Hard to predict how humans will respond to climate change

The largest uncertainty in future climate projections is the level of anthropogenic emissions expected throughout the 21st century. As Phil Duffy from Climate Central (and a Data Basin Advisor) often reminds us in his talks: it is IMPOSSIBLE to project future human emissions because human behavior is unpredictable.

Our societies may switch to renewable sources of energy, subsidize public transport, adopt a “greener” style of living... or not! As our climate changes, our activities change and, in western countries where technology is often available, our overall resource use shifts.  A typical illustration of how human behavior may change with climate was posted on Cliff Mass’s weather blog. It shows water use in Seattle this year and for 2009 as well as for two periods 1985-1991 and 1999-2008 (source: Seattle Public Utilities). Water use was down this cool wet spring. Fewer people were irrigating gardens (usage just above winter levels). More importantly, this graph clearly shows how much less water is used in Seattle use today compared to the 1980s, despite the increase in population. As Cliff notes, “clearly, water-conserving toilets and showerheads are making a major difference, and perhaps some individuals are making better landscaping decisions (drought-tolerant plants, drip irrigation, etc)”.

In the summer of 2003, when European suffered from a two-week extreme heat wave (temperature anomalies in France were ~10oC) that caused over 30,000 casualties, one little known fact was that nuclear power plants that provide about 80% of France’s electricity had to shut down to cool their cores. Not enough cold water was available to the plants to remain below risk levels. This happened at the same time demand was rising to power various cooling devices. Imagine a 10oC increase in your average summer temperature in the United States for a whole two weeks and imagine the increase in energy demand in a country where air conditioners are the norm (they are not in western Europe where ocean or sea usually moderate the climate on all sides). Will our grid be able to deal with this surge in demand? What if more than one region was affected? What if every summer the same spike in demand was to be expected? What if air conditioners stopped functioning?

Mike Dettinger recently presented a summary of ArkStorm, an exercise to test the vulnerability of the state of California to an extreme event, in this case extreme flooding both in Northern and Southern California. A useful exercise not only to realize how likely such an event could be (as Mike said, to create the climate scenario was easy, he did not have to go much beyond what had already happened in the Fall/winter of 1968-69) but how vulnerable our infrastructure is to extremes in climate.

Based on results such as these, state and city officials may take decisions that they would not have made had they not gone through the exercise. When we talk about climate projections (keep an eye out in Data Basin for soon-to-be published datasets from California Academy of Sciences, Healy Hamilton’s group) and the uncertainties associated with the climate models, one would do well to remember that the most difficult thing for modelers peering into the future is to figure out what political decisions and social choices will be made between now and the next extreme event, and how these will affect emissions and how much greenhouse effect our future will be subject to.