Measuring uncertainty in the sea

What started out as a simple love of fishing for a Victoria University of Wellington researcher has led to a complex mathematical study in measuring fish populations.


Determining the size of fish stocks is important for ensuring sustainable commercial operations, but also difficult to do with past models prone to bias or oversimplification.

Darcy Webber has developed statistical methods and modelling techniques which more accurately reflect a current environment or situation. Ironically, this can include admitting uncertainty, which is subsequently factored into the model.

“The models aim to make sense of many different sources of information, including catch estimates, fishing intensity, growth and maturity rates, and tagging, genetic, environmental and economic data,” says Darcy.

“There is uncertainty associated with all this information, which can arise due to natural variations, measurement error or model uncertainty.”

According to Darcy, developing models that reflect this uncertainty is important so that management decisions are not made on rigid estimates.

While fish stocks around some Asia-Pacific nations are being exploited to below acceptable levels, Darcy says New Zealand is managing its fish resources fairly well, but it’s important to remain vigilant.

“It’s important to be innovative with our management techniques to ensure we maintain sustainable fish stocks, despite increasing pressure from fishing, pollution and climate change.

“Fisheries are worth more than $1.5 billion per year to our economy and New Zealand’s Exclusive Economic Zone spans four million square kilometres, making it the fourth largest in the world.

“New Zealand’s marine natural resources are of great value, but a connection to the sea is also something deeply ingrained in our culture.”

Darcy will be graduating from Victoria University of Wellington this month with a PhD in statistics and is now applying his mathematical skills to his own consulting business.

Although a relative late starter with statistics, he notes the wide range of opportunities for those competent in mathematics and computer programming.

“Almost anyone can do some research and write a report. The skill that is in highest demand is the ability to sit down and make sense of data. Statistics and computer science provide a solid foundation for developing this ability.”