Aquaculture-environment interactions


Coastal aquaculture provides one of New Zealand’s biggest opportunities to generate new wealth from the primary production sector.

Uncertainty about potential environmental effects of aquaculture expansion is a major impediment to realising this potential. Conversely, there is an increasing evidence that environmental fluctuations (whether natural or human-induced) influence crop performance.  Whilst marine farms can have effects upon the coastal environment, that environment is also subject to disturbance by other human activities. Consent conditions for marine farms often impose environmental standards. In practise, however, if standards are breached, it can be difficult to determine the cause of the breach with sufficient certainty to permit enforcement action to be taken. Knowing that enforcement will be difficult, regulators adopt precautionary tactics. In some cases, these may be warranted, in others they inhibit economic development unnecessarily.

The challenge

Sustainable development of marine farming requires providing opportunities for investors while maintaining coastal ecosystem health and integrity.

When considering whether to invest in a marine farm, the farmer requires a clear understanding of:

  1. the extent to which environmental factors will influence/constrain both their long-term average returns and the variability of annual returns. 
  2. They also need to be confident that they will not penalized for environmental breaches which are not of their making.

Conversely, when considering a new application, regulators must be able to satisfy themselves that:

  1. environmental changes arising from the additional farm (in concert with those of pre-existing farms) are likely to remain within acceptable bounds,
  2. environmental monitoring regimes can detect changes that would be unacceptable large
  3. the monitoring data (in concert with other tools and appropriate follow-up studies) will enable them to determine the cause(s) of any changes that have deemed to be too large to be acceptable.

Both farmers and regulators seek monitoring methods that yield reliable results without imposing unecessary onerous costs upon either party.

The solution

Our Aquaculture-Environment Interactions program seeks to address all of these topics. It is split into three projects.

  • Aquaculture: Local-scale interactions
  • Aquaculture: Regional-scale interactions
  • Aquaculture:  Drivers of mussel performance

Local-scale: is focussed upon developing improved methodologies for better characteristing and forecasting the influence which marine farms have upon water-flows, sediment resuspension and nutrient-cycling capacity of the seabed.  We have developed novel means to measure seabed/water-column oxygen fluxes at larger spatial scales (and with fewer artefacts) than has previously been possible (Plew 2019). We are also developing very high resolution models to simulate current flows and dissolved oxygen in and around fish-pens.  Whilst these activities have focussed upon fish-farms, the methods are equally applicable to shellfish farms and will allow better forecasting and monitoring of farm-scale environmental influences upon the local environment’s capacity to support life and natural nutrient-cycling.

Regional-scale: has two foci.  The first aims to develop tools to forecast and measure the cumulative the bay- and regional-scale effects of multiple marine farms.  For example, in a world-first, we were able to use satellite data to demonstrate that development of 1200 ha mussel farming zone in the Firth of Thames has been associated with a reduction of phytoplankton concentration that extend over an area that is about 50% larger than the marine farming zone itself.  Within that larger area, the time-and-space averaged depletion is 1% - 2%. The maximum time-averaged depletion within the farming zone itself is around 6% (Figure 1).  We have developed a coupled hydrodynamic and biogeochemical simulation models of the Marlborough Sounds (Hadfield, Broekhuizen et al. 2014, Broekhuizen, Hadfield et al. 2015).  This year, we have applied a similar model to the Firth of Thames.  In both water-bodies, the model predicts greater phytoplankton depletion than is evident from field data.  Future research will focus upon determining what aspects of the model need to be modified to remedy these deficiencies. We are also collaborating with the Australian Bureau of Meteorology to develop the capacity to produce well-validated six-month forecasts of water-temperature for our marine farming zones (de Burgh-Day et al.2019 – see also Figure 2).

Drivers of Mussel performance: in this task we have developed a model to forecast the condition-score of harvest size mussels in Pelorus Sound and made this available to industry through an open website: We are now seeking to extend that study to examine how environmental fluctuations influence the neutriceutical-value of mussel flesh (Stenton-Dozey and Ren 2019).   Losses of recently seeded mussel-spat from crop-lines are variable, but often very large.  We have recently commenced studies to try to determine what characteristics of the environment influence spat retention.



Plew, D. R. (2019). "Investigating benthic impacts at salmon farms using eddy covariance measurements of benthic oxygen fluxes." Aquaculture Environment Interactions 11: 337-357.

Pinkerton, M., M. Gall, S. Wood and J. Zeldis (2018). "Measuring the effects of mariculture on water quality using satellite ocean colour remote sensing." Aquaculture Environment Interactions 10: 529-545.

Stenton-Dozey, J. and J. Ren (2019). Relating mussel farm production and environmental variation.

Broekhuizen, N., M. Hadfield and D. Plew (2015). A biophysical model for the Marlborough Sounds part 2: Pelorus Sound. National Institute of Water & Atmospheric Research Ltd. Christchurch, National Institute of Water & Atmospheric Research Ltd: 163.

Hadfield, M., N. Broekhuizen and D. Plew (2014). A biophysical model of the Marlborough Sounds: part 1: Queen Charlotte & Tory Channel. NIWA Client Report (for Marlborough District Council). Christchurch, NIWA: 183.

de Burgh-Day, C.O., Spillman, C.M., Stevens, C., Alves, O. and Rickard, G., 2019. Predicting seasonal ocean variability around New Zealand using a coupled ocean-atmosphere model. New Zealand journal of marine and freshwater research, 53(2), pp.201-221.



NIWA Contacts

Principal Scientist - Ecosystem Modelling
Page last updated: 
23 April 2020
Mussel farm in Pelorus Sound (Barb Hayden, NIWA)
Figure 1: illustration of the estimated concentration of chlorophyll in the presence of the marine farm relative to that prior to the farming-area’s development.  A value of 1 indicates no depletion.  A value of 0.94 indicates 6% depletion.  The inner-most black rectangle is the 1200 ha marine farming zone.  The middle rectangle has an area that is 50% larger than the central one.  The outer rectangle demarcates the maximum area to which the analysis initially assumed a farm influence may extend.  Ultimately, the analysis revealed that the influence extended only to about central rectangle.
  Figure 2: An ensemble forecast of temperatures in Cook Strait for the period December 2019-June 2020. 99 individual temperature-forecasts were made (grey-lines) for the period December 2019-May 2020. The black line is the average of the 99 forecasts. The green line illustrates the 1990-2012 monthly average temperatures.  The yellow line illustrates the actual water temperatures that were subsequently observed in the forecast region (Greater Cook Strait).  Produced by the Australian Bureau of Meteorology in collaboration with NIWA’s Aquaculture, Coasts and Oceans and Climate centres.   
Research subject: Coasts