Balance of drag between the middle and lower atmospheres in a global atmospheric forecast model

A global atmospheric forecast model that includes the middle atmosphere is used to investigate systematic model biases. As in many other global atmospheric models, Northern Hemispheric winter simulations from the vertically extended Navy Operational Global Atmospheric Prediction System reveal an unrealistic vertical tilt of the polar‐night jet associated with a temperature bias in the northern polar middle atmosphere. This bias has important implication for satellite data assimilation in that inaccurate model background field associated with the bias can cause a data assimilation system to reject useful observation data. Efforts are made to alleviate this bias using improved radiation and orographic gravity wave drag parameterizations, the latter of which can provide adequate amount of drag to greatly reduce the bias, but at the cost of overestimated surface polar pressure. To understand and fix this secondary bias, this study addresses the cause and effect of the bias in terms of the balance of model drag mechanisms between the middle and lower atmospheres as well as within the lower atmosphere. Disruption of these balances in the model is argued to lead to degradation of surface simulation by reasonable middle atmospheric gravity wave drag despite an improvement in the middle atmospheric circulation. An experimental treatment of limiting the parameterized gravity wave drag to the middle atmosphere, as similarly done with a spectral gravity wave drag parameterization, is shown to largely alleviate the bias. This study discusses results from a series of winter and multiyear simulations, schematically illustrates the drag balance in the model, and investigates conservation of atmospheric angular momentum.