The complex structure of a typical stratus cloud base height (or profile) time series is analyzed with respect to the variability of its fluctuations and their correlations at all experimentally observed temporal scales. Due to the underlying processes that create these time series, they are expected to have multiscaling properties. For obtaining reliable measures of these scaling properties, different methods of statistical analysis are used herein : power spectral density, detrended fluctuation analysis, and multifractal analysis. This broad set of diagnostic techniques is applied to a typical stratus cloud base height (CBH) data set; data were obtained from the Southern Great Plains site of the Atmospheric Radiation Measurement Program of the Department of Energy from a Belfort Laser Ceilometer. First, we demonstrate that this CBH time series is a nonstationary signal with stationary increments. Further, two scaling regimes are found, although the characteristic laws are quite similar ones. Next, the multi-affine scaling properties are confirmed. The scaling properties of the cloud base height profile of such a continental stratus are found to be similar to those of the marine cloud base height profiles studied by us previously. Some physical interpretation in terms of anomalous diffusion (or fractional random walk) is given for the continental case.

The scaling ranges of time correlations in the cloud base height records of marine boundary layer stratocumulus are studied applying the Detrended Fluctuation Analysis statistical method. We have found that time dependent variations in the evolution of the α exponent reflect the diurnal dynamics of cloud base height fluctuations in the marine boundary layer. In general, a more stable structure of the boundary layer corresponds to a lower value of the α -indicator, i.e. larger anti-persistence, thus a set of fluctuations tending to induce a greater stability of the stratocumulus. In contrast, during periods of higher instability in the marine boundary, less anti-persistent (more persistent like) behavior of the system drags it out of equilibrium, corresponding to larger α values. From an analysis of the frequency spectrum, the stratocumulus base height evolution is found to be a non-stationary process with stationary increments. The occurrence of these statistics in cloud base height fluctuations suggests the usefulness of similar studies for the radiation transfer dynamics modeling.

No abstract