The feasibility of a commercial-scale recirculating aquaculture system integrating sea urchins (Tripneustes gratilla) and seaweed (Ulva)

De Vos, Sebastiaan

The feasibility of a commercial-scale recirculating aquaculture system integrating sea urchins (Tripneustes gratilla) and seaweed (Ulva)

Thesis / Dissertation

2024

Abstract

It has been proposed that the highly valued sea urchin, Tripneustes gratilla, and the green macroalga, Ulva, could be suitable co-candidates for a land-based recirculating Integrated MultiTrophic Aquaculture (IMTA) system. This IMTA system is suggested to be sustainable and efficient, primarily because the Ulva would simultaneously provide bioremediation of effluent and a substantial feed source for T. gratilla. While evidence for these proposed benefits does exist, it is dispersed and incomplete. The primary aim of this thesis is to create a farm-scale model to provide clear evidence of the biotechnical feasibility of a T. gratilla-Ulva IMTA system. To develop this model, literature gaps needed to be filled. These gaps became the auxiliary initial objectives, specifically; to determine optimal T. gratilla basket depth, stocking density and quantification methodology to allow for the accurate prediction of T. gratilla nitrogen emissions. The literature indicates that production of various urchin species is reduced when cultivated in deeper baskets, in contrast to shallower baskets. Thus, if and why basket depth has this effect was investigated and an ideal basket depth for T. gratilla culture was determined. Deeper baskets (30 cm deep) resulted in significantly lower consumption of various feed types (formulated feed, Ulva lacinulata and Ecklonia maxima; W > 38, p < 0.026). This is likely the consequence of lower feed accessibility, which in turn causes the observed reduced yield. Therefore, shallow baskets (±15 cm deep) enhance production of T. gratilla and were applied in subsequent stocking density trials. The stocking density of T. gratilla expected in commercial systems has not been clarified or optimised. Two trials were conducted where production parameters were compared between urchins stocked at various densities. The first trial was a three-month grow-out trial where only fresh Ulva lacinulata was supplied and the objective was to maximise urchin size. The subsequent trial focused on maximising gonad production where predominantly a formulated feed with a 20% Ulva inclusion was provided over two-months. While higher stocking densities did significantly reduce the specific growth rates of average individual urchin mass in both trials (p < 0.044), there was no indication that mortality, cannibalism or gonad size and quality was influenced by the densities tested in this study. Thus, it was concluded that the optimal stocking density for both the grow-out and gonad enhancement phases of T. gratilla is approximately 20% coverage (surface area of urchins' tests by surface area of basket), regardless of urchin size. During the above trials it became apparent that the implementation of a commercial T. gratilla culture would not be feasible in the absence of a reliable and precise method to measure large quantities of live urchins, which is necessary for the successful development and management of such an industry. Therefore, to avoid this future bottleneck, this study also developed an accessible, accurate and efficient protocol for the reliable and precise measurement of large quantities of sea urchins using computer vision. For a larger-scale context, this open-source software could easily be incorporated into various tools, such as a grading machine, to completely automate farm processes. Additionally, this protocol can be used in a research context to greatly enhance the accuracy and standardisation of live urchin measurements.The foundation of the T. gratilla-Ulva IMTA model is a total ammonia nitrogen (TAN) sub-model, as TAN is the most limiting exchanged resource. Due to the lack of literature on T. gratilla nitrogen emissions and its complexity, an empirical ‘black box' approach using regression models was used for predicting TAN concentrations in T. gratilla effluent. Training and validation data were acquired from an extensive trial where water samples were collected from urchin systems with different treatments hypothesized to influence water quality, such as flow rates, feed type and quantity of urchins. Additionally, additional water quality parameters were observed, thereby providing an indepth understanding of how T. gratilla aquaculture systems will influence water quality. The nitrogen concentration in T. gratilla effluent was low by aquaculture standards (average 0.001 mg/lTAN) while the high dissolved CO2 concentrations were of concern (maximum of 525 μATM). This provides novel insights, from which valuable management recommendations are made. As the literature gaps were filled as described above, it became possible to create the T. gratillaUlva farm-scale IMTA model, which would not only assist justifying the integration of these species but also estimate the production capabilities of this system. To ensure its applicability and practicality, the model was developed as a digital twin of the established and extensively validated commercial abalone-Ulva IMTA farm systems in South Africa. The model suggests that, based on the proposed monthly culture cycle, the T. gratilla capacity of the 42 tanks (8.5 m3 each) is approximately 360 000 individuals of seven successive cohorts, which could result in a monthly gonad harvest between 0.31 and 0.88 t per month, depending on finishing feed. The model revealed that while the simulated 300 m2 Ulva raceway could remove all the TAN emitted by T. gratilla (average of 0.009 mg/l), it cannot be considered efficient as a biofilter as it requires considerably more farm surface area (land footprint) than necessary. The projected TAN emissions from the T. gratilla production system are insufficient to sustain the Ulva population, let alone enable substantial Ulva production for T. gratilla feeding. Hence, the farm-scale model indicates there is inadequate nitrogen emissions from the T. gratilla production unit to justify the integration of Ulva with T. gratilla production, based on the existing abalone-Ulva IMTA systems. Yet a nitrogen retention model provides evidence that the nitrogen supplied in the urchin feed is sufficient for substantial Ulva production. Most of this nitrogen is likely deposited as particulates and thus cannot be utilized by Ulva. However, this nitrogen could be converted into a dissolved form via mineralization. While further investigation is required, this indicates the potential to create a circular and efficient IMTA system if designed and managed correctly. Importantly, the model indicates that T. gratilla farming using this land-based system would not be limited by ammonia toxicity and would likely be highly productive when compared to aquaculture of other high value species. Therefore, this study indicates that this recirculating T. gratilla-Ulva IMTA system may be feasible if the correct design or management adjustments are made. The recommended basket design, stocking density, quantification method, water quality insights and the farm-scale model could greatly assist establishing this potential industry. Additionally, with further development, the model could be used as a tool for environmental impact assessors, farmers, entrepreneurs and investors to design farms or conduct economic, environmental and social analyses across various scales.

Keywords

Science

Reference:

Collections

PhD / Doctoral

Full item page