Bram Parmentier

Small (<30 cm) demersal fish, including buried sandeel (despite their usual classification as pelagic species), play an important role as prey for a wide range of predators, such as marine mammals, seabirds, and piscivorous fish. However, significant knowledge gaps remain regarding species-specific spatial distributions, abundances, energetic characteristics, and, specifically for sandeel, diurnal behaviour. Therefore, the primary objective of this thesis is to improve the understanding of these key aspects of prey availability. Accordingly, the main research question guiding this thesis is: What are the distributional, energetic, and (diurnal) behavioural characteristics of small demersal fish, particularly the two Ammodytes species, in the southern North Sea? An overarching goal is to assess whether the two most abundant sandeel species, Ammodytes marinus and Ammodytes tobianus, play different roles as prey within the ecosystem. As outlined in the general introduction (Chapter 1), knowledge gaps regarding small demersal fish persist, in part due to the inefficient sampling of standard demersal fish surveys for these species. This is because these surveys are primarily designed to monitor commercially important, larger fish. In addition, demersal surveys aimed at capturing small (juvenile) fish primarily focus on quantifying new year classes in coastal areas and therefore lack sufficient spatial coverage of offshore areas. Similarly, the North Sea-wide sandeel survey (NSSS) also provides limited spatial coverage, with only a few sampling locations in the Dutch North Sea. In this thesis, two alternative sampling methods were therefore used: the Triple-D dredge used by the Netherlands Institute for Sea Research (NIOZ), originally designed for macrobenthos sampling, and the dredge used by Wageningen Marine Research (WMR) to quantify annual trends in shellfish populations. Both gears use fine mesh sizes and cutting blades that penetrate the sediment, making them suitable for capturing small fish, particularly buried sandeel. In Chapter 2, data collected by the Triple-D were analysed to address the spatial distribution of small fish. Biomass distribution maps were created for the most abundant small demersal fish species, as well as for all species combined. The maps covered the Dutch Exclusive Economic Zone and the UK Dogger Bank and were based on 1,307 observations collected between 2006 and 2023. Biomass of all species combined was particularly high on the UK Dogger Bank, with estimates up to 100 g m⁻². In the Dutch EEZ, hotspots (<25 g m⁻²) were located west and north of the Wadden Islands, while lower biomass areas occurred in nearshore regions along the west coast of the Netherlands, the Oyster Grounds, and on the Dutch Dogger Bank. At the species level, dab and plaice showed relatively homogeneous distributions across the study area, whereas dragonet spp., scaldfish, and solenette were more broadly distributed across the entire study area. Goby spp. was restricted to the coastal areas. Sandeel spp. stood out as highly spatially heterogeneous, with locally very high biomass observations. Biomass estimates were further compared with existing trawl-based studies and were found to be at least twice as high as previously reported values. Our estimates of prey biomass may therefore reshape our current understanding of the (prey) carrying capacity of the North Sea ecosystem for predator species. Chapter 3 addresses the energetic component of the main research question by determining species-specific seasonal calorific energy densities from early autumn to early spring. Clear species-specific over-winter patterns in energy density were observed. Energy densities of dab, dragonet, plaice, and sandeel spp. declined by ~25%, sole by ~10%, while rockling and scaldfish increased by ~10%. Gobies increased from early autumn to early winter before declining, whereas solenette and whiting remained relatively stable. Spring and summer species-specific energy densities were additionally estimated for the two Ammodytes species. No significant differences were found between the two Ammodytes species, both of which showed a peak in energy density in mid-summer, representing an increase of ~40% from early spring. These results highlight the importance of accounting for seasonal variability in energy density when assessing prey quality. Species-specific end-of-winter energy densities were then applied to convert the biomass estimates from Chapter 2 into species-specific and total energy landscapes. This is important, as prey quality is even more important than prey quantity in predator foraging strategies. Chapter 4 focuses on the development of species-specific primers for real-time PCR analysis of the three most abundant sandeel species along the Dutch coast (both Ammodytes species and Hyperoplus lanceolatus). The aim was to enable reliable and cost-effective DNA-based species identification. This is necessary because morphological identification of the two sandeel species is challenging and impossible for severely damaged individuals (a substantial proportion of sandeel collected by a dredge). Yet accurate species identification is essential for understanding their species-specific ecological roles. Morphological characteristics, suitable for large sample sizes, were evaluated by comparison with DNA-based identification. DNA-based identifications were already used to analyse species-specific differences in the energy densities of the two Ammodytes species in Chapter 3. In Chapter 5, DNA-based identifications were used alongside morphologically identified species. Chapter 5 examined the distribution of both Ammodytes species in the Dutch coastal zone (<20 nm) to identify species-specific differences in spatial distribution and diurnal behaviour. We analysed seabed data collected during the annual Dutch shellfish survey in spring (2016–2018, 2020–2022). Niche segregation was assessed in relation to depth, sediment characteristics, bivalve biomass, salinity, and temporal variables (day of year and relative time of day). The species showed a largely offshore–nearshore segregation, with A. marinus dominant in offshore areas at high abundance, while A. tobianus was dominant in nearshore areas. This pattern was largely driven by differences in preferred depth ranges. However, areas where both species co-occurred were also observed, mainly in transition zones between deep and shallower regions. Even within the same sampling hauls they co-occurred. For both species, bivalve biomass was negatively associated with abundance, which suggests that dense bivalve aggregations may hinder burrowing behaviour. Because seabed sampling took place during daytime in spring—the season in which sandeel exhibit their typical diurnal foraging behaviour—we could infer species-specific diurnal behaviour from the data collected in the seabed. Both species showed similar diurnal patterns, with lower seabed densities in the morning, further reduced densities during the day, and a gradual increase towards the highest densities in the evening. Over the sampling period, seabed densities in A. marinus increased strongly, while A. tobianus showed a more parabolic pattern. The diurnal behaviour of sandeel is an important aspect of their prey availability to predator species, as many predators target sandeel either in the water column (e.g. sandwich terns) or when they are buried in the seabed (e.g. whiting, haddock, harbour porpoise). Chapter 6 provides a final discussion in which I reflect on and synthesize the insights gained throughout this thesis regarding different aspects of prey availability of small demersal fish to predator species. New research questions are addressed, including suggestions for how to link our results with (top) predators that depend on small demersal fish. This link is crucial, as it provides insight into the functioning of predator species, which are often used as indicators of the ecological state of the Dutch North Sea. A better understanding of prey–predator interactions, including the effects of human interference, can help determine whether management measures (e.g. limiting anthropogenic activities in important foraging areas when these negatively affect either prey or predators) are necessary and effective for improving the ecological status of the southern North Sea ecosystem.