Origin of lapse rate changes in the upper tropical troposphere

Abstract

Vertical motions in clouds arise from a variety of thermodynamic processes, including latent heat release, evaporative cooling, melting, and cloud radiative heating. In the Tropics, the net upward vertical mass flux from convective systems should approximately balance subsidence in clear sky regions associated with radiative cooling, provided the exchange of mass with midlatitudes can be assumed small. Tropical climatologies of temperature, water vapor, and ozone are used to calculate the clear sky radiative mass flux, and the derivative of this mass flux with respect to potential temperature, dMr (x)/dx, is used as a proxy for net convective outflow. Convective outflow increases rapidly at 345 K (~11.3 km). This corresponds to the pseudoequivalent potential temperature xe at which air parcels near the surface first attain positive convective available potential energy (CAPE). The rate at which dMr(x)/dx decreases above 345 K is similar to the rate at which the near surface xe probability distribution function (PDF) decreases. This behavior is referred to as "scaling." It suggests that the timescale for removal of an air parcel from the convective boundary layer is independent of xe (once it has positive CAPE), and that the residual vertical mass flux from convective clouds can be described as if air parcels detrain near their level of neutral buoyancy (LNB). It is also suggested that the mean tropical temperature profile above 345 K is controlled, not by mixing, but by the need for the vertical variation in net convective outflow to be consistent with the near-surface xe PDF, and that this accounts for the fact that the mean temperature profile above 345 K increasingly deviates from a moist adiabat. It is also argued that there are sufficient high xe air parcels near the surface to sustain the Brewer-Dobson circulation by detrainment at the LNB followed by radiative ascent into the stratosphere.