The Invisible Farm: Rethinking What Intensive Aquaculture Does to Fish

Farmed fish inhabit environments that bear little resemblance to the conditions under which their sensory systems, immune responses and behavioural repertoires evolved. Aquaculture science has made extraordinary progress in managing the consequences of this mismatch where it is most visible: disease, parasites, nutrition, genetics, systems, management and engineering. Three reviews published in the previous issue (Reviews in Aquaculture 18/2) direct our attention elsewhere, to the invisible, and specifically to the acoustic, sensory and chemical fabric of the rearing environment itself, and in doing so, make a case that some of the most consequential stressors in intensive aquaculture are precisely those that current production metrics are not designed to detect. Overton et al. [1] reveal that over half of Norwegian farmed Atlantic salmon possess vateritic otoliths, a skeletal deformity linked to rapid hatchery growth that occurs at an incidence far exceeding that found in wild populations. Mechanistic modelling suggests this deformity reduces hearing sensitivity by 28% to 50% across the known hearing range of the species. The welfare implications are entirely uncharted: we do not yet know whether hearing-impaired fish experience greater or lesser distress in noisy farm environments, or how altered auditory perception affects their capacity to navigate the social and operational demands of intensive rearing. What is clear is that intensive farms are noisy places, that farmed salmonids cannot escape that noise, and that a large proportion of them may be sensing it through a system already compromised by the conditions of their own production. Spiliopoulos et al. [2] provide a compelling complement to this picture. Their synthesis of environmental enrichment research demonstrates that the welfare costs of sensory impoverishment are real and measurable, but also that they are addressable through surprisingly modest interventions. Particularly striking is the evidence that partial overhead coverage in circular hatchery tanks reduced physiological stress indicators in juvenile Atlantic salmon despite the fish rarely occupying the sheltered space. The mere availability of refuge, the option of agency rather than its exercise, appears sufficient to attenuate chronic stress. This finding reframes enrichment from an optional welfare addition to an argument about environmental design: the default conditions of intensive aquaculture are not neutral, and the absence of complexity is itself a stressor. Xie et al. [3] extend this argument into the chemical domain. Micro- and nanoplastics have infiltrated aquaculture systems globally, entering fish through both water and feed and triggering cascading disruption of gut integrity, oxidative balance and immune function. Among the most novel findings is the identification of a physical bioremediation mechanism: certain probiotic strains, particularly lactobacilli, can adsorb plastic particles directly onto their cell surfaces through hydrophobic and electrostatic interactions, reducing the bioavailable fraction of plastics within the intestinal lumen before translocation into deeper tissues occurs. The gastrointestinal tract, potentially already weakened by the sensory poverty of barren rearing environments [2] and the cumulative physiological burden of chronic farm noise [1], represents a primary front line of defence against chemical contaminants that targeted dietary interventions can successfully fortify [3]. What unites these three bodies of work is not merely their shared concern with invisible stressors, but the convergence of those stressors on common physiological ground. Chronic noise exposure, sensory impoverishment and plastic contamination each operate through distinct proximate mechanisms, yet all three ultimately burden the same downstream systems: the stress axis, the antioxidant defence network, the gut microbiota and the immune response. The welfare cost of the invisible environment is therefore unlikely to be simply additive. Fish simultaneously subject to compromised hearing, barren conditions and plastic-laden feed may be navigating an interactive and cumulative physiological burden that no single intervention can fully address. Quantifying and managing that burden will require new standards, new measurement protocols and a willingness to design production systems around the full reality of the animals they contain. None of these can happen without people willing to do work that is itself largely invisible. Peer review and editorial oversight are among the least visible forms of scientific labour, and it seems fitting that an editorial arguing for closer attention to what cannot easily be seen should pause to acknowledge those who provide it. We take this opportunity to introduce a refreshed and expanded Editorial Board for Reviews in Aquaculture. Following an open global call for expressions of interest, we have assembled a larger, more diverse and more internationally representative group than ever before, spanning all major areas of aquaculture science, a broader range of geographic regions and researchers at every career stage. We are particularly pleased to welcome an expanded cohort of Junior Editorial Board members, and four new Associate Editors: Nie Pin, Zhen-Yu Du, Richard Newton and Zhangying Ye, whose complementary expertise will help ensure the journal continues to cover the full breadth of the field. The complete Board is too distinguished to list here in full; we invite readers to visit the journal homepage to meet the whole team. We also extend our sincere gratitude to those who have stepped down after years of dedicated service: their contribution to the quality and integrity of this journal has been immeasurable. To everyone on the Editorial Board, new and continuing alike, aquaculture science moves forward because of people who volunteer their knowledge and judgement, and no small number of evenings, for a discipline they care about. We are grateful for their time, their judgement and their willingness to support the journal and the discipline. Giovanni M. Turchini: conceptualization, writing – original draft, writing – review and editing. This editorial was drafted with the assistance of Claude (Anthropic) and NotebookLM (Google), two generative AI tools used to help structure arguments and synthesise source material. All content was reviewed, revised, and approved by the author (GMT), who takes full responsibility for the final text. The author declares no conflicts of interest. Data sharing is not applicable to this article as no datasets were generated or analysed during the current study.