Clouds have served up some strange challenges in climate change modelling. Photo / Getty Images
There’s a puzzle in climate change modelling that has confounded many scientists over the years.
The reality is that ice formation in clouds at high latitudes (close to the poles) is not fully explained by accepted estimates of soot, dust and aerosols.
So what else might be causing clouds to freeze? And how important is this gap in modelling the climate to making wise decisions about mitigating and adapting to the effects of climate change?
Kiwi scientist Peter Wilson is playing a leading role in answering those questions and solving this puzzle.
The puzzle is well known, as the International Panel on Climate Change (IPCC) noted it in its 2014 report. In a letter to a science journal that Wilson shared with me, he and his colleagues expressed the problem in more technical language as a: “clear lack of a proper accounting in the existing literature for nucleation of droplets of supercooled water in the upper atmosphere at high altitudes by radiation: neutron particles and others from galactic and solar cosmic rays, which have long been studied in terms of cloud formation models, but never leading to any conclusive results – instead many contradictory conclusions in existing papers”.
Recent research by Wilson, in collaboration with professors Cecilia Levy and Matthew Szydagis at the University at Albany in New York, has shown that radiation can freeze supercooled water, a new physical chemistry process in its own right.
The three of them realised that radiation from space – cosmic rays – produces neutrons that pass through cloud cover and may have the same effect on supercooled water droplets high in our atmosphere. Could this explain the missing factor? Quite likely, since it is known that there are seven times as many neutrons at the poles than at the equator, known as “muon-induced neutrons” (as opposed to “moon-induced neurosis”, which of course is lunacy on several levels).
Growing up in Dunedin, Wilson was always fiddling with radio equipment and the like. I remember my father giving him a small motor from his workshop so Peter could build a rotating aerial.
The professor has had roles with universities and other institutions in the United States, Japan, Australia and Antarctica. He’s a physicist and started studying supercooled water in 1985 when he became involved in the US Antarctic Program, working with biologists to figure out why Antarctic fish don’t freeze in water that is two degrees below zero.
The so-called antifreeze (or ice-binding) proteins that Wilson and others identified are now known to exist in many fish, as well as some insects, plants and algae.
This particular Notothenioidei fish, known as the icefish, is the only vertebrate with transparent blood (it has no haemoglobin, the protein in red blood cells). They diffuse oxygen from the water rather than transporting it around with their blood.
When this research became public, I remember seeing Wilson being interviewed on TV about whether this meant the fish protein could be repurposed to enable humans to finally embark on interstellar travel, with a quick injection and then just a defrost on arrival. Alas no.
Studying the intricacies of small ice crystals is as challenging as you would imagine. It became apparent that an instrument known as a nanolitre osmometer was needed. This enables precise temperature control (down to one-hundredth of a degree) to enable scientists to observe ice forming, growing or melting at the level of individual droplets.
This presented a new problem – you can’t buy a nanolitre osmometer at Briscoes. So, being a Kiwi, Wilson designed his own and formed a company called Otago Osmometers. He made the first few in his garage in Dunedin and has been manufacturing and exporting these very niche instruments globally since 1999.
His research has taken him to many top-tier universities, to Japan for several years, to Scripps Institution of Oceanography in San Diego and most recently to the University of California at Berkeley. It was there, earlier this year, that Wilson and Matthew Szydagis reconfirmed that neutrons (this time from a plutonium source in the basement of the engineering faculty at Berkeley) could cause deeply supercooled water to freeze.
Clouds are important for the regulation of the Earth’s surface temperature. They reflect radiation from the Sun back into space (a cooling effect), but they also act like a blanket that traps heat radiated from the Earth’s surface (a warming effect).
The following diagram shows the various roles of clouds (and the relative importance of each) in regulating the heating and cooling processes of the Earth.
Having more cloud cover can change the balance between these processes – therefore, understanding the processes for cloud formation is very important for climate change modellers and can have a significant impact on the accuracy of long-term climate change models.
Wilson’s work on neutron-induced ice nucleation and its application to cloud formation near the poles could fill an important gap in many climate models.
Many of you will be familiar with the famous quote by statistician George Box: “All models are wrong, but some are useful.”
Climate models are no exception. They will never deliver perfect information (especially when projecting decades into the future) but the better we understand the physics of how the Earth regulates its temperature, and how these processes may change as the Earth warms over time, the more useful these models will be in helping us mitigate and adapt to the effects of climate change.
Bryan Field is an expert adviser on climate change policy (and coincidentally has an MSc in physics from the University of Otago). He told me an anecdote about wrong but useful models the other day. In 2018, he helped the Ministry for the Environment develop the Zero Carbon Bill – specifically, Field worked with a group of NZIER economists to model the long-term impact on the New Zealand economy of adopting various emissions reduction targets.
NZIER used a computable general equilibrium (CGE) model of the economy to measure the impact of adopting the emissions targets. CGE models are devilishly complex at the best of times, but this task pushed the model to its limits. The model produced economic costs of the emissions targets that were eye-wateringly large compared with the status quo, but the status quo assumed that the economy could keep producing at historical growth rates (particularly in industries like agriculture) and face no negative effects from the pollution it causes (obviously, an impossible scenario).
While the economic costs produced by the model will be wrong (especially decades into the future), the purpose of the modelling was to help ministers see some of the things that could happen if they chose various targets, especially changes to the structure of the economy. Indeed, the model was wrong, but it was useful.
As the climate change threat deepens, and extreme weather events become more frequent and perhaps more severe, predicting what will happen next in a given location is becoming more critical. Climate modelling requires very powerful computers and very complex algorithms, and it is crucial that the raw data going in the front end is accurate. We’re all familiar with the “garbage in, garbage out” problem.
If the work on particles and nucleation can be further developed, we will be able to input the correct parameters for high-latitude clouds and better predict snow and ice cover.
Hopefully, Professor Wilson can keep solving these puzzles and keep the grand tradition of Kiwi insights and innovation alive.