Climate change projections

Climate is currently changing due to anthropogenic greenhouse gas and aerosol emissions. However, future changes in climate cannot be predicted exactly, because

  1. The evolution of greenhouse gas and aerosol emissions in the future is difficult to predict
  2. Different climate models show different responses to changing atmospheric composition
  3. Climate also varies because of natural processes..

These uncertainties can be explored by comparing climate model simulations made with different models and different assumptions. In our research, we have used several dozens of different model simulations, produced in large international projects such as CMIP3 (http://www-pcmdi.llnl.gov/) and ENSEMBLES (http://www.ensembles-eu.org/).

Questions of interest:

  1. How reliable / unreliable are estimates of future climate change?
  2. How will anthropogenic climate change and natural variability in combination affect the climate in the next few decades?
  3. How to best combine observation- and model-based information when estimating future climate conditions?
  4. How has the gradual improvement of climate models affected projections of climate change?


Examples of results:

tuloksia

Example 1: Changes in annual mean temperature during the next 100 years in different climate models, assuming that greenhouse gas concentration increase following current business-as-usual projections. All model show higher temperatures in northern Europe, although the magnitude of the warming varies widely. Over the North Atlantic Ocean, however, some models indicate lower temperatures due to changes in ocean currents.

tuloksia

Example 2: July 2010 was record warm in Helsinki, with a monthly mean temperature of +21.7°C. Based on earlier observations alone (blue line and shading), such hot a July should only occur once in 300 years. However, when the recent climate change is taken into account, the statistical return period decreases to 60 years (orange line and shading). If climate change follows current best model-based estimates, such hot Julys would be repeated almost once per decade in the middle of this century.

tuloksia

Example 3. Time series of maximum annual snow amount (in mm of water equivalent) in Jokioinen and Sodankylä in three model simulations. Winters with abundant snow (more than the average for 1971-2000) are identified with blue bars. Winters with very little snow (less than half the mean for 1971-2000) are marked in red, and the winters between these two extremes in yellow. Although climate change is projected to reduce the amount of snow in general, particularly in southern Finland, individual winters with large amounts of snow are still expected in the future


Further reading

Räisänen, J., 2007: How reliable are climate models? Tellus, 59A, 2-29.
Räisänen, J., 2010: Ilmastonmuutos ja heinäkuun helteet. (Climate Change and the hot July 2010, in Finnish). Ilmastokatsaus 8/2010, s. 4-6.
Räisänen, J., ja J. Eklund, 2011: 21st century changes in snow climate in Northern Europe: a high-resolution view from ENSEMBLES regional climate models. Climate Dynamics, DOI: 10.1007/s00382-011-1076-3