Maths helps predict drought

A statistics PhD student at the University of Adelaide in Australia could prove a godsend for Australia's farming sector, helping rural communities to better plan for future droughts.

Geraldine Wong is using advanced mathematical models to more accurately predict when, where, and for how long, droughts are expected to occur in Australia.

The information will assist the rural sector in determining the viability of different crops and drought mitigation measures in the presence of varying climatic conditions.

Wong, a 25-year-old postdoctoral student in the School of Mathematical Sciences, is analysing Australia's global climatic indices such as the Southern Oscillation Index (SOI) in conjunction with rainfall statistics to accurately predict future droughts in the short and long term.

'The drought in the past two years has cost Australia billions of dollars. This is a big issue for farmers and the nation. If we can forecast in advance the probability of occurrence and the severity of a drought, this will give the rural sector supporting information to help decide on investments such crops or water allocations,' she said.

The mathematical models being used by Wong, known as copulas, are similar to those employed by the financial sector to predict stock market movements.

'Not many hydrologists have used this statistical method until recently, but it can show us the relationship between the severity, intensity and duration of a drought with relative accuracy,' she said.

'If we can improve the prediction of droughts, this will help mitigate their impact, since farmers can plant more drought-tolerant crops or alter their irrigation needs. Water resource managers can also plan contingency arrangements to secure water supply in cities and country towns,' she added.

Ms Wong will present her research findings at a Hydro Predict Conference in Prague this September, assisted by a $2500 Mutual Community Postgraduate Travel Grant.