Links between atmospheric blocking and North American winter cold spells in two generations of Canadian Earth System Model large ensembles

Dae Il Jeong,Bin Yu,Alex J Cannon

doi:10.1007/s00382-021-05801-0

Abstract

Due to the significant negative consequences of winter cold extremes, there is need to better understand and simulate the mechanisms driving their occurrence. The impact of atmospheric blocking on winter cold spells over North America is investigated using ERA-Interim and NCEP-DOE-R2 reanalyses for 1981–2010. Initial-condition large-ensembles of two generations of Canadian Earth System Models (CanESM5 and its predecessor, CanESM2) are evaluated in terms of their ability to represent the blocking-cold spell linkage and the associated internal-variability. The reanalysis datasets show that 72 and 58% of cold spells in southern and northern North America coincide with blocking occurring in the high-latitude Pacific-North America. Compared to the two reanalyses, CanESM2 and CanESM5 ensembles underestimate by 19.9 and 14.3% cold spell events coincident with blocking, due to significant under-representation of blocking frequency over the North Pacific (− 47.1 and − 29.0%), whereas biases in cold spell frequency are relatively small (6.6 and − 4.7%). In the reanalyses, regions with statistically significant above-normal cold spell frequency relative to climatology lie on the east and/or south flanks of blocking events, whereas those with below-normal frequency lie along the core or surrounding the blocking. The two ensembles reproduce the observed blocking-cold spell linkage over North America, despite underestimating the magnitude of blocking frequency. The two ensembles also reproduce the physical drivers that underpin the blocking-cold spell linkage. Spatial agreement with the reanalyses is found in the simulated patterns of temperature advection and surface heat flux forcing anomalies during blocking events. While CanESM5 shows an improved representation of the blocking climatology relative to CanESM2, both yield similar results in terms of the blocking-cold spell linkage and associated internal-variability.

Highlights

Winter cold extremes have adverse impacts on various human and natural systems (e.g., Roland and Matter 2013; Sheridan and Allen 2015)
We investigate the connection between blocking and cold spells over the Pacific-North American sector in winter (December–February) based on ERAInterim reanalysis (Dee et al 2011) and National Centers for Environmental Prediction-Department of Energy reanalysis 2 (NCEP-DOE-R2) (Kanamitsu et al 2002) over the 1981–2010 period
Canadian Earth system model version 5 (CanESM5) performs better than CanESM2 in reproducing the blocking climatology, as CanESM5 and CanESM2 underestimate around 3 and 4 days of blocking, respectively, for each season when compared to NCEP-DOE-R2

Summary

Introduction

Winter cold extremes have adverse impacts on various human and natural systems (e.g., Roland and Matter 2013; Sheridan and Allen 2015). The cold wave and snowstorm that occurred over North America in early 2014 was blamed for economic losses in excess of 3 billion USD (LeComte 2015). Given the severity of the impacts of such events, it is of great importance to investigate the mechanisms that drive winter cold extremes. Blocking events that occur in the North Pacific and North Atlantic in winter contribute to cold extremes over North America and Europe, respectively, by inducing northerly cold air advection (e.g., Buehler et al 2011; Sillmann et al 2011; Pfahl and Wernli 2012; Whan et al 2016). The connection between blocking and cold extremes has often been investigated by statistical approaches, including

Objectives

Results

Conclusion