Fall Speeds of Cirrus Cloud Ice Crystals Are Faster than Thought

A new look at ARM field campaign data finds larger bullet rosette crystal masses for a given projected area than in previous studies.

The Science

The fall speeds of high cirrus cloud crystals depend on crystal mass, for which no direct measurements exist. Based on new measurements of crystal images, calculated mass yields fall speeds as a function of maximum particle dimension that are, surprisingly, roughly twice as large as derived in the most commonly used past studies.

The Impact

High ice clouds cover large parts of the Earth; past studies have found that the extent to which they warm or cool the planet is sensitive to assumed ice crystal fall speeds.


When scientists supported by the Department of Energy’s (DOE) Atmospheric System Research (ASR) program embarked on this study to prepare internally consistent ice physical and optical properties, they expected to corroborate past derivations of bullet rosette mass as a function of maximum dimension (a physical property of the ice). The team worked with in situ observations from two cirrus clouds sampled during the 2010 Small Particles in Cirrus (SPartICus) field campaign supported by DOE’s Atmospheric Radiation Measurement (ARM) Climate Research Facility. The team found that while their crystals had projected areas, which are directly measured, that were similar to those in previous studies, their calculations yielded substantially greater crystal masses than previously found. The researchers identified several likely sources of error in previous studies. First, virtually no direct measurements of individual crystal mass exist for cirrus particles, so masses must be calculated based on crystal shape and maximum dimension. Second, the maximum dimension commonly used for projected area is randomly oriented, whereas that used for idealized calculations of mass is true maximum dimension; they find that randomly oriented maximum dimension is substantially smaller. Finally, large uncertainties are expected in particle mass derived from measured particle size distributions and total ice mass, and such measurement uncertainties (in both bin-wise number concentration and total ice mass) have remained essentially uncharacterized in the literature to date.

Principal Investigator(s)

Ann M. Fridlind
National Aeronautics and Space Administration (NASA) - Goddard Institute for Space Studies


This work was supported by DOE, Office of Science, Office of Biological and Environmental Research and NASA Radiation Sciences Program under agreements DE-SC0006988, DE-SC0008500, and DE-SC0014065. This research used resources at the National Energy Research Scientific Computing Center, a DOE Office of Science user facility, under contract number DE-AC02-05CH11231; and the NASA High-End Computing (HEC) Program through the NASA Advanced Supercomputing (NAS) Division at Ames Research Center.


Fridlind, A. M., R. Atlas, B. van Diedenhoven, J. Um, G. M. McFarquhar, A. S. Ackerman, and E. J. Moyer. 2016. “Derivation of Physical and Optical Properties of Mid-Latitude Cirrus Ice Crystals for a Size-Resolved Cloud Microphysics Model,” Atmospheric Chemistry and Physics 16, 7251-83. DOI: 10.5194/acp-16-7251-2016.