Effects of anthropogenic stressors on temperate mesophotic ecosystems

Coastal ecosystems are among the most productive biomes on the planet, but also the most vulnerable. A wide range of anthropogenic impacts are threatening the integrity of coastal systems, and therefore their capability to provide goods and ecosystem services. Despite their importance and vulnerability, many coastal ecosystems remain poorly studied. Mesophotic ecosystems lie between the shallow euphotic waters and the aphotic deep-sea. While these ecosystems have been relatively well-studied in tropical regions, the importance of temperate mesophotic ecosystems (TMEs) has only recently been recognised. TMEs extend from the lower limit of the euphotic zone to the limit of benthic primary production, which usually correspond to ~30–150 m. However, in particular conditions, mesophotic ecosystems can occur in shallower waters. Lough Hyne (Ireland) is a fully marine sea lough, designated as Europe's first Marine Nature Reserve in 1981 for its extraordinary biodiversity. This reserve hosts particularly diverse sponge-dominated mesophotic communities as shallow as 10 m. Unfortunately, during the last decade, the lough’s underwater communities have undergone drastic changes, which have been attributed to changes in water quality and in particular, an increase in dissolved nitrogen. My research aimed to investigate the effects of anthropogenic stressors on temperate mesophotic ecosystems using Lough Hyne as a model system. My thesis aims to: 1) characterise the changes that have occurred in the subtidal communities at Lough Hyne; 2) investigate the possible causes of these changes through tolerance experiments; and 3) investigate community dynamics and any recovery of Lough Hyne’s mesophotic communities.

In my first chapter, I collated 30 years (1990-2019) of scientific surveys and opportunistic observations of the subtidal (6–30 m) communities at Lough Hyne to investigate the long-term stability and vulnerability of TMEs. I then explored the potential causes of the observed changes. I found significant changes in the overall biological community and sponge assemblages at all sites within Lough Hyne. However, these changes were not consistent across sites and mainly affected the innermost areas of the lough. Changes were also not consistent between taxa and functional groups, suggesting differential vulnerability of TME organisms to stress events. The main finding was a marked decline in most three-dimensional sponges in the inner part of the lough, which was likely the result of one or more mass mortality events between 2010 and 2015. These changes did not seem to be related to either thermal or rainfall anomalies. The only factor known to have changed over this period is nitrogen, which has increased threefold during the last 20-30 years. Therefore, I hypothesised that the sponge mortality at Lough Hyne is linked to eutrophication. Importantly, this chapter shows the potential vulnerability of TMEs to short-term disturbance events, and highlights the importance of monitoring mesophotic habitats globally to ensure they are appropriately conserved.

In my second chapter, I investigated the possible causes of the mass sponge mortality at Lough Hyne through laboratory experiments. One of the most common and severe consequences of eutrophication in aquatic ecosystems is hypoxia. Therefore, I investigated the response of sponges to moderate and severe simulated hypoxic events. I ran three laboratory experiments on four species from two different temperate oceans (NE Atlantic and SW Pacific). I exposed sponges to a total of five hypoxic treatments, with increasing severity (3.3, 1.6, 0.5, 0.4 and 0.13 mg O2 L-1, over 7–12 days). The main finding was that sponges are generally very tolerant of hypoxia. All sponges survived under the experimental conditions, except Polymastia croceus, which showed significant mortality at the lowest oxygen concentration (0.13 mg O2 L-1, median lethal time: 12 days). In all species except Suberites carnosus, respiration rate was unaffected down to 0.4 mg O2 L-1. Importantly, sponges showed species-specific phenotypic changes in response to hypoxic treatments, likely representing adaptive strategies for living in low oxygen water. Compared to other sessile organisms, sponges generally showed a much higher tolerance to hypoxia, suggesting that sponges may be favoured and survive in future deoxygenated oceans. These results also indicate that hypoxia alone was probably not the cause of sponge mortality at Lough Hyne.

In my third data chapter, I investigated the resilience and temporal dynamics of shallow-water (~ 17 m) mesophotic communities at Lough Hyne, following the mass mortality event. In June 2018, I established five replicate permanent quadrats (0.25 m2) on the rocky cliffs (18 m) at five sites inside the lough. My collaborators and I took photoquadrats twice a year until June 2021 (36 months, 5–7 time points), from which I extracted data on the percentage cover of sessile organisms identified to the lowest practical taxonomic level. In addition, I analysed historical photoquadrats collected with a similar methodology between 1994 and 1995, from two internal sites at Lough Hyne. Historical data were used to assess if differences in temporal patterns between sites were exclusive to impacted communities or also occurred in the past. Multivariate analysis did not detect any directional community or assemblage changes over time in the internal and innermost sites, suggesting that recovery of benthic communities and sponge assemblages is either occurring very slowly or not at all. In contrast, I found significant community changes at the entrance site, where barnacles suffered a mass mortality event in 2018, perhaps due to a heatwave. Univariate analysis revealed weak signs of recovery for some of the three-dimensional sponges and anemones that were highly affected by the disturbance, represented by a small increase in percentage cover. In general, temporal dynamics (turnover, diversity and percentage cover of benthic groups) were found to be different between: 1) sites experiencing very distinct environmental conditions; and 2) between sites that shared similar conditions and communities. Most importantly, sponges seem to be recovering only in one of the three internal sites, where I detected a positive trend in the three-dimensional sponges affected by the 2010–2015 mortality event. This finding indicates that small variations in environmental conditions can affect the dynamics and recovery of mesophotic subtidal ecosystems.

Overall, my thesis shows that mesophotic ecosystems are vulnerable to environmental stressors and slow recovery rates. Most temperate mesophotic taxa are long-lived and slow-growing organisms, likely to have limited resilience to human-induced impacts. Despite sponges being generally considered tolerant of stressors, the mass mortality at Lough Hyne shows that this is not always the case. My results suggest that three-dimensional mesophotic sponges are among the most sensitive species and the slowest growing. Any decline in these habitat-forming species will likely affect important ecosystem functions (e.g., nutrient cycling, bentho-pelagic coupling, habitat provisioning), with detrimental effects on the associated ecosystem services. My study also shows that, as some sites, recovery is underway. However, at the current rate, whole-lough recovery is likely to be in the order of decades. Given the vulnerability and importance of TMEs, research and management of these habitats should be prioritised at Lough Hyne and elsewhere. In particular, long-term monitoring of biotic and abiotic factors will be crucial to understand TME long-term dynamics, recovery and how these ecosystems respond to environmental variations and anthropogenic disturbances. Monitoring will improve our ability to make evidence-based decisions for the management of TMEs in a fast-changing world. In addition, we also need more research on how mesophotic organisms respond to stressors and how they contribute to ecosystem functioning. Better knowledge of these ecosystems will increase awareness of the value of TMEs among decision-makers and the general public, which will be essential to ensure their conservation.