Refining monitoring practices and applying novel enhancement strategies to populations of threatened diadromous species.

Abstract

Diadromous species are those which migrate between marine and freshwater environments. These can be sub-divided into anadromous (breed in freshwater but migrate to the sea to attain maturity, such as Atlantic salmon, Salmo salar) and catadromous species (breed in the sea but mature in freshwater, such as European eel, Anguilla anguilla). High socioeconomic importance has led to both species listed being subject to widespread and often large-scale commercial and recreational fisheries, with this driving extensive population monitoring and management. However, overexploitation and other anthropogenic impacts mean both species are threatened and/or declining throughout their range. Population monitoring for salmonids generally consists of electrofishing surveys to determine juvenile abundance, where accurate population estimates are critical for conservation target-setting. However, the accuracy of some current population monitoring practices remains untested, with many studies also showing current management techniques to be unproductive, especially enhancing salmonid populations with hatchery reared fish. These two issues are addressed in this research. To evaluate the accuracy of current population monitoring methods, a common population monitoring technique for juvenile salmonids, time-delineated single-pass electrofishing surveys, was tested against a more precise monitoring method (area- delineated single-pass electrofishing surveys), using Atlantic salmon as the focal species. Comparison of these two methods showed a high degree of density-dependent error in abundance estimates from time-delineated surveys. Further analysis showed that the area covered during time-delineated electrofishing surveys is highly variable, leading to imprecision around abundance estimates. The application of time-delineated single-pass electrofishing was then reviewed by analysis of an electrofishing dataset with over 11,000 observations, used to monitor Atlantic salmon fry (age 0+) abundances. Currently these surveys are carried out over a three month summer sampling period. Linear models revealed this is leading to substantial under- or overestimation of salmonid abundances by not accounting for high mortality experienced by salmonid populations during this time, with model predictions showing a decline in fry numbers of 35 % across this sampling window. With mounting scientific evidence showing salmonid stocking is ineffective at enhancing existing populations, alternative approaches are needed. This research tested the application of a novel management tool (intra-catchment translocation) on salmonids, using Atlantic salmon as the model species. Additionally, given the potential broader applicability of this management technique and need to conserve other diadromous species, this research also demonstrated the efficacy of a similar technique (inter- catchment translocation) on European eels. Using an existing dataset from a translocation experiment, the effectiveness of translocations in enabling critically endangered European eels to colonise headwater stream habitats was tested. Juvenile elvers were released into 13 streams in Wales under two treatments: (1) streams populated by brown trout (Salmo trutta), and; (2) brown trout-free streams. Electrofishing surveys conducted three months post-release showed that elvers successfully colonised all streams but that trout presence reduced eel recapture rates by approximately 70 % and predicted eel survival from 4 to 2 %, with this primarily related to the density of 2+ age class trout, most likely via predation. To test the efficacy of intra-catchment translocations on Atlantic salmon, a series of large- scale field experiments were conducted across three years within the River Erriff catchment, Ireland. Atlantic salmon undergo a period of density-dependent mortality immediately following their larval emergence, leading to high mortality rates at sites with an abundance of spawning gravel. These experiments tested whether freshwater production of Atlantic salmon may be increased by removing surplus fry from such sites and releasing these fish into areas of lower fry density, hence reducing the level of density-dependent mortality. The methods employed for these experiments underwent significant refinements throughout the study period, with the results of the first two experiments considered together and the third experiment considered separately. During years 1 and 2, post-emergent fry densities were recorded through area-delineated single pass electrofishing surveys. Fish were then captured from high fry density sites, adipose fin clipped to enable identification, and released at low density sites. These two treatments were compared to high- and low density control sites where no salmon fry were removed or added. Electrofishing surveys were carried out at all sites in autumn to determine autumn fry densities, with habitat variables recorded and accounted for within analyses. Testing of the efficacy of the translocations revealed no significant effect on autumn fry densities at sites where fry were removed or released. However, as sites which received additional fry were relatively small (20 m instream length), the study design was highly vulnerable to the effects of fry dispersal, with electrofishing surveys of suitable habitats upstream and downstream of fry release sites confirming the presence of translocated fish. Substantial refinements to the experimental design were made for year 3. To account for dispersal, fry were released at the centre of two larger stream reaches, approximately 500 m long, with approximately 1500 fry released per reach. These were compared against two stream reaches which received no salmon fry. Post-emergent fry densities were recorded through electrofishing surveys at all high- and low density sites prior to translocations, with low density stream reaches split into individual sites based on habitat units (i.e. pools and riffles). Electrofishing surveys were conducted at all sites in September to determine the effect of translocations on the two treatments versus their controls, with habitat variables used to account for additional variation in fry densities. Testing revealed that removal of post-emergent fry from high density sites had no significant effect on autumn fry densities. Additionally, translocated fry densities were significantly and positively correlated with total density of autumn salmon fry. Translocated fry had a small negative impact on population depletion of native salmon, which was offset by the increase in fry densities from translocations. Moreover, translocated fry densities had no effect on native fry lengths in autumn. Thus, translocation shows promise as a management tool to increase freshwater production of Atlantic salmon. Considered together, these findings highlight the need to refine current monitoring practices for salmonids and to further test novel management strategies on salmonids and other diadromous species, ensuring that viable management techniques are available to policy makers and fisheries managers, and can be applied based on robust abundance data.