From saving the planet to making the perfect cocktail

I am now at my second workshop of the week, also at Princeton. It is on ice nucleation, the process that kicks off the formation of an ice crystal. It included an inspiring, and scary, talk on modelling our Earth’s climate by Yi Ming, a scientist at an American government lab. Modelling our climate is hard, very hard.

Consider clouds, they are beautiful, but complex. At high altitudes they are mainly ice crystals, and how ice nucleates is, despite a lot of work, still quite mysterious. This is mainly because the key step occurs in maybe less than a microsecond, and when the growing ice crystal is perhaps only a 100 molecules, or 10 nanometres across – and so impossible to see.

So ice nucleation is hard to understand as the length and time scales are so tiny. Clouds of trillions of ice crystals are also hard to understand as they are so big, many kilometres long and high. This is bad enough but clouds flow and swirl in complex patterns from centimetres across to weather patterns much bigger than the UK.

All this has to be approximated in the models that make climate predictions*. That was the scary part of the talk. Of course these models have to be make a lot of approximations, to give a practicable model. Climate modellers realise this of course, and it is up to us scientists studying ice nucleation to do our bit to make our contribution to making their models better.

This is hard work, and so maybe it would be good to get away from the problem, with a relaxing cocktail. But there is no escape from ice nucleation there. I also learnt today that companies are selling the perfect ice to posh cocktail bars. Not for these bars the cloudy ice cubes we pull from our freezers and plonk in our drinks. They want perfectly clear, beautifully transparent ice cubes, for their discerning customers, and to do this requires controlling ice nucleation. It is fun, and a privilege, to work on a problem important from the stratosphere to a cocktail bar.

* As an aside, many of the huge computer codes that run these models are Fortran, a fine language that we teach our students. There are other languages used elsewhere, and that have other strengths, but it is Fortran that is telling us what we are doing to Earth’s climate.