Interactions between the atmospheric large-scale and the small-scale tropical convection using 15 years of data

Description

Convective clouds occur in a large-scale atmospheric disturbance at spatio-temporal scales that are much smaller than the disturbance itself. Because of this scale separation, numerical weather prediction and climate models use convective parametrization schemes to represent the small-scale processes as a function of the large-scale variables. Yet, although important progress were made in the development of parametrization over the last few decades, modelling of large and small scale phenomena are still a long-standing issue. We investigate the relationships between large and convective scales using 14 years (2001-2015) of observations in the Tropics. For this, we generate two parallel data sets that describe characteristics of both the small-scale convection and the large-scale atmospheric state. The first data set are radar observations from the region of Darwin, northern Australia, while the second one is derived from a variational analysis of that same period and domain. The data sets used herein contain more than three millions individual convective cells over more than 300,000 time steps. We study the correlation of the small-scale variables against the large-scale variables, like the convective available potential energy, the convective inhibition, the relative humidity and the vertical velocity at 500 hPa. While previous studies did not find any correlation between the small scale and the large, the use of the number and the mean area of convective cells as descriptors of the small scale convection instead of the convective area fraction allows a better representation of the relationships between the two scales. We find that CAPE impact the mean area of the larger convective cells and that CIN anti-correlates (r=-0.72) with the number of convective towers. Strong relationships are found when looking at the phase-space of one scale as a function of the other scale. These phase-space diagrams can be used to describe more accurately the interrelationships of the large-scale and the small-scale.