Challenges in Analyzing and Representing Cloud Microphysical Data Measured with Airborne Cloud Probes

Abstract:

There are a variety of in-situ instruments that are deployed on aircraft for measuring cloud properties, some of which provide data which are used to produce number and mass concentrations of water droplets and ice crystals and their size and shape distributions. Each of these instruments has its strengths and limitations that must be recognized and taken into account during analysis of the data. Various processing techniques have been developed by different groups and techniques implemented to partially correct for the known uncertainties and limitations. The cloud measurement community has in general acknowledged the various issues associated with these instruments and numerous studies have published processing algorithms that seek to improve data quality; however, there has not been a forum in which these various algorithms and processing techniques have been discussed and consensus reached both on optimum analysis strategy and on quantification of uncertainties on the derived data products.

Prior to the 2014 AMS Cloud Physics Conference, a study was conducted in which many data sets taken from various aircraft (NCAR-130, North Dakota Citation, Wyoming King Air and FAAM BAE-146) and many instruments (FSSP, CDP, SID, 2D-C/P, CIP/PIP, 2D-S, CPI, Nevzorov Probe and King Hot-wire LWC sensor) were processed by more than 20 individuals or groups to produce a large number of derived products (size distributions, ice fraction, number and mass concentrations, CCN/IN concentrations and median volume diameter). Each person or group that processed a selected data set used their own software and algorithm to produce a secondary data file with derived parameters whose name was encoded to conceal the source of the file so that this was a blind comparison.

The workshop that was convened July 5 and 6, 2014, presented the results of the evaluation of the derived products with respect to individual instruments as well as the types of conditions under which the measurements were made. This comparison will ultimately allow quantifying the error bars of derived parameters as a function of the conditions in which the observations are made. The results of this evaluation and the recommendations that evolved from the workshop will be summarized in this presentation.