Ross Vennell

This study presents field measurements for a non-conventional mobile square cylinder fish enclosure, with permeable ends, evaluated under both towed and moored conditions at a semi-open ocean test site. Novelty lies within the enclosure design that enables both mobility and control of internal flow through the incorporation of an adjustable aft-end perimeter constriction to regulate internal flow and support fish welfare by enabling control of swimming conditions. Enclosure motion, internal flow speeds and hydrodynamic loads were measured for three constrictions (0%, 40% and 60%). The primary objective was to assess the effectiveness of aft-end constriction in regulating internal flow to levels compatible with sustainable swimming speeds for finfish culture. The enclosure remained stable at approximately 9 m depth across all vessel speeds and constriction settings in both towed and moored scenarios. During towing, increasing constriction to 40% reduced time-averaged internal flow by up to 24% without a significant increase in hydrodynamic load. A 60% constriction achieved a larger reduction (~51%) but resulted in a substantial load increase (~90% relative to 0% constriction), indicating a trade-off between flow control and towing resistance. Under moored conditions, aft-end constriction had minimal influence on both internal flow and hydrodynamic load. Mooring loads showed no clear relationship with wave height and only a weak correlation with ambient current speed. Overall, the results demonstrate that aft-end constriction is an effective mechanism for controlling internal flow during towing, but has limited impact when moored. The enclosure’s stability and controllability highlight its potential advantages over conventional gravity cages for mobile open-ocean finfish aquaculture applications.

The aquaculture sector in New Zealand is experiencing rapid expansion, with a particular emphasis on mussel exports. As the demands of mussel farming operations continue to evolve, the integration of artificial intelligence and computer vision techniques, such as intelligent object detection, is emerging as an effective approach to enhance operational efficiency. This study delves into advancing buoy detection by leveraging deep learning methodologies for intelligent mussel farm monitoring and management. The primary objective centers on improving accuracy and robustness in detecting buoys across a spectrum of real-world scenarios. A diverse dataset sourced from mussel farms is captured and labeled for training, encompassing imagery taken from cameras mounted on both floating platforms and traversing vessels, capturing various lighting and weather conditions. To establish an effective deep learning model for buoy detection with a limited number of labeled data, we employ transfer learning techniques. This involves adapting a pre-trained object detection model to create a specialized deep learning buoy detection model. We explore different pre-trained models, including YOLO and its variants, alongside data diversity to investigate their effects on model performance. Our investigation demonstrates a significant enhancement in buoy detection performance through deep learning, accompanied by improved generalization across diverse weather conditions, highlighting the practical effectiveness of our approach.

Due to their role as primary producers, phytoplankton are essential to the productivity of estuarine ecosystems. However, it is important to understand how these nearly passive organisms are able to persist within estuaries, when river inflow results in a net outflow to the ocean. Estuaries are also representing challenging habitats due to a strong salinity gradient. So far, little is known about how phytoplankton are able to be retained within estuaries. We present a new individual-based Lagrangian model of the Elbe estuary which examines possible retention mechanisms for phytoplankton. Specifically, we investigated how reproduction, sinking and rising, as well as diel vertical migration may allow for populations to persist within the estuary. We find that vertical migration especially rising favors the retention, fast sinking does not. We further provide first estimates on outwashing losses. Our simulations illustrate that riverbanks and tidal flats are essential for the long-term survival of phytoplankton populations, providing refuges from strong downstream currents. These results contribute to the understanding needed to advance ecosystem-based management of estuaries.

Aquaculture is an important industry in New Zealand (NZ). Mussel farmers often manually check the state of the buoys that are required to support the crop, which is labour‐intensive. Artificial intelligence (AI) can provide automatic and intelligent solutions to many problems but has seldom been applied to mussel farming. In this paper, a new AI‐based approach is developed to automatically detect buoys from mussel farm images taken from a farm in the South Island of NZ. The overall approach consists of four steps, i.e. data collection and preprocessing, image segmentation, keypoint detection and feature extraction, and classification. A convolutional neural network (CNN) method is applied to perform image segmentation. A new genetic programming (GP) method with a new representation, a new function set and a new terminal set is developed to automatically evolve descriptors for extracting features from keypoints. The new approach is applied to seven subsets and one full dataset containing images of buoys over different backgrounds and compared to three baseline methods. The new approach achieves better performance than the compared methods. Further analysis of the parameters and the evolved solutions provides more insights into the performance of the new approach to buoy detection.

A simple diagnostic narrow channel shallow water model for the cross‐channel distribution of the depth average velocity is developed. It is diagnostic, because the slope of the stream‐wise free surface is specified and it models the cross‐channel distribution of the depth average velocity without any knowledge of the velocity field. A non‐dimensional parameter that encapsulates the channel dimensions and the relative importance of bottom friction to horizontal eddy viscosity is defined. Numerical solutions for the cross‐channel distribution of the depth average velocity are obtained using values of this non‐dimensional parameter for different cross‐channel topographies. Numerical solutions for the depth average velocity are in turn used to find the cross‐channel distribution for secondary flow and vertical velocity. Two‐dimensional distributions of the vertical velocity in the n – z ‐plane show that changes in topography dominate the vertical velocity distribution. Three‐dimensional tracers in a hypothetical open channel using scaling values from geophysical observations show that secondary circulation has a loose helical flow pattern of approximately one full rotation for every of curvature. It is not unreasonable to suggest that this 1:1 relationship describes the looseness of secondary circulation in geophysical data.

In this paper, periodic tide‐current‐driven banding in a sea‐ice core is demonstrated as a measure of the growth rate of first‐year sea ice at congelation‐ice depths. The study was performed on a core from the eastern McMurdo Sound, exploiting the well‐characterized tidal pattern at the site. It points the way to a technique for determining early‐season ice growth rates from late‐season cores, in areas where under ice currents are known to be tidally dominated and the ice is landfast, thus providing data for a time of year when thin ice prevents direct thickness (and therefore growth rate) measurements. The measured results were compared to the growth‐versus‐depth predicted by a thermodynamic model.