Trout and native fish need more water than we think, research finds
Science has now provided evidence for what many anglers have suspected for years: taking water from rivers is risky for some fish, and we may have been short-changing them and their food sources in the past. The new knowledge, based on 15 years' research led by Cawthron Institute, has global implications for irrigation and hydro-electric development, and recreational fishing. In New Zealand, regional councils may need to revise minimum flows upward and water allocation limits downward.
The value of water
Southland’s Mataura River is an angler’s paradise. Often dubbed the world’s best dry fly brown trout fishing river, it begins south of Lake Wakatipu and winds southeast through farmland and rural towns to the Southern Ocean. Not only is it renowned for fly fishing, but also as an important bird habitat, which enjoys our highest level of environmental protection – a National Water Conservation Order.
Recent estimates show the international fly fishing market is worth at least $1 million to $2 million a year to Southland. The Mataura is also valuable for its irrigation water and there is increasing demand for more. But new research from the Cawthron Institute supports what many anglers have suspected for years: taking water from rivers is risky for fish, and we may have been short-changing them and their food sources in the past.
The 15-year project investigated the impact of stream-flow on trout, which drift-feed on small aquatic and terrestrial invertebrates (such as mayflies) that drift in and on the water. It reveals that these fish have higher flow requirements than present models allow for. The principles learnt also apply to other drift-feeding fish, including juvenile salmon and native species – such as some whitebait.
Wide implications for water use
Project leader and Cawthron freshwater fisheries scientist Dr John Hayes says the new knowledge has global implications for irrigation and hydro-electric development, and freshwater fisheries conservation and management. In New Zealand, regional councils may need to revise minimum flows upward and water allocation limits downward.
“A river acts like a conveyor belt delivering the drifting food to the waiting fish,” Dr Hayes says. “We’ve shown that as flow declines, the diminished power and transport capacity of a river results in less drifting food. A new computer model that our team developed predicts that this translates to fewer, or more slowly growing, fish.”
He says the insights from the model are a caution to regional councils and the public to be more careful when allocating water.
“A lot of current minimum flows and allocation rates around the country could be having adverse effects on trout and native fish.”
“The environmental, social and economic consequences are far-reaching,” he says. “Fish & Game and the Department of Conservation have a better case for arguing for precautionary flow decisions, but tighter limits on taking water will be challenging for farmers and a government committed to sustainable economic growth.”
Listen to John Hayes' interview on Radio New Zealand's Checkpoint programme (17 June 2016)
A delicate balance
Dr Hayes says balancing all needs is difficult, but crucial.
“It’s all about how the changes in flow affect the productivity of a fishery. On one side of the coin is the water the farmers want, to increase the productivity of the land. And what’s been happening in the past – and the new model can show this – is that for an increase in productivity on land brought about by taking water out of rivers, you reduce the potential productivity of the fishery. There's no getting away from that, and that’s what you’ve got to balance.”
It was on the Mataura River that Dr Hayes had what he calls his “Eureka moment”. The team was running the new model on the river, and up until then he hadn’t appreciated that the concentration of drifting invertebrates increases with water flow, in a similar way that fine sediment concentrations increase with flow.
Updating a 40-year-old model
Traditional habitat modelling for estimating flow requirements was developed in Colorado in the 1970s. The model is based on fish preferences for physical habitat, including water depth and velocity, and substrate composition and cover. It’s become the most widely used and accepted method of assessing the flow requirements of fish populations, but accounting for the effects of flow on physical habitat alone is not enough.
The missing ingredient in that model is food availability, in the form of drifting invertebrates.
“It doesn’t take into account the flow-related processes influencing habitat selection and foraging by drift-feeding fish or their drifting invertebrate food,” Dr Hayes says.
What's the difference?
Although ecologists have long known that both the physical habitat and food supply are important regulators of fish populations, the latter is rarely considered in North American assessments (though it is usually included in New Zealand). Dr Hayes and his team were able to demonstrate that traditional modelling substantially underestimated the flow requirements of brown trout in the Mataura River.
They developed what they’ve dubbed the ‘drift-NREI’ (net rate of energy intake) model. NREI is the difference between how much energy the fish consumes versus how much it expends, when foraging for food. The computer model simulates how water flow dislodges and transports aquatic invertebrates, how trout forage in the current on the drifting prey, and how this can be quantified in the currency of energy to predict fish numbers and growth rates. The model can be used to test hypothetical scenarios of stream-flow and habitat, and also takes into account depletion of the drifting invertebrates as fish eat them.
