Study on the “Warming Hole” in America Based on CMIP5 Models

Xing-xing GAO, Yan CHEN, Lei ZHANG, Wu ZHANG

Abstract


Studying the “warming hole” in the eastern United States can help us understand the warming’ mechanisms and its relationship with nature climate variability. Here we explored the relationships between the calculation of trends and the selected time window and between the “warming hole” and AMO, PDO and IPO, along with the temporal variability for 30 years running temperature trends in America and the North Atlantic. We also evaluated the ability of 21 CMIP5 climate models to simulate the “warming hole” and AMO in the twentieth century and conducted the twenty-first-century “warming hole” and AMO projections for two future emission scenarios based on the known uncertainty in CMIP5 twentieth-century simulations. The results indicate that the selected time window has a great impact on the calculation of trends, the observed United States warming hole is closely associated with AMO, PDO and IPO, the temporal variability for 30 years running temperature trends in America and the North Atlantic is sinusoidal and has a 67-yr cycle and that most CMIP5 models simulate significantly lower relative power in AMO than observed. The results also show that GFDL-ESM2G, MIROC-ESM and CanESM2 from 19 CMIP5 models have the best performances in capturing features related to the twentieth-century “warming hole” and that there is about 10% chance in a PCR4.5 scenario and no chance in a PCR8.5 scenario for the warming hole’s reappearance in the twenty-first century.

Keywords


Multidecadal climate variability, Warming hole, AMO, CMIP5


DOI
10.12783/dteees/eesd2017/11982

Full Text:

PDF

Refbacks

  • There are currently no refbacks.