Relative importance of surface air temperature and density to interannual variations in monthly surface atmospheric pressure.

The spatial inhomogeneity of air pressure drives the atmospheric circulation. The interannual variation of the pressure can be affected by both air temperature and density. The goal of this study is to identify, for the near‐surface air, whether the variation of the density or temperature is more important in the interannual variation of the pressure. The physical relation of the pressure (P) with density (D) and temperature (T), denoted as P = D ⋅ T for simplicity, is nonlinear. To be convenient for estimating the relative importance, a normalized linear regression is used to fit the relation, which is finally written as P = AD + BT. Tests show that the linear fitting is robust; it is perfect everywhere in the globe. Because of the normalizations, the D and T are in magnitude of 1, but can be positive or negative. Derivations and calculations indicate that the coefficients A and B are both positive. What they reflect are the partial correlations of the pressure with the D and T, which are both equal to 1 across the entire field. For the normalized fitting, the A and B can be used to estimate the contributions of the D and T to the variation of the pressure. The linear fitting method can provide quantitative results for the dominance. The original nonlinear relation, due to the special form for the issue, can be used to qualitatively deduce the dominance. Comparisons suggest that the results obtained from the linear fitting are consistent with those qualitatively reached from the nonlinear physical relation. The consistency is true, no matter what the relation is between the two influencing factors. Corresponding to the negative relation between the density and temperature, there are two patterns. One is that the pressure increases with density but decreases with temperature. In this case, density dominates the variation of pressure. The second is that the pressure increases with temperature but decreases with density, and thus temperature dominates the variation of pressure. The third pattern is for the positive relation between density and temperature. In this case, density and temperature both have positive contributions to the variation of pressure, and which of them is more important can further be assessed with the linear fitting method. [ABSTRACT FROM AUTHOR]