Characteristics of boreal winter cluster extreme events of low temperature during recent 35 years and its future projection under different RCP emission scenarios

Cluster extreme event (CEE), which is characterized by large affected area and long duration of extreme weather, has been arousing the worldwide serious concerns due to its severe impact on society. The winter cluster extreme events of low temperature (LT_CEEs) in the northern hemisphere from 1979 to 2013 are identified with a simplified objective method based on ERA-Interim and JRA-25 daily minimum surface air temperature. The probability density function (PDF) distributions of most indices in the winter LT_CEEs derived from the two datasets are well consistent with each other, especially in the occurrence frequency, duration, and affected area. As expected, the downward trend of all indices in recent 35 years under global warming is congruously revealed from both reanalysis. Nevertheless, the various indices of the winter LT_CEEs after 1998 are generally stable accompanied with slight upward trend, which might be closely related to the speed slowdown of global warming but require further investigation. Similar analysis was carried out with the simulated results of BCC_CSM1.1 model under four scenarios of RCP2.6, RCP4.5, RCP6.0, and RCP8.5. The occurrence frequency of the winter LT_CEEs under RCP2.6 remains stable after 2050, but significantly decreases under RCP4.5 and RCP6.0, and disappears under RCP8.5 scenario. Overall, the descent rate of the winter LT_CEEs accelerates with the emission rise.