Impacts of climate change on river flows and floods

This project demonstrates how to use the latest global climate forecasts to predict changes in river flows in New Zealand, including changes in size or frequency of future floods.

The problem

The climate in New Zealand is changing: the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report found that increased temperatures are “virtually certain”. An important consequence of this finding is that floods in New Zealand are “very likely” to become more frequent and intense, resulting in increased risk to major infrastructure including failure of flood protection measures. The pattern of rainfall over New Zealand will also change, with eastern areas expected to become drier leading to greater pressures on water resources. As well as new risks, the changing climate will also bring opportunities, and early uptake of adaptation measures can have immediate benefits for land managers.

The challenge for hydrologists at NIWA is to use the complex information on likely future changes in New Zealand’s climate, to make predictions about changes to water resources and water hazards for individual river basins. This project is funded by the Ministry of Agriculture and Fisheries (now part of the Ministry for Primary Industries) initiative: ‘Sustainable Land Management Mitigation and Adaptation to Climate Change’, and investigates the changes in river flows, including extreme flood events, under climate change.

The solution

This project will demonstrate how to use the latest global climate forecasts to predict changes in river flows in New Zealand, including changes in size or frequency of future floods. Our method will be tested in two pilot catchments.

The key steps are listed below.

• Using the latest IPCC global model projections (2007) which show significant changes in likely seasonal and geographic rainfall patterns compared to the projections from 2001.
• Using dynamic downscaling of the global predictions to the river basin scale, through our high resolution regional climate model.
• including natural climate variability in our predictions using a stochastic rainfall model.
• Using NIWA’s hydrological model TOPNET to understand how changes in rainfall affect soil moisture and river flow.
• Making predictions for changes in flood frequency from the hydrological model results.

The results

This 1-year project concluded in December 2009. Full results are available to download as a report below.

Flood Risk Under Climate Change (PDF 6.3 MB)

Study catchments

The framework was tested in two catchments. These are the Uawa flowing into Tolaga Bay (Gisborne district) and the Waihou flowing into Hokianga Estuary (Northland region). These catchments were suggested by the respective regional councils after a discussion at the River Managers Forum meeting in Wellington in March 2009.

Research results

We made the following steps to complete the Climate Impact Framework.

• The output from the Regional Climate Model simulations was processed to correct biases in the model and ‘ground truth’ the model against observed data (see Figure 2)
• The stochastic rainfall model was used to simulate longer series of rainfall typical of future climate and including natural climate variability.
• The hydrological model was set up for the Uawa and Waihou, we created the spatial data files e.g. topography and land cover, processed Observed rainfall and flow data, and calibrated the models ready for use.
• The hydrological models were used to simulate future river flows under climate change. 
• Flood frequency curves for current and future climates were constructed directly from the model output (see Figure 3). The uncertainty in these flood predictions was also assessed.

The framework was successfully tested at the two case-study catchments.

University collaboration

NIWA worked in collaboration with Dr Bethanna Jackson at Victoria University Wellington on detailed testing of our hydrological model. We compared the hydrological model against field data and against a more detailed 3D model of the soil zone. We found that the flow prediction behaviour that is important for TopNet streamflow forecasts was consistent with both field observations and with the detailed model, and were able to put physical constraints on model parameters such as saturated hydraulic conductivity and soil depth. This work gave us increased confidence that TopNet correctly models the dynamics of water movement in the soil zone and that we can trust our model under conditions of climate change in the future.

Presentation of research

We have publicised the results of our research by presenting a poster at the New Zealand Climate Change Conference in May 2009, a poster at the American Geophysical Union Fall Meeting in San Francisco in December 2009 and an oral presentation at the New Zealand Hydrology Society Conference in Whangerei in November 2009.