From atoms to clouds: bridging the gap between atomistic simulation, surface science, atmospheric observation and climate modelling

April 2, 2014 to April 4, 2014
Location : CECAM-ETHZ, Zurich, Switzerland
   Room plan


  • Ben Slater (University College London, United Kingdom)
  • Angelos Michaelides (University College London, United Kingdom)
  • Christoph Salzmann (University College London, United Kingdom)
  • Ulrike Lohmann (ETH Zurich, Switzerland)



   UK CCPs


Ice nucleation is a ubiquitous process that has tremendous technological, biological and environmental impact. For example, crystallization of ice typically results in the growth of thin needles which can puncture and irreversibly damage cells, and thus present a hazard to organ preservation, polar sea life and crop yields. Suppressing crystallization of water in jet fuel through the use of inhibiting additives is crucial to avoid blockages in fuel lines in airplanes, to avert engine failure. Ice crystallization occurs in the upper and mid-level atmosphere resulting in changes to cloud reflectivity. Since global cloud cover plays an important role in the Earth’s albedo, the extent of ice nucleation influences the climate and the hydrological cycle.

Despite being an everyday event for most of us, the details of the atomic scale mechanism of homogenous and heterogeneous ice nucleation are very unclear. For example, pure water can be supercooled to nearly -40 degrees C without crystallizing but common atmospheric ice nucleation agents such as soot, bacteria and clays can greatly increase the temperature of crystallization to -10 degrees C or higher. Yet the origin of the variation in ice nucleation efficacy is still largely unexplained. Moreover, there is relatively recent evidence that much of the ice in the upper atmosphere, at temperatures below 190K, is not the ordinary hexagonal form typically seen at ground level but a distinct phase referred to as cubic ice [1]. But the nature of cubic ice has been recently put in doubt, by strong evidence indicating that “cubic ice” is a stacking-disordered material containing both cubic as well as hexagonal stacking sequences [2]. Very recent work [3] has demonstrated that a relatively minor component of atmospheric dust, feldspar, is a potent ice nucleation agent in mixed-phase clouds that form from freezing of supercooled droplets at temperatures between 0 and -38ºC but the mechanism which triggers crystallization is completely unknown. This latter finding exemplifies the need to probe and understand the molecular structure of interfaces between water and substrates, requiring atomistic modeling [4] and insights from surface science techniques [4, 5, 6] Density functional methods are amenable to the problem of examining the complex mineral interface and water. But to shed light on long timescale nucleation events, force-field based simulations play a crucial role including coarse grained models, such as the mW model developed by Molinero et al. [7]. Stepping up the length-scales to examine the complex interplay between supersaturation and particulate size in clouds requires yet coarser numerical models to be employed[8] such as the Kärcher et al., 2006 scheme [9] for cirrus formation used in the ECHAM6 global climate model (GCM) and the Hoose et al. 2010 [10] scheme for mixed-phase cloud formation used in the CAM-Oslo GCM.

Combining this spectrum of modeling approaches with laboratory and atmospheric observation presents an unprecedented opportunity to advance our understanding of one of most fundamental processes on Earth. A strong motivation for this workshop is to bridge the gap between the “model” systems (such as atomically flat metal surfaces) studied by surface scientists and the real (relatively poorly characterized) materials studied by atmospheric scientists. The aspiration of the meeting to identify key systems that modellers, surface scientists and atmospheric scientists could focus on to bring the atomic resolution and understanding to the question of why some materials are more effective nucleation agents than others.

The workshop will focus strongly on discussion of and critical assessment of our basic understanding of ice nucleation and the impact of this most common and yet complex nucleation process on the Earth’s climate. The ambitious aspiration of the workshop is to bring together, facilitate discussions, and debate between modellers and experimentalists with an interest in ice and that work on the full range of relevant length-scales and timescales. We believe that bringing these groups of scientists together will help to identify the most crucial aspects of ice nucleation and its subsequent impact on climate.

The 23 participants that have agreed to attend the workshop are drawn from atomistic modellers, coarse grain modelling specialists, climate modellers, surface scientists and scientists working on atmospheric observation and measurement and have made substantial contributions to the literature in their specialist fields.


[1] B. J. Murray, et al., Nature, 434, 202-205, 2005
[2] T.L. Malkin et al., PNAS, 109, 1041-1045, 2012
[3] J. D. Atkinson et al., Nature, 498, 355-358, 2013
[4] J. Carrasco, A. Hodgson. A. Michaelides, Nature Mater. 11, 667, 2012
[5] A. Verdaguer, G. M. Sacha, H. Bluhm & M. Salmeron, Chem. Rev. 106, 14781510, 2006
[6] K. Thurmer and S. Nie, PNAS, 2013 (doi:10.1073/pnas.1303001110)
[7] E. B. Moore and V. Molinero, Nature 479, 506-508, 2011
[8] A. Gettelman et al., Journal of Geophysical Research Atmospheres, 117, D20201, 2012
[9] B. Karcher et al., Journal of Geophysical Research, 111, D01205, 2006
[10] C. Hoose et al., Journal of Atmospheric Science, 67, 2483-2503, 2010