An entropy-based measure of mixing at the tropopause

Qualitative analysis of tracer correlations is now a well established technique for investigating mixing in the tropopause transition layer. Generally, these studies rely on in situ data retrieved over short distances and time-scales. Using satellite data from HALOE, we introduce a simple quantitative analysis technique, based on entropy, to provide a climatology of mixing, from H2O:O-3 tracer correlations. Results compare well with previous studies using Lagrangian-based estimates of effective diffusivity and equivalent lengths and also climatologies of tropopause folding. Further analysis techniques reveal characteristics of the transition layer, including its depth and position, leading to the distinction of three primary forms of mixing. Within our analysis domain, we estimate that half of total entropy can be attributed to subtropical 'Rossby-driven' tropopause folding events. The majority of the remaining entropy is associated with tropical convective mixing and a third form, possibly linked to shear-induced mixing at subtropical jet streaks.