Supercooled cloud scale length and correlative relationships

Characterizing spatial variations of supercooled cloud icing environments is critical to improving understanding of icing effects on aircraft aerodynamics. In addition, cloud spatial properties and correlative relationships between variables impact information that can be retrieved from icing remote sensing systems. Properties of in-cloud icing environments have been a continuous topic of study since the 1940s when initial work was conducted at Mount Washington Observatory (MWO), and by the National Advisory Committee for Aeronautics (NACA) in Cleveland, Ohio. From these early studies and more recent work characterizations of supercooled clouds have occurred for establishing test procedures and predicting ice accretion. However, these recommendations often do not completely characterize the fluctuation of microphysical conditions within clouds.

Aircraft traversing clouds accumulate ice as a function of the spatially changing microphysical conditions within the clouds, with ice shape changing as cloud conditions change. Recent studies have demonstrated that ice shape, which increases drag, is affected by the spatial variation of cloud microphysical conditions. And, since icing remote sensing devices scan within and through clouds, it is essential that a better understanding be developed of the fluctuation of cloud microphysical properties for improving retrievals.

In this report we characterize and compare scale lengths of supercooled cloud liquid water content (LWC) using cluster analysis and thresholding methods for clouds ranging from nearly all liquid to nearly fully glaciated. We also assess correlative relationships between LWC and other cloud properties. Finally, we compare scale lengths of LWC with scale lengths of other cloud properties. All analyses were performed using measurements made by the NASA Glenn Research Center (NASA-GRC) Twin Otter research aircraft during the Supercooled Large Drop Research Program (SLDRP) in the southern Great Lakes in 1997-1998. All results are from flight segments conducted at near constant altitudes to allow better assessment of horizontal cloud variation.

Scale lengths of cloud LWC were computed using methods presented by Jameson and Kostinki for describing cluster intensity and coherence length of cloud properties. Scale lengths were also created by thresholding LWC magnitude and computing distances for LWC greater than the threshold magnitude. These techniques were used for over thirty flight segments isolated from the SLDRP flight program. Coherence lengths varied widely, ranging from less than 100 m to over 38 km in length. Most coherence lengths were a few kilometers in length. Results from the thresholding methods were more varied because scale length was determined for specific LWC magnitudes and are not “bulk” values as are the coherence lengths. In general, however, scale length decreases as LWC increases.

Correlations between LWC and other cloud variables were made using measurements recorded at 1-s intervals. Variables related to LWC include static temperature, total temperature, ice detector icing rate, dew point temperature, particle concentration, and median volume diameter (MVD). In general, correlations are quite low.

We will continue to refine our methods and seek relationships between scale lengths of LWC and other variables. In addition, in future work we will explore mixed phase conditions.