Michael Sievers

The condition of fish relates to their energy reserves, and a suite of proxies exist to approximate condition, including biochemical and morphometric indices. Biochemical indices directly measure energy stores but are expensive and sometimes lethal. Morphometrics offer several advantages, but their utility as condition proxies is debated and largely untested experimentally. Here, we manipulated the condition of yellowfin bream ( Acanthopagrus australis, Günther, 1859) via food reduction to simulate the effect of poor habitat quality and calculated a variety of condition indices through time. We measured four non‐lethal morphometrics (Fulton's K, Le Cren's relative condition, width‐to‐length ratio and girth), the hepatosomatic index and a biochemical benchmark (liver lipid content). Girth and width‐to‐length ratio were reasonably well correlated with lipid content ( R 2 = 0.74 and 0.56, respectively). The hepatosomatic index was only weakly or uncorrelated to other indices, including lipid content ( R 2 = 0.35). Where precise estimates of body condition are not needed or repeat measures on the same individual are useful, non‐lethal morphometrics provide a fast, cheap and non‐lethal alternative to biochemical and lethal morphometric methods for this species. We finish by outlining how artificial intelligence‐based automation can be combined with morphometrics to further enhance ethical monitoring by eliminating the need to capture and handle fish entirely.

The loss and degradation of mangrove forests have triggered global restoration efforts to support biodiversity and ecosystem services, including fish stock enhancement. As mangrove restoration accelerates, it is important to evaluate outcomes for species that play functional roles in ecosystems and support services, yet this remains a clear knowledge gap. There is remarkably little information, for example, about how fish use of mangroves varies as restored vegetation matures, hampering efforts to include fisheries benefits in natural capital assessments of restoration. We used unbaited underwater cameras within two distinct zones of mangrove forests—fringe and interior—at five pairs of restored‐natural mangrove sites of increasing age from restoration in southeast Queensland, Australia. We used deep learning to automatically extract data for the four most common species: yellowfin bream ( Acanthopagrus australis ), sea mullet ( Mugil cephalus ), common toadfish ( Tetractenos hamiltoni ), and common silverbiddy ( Gerres subfasciatus ). The abundance of these species varied among sites and zones, but was equal or greater in restored sites compared to paired natural sites. Despite younger restored sites having dramatically lower structural vegetation complexity, abundances did not increase with restoration site maturity. Furthermore, while yellowfin bream and sea mullet were more abundant in the fringe zone, we observed similarities in how fish used fringe and interior zones across all sites. Our paired, space‐for‐time design provides a powerful test of restoration outcomes for fish, highlighting that even newly restored sites with immature vegetation are readily utilized by key fish species.