Biological and Environmental Research - Earth and Environmental System Sciences
Earth and Environmental System Modeling
27 April 2020

Predicting the Particles that Initiate Freezing in Clouds

Accounting for both dust and sea spray particles improves modeled number concentrations of particles that cause ice to form within marine clouds.

Sea spray
Science

Freezing of liquid cloud droplets into ice is often triggered by solid particles in the droplets, or ice-nucleating particles (INPs). Most global models of the atmosphere predict the number of INPs based on temperature or sometimes the amount of desert dust or total particles in the air. Windblown dust is an important source of INPs, but recent observations have shown that in clean marine air, sea spray particles can also contribute. Global models that do not account for different contributions of INPs from dust, sea spray, and other sources may struggle to accurately predict observed INP numbers in many regions of the Earth. Researchers at Colorado State University and the U.S. Department of Energy’s Pacific Northwest National Laboratory accounted for both dust and sea spray particles within an atmospheric model. By accounting for both types of particles, the team saw improved model predictions of INP number concentrations in marine air.

Impact

This study demonstrates the potential of global atmospheric models to accurately predict INP number concentrations in the lowest atmospheric level that feeds clouds over the ocean. It is an important step toward incorporating improved representations of INP number concentrations into global atmospheric and Earth system models. This work also tests a potential new approach for predicting INP number concentrations in models. Researchers can use these predictions to explore interactions between atmospheric particles and mixed-phase clouds—composed of both liquid and ice—in remote marine regions. With 70% of the Earth covered in water, clouds in remote marine regions play an important role in the Earth’s global energy budget. More accurate predictions of INP number concentrations will help improve model predictions of precipitation and radiative properties of clouds.

Summary

INPs are required for cloud water to freeze at warmer temperatures, affecting the formation of precipitation and the amount of sunlight and heat those clouds reflect. INPs originate from many sources, including windblown dust, pollen, and fungal spores. Around the ocean, sea spray particles, which contain salt and marine organic matter, can also be an important source of INPs. However, global models struggle to accurately simulate ice formation in clouds over remote oceans. One obstacle to improving cloud-phase simulations in remote marine clouds is a poor understanding of INP sources and amounts in these clouds.

To help close the gap between observed and modeled INPs, researchers simulated concentrations of dust, sea spray, and other types of atmospheric particles within a global atmospheric model. Then they calculated the number of INPs that would be predicted based on the dust and sea spray amount, using relationships derived from previous observational studies. Finally, they evaluated how well the simulated INPs matched the INPs in samples from two remote locations: a coastal research station in Ireland and a scientific research vessel traveling south of Australia. For these two sets of measurements, the model predictions of INPs improved when both sea spray and dust particles were considered. These findings will help researchers improve predictions of atmospheric INPs, and ultimately, precipitation and radiative properties of clouds. Future work will focus on improving model representations of particle sources acting as INPs, as well as evaluating these models with additional observations.

Contact
Susannah Burrows
Pacific Northwest National Laboratory
Publications
McCluskey, C, P DeMott, P Ma, and S Burrows.  2019.  "Numerical Representations of Marine Ice‐Nucleating Particles in Remote Marine Environments Evaluated Against Observations."  Geophysical Research Letters 46(13): 7838-7847.  https://doi.org/10.1029/2018gl081861.