Height-Dependent Convective Entrainment Rate Estimates Based on a Plume Model Constrained by Satellite Observations

Convective entrainment rate is one of the most important and sensitive parameters in climate models and contributes significantly to the uncertainties in modeled cloud feedbacks and climate sensitivity. Entrainment rate varies with height, as shown in both CRM simulations and in situ observations. In this study, we estimate height-dependent entrainment (HDE) rate using satellite observations of convective cloud top height, cloud-top buoyancy, and carbon monoxide, following the method described in Masunaga and Luo (2016). CloudSat data are used to identify deep convection and cloud top heights. Aura Microwave Limb Sounder (MLS)/Tropospheric Emission Spectrometer (TES) carbon monoxide measurements are collocated with CloudSat to obtain ambient carbon monoxide profile. A single plume model is applied to Atmospheric Infrared Sounder Advanced Microwave Sounding Unit (AIRS/AMSU) sounding data to produce a series of candidate profiles of HDE. These candidate HDEs are then weighted by the cloud and carbon monoxide observations in a Bayesian manner to produce the final estimate of HDE. To compare with prior studies and GCM simulations that assume entrainment rate is height invariant, the HDE profiles are further converted to equivalent height-independent entrainment rates, which are then sorted by different environmental conditions. We identify two modes of entrainment rates from our observations: deep convective entrainment rates of 5-10% km^-1 and cumulus congestus entrainment rates of 10-20% km^-1. Entrainment rates decrease with could top height and their relationships with environmental relative humidity and convective available potential energy (CAPE) are examined. Observed convective detrainment rates are also derived. Comparisons against the entrainment and detrainment profiles in GEOS-5 convective parameterizations are conducted.