An integrative study on fisheries and environmental impacts on demersal communities in flemish cap
- Autores: Krerkkrai Songin
- Directores de la Tesis: Juan Francisco Saborido Rey (dir. tes.), Graham John Pierce (dir. tes.)
- Lectura: En la Universidade de Santiago de Compostela ( España ) en 2025
- Idioma: inglés
- Tribunal Calificador de la Tesis: Alexandre Alonso-Fernández (presid.), Izaskun Preciado Ramírez (secret.), Manuel Hidalgo Roldán (voc.)
- Programa de doctorado: Programa de Doctorado en Ciencias Marinas, Tecnología y Gestión por la Universidad de A Coruña; la Universidad de Santiago de Compostela; la Universidad de Vigo; Universidade de Aveiro(Portugal); Universidade de Trás-os-Montes e Alto Douro(Portugal) y Universidade do Minho (Portugal)
- Materias:
- Enlaces
- Tesis en acceso abierto en: MINERVA
- Resumen
Fish stocks around the world are under threat from both direct and indirect human impacts. Among the most significant of these is overfishing, which has led to severe population declines and stock collapses in many regions. In addition to direct exploitation, climate change also have widespread effects on fish populations, altering life history traits, disrupting food availability, and impacting overall ecosystem dynamics. Traditional fisheries management, which often relies on single-species stock assessments and overlooks environmental influences, risks missing critical threats to fish populations and marine ecosystems. This thesis explores the combined effects of fisheries and environmental changes on demersal fish communities in the Flemish Cap, examining various aspects including abundance, distribution, growth, maturity, and broader ecological shifts.
Flemish Cap is an underwater mountain located outside the Canadian Exclusive Economic Zone (EEZ), near Newfoundland and separated from the Grand Banks by Flemish Pass. It hosts a diverse array of demersal species, including commercially important ones such as Atlantic cod (Gadus morhua), redfish (Sebastes spp.), and northern shrimp (Pandalus borealis). Historically, the high abundance of these species attracted multinational fishing fleets, leading to intense exploitation. This overfishing resulted in the collapse of the Atlantic cod stock by the mid-1990s, followed closely by the collapse of redfish. In the aftermath, fishing efforts shifted toward northern shrimp due to the decline of their predators. However, intensified shrimp fishing eventually led to the collapse of that stock as well. These sequential collapses and shifting targets clearly demonstrate that, without effective regulation, overfishing can escalate beyond individual species to impact the entire ecosystem.
In addition to fishing pressures, environmental variability and complex trophic interactions among species in the Flemish Cap have also been linked to the health of fish populations. For instance, the delayed recovery of Atlantic cod which extended the moratorium from 1999 to 2010, was believed to be at least partly due to unfavourable environmental conditions. Another example is the increased natural mortality of redfish in the late 2000s, which coincided with a rise in cod individual growth rate. These cases highlight the importance of incorporating both environmental factors and species interactions into stock assessments. A more holistic approach is essential for fisheries management that aims to sustain not only individual stocks but the broader ecosystem as a whole.
In Chapter I of this thesis, the influence of bottom temperature on the density and occurrence of various demersal species at different life stages (trophic species) was investigated. Trophic species for which there were significant relationships between occurrence and temperature share a common pattern whereby occurrence increases with temperature up to 3ºC. Trajectories at warmer temperatures diverge and can be categorized into three types. The first type shows a continuous increase with low deceleration, exemplified by longfin hake (Phycis chesteri), large Acadian redfish (S. fasciatus) and northern shrimp. The second type reveals a trajectory that plateaus, as demonstrated by marlin-spiked grenadier (Nezumia bairdii), small beaked redfish (S. mentella) and large Greenland halibut (Reinhardtius hippoglossoides). The third type features a trajectory that re-accelerates from a plateau when the temperature surpasses a certain threshold (around 5ºC), as seen in thorny skate (Amblyraja radiata), juvenile redfish, small Acadian redfish, small golden redfish (S. norvegicus) and large Atlantic cod.
Temperature effects on density varied across trophic species. Only Atlantic wolffish (Anarhichas lupus) and large cod show a continuous increase in density with temperatures warmer than 3ºC. Multiple trophic species show a clear negative effect of warming temperature on density. These include small roughhead grenadier (Macrourus berglax), large marlin-spiked grenadier, American plaice (Hippoglossoides platessoides), northern wolfish (A. denticulatus) and large spotted wolffish (A. minor). In some trophic species, a distinct peak in density can be identified within commonly observed temperature ranges. These distinct peaks appear in juvenile redfish between 4 5ºC, and golden redfish between 3 4ºC.
