New Method Simulates 3D Ice Crystal Growth Within Clouds

Bhattacharya, A., Pacific Northwest National Laboratory

Radiation Processes

Cloud Life Cycle

Harrington JY, K Sulia, and H Morrison. 2013. "A Method for Adaptive Habit Prediction in Bulk Microphysical Models. Part I: Theoretical Development." Journal of the Atmospheric Sciences, 70(2), 10.1175/jas-d-12-040.1.

Harrington JY, K Sulia, and H Morrison. 2013. "A Method for Adaptive Habit Prediction in Bulk Microphysical Models. Part II: Parcel Model Corroboration." Journal of the Atmospheric Sciences, 70(2), 10.1175/jas-d-12-0152.1.

A close-up of ice crystals.

A close-up of ice crystals.

Ever noticed the different shapes of snowflakes sticking on the windowpane on a snowy day? The flakes are mostly just several crystals sticking together. Each ice crystal, however, is fragile and comes in unique, often intricate, three-dimensional shapes. Scientists find it challenging to simulate the growth of ice crystals, especially when they are suspended within clouds.

Ice crystals grow along not one, but several directions, which scientists refer to as ‘crystal growth axes’. The growth of ice along these axes is not uniform, especially within clouds where humidity and temperature change very quickly. The crystals grow at different rates along different axes, and even along a single axis, and the rates change over time. Thus, different growth rates along these axes determine how big and how dense these ice crystals become and also the rate at which they ultimately fall out, either as snow or rain.

Cloud models, however, operate on the assumption that within clouds, ice crystals grow along a single axis and at a constant rate. Such assumptions, purely for the ease of calculations, result in errors—often as large as 20–40 percent—in predicting the ice and liquid water content of the clouds and ultimately how these clouds rain out or snow out.

With funding from the Department of Energy’s Atmospheric System Research (ASR) program, a team of scientists developed a method to account for differences in growth rates of ice crystal along two axes. Led by Jerry Y. Harrington of the Department of Meteorology at The Pennsylvania State University, their model is closer to what scientists think actually occurs within clouds.

The researchers calculated how such ‘nonlinear’ or varying growth patterns affect the mixing of water and ice within cloud and ultimately the speed at which they fall out. The team published their results earlier this year in two related papers in the Journal of the Atmospheric Sciences. Their method reduced errors in estimating physical parameters of ice crystals within clouds to 5%. The technique appeared to be particularly efficient in mixed-phase clouds, or those which contain both ice crystals and water droplets and produce maximum precipitation.

However, more work needs to be done in order to more completely represent ice crystal formation within clouds. “Though the method presented here has numerous advantages, it has yet to be tested more thoroughly for situations in which ice crystals are exposed to varying temperatures and lower ice super saturation,” says Harrington.