Perceived midlatitude jet waviness response to polar warming is sensitive to warming structure and metric choice

Arctic amplification has been proposed to influence midlatitude weather via a range of pathways. One hypothesis that has garnered wide public interest is that the weakening of the midlatitude temperature gradient due to Arctic warming causes larger meanders in the jet stream and so more intense weather extremes. However, previous work with idealized model simulations indicated that polar warming reduces waviness.

A variety of metrics have been developed to describe jet waviness, with some focused on jet geometry and others on the magnitude of the associated geopotential anomalies. Recent studies analyzing sea-ice loss and global warming simulations indicate that the response to polar warming may depend on its depth, with deeper warming having a larger effect on midlatitudes. Here we perform a variety of dry idealized model simulations in which we apply polar warming of different depths and latitudinal extents and assess the behavior of these different metrics.

Eddy heat transport decreases regardless of how polar warming is applied. However, unexpectedly we find that metrics relating to geometry (jet sinuosity, meander area) indicate a robust increase in jet waviness with polar warming. Meanwhile Local Wave Activity suggests increased waviness if warming is confined to the poles, but decreased waviness if warming extends into the midlatitudes. The apparent disagreement between metrics can be reconciled by assessing the changes in midlatitude geopotential gradient in the different simulations. Overall, these idealized simulations indicate that polar amplification could cause an increase in jet meandering, characterized by a shift of geopotential anomalies to smaller spatial scales and lower frequencies. However, the heat transport achieved and magnitude of pressure anomalies generated depend predominantly on the equator-to-pole temperature difference and geopotential gradient. Implications for weather extremes are discussed.