In Chapter III, we found that, in addition to temperature, other environmental factors including salinity and dissolved oxygen also influenced Atlantic cod across all size classes. Chlorophyll concentration appeared to also have a minor effect on smaller individuals. These environmental influences were most pronounced during periods when the cod stock was particularly vulnerable, whereas their effects were less evident when stock abundance was high. A prolonged period of low bottom temperatures in the early to mid-2000s may have contributed to the delayed recovery of the cod population. In contrast, a notable increase in both abundance and spatial distribution became evident in the late 2000s, coinciding with a period of rising temperatures.
Several demersal species exhibited density-dependent distribution patterns, with their occupied ranges contracting during periods of low abundance and expanding only when stocks recovered. For Atlantic cod, the distribution was largely confined to the summit of Flemish Cap from the mid-1990s to the late 2000s, with expansion into deeper areas occurring only as the population rebounded. In contrast, species such as wolffishes were observed primarily in the deepest parts of their distribution during periods of low abundance. These patterns highlight the importance of incorporating spatial dynamics into fisheries management and suggest that preserving potential refuge areas could be critical for the conservation of these species.
Strong negative correlations were observed between Atlantic cod and its competitors, both in terms of abundance and spatial distribution. A notable shift was seen in the distribution of large Greenland halibut, a species typically found in deeper waters. Following the collapse of the cod population, halibut densities increased markedly in shallower regions, contributing to significant changes in the overall assemblage structure. In contrast, species such as American plaice, which are neither direct competitors nor prey of cod may have benefited from high cod abundance, potentially due to cod deterring their predators. However, it is important to recognize that some of the observed correlations may reflect species' responses to changes in bottom temperature or other unmeasured environmental variables.
For cod's primary prey species, northern shrimp and small redfish showed negative correlations with cod, which can likely be attributed to their direct predator-prey relationship. In contrast, large redfish exhibited either insignificant or positive correlations with cod. The positive correlations between large redfish and cod may reflect similar fishing pressures on both species, as their collapses and subsequent recoveries occurred during similar timeframes. These overlapping trends could obscure the trophic interactions between the two species, making it difficult to isolate the effects of their predator-prey relationship.
In conclusion, Chapters II and III highlight the significant influence of both fisheries and environmental factors on demersal communities. Environmental impacts were particularly pronounced during periods of overfishing, which could hinder stock recovery in unfavourable conditions. As a keystone species, the fisheries and environmental pressures on cod can have unforeseen cascading effects throughout the entire ecosystem. Effective management strategies must incorporate environmental factors and trophic relationships, especially in the context of climate change, to prevent ecological damage that may arise from overfishing a single species.
This thesis also examines the effects of temperature on two critical biological processes in fish: growth (Chapter IV) and maturity (Chapter V). The study focused on species selected for their commercial importance and the availability of biological data. These species include Atlantic cod, American plaice, Greenland halibut, roughhead grenadier, Acadian redfish, beaked redfish, and golden redfish. Analyses were conducted separately for females and males to account for potential sex-specific differences.
Before analysing the effects of temperature on life history traits, long-term data revealed clear temporal trends in maximum age for several species. For Atlantic cod and American plaice, there was a marked increase in maximum age and size from the mid-1990s to the mid-2010s. This trend likely reflects reduced fishing pressure following the severe depletion of cod stocks in the late 1980s and early 1990s, which allowed more individuals to survive longer and attain larger sizes. In contrast, beaked redfish exhibited a declining trend in both maximum age and length from the mid-1990s to the late 2000s, despite a sharp drop in fishing mortality in 1997 that remained low thereafter. This opposing trend, relative to cod, could be explained by their predator-prey relationship, suggesting that changes in predation pressure can counteract the expected effects of reduced fishing.
