Achieving sustainable and resilient river ecosystems in Aotearoa under climate change

Leveraging data and models to identify solutions to increase the resilience of river ecosystems to uncertain futures both nationally and globally.

The need and impact

River ecosystem management is at a tipping point: as river ecosystems continue to degrade under pressures of increasing human demand and global change, sustaining them is imperative.

Contemporary ecosystem management is guided by the principle of embracing the historical range of environmental variation when setting restoration targets for degraded systems. However, there is now consensus that climate change is pushing hydro-climatic and ecological systems out of their historical domains.

We are seeing major catastrophic events occurring globally, from unprecedented fish die-offs to prolonged droughts and extreme floods. Thus, the reliance of management on historical benchmarks and the current practice of extrapolating future river ecosystem states from contemporary trends is destined to fail.

Current modelling approaches based on past correlations are linear and do poorly looking into the future for how species may respond to unprecedented changes. In reality, river ecosystems exhibit clear nonlinearities and feedbacks between different sectors, including hydrological, geomorphological, and biological.

This project will impact environmental management and planning in Aotearoa New Zealand through a commitment to ensuring that all research results and findings are developed both for high-impact journals and for research translation, via a series of workshops. This project is of great utility to decision-makers tasked with managing ecosystems experiencing rapid change, including the Department of Conservation, who are instigating a nationwide catchment-based restoration scheme.

The approach

By explicitly modelling the biology underpinning the distribution of species, rather than correlating the presence of species with particular conditions (the emergent properties), mechanistic models offer a greater capacity to forecast species responses to novel combinations of environmental conditions, such as flood and drought regimes. These models address key nonlinearities and feedbacks–critical elements of complex systems ecology.

However, there are major gaps in the mechanistic data required to parameterise such models for Aotearoa New Zealand’s native fish species. Leveraging observational (correlative) and mechanistic data to develop hybrid models that are robust to changing hydro-climatic conditions to forecast NZ’s freshwater fish species responses to future climate change and management scenarios, including integrated population models (simultaneous analysis of individual-level demographic and population-level data to infer vital rates and population dynamics), hierarchical models (e.g. hidden process models), and inverse modelling (used for parameter estimation). Such approaches ‘borrow strength’, fill gaps, and fuse data to get the best from different forms of data.

Through a series of expert working groups that expand on a recent paper by the lead investigator in Nature, and incorporating our existing models, we will identify solutions to increase the resilience of river ecosystems to uncertain futures both nationally and globally.

Research aims

  • Develop and apply hybrid models that leverage broad-scale occurrence data with mechanistic data. We will source, aggregate, and synthesise data: biological mechanisms from the literature (e.g. physiology, phenology, demography, life history, evolutionary potential, species interactions, dispersal, range dynamics); hydrology, including present-day from flow gauges and climate projections; Long-term fish occurrence data from NZ Freshwater Fish Database; >34,000 occurrence records NZ-wide.
  • Combine these data into models for several target species spanning a range of life-history strategies (including migration to sea) and life-cycle speeds. Models will link population dynamics, species occurrences, and hydrology.
  • Make predictions of population responses to various future climate-change (RCP2.6, RCP4.5, RCP6.0, RCP8.5) and intervention (environmental flows, nature-based solutions to alter hydrology) scenarios.


  • Dr Jonathan Tonkin (Project Lead)
  • Associate Professor Michael Plank
  • Dr Rachelle Binny
  • Dr Andrea Tabi