Clean Air and Tropical Storm Frequency

earth, aerosols, atmosphere, climate change

Using the word ‘aerosol’ in everyday language might generally be associated with spray cans. However besides styling our hair or odourising a room, aerosols are known to influence various physical processes in our atmosphere.

Despite their small scale they can have a huge effect on our climate system. Naturally occurring – such as suspensions of liquid water droplets as clouds or fine ash hanging in the atmosphere from volcanic eruptions, or man-made, like soot or smoke emitted from the burning of fossil fuels – all aerosols contribute to the composition of the atmosphere and the dynamic workings of the climate.

Aerosols and the impact on our climate

Whether aerosols are natural or man-made, they are everywhere and leading science suggests that they are having a significant impact on our climate. For example, aerosols encourage clouds to form in the sky and provide a surface (albeit a very small surface) around which condensation can form. High concentrations of aerosols at ground level can lead to smog.

Lately, you may have read some stories in the press about recent studies that suggest man-made aerosols in the atmosphere could be a driving force influencing tropical storm frequency in the North Atlantic (Dunstone et al, 2013).

This paper uses an array of climate models to show that increases in anthropogenic aerosols (those originating from human activity) actually reduced the frequency of tropical storms over the twentieth century, and subsequent sharp decreases in these same man-made aerosols have led to a rise in tropical storm frequency.

It is worth noting that these results are heavily discussed within the climate and tropical storm research community with no consensus in sight. To the uninitiated in climate science, these findings may sit uneasily. For a long time, we’ve been told that reducing air pollution is a good thing. Various environmental regulations around the world have helped us reduce “harmful” emissions of man-made aerosols.

This new research suggests we are encouraging tropical storms to occur more frequently. If this research is to be believed – it could have huge implications for anyone living in hurricane prone areas, such as around the east or south coast of the US or in the Caribbean.

The science part

Cynics might say this research is overly political or supported by vested interests within the extractive industries. Or maybe it feels like a statistical quirk. If the data or modelling methods are flawed it could lead to a misconception that this is correlation and not just a causal link (a view evident in the comments beneath some of the press articles).

What we do know, is that this research was conducted by the Met Office using their highly sophisticated climate modelling techniques which simulate the fundamental physical interactions within the climate system. Past climate data is used to train these models and climate variability is largely reproduced when past observations are forced by sea surface temperatures (a key driver of our climate system).

Uncertainty in the observations is ever-present so any biases created by this uncertainty are painstakingly corrected to try to reduce margins of error in the model simulations. Limitations on computing power leads to compromises in the model algorithms and their attempt to model the atmosphere, and as the climate models are tested, studied and improved, the effects of these compromises are accounted for using complex recalibration techniques. Also, due to subtle differences between climate models from different organisations around the world, multiple models are examined to get a broader view, essentially trying to avoid being too blinkered.

These optimized physical climate models are used to experiment with the modelled atmosphere and in doing so they can test the effects of increasing, reducing, or even removing the man-made component of atmospheric aerosols all together. The result gives convincing grounds for comparison that – if you trust what the models are telling you – is a useful method for exploring the effects of a change atmosphere.

Not to be ignored

There will be areas that can be improved within the study but these findings should not be dismissed, if only for the fact that it has been fully peer-reviewed and published in an established scientific journal. These results will no doubt undergo further testing and will be replicated, and undergo further comparison to other models, but these findings from Dunstone et al may help to disentangle more of the interwoven climate signals that overlap to make our wonderfully complex climate system.

The implications in this study of cleaning up our atmosphere through trying to reduce the levels of man-made aerosols, and returning these anthropogenic pollutants to more natural levels, are perhaps uncomfortable. We may be lifting a dampening effect on tropical storm development and therefore putting communities in hurricane prone regions at greater risk from these storms.

However, we need to put this into context. It is only one paper which is experimenting with a modelled version of the climate system. A sandbox, if you will, where scientists can play and learn. The scientific community will debate papers such as these, and evidence will be presented either side until the most convincing case finds consensus.

Whether the science is fully supported or not, if we look purely at impacts of aerosols in these models, there is also evidence through a similar study by Booth et al last year, that the increase in man-made aerosols in the atmosphere, may have contributed significantly to Sahel droughts during the 1980’s. There is room for uncertainty and debate about both of these quite profound findings, but we shouldn’t ignore the big picture.

As we try to reduce the impact of our industrial activities on the natural environment, there will inevitably be challenges along the way. With this in mind, it is important not to get carried away with media frenzy. We have to explore each study in the context of wider research.


Dunstone N.J., D.M. Smith, B.B.B. Booth, L. Hermanson, & Eade, R. Anthropogenic aerosol forcing of Atlantic tropical storms. Nature Geoscience 6, 534-539 (2013).

Booth, B. B. B., Dunstone, N. J., Halloran, P. R., Andrews, T. & Bellouin, N. Aerosols implicated as a prime driver of twentieth-century North Atlantic climate variability. Nature 484, 228–232 (2012).

About Geoffrey Saville

Geoffrey Saville is a member of Willis Towers Watson's Analytics Technology Team, having joined the company in 2013…
Categories: Analytics, Natural Catastrophe, Reinsurance | Tags:

Leave a Reply

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