The third in a series of recreational catch surveys for Kaikōura and North Canterbury was carried out over the entire fishing year 2012–13 and was also expanded to include rock lobster with the sea perch and blue cod previously monitored. The results describe increases and shifts in effort, increases in harvest, and some decline in catch rates. Smaller fish evident in the harvest may indicate recent improvement in recruitment.

Catch (in number of fish) and effort (CPUE) of juvenile albacore (Thunnus alalunga) in the New Zealand troll fishery are reported by fishers and the catch rates are standardised for shifts in the fleet and fisher behaviour to identify any underlying trends in abundance that may inform the stock assessment. The conclusion of this study is that CPUE of troll-caught albacore in New Zealand waters is unlikely to be a useful index of abundance. There is probably no recruitment signal in these indices.

8 July 2021

This project aimed to evaluate potential agricultural land uses that could provide an alternative to livestock production. In all scenarios, changing the land use to the selected crops increased the profitability. Converting nearly 200,000 ha of pastoral land to crops and mānuka plantations will have a relatively small effect on CH4 emissions. Adoption and development of mitigation technologies is more likely to have a large effect towards meeting future CH4 emission reduction targets.

This research strongly supports the idea that soil C stocks are stable under permanent pasture. They are not influenced by P fertiliser inputs and sheep grazing but are influenced by slope and aspect. Understanding landscape attributes is an important factor in assessing soil C stocks. A spatial nutrient budget model was developed. The model could form the basis of a farm scale C accounting tool by linking the soil C measurements to the animal production system.

Keywords: soil carbon, slope, phosphorus, sheep, ballantrae, greenhouse gas, fertiliser

A machine learning algorithm was developed to analyse underwater videos and to detect the presence of macroalgae. Three habitat-forming algae Ecklonia radiata, Lessonia variegate, and Carpophyllum spp. were successfully identified. The machine learning models can be readily applied to ongoing monitoring programmes to rapidly determine and map the distributions of key macroalgal indicator species along coastlines. Monitoring data are critical to documenting changes in our coastal communities.