Pirikahu, Sarah

Bayesian networks as a real-time prediction tool for swimmability

Sarah Pirikahu1, Beverley Horn1, and Roger Hodson2

1. Institute of Environmental Science and Research, Christchurch

2. Environment Southland, Invercargill

New Zealand government have set a national target of making 90% of New Zealand’s large rivers and lakes swimmable by 2040. In order to achieve this goal, it is important to investigate the interaction of environmental and microbiological factors over the entirety of a river catchment. Bayesian networks can act as a tool for analysing the interactions between variables in a complex system, whilst providing a framework that can be easily understood by a non-specialist audience. In this talk we share our experiences of developing a Bayesian network to describe the swimmability of the Aparima River at Thornbury, Southland, and discuss the potential for Bayesian networks as a predictive tool for swimmability.

LightningTalk This is a lightning talk.