60 seconds with ... Doug Smith
Wed 24 Nov 2010
Met Office research has shown for the first time that accurate long range hurricane predictions are a real possibility. The breakthrough will help insurers and business better understand climate risk, says Doug Smith, winner of the inaugural Lloyd’s Science of Risk Prize and author of the research.
Doug Smith was awarded the Lloyd’s Science of Risk Prize for his research paper Skilful Multi-year Predictions of Atlantic Hurricane Frequency, which was published in the journal Nature Geoscience. He currently works at the UK’s Met Office where he is responsible for developing the decadal climate prediction system. He studied mechanical engineering at Imperial College, gaining a PhD in computational fluid dynamics. As a research scientist at Bristol University and UCL he used space satellites to measure sea ice and rainfall, identifying an increasing trend in the length of the summer melt season of Arctic sea ice.
What did you set out to achieve with your research paper on predicting hurricane frequency?
The research was part of a broader Met Office study into decadal climate prediction, which is currently one of the most exciting areas of climate research. We have been investigating where it might be possible to predict climatic variables over a longer time scale. Climate models are routinely used to forecast hurricane frequency for the season ahead, but skill beyond the seasonal range had not previously been achieved.
The Met Office described the research as a major break through, why was this?
For the first time we have been able to demonstrate that it is possible to predict hurricane frequency using a physical model beyond the seasonal time scale. We have also been able to understand where some of the skill is coming from. This should help to improve the forecasts in future.
Is this valuable to business and insurers in particular?
Atlantic hurricanes are the biggest single cause of insured loss and account for seven of the ten largest losses ever incurred by Lloyd’s. Predictions and research can help the insurance industry better understand hurricane risk and take appropriate action. The study also has wider applications. For example, forecasts could help the energy sector prepare for unusual periods of hot or cold weather, or be used to predict droughts. There is also a benefit in demonstrating that models can make accurate predictions because they can help to build trust in climate change projections.
What challenges did you have to overcome to reach your findings?
Sea temperature is an important environmental factor driving hurricane formation, but observations are historically very limited, especially for temperatures below the sea surface. Also, as with all models, there is the potential for model errors. These two problems were overcome by using a novel approach to reconstructing historical ocean properties and by running several variations of the model to cancel out errors.
What is behind the increase in hurricane activity in recent year?
There has been a dramatic increase in hurricane activity since the mid-1990s compared with a quiet period in the 1970s. The research showed that this was at least partly because of external factors like greenhouse gases and aerosols and natural variations of solar radiation and volcanic activity. This is an important finding because there is much controversy about the cause of the increase in hurricane frequency. However, we still need to do more work to pin down the relative importance of the different external factors.
Do you expect our ability to predict hurricanes to improve further?
I certainly see it improving as greater computing power increases the resolution of models and allows us to better simulate the intensity of hurricanes and possibly where they may make landfall. We don’t know what the ultimate limits of modelling will be at present but we see areas where there could be developments – for example there is evidence that it may be possible to use decadal models to predict African and US rainfall and European summers.
Do you support initiatives like the Science of Risk Prize and do they help foster cooperation between science and business?
We do see science working closer with business, and it has been great that Lloyd’s has been encouraging research in areas that are relevant to business. Such partnerships are a benefit to both science and business because we need to know the key areas to focus our research. The Science of Risk Prize will encourage further research with real societal benefit.
How does it feel to have come top and have your work recognised?
I am delighted and honoured to have won and it is always good to have your achievements recognised. I would also like to acknowledge the many scientists both at the Met Office and around the world whose work paved the way for our results.