Most of the time, our lives run smoothly and predictably, but they are also prone to intermittent instability with disastrous consequences. Can we do more to predict and even intervene to prevent these periods?


For many of us, life appears to flow smoothly and predictably most of the time. Indeed, it appears that one of our major concerns is becoming stuck in a rut. Our world is then turned upside down, seemingly out of nowhere. A global pandemic strikes, killing millions and putting entire countries on lockdown. Then comes the inflation, and the economic downturn threatens our livelihoods. And (not unrelated), one country invades another, resulting in a war that affects us all. Whoa! Where did all of that come from?

There is a general cause for these unexpected disasters. Intermittent instability is common in many “nonlinear systems” in the physical, biological, and social sciences. To put it another way, These systems suddenly become wildly unpredictable, exhibiting extreme fluctuations after long periods of boring and predictable behaviour.

What exactly is a nonlinear system? It is one whose outputs are not proportional to inputs. We humans are nonlinear systems: if we win $1 million (or £1 million, if you prefer) in the lottery, we will most likely be overjoyed. But we wouldn’t be four times as happy if we won $4 million/£4 million. To put it another way, if winning the lottery would make us happy, losing our wealth in some stupid gamble would not only make us sad, but would devastate us.

Nonlinearity is a key component in the phenomenon of chaos – the process underlying the well-known butterfly effect – which describes how a small uncertainty can grow and make the entire system unpredictable. However, in chaotic systems, the butterfly effect is not always fully operational. A chaotic system can sometimes be predicted for a long time in the future. In other cases, near an intermittent instability, the flap of a butterfly’s wings can destroy predictability in a matter of seconds. It’s all a result of nonlinearity.

The weather is a chaotic nonlinear system. It’s mostly boring and predictable, but it occasionally becomes wildly unpredictable and extreme. In the month of October 1987, Just the day before, BBC weather forecaster Michael Fish advised viewers not to be concerned about strong winds that turned out to be the worst storm to hit southern England in 300 years. Meteorologists learned from their mistakes. It was discovered that the evolution of what has come to be known as the “Fish storm” was extremely sensitive to the butterfly effect – much more so than usual. As a result, when attempting to predict the weather, forecast centres now run ensembles of 50 simulations, each with slightly different initial conditions (by flaps of meteorological butterfly wings).

When the atmosphere is extremely unstable, as it was in October 1987, the various forecasts within the ensemble will diverge dramatically: some will show hurricane-force extreme weather, while others will show much more benign weather. In such a case, all a forecaster can do is estimate the probability, or likelihood, that the extreme event will occur.


However, forecasting a significant probability of an extreme event is preferable to forecasting nothing significant, as occurred in 1987. And ensemble weather prediction is currently transforming the way humanitarian and disaster relief organisations send emergency food, shelter, medicine, and even finance to areas at risk of extreme weather. When the ensemble-based probabilities for extreme weather exceed a predetermined threshold, such anticipatory action is taken. Can we, as humans, influence the likelihood of such instabilities? We certainly can.

We are tilting our nonlinear climate system towards greater intermittent instability by continuing to emit greenhouse gases into the atmosphere, warming and moistening the air. One of the most devastating types of instability is associated with a tipping point, such as the disintegration of a large ice sheet, which results in significant sea-level rise. In this situation, simply reversing our emissions will not undo the damage; that would be like locking the stable door after the horse has bolted.

We have the ability to reduce the likelihood of these climatological instabilities by regulating greenhouse gas emissions. Which actions, however, will be the most effective?

In the 1950s, an idea was mooted to explode nuclear bombs in suitably targeted parts of the atmosphere, to move hurricanes away from their predicted paths if they threatened populated regions


Climate scientists attempt to model the climate system and estimate how climate will change as a result of various future emission scenarios to help guide these decisions. However, there are uncertainties in these estimates, including the effect of these emissions on internal atmospheric processes such as cloud cover.

Climate scientists attempt to account for these uncertainties by running ensembles of climate models in which uncertain cloud processes, for example, are represented in various ways. It is critical that our ensembles provide reliable probabilities of harmful climate change. In my opinion, we need to do more on the latter and should pool our resources internationally to create a new type of “Cern for climate change” – an intergovernmental research organisation that, unlike Cern, will focus on climate science rather than particle physics.

Even our solar system, which is often thought to be the most orderly and predictable of all, is chaotic and prone to intermittent instability.


Over a century ago, the French physicist Henri Poincaré demonstrated that there was nothing in principle preventing planets from being ejected abruptly from the Solar System. Fortunately, based on solar system ensemble forecasts, we can be confident that this type of intermittent instability will not occur in the near future (in this case many millions of years).

A mini-version of this instability, however, is much more likely: a small asteroid will be unexpectedly ejected from the asteroid belt and will impact Earth with the effects of one or more nuclear bombs.


As a result, Nasa is improving its near-term asteroid prediction capabilities and developing methods for diverting rogue asteroids if they suddenly approach Earth. The Dart mission, in which a spacecraft collided with the asteroid Dimorphos, successfully tested our ability to deal with the asteroid belt’s intermittent instabilities by altering the course of a potentially Earth-bound space rock. In this case, we have the ability to control the instabilities and prevent their worst consequences. Could similar types of intervention prevent the occurrence of individual weather events, such as the 1987 storm that hit England or the hurricanes that ravage the US east coast every year? In the 1950s, it was proposed to drop nuclear bombs in precisely targeted areas of the atmosphere in order to divert hurricanes away from their predicted paths if they threatened populated areas. Despite the obvious problem of radioactive fallout, the weather is far too complex to reliably perform this type of weather control.

Leave a Reply

Your email address will not be published. Required fields are marked *