Introduction
Robust river flow forecasts rely on accurate estimates of runoff into river catchments across much of Aotearoa New Zealand. An important part of such runoff is due to snowmelt.
During dry periods, snowmelt sustains river flows, while during extreme rain-on snow events, snowmelt significantly increases runoff above that expected from rainfall alone. This is especially the case in maritime regions like New Zealand, where atmospheric rivers can deliver intense rain to alpine areas during winter and spring, when snow cover is extensive.
For example, during the destructive March 2019 flood event that destroyed the state-highway bridge at Franz Josef Village, snowmelt and ice-melt generated an additional 400 mm of rainfall-equivalent runoff and increased river flow in the Waiho catchment by 20%.
New Zealand currently lacks a modelling system to accurately simulate snowmelt, especially during rain-on-snow events. Accurate snowmelt forecasts rely on a robust understanding of how fast snow will melt, along with the extent and depth of snow cover across the landscape.
Developing a state-of-the-art snowmelt forecast system
NIWA-led researchers are therefore working to develop a state-of-the-art snowmelt forecast system to enable more accurate and confident forecasts of river flow and alpine hazards across New Zealand.
The forecasts will enable people working in hazard management, energy, agriculture and tourism to better respond to rain-on-snow impacts on river flows and alpine hazards.
The research will also support New Zealand’s reliance on hydro-electricity generation through better predictions of snowmelt contribution to lake inflow, which will lead to increases in generation efficiency.
The project will use hydrometeorological and snow data from NIWA’s high-elevation weather stations to test the system and forecasts will be made available to end users through NIWA’s operational multi-hazard forecasting system.
The project brings together a team of specialists from NIWA and Otago University, supported by international experts in snow modelling and observation.
It will also be guided by a group of Māori, industry and government representatives to ensure that forecast outputs will enable better decision-making across Aotearoa New Zealand. Engaging with these people throughout the project will be important to guide its direction help ensure that outputs are usable and used.
The work is funded through the MBIE Endeavour Fund Smart Ideas and runs from October 2023 to October 2026.
Impacts: supporting the transition to a low emissions and climate resilient economy
The research is expected to have two major impacts:
- Supporting New Zealand’s reliance on hydro-electricity generation through better predictions of snowmelt contribution to lake inflow, which will lead to increases in generation efficiency (less spill and better timing of generation to meet demand). This will maximise economic returns and help New Zealand meet its 100% renewable energy goals.
- Improving resilience to more frequent future extreme events through better forecasts of extreme rain-on-snow floods. These forecasts will be used by councils, civil defence, and mountain safety organisations to issue improved warnings and enhance operations during weather-related emergencies. This will lead to improved readiness and reduced losses from extreme weather events.
Methods: Ensemble forecasting to estimate snowpack and predict snowmelt
The research team will use the output from an ensemble convective-scale numerical weather prediction model (https://niwa.co.nz/climate-and-weather/weather-and-climate-forecasting-services/weather-and-climate-forecasting-data-services) to drive an ensemble of physics-based snow energy balance simulations.
This will include input from novel satellite snow datasets to produce robust operational forecasts of snowmelt. This represents a step change in the ability to forecast snowmelt and resolve the influence of snowmelt on hydrology, both during extreme rain-on-snow events and during low flow periods. The team hypothesises that this ‘super-ensemble’ approach will allow full characterisation of operational snowpack forecast uncertainties, enhance satellite data assimilation, and significantly improve snowmelt estimates.
