Incidental captures of marine mammals in New Zealand waters have been documented for a range of commercial fisheries, including trawl, longline, set net, pots (and traps) and purse seine. The current study used existing information to characterise marine mammal interactions with these fisheries from 1992–93 to 2017--18, and reviewed potential mitigation techniques for reducing incidental captures in New Zealand waters. Additional analyses were conducted for species with high numbers of captures or for which additional information was available; these species were common dolphin (Delphinus delphis), Hector's and Māui dolphins (Cephalorhynchus hectori hectori and Cephalorhynchus hectori maui), New Zealand sea lion (Phocarctos hookeri) and New Zealand fur seal (Arctocephalus forsteri).

Across fisheries, trawl had the highest number of observed captures when aggregated by species and gear, followed by surface longline. Post-capture survival (whether an individual was recorded as alive upon release) was the highest for surface-longline fisheries, and almost all individuals caught in these fisheries were recorded as live releases. In contrast, post-capture survival was lowest for trawl fisheries, where almost all individuals observed caught were recorded as dead.

The highest number of observed captures were pinnipeds, with New Zealand fur seal featuring the most frequently in observer records. Observed captures of this species were high in trawl fisheries and also in surface longline. New Zealand sea lion was almost exclusively caught in trawl. Common dolphin was the most frequently observed cetacean species, followed by long-finned pilot whale and dusky dolphin.

The current characterisation also identified four specific associations between species and fisheries that resulted in high numbers of captures: common dolphin and large-vessel trawl fisheries targeting jack mackerel, Hector's and Māui dolphins and set-net fisheries, New Zealand sea lion and trawl fisheries targeting squid, and New Zealand fur seal and trawl fisheries targeting southern blue-whiting. Most of these fisheries have implemented mitigation measures over the study period, and reductions in observed and estimated captures were evident for all of these species in the time-series data.

The ability to assess and estimate the extent of captures relies on comprehensive observations of fishing effort via the fisheries observer programme. High observer coverage is crucial for the recording of captures of rare species, for which even a low number of captures can have a significant impact on the population (e.g., Hector's and Māui dolphins). In addition, observer coverage needs to be adequately high to provide reliable capture estimates for species that are observed sufficiently often to inform a model in systematic bycatch assessments. The current analysis revealed that observer coverage was low overall across gears, although there were increases in some fisheries with high marine mammal capture rates in recent years. Low observer coverage in inshore trawl fisheries is concerning, given the high level of effort overlap with the habitat of many coastal marine mammal species.

Mitigation efforts in New Zealand and elsewhere have focused on technical and other approaches to reduce or prevent incidental captures of marine mammals in commercial fisheries. Findings from this research highlight the challenges of testing the efficacy and effectiveness of different mitigation measures, and document the limitations of many approaches; they also show that successful mitigation techniques are often species- and fishery-specific.

Mitigation techniques for New Zealand fisheries include exclusion devices that are currently in use in subantarctic trawl fisheries to mitigate the bycatch of New Zealand sea lion. Exclusion devices have also been trialled in other New Zealand trawl fisheries to reduce incidental captures of New Zealand fur seal. Although these limited trials did not indicate the suitability of exclusion devices, the high number of incidental captures of New Zealand fur seal in trawl fisheries warrants further research into this mitigation technique.

In northern North Island trawl fisheries, acoustic deterrent devices are currently being used on night-tows to prevent common dolphin bycatch. The lack of systematic data of their use and of associated operational aspects means that their effectiveness remains untested. Furthermore, the acoustic deterrent devices are used in combination with other measures in this fishery, precluding the assessment of individual measures.

For longline, set-net, pot (and trap) and purse-seine fisheries, bycatch mitigation options remain limited, and research to date has been unable to identify technical approaches and gear modifications that would have potential in a New Zealand context. Although acoustic deterrent devices have been shown to be effective in preventing bycatch of some small cetacean species in set-net fisheries, research to date does not support their use as a mitigation device for Hector's dolphin bycatch.