To investigate the effects of temperature on growth, fish size-at-age data were analysed using Generalized Additive Mixed Models (GAMMs), incorporating bottom temperature as a predictor. Size-at-age predictions from the best-fit GAMMs in 3, 3.5, 4, 4.5 and 5ºC were subsequently fitted to the von Bertalanffy Growth Function (VBGF) to estimate growth parameters. The results indicated temperature-related effects on growth for all studied species except American plaice. In general, increasing bottom temperature was associated with a decrease in asymptotic length (L). For some species-sex combinations, growth rate (k) increased with temperature. VBGF parameters were then used to predict size at maximum age (Lmax) across five temperatures. Most species and sexes showed a decline in Lmax at higher temperatures, with beaked redfish of both sexes exhibiting the most pronounced reductions. Interestingly, for male Atlantic cod, although L decreased at warmer temperatures, Lmax increased. This was attributed to a substantial rise in k, which accelerated early growth, allowing individuals to reach larger sizes at younger ages. While size differences diminished in later life stages, Lmax remained slightly higher under warmer conditions.
Male Greenland halibut, male Acadian redfish, and female golden redfish exhibited opposing trends for which L increase and k decrease at higher temperature. However, for female golden redfish, the von Bertalanffy growth trajectories were nearly identical across all temperatures, indicating minimal temperature driven changes in growth. More noticeable differences were observed in male Greenland halibut and male Acadian redfish. Still, it is important to note that the 95% confidence intervals (CIs) of their GAMM predictions overlapped considerably across all temperatures, suggesting limited differences in these apparent trends. These results may indicate that the temperatures analysed in this study remained within the optimal range for these species-sex combinations.
The observed trend of faster growth rates at higher temperatures in many species is concerning, as higher k typically associated with shorter lifespans and increased natural mortality. Among the species examined, Atlantic cod showed the most pronounced shift toward a higher k. Notably, the projected increase in Lmax for male cod with warming was based on a constant maximum age across all temperatures. In reality, a potential temperature driven reduction in lifespan could offset this increase, ultimately leading to a lower Lmax. Declines in fish size, coupled with elevated natural mortality, may have serious consequences for fisheries. Populations composed of smaller individuals, even at similar abundances, contribute less total biomass and amount that could be sustainably fished. Furthermore, increased natural mortality may weaken stock resilience to fishing pressure. Beyond fisheries, uneven growth responses among species may have broader ecological consequences. Species that grow faster at higher temperatures would attain larger sizes earlier in life, potentially gaining a competitive advantage. In Flemish Cap, for instance, increased growth in cod has already been suspected to contribute to elevated natural mortality in redfish.
In addition to growth, analysis using Generalized Linear Mixed Models (GLMMs) revealed significant temperature effects on maturity in several species. The size-at-maturity analysis showed that the best-fit models included temperature as a predictor for females of all species except American plaice. For males, significant temperature effects were detected only in Atlantic cod and golden redfish. In the age-at-maturity analysis, temperature influenced maturity in both sexes of Atlantic cod, roughhead grenadier, and golden redfish. Acadian and beaked redfish showed significant temperature effects only in females, while American plaice and Greenland halibut exhibited no detectable temperature influence on age-at-maturity for either sex.
A general trend of decreasing size at 50% maturity (L50) with increasing temperature was observed in females of most species, including Greenland halibut, roughhead grenadier, and all redfish species. In addition, female roughhead grenadier and female redfish also exhibited a decrease in age at 50% maturity (A50) at higher temperatures. However, some species-sex combinations showed the opposite pattern. For instance, both sexes of Atlantic cod and male golden redfish exhibited an increase in L50 with rising temperature. These groups, along with male roughhead grenadier, also showed an increase in A50 under warmer conditions. Consistent with the growth modelling results, American plaice was the only species in which temperature had no significant effect on maturity in either sex.
The best-fit GLMMs were used to predict size and age at maturity across different temperatures. The limitations of model predictability were most evident for female roughhead grenadier and male golden redfish, as reflected in their wide 95% CIs, which overlapped substantially across temperature ranges. Excluding these high-uncertainty cases, temperature increases from 3ºC to 4.5ºC resulted in notable changes in maturation patterns. L50 decreased by as much as 18% in female golden redfish, while increasing by up to 18% in female Atlantic cod. Female golden redfish also showed the most substantial decline in A50, maturing approximately three years earlier at higher temperatures. In contrast, Acadian redfish exhibited the smallest shifts in maturation, with an 8% reduction in L50 and less than 1 year decrease in A50.
