National Electrofishing Programme for Scotland (NEPS) 2023: status of juvenile Atlantic salmon and brown trout populations

Executive summary

Atlantic salmon and brown trout are iconic species that are important to the rural economy, biodiversity and cultural heritage of Scotland and valuable indicators of overall ecosystem health. However, wild salmonid populations are in decline across much of their native range. Reliable methods are therefore required to understand the health (status) of populations, identify issues requiring management (pressures), inform regulation and policy development, and meet legislative reporting requirements.

In 2014 a major review of freshwater fisheries management in Scotland provided the catalyst for the development of new juvenile salmonid assessment methods. Electrofishing data are one of the most commonly collected sources of information on fish populations. However, their analysis and interpretation can be challenging, requiring robust sampling methods, an appropriate survey design and models of expected densities for healthy populations that vary among habitat types. In 2018 the National Electrofishing Programme for Scotland (NEPS) was established providing the first ever national scale statistical survey of Scotland’s salmon rivers. A new assessment benchmark for salmon was published in 2019, immediately followed by the introduction of a new juvenile salmon assessment method. A third NEPS sampling programme was undertaken in 2021 employing a revised survey design.

This report presents data and analyses from a new survey design (NEPS 2023). It extends the available data on salmon and provides the first ever national assessment of the status of trout. Comparisons are also made with previous NEPS surveys through careful harmonisation of data (aggregation and post-stratification) to account for changing sample frames (rivers in the survey design) and strata (base reporting units for the surveys). At the national scale, the mean density of salmon fry (0+) in 2023 was only 74% of the benchmark (expected density for healthy populations). When compared across years the mean salmon fry density for 2023 was also lower than 2018 or 2021. The mean density of salmon parr (≥1+) for 2023 was only around 56% the benchmark, and was also the lowest recorded across all four NEPS surveys. Although there was substantial regional variability in performance, healthy (overall Grade 1) salmon populations were only found from the Moray Firth north. The national mean density estimate for trout fry in 2023 was 86% of the benchmark. Having accounted for differences in survey designs between years, trout fry densities in 2023 were higher than 2021 but broadly comparable to 2018. The national mean parr density estimate for 2023 was only 64% of the benchmark. Densities were slightly higher than 2019 and 2021, but remained lower than 2018. At a regional scale, there were only three overall Grade 1 trout populations in 2023, the Spey, Ness and Caithness.

Juvenile salmon and trout densities were compared to lagged rod catches (as a proxy of adult abundance) for salmon and sea trout to explore consistency in different abundance metrics and assess evidence of stock-recruitment (S-R) relationships that could provide independent validation of NEPS Grades. Consideration of NEPS Grades was further supported by assessment of long-term temporal trends in juvenile salmon densities in three rivers with contrasting NEPS Grade profiles. There was evidence of stock-recruitment relationships for salmon and trout, with Grade 1 densities being broadly consistent with maximum juvenile production. There was evidence of significant long-term trends in juvenile densities in the Tweed and Dee, but not in the Spey which was predominantly classified as Grade 1 across NEPS surveys. This suggests that NEPS Grade 1 is broadly consistent with a conservative estimate of maximum freshwater production.

Relationships between mean density and occupancy (the proportion of the river length occupied by fish) were examined at a regional scale to assess the likely consequences of declining populations for habitat use and assessment. Occupancy declined as regional densities declined, with much of the response observed at intermediate densities. Occupancy therefore provides a metric of population status that could be complimentary to mean abundance. However, this would require the development of a new benchmark model to define expected occupancy under healthy conditions in different habitats. As occupancy declines, relative inter-site differences in density also increase. This can make it difficult to precisely estimate population sizes with consequences for confidence of assessments. It is unclear whether population declines will be manifested as proportional reductions in density across the river network or the concentration of fish within the highest quality habitats at the core of the network. This has important consequences for interpreting ad-hoc data.

Water quality data continued to show broadly consistent spatial patterns of variability reflecting natural variability in geology and soils and human impacts. Incorporation of these data in future models would improve the precision of assessments for individual sites and assessment of the effects of water quality pressures.

NEPS provides a scientifically robust, catch independent, sampling, data storage and analytical framework for the collection of data on the status of freshwater fish populations. This can support numerous policy, management and reporting requirements from a single endeavour, consistent with the ethos of “collect once, use many times”. However, it is important that future reporting requirements are fully documented and understood before further development work is undertaken given the analytical challenges involved. Depending on resource, the NEPS survey design is capable of providing full national coverage and assessment, or of supporting more modest regional data collection initiatives while maintaining scientific rigour that would allow data to be analysed together at larger scales as required. Future application of NEPS will therefore be necessarily defined by reporting requirements and the availability of sampling resources.