are "below resolution" (such as the local convection that produces thunderstorms) must be simulated in some way so as to retain the functions they provide for heat transfer, evaporation and precipitation (more on this subject later in this lesson). In a manner analogous to the “resolution” of a computer monitor, the better resolution we have in our model (that is, the smaller the distance between points on our Earth), the more detailed the image of the Earth’s climate we see. However, such high resolution climate models require high speed computers.

An example of climate model design, illustrating the potential factors that can be added in order to get realistic results. Different models may contain different factors depending on their complexity and the computing power that is available.
Progress in Climate Models
One of the great steps forward in this field was the replacement of the so-called "swamp ocean" with a more realistic construct. The “swamp ocean” modeled the ocean as a layer of water of some modest thickness and was a device of the climate machines of the 1970’s to provide a means for winds to pick up water vapor, depending on their temperature and speed. Today, this transfer of water to winds is achieved by a full coupling of the ocean and atmosphere models, whereby one can influence the other. Generally, over time, simple devices such as the swamp ocean have been replaced by fully responsive climate subsystems whose properties are constantly recalculated as a function of conditions elsewhere.

Using such simulations, we can test the responses of the models by entering disturbing factors into their mathematical machinery and observing the results. The technical term for such experiments is "sensitivity analysis." The building of "scenarios" of the future is a special case of such analysis, which provides hints at what might happen in the real world, according to the various models, run with a range of plausible input scenarios. When generating "scenarios," we pool all our experiences and knowledge in the different sciences bearing on climate and produce a best guess for the consequences resulting from a particular political strategy or economic development using the most powerful computers. To demonstrate the ramifications of human-induced climate change, we should be able to predict what happens to climate with and without continued input of manmade trace gases and to check whether differences in the predictions are sufficiently great to impact life in the real world.


back to top
© 2002 All Rights Reserved - University of California, San Diego