The earlier onset of sexual maturation in warmer waters, observed in females of most species, may interfere with other critical biological processes, particularly growth. Many of the species and sexes that exhibited reduced size and/or age-at-maturity under higher temperatures also showed smaller maximum sizes. This pattern was especially pronounced in female beaked redfish. In many fish species, smaller mature females are known to be disproportionately less fecund than their larger counterparts, which can have significant implications for reproductive output. The reduction of both size-at-maturity and maximum size is a concerning trend, as it may lead to decreased population productivity. This is particularly critical for species like redfish, which are subject to both predation (e.g., by Atlantic cod and Greenland halibut) and fishing pressure. A warming-induced decline in reproductive potential could increase their vulnerability and compromise the resilience of their populations.
In Chapter VI, outputs from the distribution, growth, and maturity models were integrated into a multispecies ecological model, OSMOSE (Object-oriented Simulator of Marine Ecosystems). The model was calibrated using ten years of biomass data (2010 2019) for 14 demersal species, along with catch records for seven of those species over the same period. Once calibrated, the model was employed to simulate ecosystem dynamics under three temperature scenarios (3.5ºC, 4.0ºC, and 4.5ºC), by modifying species-specific growth and maturity parameters. The simulations focused on seven key species: Atlantic cod, American plaice, Greenland halibut, roughhead grenadier, and three species of redfish.
The simulations revealed an overall decline in the combined biomass of the main species as temperature increased from 3.5ºC to 4.5ºC, although individual species responded differently. Atlantic cod biomass remained stable between 3.5ºC and 4.0ºC but showed a notable increase at 4.5ºC. Greenland halibut experienced a substantial biomass increase between 3.5ºC and 4.0ºC, followed by a moderate gain at 4.5ºC. In contrast, redfish biomass remained relatively unchanged from 3.5ºC to 4.0ºC but declined sharply at 4.5ºC. American plaice exhibited a consistent decline, losing over 90% of its biomass by 4.5ºC. Roughhead grenadier biomass initially increased from 3.5ºC to 4.0ºC but declined at 4.5ºC to levels below those at 3.5ºC. Simulated catch patterns closely mirrored the biomass trends across temperature scenarios.
The simulations indicated notable shifts in trophic structure as temperature increased from 3.5ºC to 4.5ºC. Atlantic cod became significantly more dominant within the ecosystem, intensifying their predation pressure on other species. This increased dominance could threaten prey populations and reduce food availability for cod competitors. Additionally, the trophic levels of several species declined slightly at higher temperatures, largely due to reduced availability of high-trophic prey such as redfish. An ecosystem where one species benefits disproportionately from environmental change, while others struggle to adapt, may become more sensitive to external stressors, including fishing pressure and further environmental fluctuations. Notably, the availability of background low-trophic preys, phytoplankton and zooplankton were held constant across all temperature scenarios. If these foundational resources were to be negatively impacted by warming, the consequences for the ecosystem could be significantly more severe The simulation results also revealed increased predation-driven natural mortality for key prey species such as redfish at higher temperatures. In single-species stock assessments, commonly used in Flemish Cap and globally, natural mortality rates are often arbitrarily assigned and treated as constant. This assumption can lead to significant underestimation of natural mortality, particularly under changing environmental conditions. Consequently, stock resilience to fishing pressure may be overestimated. In the simulations, fishing mortality rates were held constant across all temperature scenarios. In practice, fisheries management may need to reduce fishing mortality for species negatively impacted by rising temperatures. While such measures would result in lower catches, they could prove essential to preserve population and safeguard overall ecosystem stability.
In conclusion, the findings across all chapters demonstrate that both fisheries and environmental conditions can profoundly affect fish populations in diverse and interconnected ways. The same environmental variability can have vastly different impacts across species, and changes in the abundance or life history of key species can reverberate throughout the entire ecosystem. Relying solely on single-species assessments and management, without fully accounting for trophic relationships, risks triggering stock collapses and ecosystem alteration such as those previously witnessed in the Flemish Cap. Population recovery, if it occurs, can take decades and it needs the alignment of multiple favourable conditions. To prevent the recurrence of stock collapses and to protect demersal communities, the development and application of multispecies assessment approaches are essential. Modelling tools that incorporate interspecific interactions and environmental drivers should be, at a minimum, integrated as complementary components of stock assessments and fisheries management. Climate change is already underway and unlikely to reverse in the near future. Proactive strategies are urgently needed to mitigate its effects or, at the very least, to prepare for the ecological transformations that lie ahead.
