Home / Our Science / Environment and Climate / Our stories / Better modelling to support groundwater decisions

Better modelling to support groundwater decisions

Groundwater models provide essential information for the management of New Zealand’s freshwater resources. However, existing models too often suffer from being too data hungry, too slow, and too costly. When designed inappropriately models can provide inaccurate information, undermining any management decisions relying on their output.

clear blue-green water pool. Photo: Margaret Low

Groundwater accounts for one-third of all freshwater usage. Photo: Margaret Low

"

“The whole process has been invaluable for many reasons. I’m looking forward to the results further unfolding.”

Dr Tim Ellis, Senior Science Co-ordinator, Environment Southland

"

The GNS Science-led Smart Aquifer Management programme is a collaboration with many other organisations. It has led to the development of a world-leading framework, strategies and tools to improve the cost-effectiveness and reliability of the management of New Zealand’s aquifers via groundwater models. 

There are about 200 known aquifers in New Zealand and groundwater currently accounts for about one-third of all freshwater usage. Councils are increasingly relying on models for decision-making, so our new outputs from this programme are timely. 

These outputs provide new guidance on modelling design and uncertainty quantification for decision support applications. Applications include the management of land and water use to minimise the impact of human activity on systems of ecological, environmental, cultural and economic importance. 

During this three-year Ministry of Business, Innovation and Employment funded project, we designed, developed and tested the performance of a range of case-study numerical models for Wellington, Southland and Waikato regional councils. These models aim to help councils address real management challenges when setting allocation limits for land and water use. The results from these studies provide insights into the source of uncertainty in numerical groundwater models. They clarify how to design models that retain salient components and remove superfluous components, to allow quicker and cheaper decision-making support. 

The programme provided guidance for selecting the most appropriate modelling approach for addressing management problems, with an aim to reduce the duration and cost of modelling studies. Importantly, it reduces the number of erroneous modelling predictions. Central government and iwi also provided input. 

Our groundwater scientists estimate the new developments will deliver annual improvements valued at about $350 million to New Zealand. Partner organisations throughout the project included NIWA, ESR, Victoria University of Wellington, Market Economics Ltd, the US Geological Survey, and Watermark Numerical Computing in Australia. This project is a good example of GNS Science’s leadership in groundwater modelling and science.