Abstract
In the 2020 assessments of Atlantic cod (Gadus morhua) and European plaice (Pleuronectes platessa) stocks from the western Baltic Sea, a marked reduction in numbers of fish caught in the standard survey was observed. Survey reports indicated that this was probably due to widespread hypoxia. Thus, this project HypCatch, was initiated to investigate the effects of hypoxia on cod and plaice from these stocks and to further investigate the observed impacts on fish distribution, the potential mechanisms of impact on the population and small scale fisher’s perceptions of how hypoxia in the south-western Baltic Sea impacts upon their decisions, effort and catches.
We describe how different levels of hypoxia severity in the bottom-waters elicit different responses from cod and plaice, in terms of their catchability in the scientific surveys. Cod catchability decreases at moderate levels of hypoxia (between two and four milligrams of oxygen per litre) before rapidly declining at acute levels of hypoxia (below two milligrams of oxygen per litre). Similarly, plaice catch levels decline rapidly under acute hypoxia, however, they appear unresponsive to moderate hypoxia. This finding in the offshore survey was mirrored by increasing (albeit, non-significant) catch rates in coastal monitoring with increasing extents of moderate (cod only) and acute (cod and plaice) hypoxia across the study area. In spite of these responses, we found that accounting for oxygen in the modelled stock indices modified these indices very little, and the subsequent stock assessment outcomes even less. Hence, our recommendation for these stocks is not to include oxygen as a mediating environmental variable in the production of the survey abundance indices.
While oxygen conditions have a minimal effect on the calculation of stock indices and their subsequent use in assessment, the wide spread nature of moderate hypoxia that we document, is likely to have an effect on stock productivity via many potential mechanisms. This report’s subsequent chapters describe how fish and fishers’ distributions change in response to hypoxia.
First, a more realistic species distribution modelling framework is applied to cod and plaice in the inner Danish waters. This approach, called a soap-film smoother accounts for the complex landmasses and coastline that pepper the inner Danish waters. We document how the models we produce better predict species distributions in space and in response to hypoxic waters, than more traditional approaches.
Similarly, we describe the results of a short survey of small-scale, commercial gill-netters and their interactions with hypoxia. While only 30% of these fishers consider hypoxia in their fishing decisions, nearly 80% of fishers who experienced hypoxic waters moved their gears shallower and closer to land in response. Of those fishers in ICES sub-division 22 whose fishing was impacted by hypoxia in 2019, all noted that there were fewer fish in catches.
While we are currently unable to quantify if such responses to hypoxia are general trends across all small-scale fisheries in the area, this report documents a new, greatly improved method for attributing small scale fisher effort in space, using AIS data via a random forest modelling approach. These methods can be applied in the future to extract catch rates for various species in different areas and in relation to hypoxic waters, to strengthen further the circumstantial evidence we have of cod and plaice using coastal waters as a refuge from basin-scale hypoxia.
From a collaboration between this project and another, SEAwise, we investigated hypothetical scenarios of impact from hypoxia in the deep basin spawning grounds and in the coastal juvenile habitats and the relative effect these different mechanisms of action have on population productivity. We found that juvenile habitats act as a filter determining future class strength, but that adults aggregating in areas of hypoxia for spawning, are the greatest risk for loss of population productivity. Furthermore, we found that there is a larger scope for the positive effects coming from restoration (reducing hypoxia) than there is for continued negative effects of broad-scale hypoxia.
Finally, we describe how all of these findings were distributed and disseminated amongst working groups with both advisory and scientific tasks.
We describe how different levels of hypoxia severity in the bottom-waters elicit different responses from cod and plaice, in terms of their catchability in the scientific surveys. Cod catchability decreases at moderate levels of hypoxia (between two and four milligrams of oxygen per litre) before rapidly declining at acute levels of hypoxia (below two milligrams of oxygen per litre). Similarly, plaice catch levels decline rapidly under acute hypoxia, however, they appear unresponsive to moderate hypoxia. This finding in the offshore survey was mirrored by increasing (albeit, non-significant) catch rates in coastal monitoring with increasing extents of moderate (cod only) and acute (cod and plaice) hypoxia across the study area. In spite of these responses, we found that accounting for oxygen in the modelled stock indices modified these indices very little, and the subsequent stock assessment outcomes even less. Hence, our recommendation for these stocks is not to include oxygen as a mediating environmental variable in the production of the survey abundance indices.
While oxygen conditions have a minimal effect on the calculation of stock indices and their subsequent use in assessment, the wide spread nature of moderate hypoxia that we document, is likely to have an effect on stock productivity via many potential mechanisms. This report’s subsequent chapters describe how fish and fishers’ distributions change in response to hypoxia.
First, a more realistic species distribution modelling framework is applied to cod and plaice in the inner Danish waters. This approach, called a soap-film smoother accounts for the complex landmasses and coastline that pepper the inner Danish waters. We document how the models we produce better predict species distributions in space and in response to hypoxic waters, than more traditional approaches.
Similarly, we describe the results of a short survey of small-scale, commercial gill-netters and their interactions with hypoxia. While only 30% of these fishers consider hypoxia in their fishing decisions, nearly 80% of fishers who experienced hypoxic waters moved their gears shallower and closer to land in response. Of those fishers in ICES sub-division 22 whose fishing was impacted by hypoxia in 2019, all noted that there were fewer fish in catches.
While we are currently unable to quantify if such responses to hypoxia are general trends across all small-scale fisheries in the area, this report documents a new, greatly improved method for attributing small scale fisher effort in space, using AIS data via a random forest modelling approach. These methods can be applied in the future to extract catch rates for various species in different areas and in relation to hypoxic waters, to strengthen further the circumstantial evidence we have of cod and plaice using coastal waters as a refuge from basin-scale hypoxia.
From a collaboration between this project and another, SEAwise, we investigated hypothetical scenarios of impact from hypoxia in the deep basin spawning grounds and in the coastal juvenile habitats and the relative effect these different mechanisms of action have on population productivity. We found that juvenile habitats act as a filter determining future class strength, but that adults aggregating in areas of hypoxia for spawning, are the greatest risk for loss of population productivity. Furthermore, we found that there is a larger scope for the positive effects coming from restoration (reducing hypoxia) than there is for continued negative effects of broad-scale hypoxia.
Finally, we describe how all of these findings were distributed and disseminated amongst working groups with both advisory and scientific tasks.
| Original language | English |
|---|
| Place of Publication | Kgs. Lyngby, Denmark |
|---|---|
| Publisher | DTU Aqua |
| Number of pages | 56 |
| ISBN (Electronic) | 978-87-7481-369-9 |
| DOIs | |
| Publication status | Published - 2023 |
| Series | DTU Aqua-rapport |
|---|---|
| Number | 436-2023 |
| ISSN | 1395-8216 |