Dynamically-downscaled projections of changes in temperature extremes over China
Junhong Guo,Guohe Huang,Xiuquan Wang,Yongping Li,Qianguo Lin
DOI: https://doi.org/10.1007/s00382-017-3660-7
IF: 4.901
2017-01-01
Climate Dynamics
Abstract:In this study, likely changes in extreme temperatures (including 16 indices) over China in response to global warming throughout the twenty-first century are investigated through the PRECIS regional climate modeling system. The PRECIS experiment is conducted at a spatial resolution of 25 km and is driven by a perturbed-physics ensemble to reflect spatial variations and model uncertainties. Simulations of present climate (1961–1990) are compared with observations to validate the model performance in reproducing historical climate over China. Results indicate that the PRECIS demonstrates reasonable skills in reproducing the spatial patterns of observed extreme temperatures over the most regions of China, especially in the east. Nevertheless, the PRECIS shows a relatively poor performance in simulating the spatial patterns of extreme temperatures in the western mountainous regions, where its driving GCM exhibits more uncertainties due to lack of insufficient observations and results in more errors in climate downscaling. Future spatio-temporal changes of extreme temperature indices are then analyzed for three successive periods (i.e., 2020s, 2050s and 2080s). The projected changes in extreme temperatures by PRECIS are well consistent with the results of the major global climate models in both spatial and temporal patterns. Furthermore, the PRECIS demonstrates a distinct superiority in providing more detailed spatial information of extreme indices. In general, all extreme indices show similar changes in spatial pattern: large changes are projected in the north while small changes are projected in the south. In contrast, the temporal patterns for all indices vary differently over future periods: the warm indices, such as SU, TR, WSDI, TX90p, TN90p and GSL are likely to increase, while the cold indices, such as ID, FD, CSDI, TX10p and TN10p, are likely to decrease with time in response to global warming. Nevertheless, the magnitudes of changes in all indices tend to decrease gradually with time, indicating the projected warming will begin to slow down in the late of this century. In addition, the projected range of changes for all indices would become larger with time, suggesting more uncertainties would be involved in long-term climate projections.