Hi all, and apologies for a long-delayed posting. I've been swamped with "science" work, the details of which are probably not exciting enough for a blog post! But, I'm back.
Someone raised the opinion to me the other day that a certain former Vice-President must be wrong; if we can't predict the weather tomorrow, how can we possibly predict the climate 50 years from now? Excellent question, with two answers. First, weather is not climate. Weather describes daily atmospheric conditions; rainy, foggy, sunny (not in SF!). Climate could be described instead as a long-term average of daily weather. For example, the Californian climate is Mediterranean, meaning roughly that we have rainy winters and very dry, warm summers. So, January 2008 will have more rainfall than July 2007. But do you know which January days will be the rainy ones?
Why do we know more about the climate next year than we do about next week's weather? That brings me to the second answer. Scale and uncertainty matter. Look at the table that your computer is resting on. Looks relatively smooth, doesn't it? But magnify the surface 10,000 times and the landscape would probably be more reminescent of Nevada's Great Basin. Or, roll a tennis ball down a steep street (please check for traffic first!). You have no doubt of the outcome, but can anyone predict the precise path of the ball, given the roughness of both ball and road surface, wind speed, etc.? Uncertainty at small scales very often average out at larger scales, both in space and time. The table is smooth enough for your computer, balls roll downhill, and increasing atmospheric concentrations of greenhouse gases cause global warming. Of these things we are certain.
There are small-scale components and interactions within systems, however, that can matter to the overall behaviour of larger systems. Uncertainty at these scales may generate uncertainty at larger scales. Some of you have probably heard of the anecdotal butterfly flapping its wings in China, and via the wonders of chaotic dynamics, causes a storm over the Atlantic. Well, it isn't quite that simple, but concepts such as chaos, nonlinear dynamics, positive feedback, thresholds and tipping points, become very important when predicting Earth's future climate. More on these next time, I promise!
Powered by ScribeFire.