SCIENCE OF SEASONAL CLIMATE PREDICTION
Communicating skilful but uncertain seasonal climate forecasts
by Peter Hayman, SARDI
Dr Peter Hayman is Principal Scientist, Climate Applications with the South Australian Research and Development Institute, prior to this he was an agricultural adviser and researcher with NSW Department of Primary Industries. Since the early 1990s he has worked with farmers and their advisers to apply the advances of climate science to improve the management of climate risk.
Whatever may be the progress of the sciences, never will observers who are trustworthy and careful of
their reputations venture to foretell the state of the weather.
The Times, 18 June 1864; www.bom.gov.au/bmrc/clfor/cfstaff/nnn/nnn_climate_quotes.htm
We are confronted with unprecedented opportunities to tune our agricultural systems in a way that
improves their sustainable land use. We have a seasonal forecasting capability. We have started to
think through how we can best use the knowledge that the next season is not a total unknown.
Hammer and Nicholls, 1996.
Information generated by climate science is only valuable if it is used. However using probabilistic seasonal climate forecasts in decision-making is proving to be harder than some of us first thought. The problem is not that people are unaware of climate forecasts. Climate science gains ample media attention. Words like El Niño have moved from oceanography to headlines, advertising copy and parliament.
A major challenge is that most of us are poor intuitive statisticians and this has implications in how we make general decisions under uncertainty (Burgman, 2005) and how we respond to seasonal climate forecasts in particular (Nicholls, 1999). A further challenge is that assessing and making risky decisions is deeply embedded in the social setting and psychology of the decision maker (Hayman and Cox, 2004).
Communicating seasonal climate forecasts as a means to the end of better risk management is more than being clear about the message, using words carefully (eg. frequency rather than percent chance) and designing better graphics such as box plots and pie charts. Nor is it simply a case of improving forecasts and improving the delivery of forecast information through tools such as Yield Prophet. These developments have all been helpful in the Australian context however they should be viewed as necessary but not sufficient. If communication is defined as ‘the reciprocal construction and clarification of meaning by interacting people’ it will involve an ongoing engagement and dialogue with users.
References
Brugman, M A (2005) Risks and Decisions for conservation and environmental management. Cambridge University Press.
Hammer, G L, and Nicholls, N N (1996). Proceedings of the 2nd Australian conference on agricultural meteorology, Brisbane, 19–27.
Nicholls, N N (1999). Bulletin of the American Meteorological Society 80: 1385–97.
Hayman, P T and Cox, P G (2005) Drought risk as a negotiated construct, 113–126 in Botterill LC and Wilhite DA From Disaster Response to Risk Management. Springer.
Acknowledgements
Funding from GRDC and ACIAR, ideas from many colleagues in Australia and elsewhere who have risked reputations to promote climate forecasts as a means to the end of better management of climate risk.
