Detection of anthropogenic influences on the evolution of temperature records over Europe
Abstract
Extreme events have significant impacts and are among the most serious challenges to society. Here we focus on extreme temperature events, which can cost human lives and affect several sectors (energy, agriculture, water resources, etc), as it happened during the 2003 heat wave in France. One way to study extreme temperature events is the analysis of the occurrence and values of temperature records. By definition, an upper or lower record breaking occurs when a higher or lower value of maximum or minimum surface temperature appears in the annual time series, recorded since an initial date and dependent of the calendar days. This study focuses on the mean upper and lower record occurrence changes over Europe during the 20th and 21st century. In a stationary climate, the occurrence of record breaking temperatures can be expressed by simple probabilistic laws. However, it has been shown using observations that since the beginning of the 1980s the number of record breaking temperature in Europe no longer follows these laws (Wergen and Krug 2010, among others). An increase or decrease in the mean number of upper or lower records is found compared to the stationary climate breaking rate. To understand these changes we use a set of historical (1850-2005) simulations performed with the CNRM-CM5 model. We first use a 10-member ensemble with anthropogenic and natural forcings to calculate the evolution of daily temperature record occurrences. We find the same behavior as in the observations: the upper and lower records diverge from the stationary climate record breaking rate from the 1980's. We then use two other ensembles to attribute these changes to a given set of forcings (either anthropogenic or natural). We show that the simulated record changes from 1980 onwards are mostly due to anthropogenic effects. Using a long (850-year) preindustrial simulation with constant forcings, we also assess these changes with regards to the model internal variability. We then analyze changes in record occurrence in the 21st century climate. We calculate record breakings in simulations following the 3 RCP scenarios (RCP2.6, RCP4.5 and RCP8.5). Under RCP8.5 (and to a lesser extent RCP4.5), the projected records show a high increase in upper record occurrence along the 21st century as well as a significant decrease in lower record occurrence. However, under RCP2.6, the projected upper and lower records tend towards a stationary climate behavior from the early 21st century. We finally investigate the projected temperature extreme events in Europe, in comparison with those of the current climate. One of the key points here is to estimate the maximum temperature that an upper record can reach at the end of the 21st century under the RCP8.5 scenario. In addition, we analyze the environmental conditions leading to the occurrence of the hottest summer temperatures such as the soil moisture preconditioning, nebulosity and circulation changes and their related feedbacks. We study whether the relative weight of these factors does indeed vary along the 21st century.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFMGC51B0955B
- Keywords:
-
- 1626 GLOBAL CHANGE Global climate models;
- 3337 ATMOSPHERIC PROCESSES Global climate models