The research team found the concentration of drifting invertebrates decreased as stream-flow declined because the river had less power to pick them up from the river bed and keep them in suspension. This means not only less drifting food for the fish, but also less for the fish’s prey, too, as filter-feeding invertebrates receive less fine particulate food.
Showing the variances between rivers
Southland Fish & Game manager Zane Moss says fisheries managers have long held concerns that the method most frequently used in New Zealand to determine appropriate flow for trout – physical habitat modelling – comes up with similar results across a range of rivers that appear to be manifestly different in their nature.
“At a fundamental level this just doesn’t make sense ecologically,” he says, adding that the energy intake model shows “great promise”.
“It provides results that seem more sensible in their predictions. This research really is outstanding in an international context and greatly advances the confidence managers have in ensuring trout habitat is not sold down the river in over-allocation.”
Confirming anglers’ gut feelings
The research has borne out the gut feeling of many anglers. A 2003 NIWA study into anglers’ perceptions of New Zealand’s lowland rivers and trout fisheries, commissioned by Fish & Game, shows there was strong consensus that angling quality had declined over 70 years of record. The survey, which ran from December 2000 to June 2001, targeted anglers with more than 20 years’ experience. Two-thirds of the assessments indicated that angling had become “generally or markedly worse”, with the most rapid decline in the 1940s and 50s, and again since the 1980s.
The report noted that the decline was perceived to be most marked in Nelson, Hawkes Bay, Marlborough and Canterbury. These are the dryland regions of New Zealand where there’s the greatest demand for water for irrigation.
New model already being used nationally and internationally
Southland and Otago regional councils have already begun to use the drift-NREI model to revise their minimum flow rules. Dr Hayes says freshwater fisheries scientists in the United States have also been quick to realise its potential, using it in a multi-million dollar research programme on endangered salmonid populations, and how to restore them, in the Columbia River catchment.
The research has attracted praise from other stakeholders around New Zealand. Environment Southland water resources scientist Lawrence Kees says freshwater and the species that it supports are key resources in Southland’s lifestyle and economy.
“So, managing water for the interests of everyone in Southland is our highest priority.”
He adds that the last decade has seen an increasing demand on water resources across Southland, which has meant methods determining its allocation have undergone closer scrutiny.
“It’s vital that we employ the most biologically realistic and defensible methods available, to understand how changes in river flow affect aquatic life and to set robust allocation limits for the future,” he says.
Neil Deans, technical policy advisor to the Minister for the Environment, says up-to-date technical information is vital for policy development, and such research can help inform water resource management.
He says the quality of Dr Hayes’ and his collaborators’ research is internationally recognised, and he would recommend it be evaluated for the national limit-setting process.
“If we are going to set limits, then they should be effective in meeting community aspirations and as up-to-date as they can be,” he says. “This has implications for both environmental protection and for the allocation of water and its quality.”
Model a team effort
The research was conducted in partnership with NIWA’s Sustainable Water Allocation Programme (SWAP). Manager Dr Murray Hicks says the drift-NREI model developed and tested by the team is an important part of SWAP’s mission to provide tools that more reliably predict environmental flow requirements, particularly minimum flows.
“Traditional methods that simply consider the availability of in-stream physical habitat at given flow rates are falling from favour, since they don’t consider key factors such as population dynamics and trophic [position in food chain] interactions – which control, for example, the food supply for fish,” he says.
“So the SWAP vision is to develop a ‘numerical river’ that couples physically based models (which predict water depths and velocities, substrate mobility and channel morphological change) with ecosystem models (which provide a much more realistic understanding of biota [animals and plants] response to flows and trophic interactions).”
He says the drift-NREI model is a key part of that vision.
The project required the input of many people; Dr Hayes worked with Cawthron colleagues Karen Shearer, Eric Goodwin, and Joe Hay, as well as scientists from the United States. The research has been supported by the Ministry of Business, Innovation and Employment; Fish & Game New Zealand; University of Alaska Fairbanks, Bureau of Land Management Fairbanks; Environment Southland; Cawthron Institute; and more recently NIWA has been a major funder under its core-funded Sustainable Water Allocation Programme (SWAP). Cawthron and NIWA are currently collaborating in the SWAP on further model development.
The research was published in April this year in the international science journal – Transactions of the American Fisheries Society. The American Fisheries Society is the world’s oldest and largest organisation dedicated to strengthening the fisheries profession, advancing fisheries science, and conserving fisheries resources.
Support our research
Find out more
Access the journal article Can WUA correctly predict the flow requirements of drift-feeding trout? published by the American Fisheries Society
Listen to Dr Hayes' interview on Cawthron's radio show
View related research by John Hayes et al: