Essential Fish Habitats (EFH), i.e. "those waters and substrata necessary to fish for spawning, breeding, feeding, or growth to maturity", are amongst the marine assets for which uncertainty exists regarding their spatial distribution. The Scottish Government commissioned this project with the aim of developing the evidence base on EFH to inform future planning and project level assessments in Scottish waters, with consideration to providing outputs that can also assist with wider UK planning and project assessments. The results of this study will help the regulators to understand where further work is required to inform consenting decisions.
EFH in this project were identified as those habitats that function as refuge for individuals of a species, as nursery for its juvenile stages or as spawning ground for the spawning adults and the spawn products (eggs) ('functional habitats'), and where these individuals are present in higher abundance (aggregations) compared to other habitats where they also occur. Two separate approaches were applied to the EFH assessment in offshore and inshore waters (distinguished by the 12 nautical miles limit): EFH data-based modelling and a semi-quantitative assessment (habitat proxy), respectively. The lower data availability in inshore waters (due to higher survey fragmentation, lower method standardisation and coverage gaps) prevented the application of EFH modelling in this region.
In offshore waters, Decision Tree models were calibrated based on catch data from broad- scale fish surveys (within the selected study period 2010 – 2020) and the associated environmental conditions (derived from publicly available data layers). These EFH models identified the set of environmental conditions where aggregations of a species/life stage were more likely to occur. Data layers for the environmental predictors of the model (derived as mean environmental conditions over the study period) were then used to extrapolate and map the UK-wide spatial distribution of the species/life stage aggregations, considered as indicative of EFH. A confidence value was calculated for each EFH model as a whole and its spatial prediction by combining model performance (statistical confidence) and confidence estimates for the input data used to calibrate and predict the model. The spatial output provided the integrated results on the prediction of the distribution of EFH and the associated confidence.
In inshore waters, habitat proxies for species/life stages that may have their EFH inshore were identified and mapped based on data layers for pre-defined habitat types (EUNIS habitat classification). A literature review on the species-habitat associations relevant to key EFH functions, and expert assessment by the project team and consulted external experts were used to score the habitat types according to their ability to provide EFH function to a species. As the evidence used for this assessment was most often based on mere species- habitat association, rarely accounting for abundance, the habitat proxies thus identified were used as indicators of the wider 'functional habitat' of a species, which is also likely to include EFH. A confidence level (low/medium/high) was allocated to the habitat scoring
based on the expert assessment of the supporting evidence/knowledge. Due to time limitations in the project, habitat proxy maps were produced for the west coast of Scotland for most of species as a demonstration site for the application of the approach and to ensure that the detail of biotopes at the EUNIS habitat Level 4 could be viewed at the appropriate scale. Only for herring spawning grounds, habitat proxy maps were produced covering all Scottish inshore waters. The spatial outputs integrated the information about the habitat scores and the confidence level.
Details on the methods used for both EFH models and habitat proxy assessments (including the confidence assessment) are in the main report.
EFH models were developed for 16 species/life stages in offshore waters, including: lesser sandeel and Norway lobster refugia; juveniles of plaice, lemon sole, common sole, anglerfish, whiting, blue whiting, hake, sprat, mackerel, and long finned squid; spawning adults of whiting, cod, Norway pout; and mackerel eggs. Habitat proxies were assessed for 19 species/life stages in inshore waters, including: small sandeel refugia; spawning herring; juvenile plaice, common sole, whiting, cod, saithe, sprat, spotted ray, European lobster, and brown crab; spawning grounds/egg nursery for thornback ray and spurdog; generic habitat for velvet crab, common cockle, dog cockle, razor clam, common whelk, and dog whelk.
Some species, which were initially considered, could not be modelled (offshore) or assessed (inshore) due to insufficient data and/or evidence on the species-habitat association from the literature obtained in the project (common skate and sandy ray, inshore and offshore; queen and king scallops and surf clam, offshore).
The spatial products thus obtained were validated by comparison with additional actual observations of the species/life stages in survey catches, where these data could be obtained. The maps were also shared with the Project Steering Group and additional external experts for review to obtain feedback on possible EFH areas that were not adequately represented by the model maps. These additional validation results were graphically incorporated into the final spatial products, of which they constitute an integral part, along with the model/assessment predictions and confidence. However, due to time limitations in the project, the feedback received was often limited and in some cases validation survey data and stakeholder feedback could not be obtained (e.g. for most of the inshore habitat proxy maps). A summary of the results and validation of the individual spatial outputs is given in section 3.1.30.
The spatial products of this study represent a step forward for an evidence-based understanding of the EFH distribution in UK inshore and offshore waters and the confidence associated with it. However, further validation is required before they can be used in assessments (as discussed in sections 4.4, 4.5 and 4.7). Further engagement with stakeholders from the fishery industry would be particularly valuable in supporting this.
This study also allowed to identify important data and knowledge gaps that need further work, including the species mentioned above and for which the EFH could not be modelled,
as well as other species for which the obtained results had a low confidence due to poor available knowledge for inshore waters (thornback ray, spotted ray, spurdog, sprat, common sole, saithe, cod, whiting, European lobster, brown and velvet crabs). Data gaps were also identified for shellfish inshore and offshore, and fish and habitat maps inshore. These gaps were relevant to both the calibration and application of EFH models, as well as the validation of both EFH model and habitat proxy maps. As some of these gaps were due to the time-limited data collation in this project, possible additional data sources that could not be obtained in this project but that could help in filling these gaps were identified.
Based on the limitations identified for the different spatial outputs produced in this study (e.g. factors affecting model confidence, spatial gaps in the map/data coverage, etc), and the data and knowledge gaps mentioned above, recommendations were made to provide more robust and validated models (see section 4.7). In summary, the recommendations were around:
A. The correct use of the overall (integrated) spatial outputs produced in this project.
B. The improvement of the overall (integrated) spatial outputs produced in this project.
C. The improvement of EFH models and their prediction maps.
D. Improvement of habitat proxy assessment and the resulting maps.
Finally, in light of the possible influence of climate change on the distribution of fish and shellfish populations (through changes in the environmental conditions and effects on the fish physiology, behaviour etc.), the sensitivity of EFH model predictions to changes in environmental predictors that are of the same scale and direction as those from climate predictions was explored. Although this was not by any mean an accurate assessment of climate change effects, it showed that these may potentially have variable influence on the distribution of EFH in Scottish waters, depending on the species, the environmental variable subject to change and its importance as environmental predictor of the species' EFH. The appropriate interpretation and use of results are further discussed in the report, where also the general implications of climate change for EFH distribution in inshore and offshore Scottish waters are briefly reviewed and discussed (see sections 2.5, 3.2 and 4.6).