Study of scale controlling factors of of marine boundary layer mesoscale cells observed from AMSR-E using wavelet analysis

Marine boundary layer (MBL) clouds tend to organize into closed or open mesoscale cellular convection (MCC) with cell sizes greater than 10 km. A conventional way of quantifying the MCC scale uses the Fourier transform of cloud water path (CWP). This approach works well for cells with uniform scales in space. However, the MBL cloud scales typically violate the uniform assumption. Under this circumstance, the Fourier analysis only returns a multi-frequency spectrum while is incapable of identifying the respective scales in the spatial domain without proper windowing. Here we apply wavelet analysis to passive microwave retrievals of CWP from AMSR-E (~10 km resolution) and a machine-learning classification of cell type to calculate the local scale of open and closed MCC. We bring in AMSR-E retrieved water vapor path and precipitation to obtain composite cell structures.

We use a full year of the CWP measurements from AMSR-E in 2008 over the Southeast Atlantic. CWP is decomposed to four mesoscale wavelength octaves (20, 40, 80, and 160 km). In each 160 x 160 km² domain, we calculate the CWP variances of the four wavelength band, and identify cell scale as the wavelength of the peak variance in the spectrum.

Our current results show fairly constant correlation between wavelet decomposed CWP and WVP (~ 0.05 mm increase in CWP per mm increase in WVP) across cell scales regardless of cloud type, and there is no obvious dependence of cell scales on potential controlling factors such as boundary layer depth, precipitation, SST, free-tropospheric humidity, wind speed, etc. This suggests that the increase of cell scale might stem more from its historical evolution than from the environmental conditions. We will expand our analysis to the Northeast and Southeast Pacific to get a global view.