By all accounts, 2011 was a busy year for climate extremes. Readers from Australia will no doubt remember the year starting with extremely heavy rainfall drenching much of the state of Queensland, followed by the catastrophic flash floods in Toowoomba in the middle of January, the Brisbane floods only a few days later, and cyclone Yasi devastating northern Queensland in early February. Taking a more global perspective, it can be seen in the figure below that this was only one of many major climate disasters which occurred that year, with other notable extremes in Thailand, the Philippines, the USA, parts of Africa and elsewhere (figure prepared by Michael Leonard based on information from the International Disaster Database).
What causes these extremes, and how can we evaluate the risk that they will recur or become even more severe in the future? These questions are much more complex than is commonly appreciated, and numerous research groups around the world are working at finding answers. We now understand that most extremes of importance to society occur because of the anomalous state (either very high or very low) of combinations of atmospheric variables such as temperature, rainfall, humidity and wind, rather than individual variables acting in isolation. Furthermore not all of these variables need to be extreme for the impact to occur. For example, a bushfire might occur because of high temperatures, strong winds and low humidity, but having a record-breaking heatwave is not necessary for a record-breaking bush fire to occur. This complicates things, because just having projections of how extreme temperature will increase will not be sufficient to develop projections of future bushfire risk.
Such combinations are referred to as “compound extremes” by the Intergovernmental Panel on Climate Change Special Report on Extremes, because of the growing understanding that each extreme is in one way or another unique, and will all involve a large number of interactions in both space and time between multiple meteorological processes. This focus on interactions explains why climate scientists are typically much more confident about projections of variables such as temperature and rainfall, than they are about projections of future flooding, storms, droughts, and wildfires.
So are extremes increasing? The answer is that they probably are in most places around the world, but that this increase will not be uniform and will depend on a diversity of meteorological processes. In most cases, however, recent research has been telling us that it is the regions that are least well equipped to deal with the extremes that are likely to experience the sharpest increase in risk. With continued population increases and mass migrations of people to urban environments, it therefore has never been more critical to understand the causes of climate extremes and the regions where the risks are likely to be greatest. Without this understanding, it will be nearly impossible to develop the diversity of measures that are required to ensure we minimise the most serious consequences to society.