NIWA has developed a rapid, desktop model which assesses the potential impact of introducing new fish species to New Zealand. The model is customised to New Zealand's unique environment and endemic fish.
It provides a preliminary screening tool to determine whether a species should be denied entry or subjected to a more formal Assessment of Environmental Effects. It also allows biosecurity managers to identify which of the introduced species already present in New Zealand have the highest potential for causing problems should they spread further.
The problem
Non-native freshwater fish species were first introduced to New Zealand by early European settlers in the mid-1880s and the rate of new introductions has not changed since then. Some introduced fish, such as koi carp, gambusia, perch, and trout, have reduced the biodiversity of native species and/or affected water quality in lakes.
To prevent the introduction of additional harmful fish species, the Hazardous Substances and New Organisms Act (HSNO 1996) provides decision-makers with criteria for accepting or declining applications to introduce new fish species. These criteria are necessarily broad - e.g. a new species will not displace a valued or endemic species, or result in habitat deterioration - and it is impossible to determine whether such impacts will occur with scientific certainty until the new species is tested in a realistic environment such as New Zealand waters.
As this test is not allowed under the Act, scientists must use models to predict whether or not an impact could occur in New Zealand waters. Any such model would need to be validated using actual knowledge of fish impacts in New Zealand.
It would provide a tool for biosecurity managers to identify which fish species should not be allowed entry to New Zealand, and to identify the species already present that require most attention to stop them spreading further.
A range of risk assessment models to assess the likely consequences of introducing a new species to a country have been developed overseas. Initial models dealt mainly with terrestrial species, especially plants, and risk assessments for fish evolved from these.
Risk assessments for freshwater fish have been developed for Australia, the USA and the UK, but these could not be applied to New Zealand because of the unique nature of New Zealand freshwater environment and fish.
These models were also limited because they were focussed on risks to biodiversity and or human health, and did not deal adequately with ecosystem effects related to water quality such as the impacts that herbivorous or planktivorous species might have on lake ecosystems.
Further, the risk that a species would invade and spread widely was emphasized over its potential to cause an impact, whereas in New Zealand with its smaller, geographically isolated rivers and lakes, the risk of site-specific impacts is more important.
The solution
The models developed for the UK and Australia were disassembled, modified, added to and re-assembled to develop a model suited to New Zealand.
This model consists of a series of questions about the biology and ecology of a fish species including its impact in other countries. The responses to each question result in a score for that question, with higher scores indicating increased risk. Where knowledge is limited, a precautionary approach also results in a higher score. The sum of the scores for all questions then provides an overall measure of risk.
The questions are organised to provide:
- a measure of the risk of a new species' likely establishment and spread throughout New Zealand (i.e., its invasiveness)
- a measure of the risk that a new species will cause an impact in an aquatic environment.
For example, species that can tolerate low salinity and/or have a marine life phase could spread more rapidly and widely than species that don't migrate, are confined to freshwater, and which cannot breed in the wild.
Furthermore, some species that have spread widely throughout New Zealand have had little known impact (e.g. goldfish), whereas others that are not as widely spread can cause a severe impact in some waters (e.g. koi carp).
The model was validated using knowledge of the impacts of the 21 introduced species now present in New Zealand.
The result
The model results in a score for the risk of impact (0-35), the risk of establishment (0-20), and the overall risk of introduction to the New Zealand environment (0-55).
Ground-truthing indicated that species scoring over 22 for risk of impact, and over 12 for risk of establishment, should be denied entry to New Zealand.
Species scoring less than 14 for risk of impact, and less than 6 for risk of establishment, would be expected to have little impact and so could be allowed entry with appropriate conditions.
Species lying between these extremes would require a robust and formal Assessment of Environmental Effects (AEE) before introduction could be considered.
Of the species already introduced to New Zealand, perch, koi carp and gambusia had the highest overall scores (Fig.2). Koi carp and gambusia are 'unwanted species' under the Biosecurity Act (1993), but perch are designated as an acclimatised fish, and although they are controlled by Fish & Game Councils, they are still spreading.
Both brown and rainbow trout also had high scores. If trout were proposed for liberation today their introduction may have been allowed on economic grounds but, given their potentially high impact, strict conditions would have been imposed to ensure their spread was closely monitored and restricted so that they were only stocked into certain waters, not all. This Fish Risk Assessment Model (FRAM) is designed to prevent such problems occurring in the future. For example, perch is also a popular species for angling in some regions, but its further spread needs to be tightly controlled to ensure further environmental damage is avoided.