Improving Estimates of Cloud Properties Through the Application of an Adiabatic Spectrally Consistent Retrieval to the MODIS Cloud Product

Abstract:

We attempt to infer cloud vertical structure and improve estimates of cloud microphysical properties through the application of an Adiabatic Spectrally Consistent Retrieval (ASCR) to Moderate Resolution Imaging Spectroradiometer (MODIS) observations. The MODIS Cloud Product provides estimates of cloud optical thickness and droplet effective radius for three near-infrared absorption wavelengths (1.6, 2.1 and 3.7 mm) under the assumption of a plane-parallel, vertically homogeneous (VH) cloud. This is not a physically realistic assumption for boundary layer clouds, where an adiabatically stratified liquid water content profile conforms better. ASCR transforms VH retrievals of optical thickness and droplet effective radii into adiabatically stratified retrievals, exploiting the varying photon penetration depth of each absorption channel. Taking advantage of the data screening and quality controls applied to the MODIS Cloud Product, existing retrievals of optical thickness and droplet effective radii are inverted to obtain equivalent scene reflectances from which two-channel and four-channel adiabatically stratified retrievals of cloud geometrical thickness (H) and cloud droplet number concentration (N) are performed using an optimal estimation framework. Through a comparison of the 2-channel and 4-channel N and H retrievals, we attempt to estimate the degree to which a cloud conforms to an adiabatically stratified model, near cloud-top. Results will be presented, demonstrating ASCR’s performance relative to VH retrievals from the cloud product through an analysis of one year’s observations of marine stratocumulus from MODIS near the South American and African Continents.