The Global Diabatic Circulation in Reanalyses: A Comparison to the Residual Circulation, Water Vapor, and Ozone

Recent work has shown that differences in vertical velocity due to calculation method can be greater than the differences among the reanalyses themselves. By comparing the three typical methods for calculating the residual circulation vertical velocity with the global diabatic overturning circulation on isentropes, we explore advantages and disadvantages of these different methods. The methods that rely on eddy fluxes have large vertical autocorrelation, making them less appropriate for determining changes in the circulation that differ between the upper and lower branch. The global diabatic circulation is theoretically related to tracer distributions, making it a good metric. However, there are discontinuities in the radiative heating rates that make the diabatic circulation and radiatively determined vertical velocity also potentially problematic, especially for determining trends. The water vapor tape recorder and ozone are also examined, and it is shown that the diabatic circulation determined from the radiative heating is a better predictor of the tracer distributions than the residual circulation in a comprehensive climate model. The reanalyses have the same qualitative relationship with observed ozone as in the comprehensive model. These results suggest that when analyzing the stratospheric circulation from reanalysis products, it is necessary to use at least one momentum derived vertical velocity and one metric determined by the radiative heating rates to get a complete picture of the variability and trends.