Regional Climate Modeling: The Next Frontier                   (home)

It is reasonable to consider the continued increase in greenhouse gas emissions as a "one-time experiment." We started this experiment with the Industrial Revolution, we continue the experiment today with greenhouse gas emissions increasing at a ~2% annually, and we will, in all likelihood, continue the experiment well into this century. The long life of electric power plants, the multi-decade residence time of CO2 in the atmosphere, and the ever-increasing need for energy in developing nations all combine to make this an experiment that is nearly impossible to stop.

What will be the outcome of this "experiment" in 2025, 2050, and 2100? How prepared is society to deal with the consequences of this experiment? Questions like these are the primary driver for the field of climate system modeling. Climate system models offer the best opportunity to explore the consequences of our fossil-fuel based energy portfolio. Unfortunately, when we start to explore the dynamics of global climate change we find that the majority of the consequences are regional. The questions we need to answer in order to quantify the integrated global threat are highly local in nature. Examples include: How will the occurrence and severity of drought in the Western United States change with increasing greenhouse gas emissions? How will the number and intensity of Atlantic hurricanes change with increasing greenhouse gas emissions? How likely is it that the ocean thermohaline circulation (that is largely responsible for the mild climate of northern Europe) will shut down? How likely is it that increasing drought in central Africa will lead to mass migration and political unrest? In order to quantify the global risk of increasing greenhouse gas concentrations we need to answer a long and difficult list of regional-impact questions.

Our understanding of the relationship between global climate change and regional climate change is tenuous. The global signal of temperature and precipitation change is the sum of the regional changes. Unfortunately the relationship, as we currently know it, does not extend much beyond that. The global signal is much more constrained than the regional signals; when greenhouse gases reduce the energy radiated to space, the climate system, as a whole, has to warm. Knowing the spatial pattern of that warming and the local feedbacks that the warming can induce is at the heart of regional climate modeling.

The basic scientific questions we need to answer here are fairly daunting: Can we predict regional climate change with sufficient accuracy and certainty to be useful to society? Can we predict, in a statistically-meaningful way, changes in extreme events due to increasing greenhouse gas forcing? How do we couple our climate system models to infrastructure models that are used to evaluate economic, water, and energy security? Needless to say, the basic science here is both deep and broad.

My approach to this daunting problem is to try to answer these questions in the context of individual regions, such as the Southwest United States. Region-by-region we need to try to link large-scale greenhouse gas forcing to regional climate change to impacts on local water/energy/agricultural/eco- systems. The global risk of climate change can then be evaluated as the "sum" of regional risks.