Emergent Relationships Among Sea Ice, Longwave Radiation, and the Beaufort High
This work uses a perturbed parameter ensemble of coupled climate model simulations to identify emerging relationships between sea ice area, net surface longwave radiation, and the atmospheric circulation over the Beaufort Sea in the Arctic. The study found a strong positive correlation between sea ice area and the net longwave radiation over the ocean‐ice surface during the melting season and a negative correlation during the freezing season. The mechanisms responsible for the longwave radiation balance at the surface are mainly driven by sea ice variations in the freezing season and by clouds in the melting season. A strong positive (negative) correlation was also found between the fall (summer) total sea ice area in the Arctic and the sea level pressure over the Beaufort High region. We argue that as sea ice coverage is lost, static stability losses are severe in the fall, resulting in enhanced evaporation, vertical motions, and weakening of the general large‐scale anticyclonic circulation of the Beaufort High.
The novel approach of identifying emergent constraints in an ensemble of climate model simulations was used to examine the response of the Arctic climate simulated by E3SMv0-HiLAT across different climate states. Two emergent constraints have been identified which could guide further research to refine our understanding of high latitude feedbacks. These results suggest the need to better characterize the role of moisture in coupling ocean, sea ice, and atmospheric interactions. Moreover, this research constitutes one of the few attempts to date to use more efficient space-filling sampling methodologies to examine sensitivities to model parameters in fully coupled climate simulations, which can help constrain model calibration activities.
The study examines responses of E3SMv0‐HiLAT climate model simulations to parameters previously identified as driving uncertainty in stand‐alone model components. In particular, we examine the joint effect of two cloud parameters, two snow-on-sea-ice parameters, and an air‐ocean coupling parameter. A model ensemble consisting of 24 E3SMv0‐HiLAT model configurations was designed in which the five model parameters were perturbed using a sampling methodology aimed at more efficiently exploring the five‐dimensional parameter space. The study found relationships that suggest different mechanisms mediating the coupled climate interactions in the melting and the freezing seasons. Of particular interest were (1) the seasonal relationships between the net surface longwave radiation and the total ice area in the Arctic and (2) the relationship of the atmospheric circulation over the Beaufort Sea and the total ice area in autumn. This study highlights intertwined cloud and sea ice interactions in the Arctic and how these mechanisms act differently during the melting and freezing seasons.