Predicting Climate Conditions


This summer I'm working at the Geophysical Fluid Dynamics Laboratory (GFDL) in Princeton, which is one of the country's main facilities engaged in climate change research. Scientists at GFDL don't go to Greenland to extract ice cores, they don't visit paradise to observe the response of coral reefs to their acidifying ocean home, and, in fact, they don't really have to go outside at all; instead, they make models. Not physical models, but rather computer simulations – really big computer simulations of the earth's entire climate system, involving the oceans, atmosphere, and the activity of living things, including humans.

These models have evolved over about fifty years into monstrous software packages that require supercomputers to run them, but the core idea of a climate model is simple.  The climate is a description of the weather patterns that occur repeatedly over long periods of time.  These weather patterns are governed by a relatively small set of physical laws, like those that determine how air moves in response to a difference in pressure or how much and how quickly air temperature increases as solar energy is absorbed.  The goal of a climate model is to apply these known physical laws to a virtual atmosphere, perhaps changing a factor like the greenhouse gas concentration, and observe the climate that emerges.  The complexity comes from the fact that the atmospheric system is fluid and dynamic (hence the lab's name), interacts with the land and ocean, and is influenced by life, geologic processes, and an enormous number of other factors.  Think of trying to predict the location at any moment in the future of a small bit of water flowing into turbulent river rapids – and also the behavior of the rapids based on the underwater rock formations and recent precipitation.  It gets complicated rather quickly.  Despite this difficulty, the models developed at GFDL and elsewhere have become remarkably successful at simulating the behavior of the atmosphere over long periods of time, aided immensely in recent years by the explosion in computing power.

You might wonder how successful prediction of climate conditions far into the future is possible, given the seeming inability of weather forecasters – who also use computer models – to predict whether it will rain even a day or two in advance.  This is a valid question, and the answer is mainly that climate researchers have a few advantages.  The most significant is that climate predictions are not made for a specific place and time; they are made for entire regions and describe expected long-term average behavior.  The removal of the extreme specificity needed for weather forecasting makes the whole enterprise possible.  In climate research, we're interested in how temperature and precipitation over continent-sized areas will change over decades and centuries, not minute-by-minute conditions in specific cities.

The results of this work have been spectacular.  Climate models are now our primary predictive tool, and we're relatively confident in our forecasts for the coming decades.  We can't say exactly how much temperatures will change or exactly where and to what degree precipitation will increase or decrease, but we can give relatively narrow ranges with confidence and determine the general trends with certainty.

And the trends are not good.  With no significant reduction in greenhouse gas emissions even conceivable in the near future, it seems likely that the more pessimistic model predictions will be the most accurate.  These changes will likely have far-reaching and broadly disruptive effects on human societies across the planet, as we implicitly rely on a static climate in almost everything we do, from growing crops to choosing building materials for highways.  Any change in this fundamental underpinning of civilization will impede our ability to make long-term plans and will decrease the efficiency of activities that we've optimized for our current environment.

An understanding of the new reality that will soon be upon us is essential if we intend to continue our progress towards a world under our control.  Fortunately, the same technological prowess that we've used to modify something as gigantic and seemingly immobile as the planetary climate can also be utilized as an early warning system.  My experience thus far at GFDL has shown me the extent to which science has given us the tools we need to protect ourselves; the hard part now is choosing to use them.




Very interesting and well written Ethan! I'm glad Cindy sent it to me. I have faith that humankind will band together (through their scientists) and find solutions to challenges due to major climate change. And you may be a one of those scientists!

Hi Ethan, thanks for your article. I'm busy with the study of the weather in Germany and the impact of pollution on our local weather. Therefore, your article has me very interested. We observe larger context, so that we assume that the weather will change fundamentally. Sorry. Best regards, Rolf

Add new comment

Filtered HTML

  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd> <p> <div> <br> <sup> <sub>
  • Lines and paragraphs break automatically.

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.