Spatial clustering of cloud microphysical conditions, such as liquid water content (LWC), particle concentration, and temperature can affect the rate of growth and shape of in flight ice accretions that form on aircraft, thus affecting their airworthiness. Clustering may also affect the ability of radars and microwave radiometers to accurately map icing conditions ahead of aircraft and provide information for avoidance or escape. In this paper we evaluate the spatial patterns of cloud microphysics at the sub-kilometer scale in cold clouds measured by NASA research flights in the Great Lakes region.

Our approach, using fundamental techniques formulated by Jameson and Kostinski, determines the correlation length, or ‘typical size,’ of clusters and computes a clustering intensity parameter for a transect of cloud. The NASA flight data is composed of one-second LWC, particle concentration, and temperature measurements for 33 flight segments at constant altitude ranging from 12 minutes (52 km) to 86 minutes (290 km) in duration. Flight segment cluster intensities range from nearly Poissonian, at a cluster intensity of 0.06, to highly clustered at a cluster intensity of 1.77. Correlation lengths range from 0.07 to 38.6 km. We found no statistically significant relationships between cluster intensity and correlation length. We also present the clustering characteristics of particle concentration and temperature in the report.

Our need was to describe the spatial patterns of cloud microphysics for modeling the potential performance of remote sensors scanning horizontally ahead of aircraft for mapping icing conditions. Therefore, we also evaluated clustering over distances that might represent the range of a radar or microwave radiometer. We chose ranges of 20 and 40 km. Plotted as running, overlapping series, the range plots indicate that cluster intensity and correlation length variability along these short transects can be extreme.

Additional flight data would allow us to determine if climatologies can be developed to characterize microphysics cluster conditions by region, time period, or weather type. We are currently developing inversion methods for creating, from cluster climatologies, synthetic cloud microphysics spatial series that would be useful to modelers.