READ ME File For 'DATASET’ for "Quantification of the Response of Brown Trout (Salmo trutta) to Habitat Modification by Reintroduced Eurasian Beaver (Castor fiber): Implications for River Management in Great Britain"

Dataset DOI: https://doi.org/10.5258/SOTON/D3116

ReadMe Author: Robert J. Needham, University of Southampton ORCID ID - 0000-0003-4639-6031

This dataset supports the thesis entitled "Quantification of the Response of Brown Trout (Salmo trutta) to Habitat Modification by Reintroduced Eurasian Beaver (Castor fiber): Implications for River Management in Great Britain"
AWARDED BY: University of Southampton
DATE OF AWARD: [2024]

DESCRIPTION OF THE DATA
All data are presented in excel files, collected using electrofishing surveys and subsequent biometric measurements, Passive Integrated Transponders Telemetry studies and remote camera monitoring. Water temperature data was collected using OTT water level loggers. Rainfall data was obtained from the Lentran weather station from the MetOffice The excel files provided contains all data used for generating the following figures. 

The figures are as follows:

Figure 5.3 Mean monthly temperatures for the modified (top) and control (bottom) streams, averaged every 15 mins. Solid and upper/lower dotted lines represent mean and maximum/minimum temperatures, respectively, while the dashed line denotes the ambient air temperature. The modified stream temperature represents the mean of data provided by three loggers. The control stream temperature is based on outputs of a single data logger. Ambient air temperatures were obtained from the Met Office from Lentran Weather Station located ca. 6 km NE of field site.	

Figure 5.4 Mean ± SD brown trout abundance (trout m⁻1) during spring (black), summer (grey) and autumn (white) 2016 in the beaver modified and control streams with all size class of trout included (A) and (B) fry ≤ 30 mm and parr 31-60 mm omitted.	

Figure 5.5 Mean ± SD brown trout density (m⁻2) during spring (black), summer (grey) and autumn (white) 2016 in the modified and control streams with all size class of trout included (A) and fry ≤ 30 mm and parr 31-60 mm omitted (B).	

Figure 5.6 The distribution of length class frequency between the modified (white bars) and control (black bars) streams in spring (2015 and 2016), summer (2016) and autumn (2014, 2015 and 2016). Seasonal data has been pooled for the years.	

Figure 5.7 Fork length of trout captured in the modified ‘Mod’ (white boxes) and control ‘Con’ (black boxes) streams during autumn, spring and summer 2014, 2015 and 2016. The box plots illustrate the median (horizontal line), interquartile range (boxes) and overall range up to 1.5 times the interquartile range (whiskers). All outliers are depicted (clear circles).	

Figure 5.8 Differences in mass (g) of (A) adult trout (FL 61-121 mm) and (B) parr (FL 31-60 mm) between the modified (white boxes) and control (black boxes) streams. The box plots illustrate the median (horizontal line), interquartile range (boxes) and overall range up to 1.5 times the interquartile range (whiskers). All outliers are depicted (clear circles).	

Figure 5.9 The linear relationship between growth in mass (g) and FL (mm) of trout captured in the modified stream. Grey shading indicates 95% confidence intervals.	

Figure 5.10 Relationship between observed seasonal growth rates of brown trout inhabiting a beaver modified stream and that predicted by an optimal growth model that assumes fish are fed to satiation under laboratory conditions (Elliott et al., 1995).  ✕ - winter 2014/2015 (October, November and December 2014 - March/April 2015 [n = 16]), ◯ - spring/summer 2015 (March/April 2015 - October 2015 [n = 16]), △ - winter 2015/2016 (October 2015 – April/May 2016 [n = 12]), ＋ - spring 2016 (April/May 2016 – July 2016 [n = 58]), ◇ - summer 2016 (July 2016 – October 2016 [n = 17]).	
 
Figure 5.11 Mean invertebrate abundance between the beaver modified stream and the control stream. Error bars denote SD.

Figure 5.12 NMDS ordination plot of invertebrate community composition between the modified (black diamonds and blue polygon) and control streams (grey circle and red polygon).

Figure 6.2 Coarse resolution environmental data of mean [ SE] daily rainfall obtained from the Lentran Meteorological station and mean [ SE] daily river discharge for the river Enrick at Mill of Tore (NH4504429976, ~ 14km southwest of study site) for the Study Period (October, November and December) from 2006 – 2020. Rainfall data is missing for October and November 2007 and December 2010 and 2017.

Figure 6.3 Environmental data for the Allt Coire an t- Seilich (beaver modified) and Allt a’ Choilich (control) streams in Inverness-shire collected during the Study Period (October - December), 2015 and 2016. The study investigated the influence of beaver dams on the spawning migration of brown trout. Solid, dot-dashed, dashed and dotted lines denote the daily means for rainfall, ambient air temperature, modified stream water temperature (based on an average of all four data loggers), and control stream water temperature, respectively.	

Figure 6.4 Movements of an individual PIT tagged trout over an ~ca 10-hour period (between 18:22 -25 October and 04:32 - 26 October 2016) exhibiting ‘abnormal’ movement patterns. Columns bordered by the dashed lines A, B and F represent ‘feasible’ movements in which the fish likely used the by-pass channels to be detected (circles on the solid black line) at antennas 1 and 3 (but missing antenna 2), and 5 and 7 (missing antenna 6 at Dam 3. Grey columns C, D and E illustrate ‘abnormal’ behaviour when multiple dams are passed with the absence of detection at multiple antennas, suggesting overland movements. The PIT tag of this individual was subsequently found on the bank of the modified stream using a handheld PIT antenna suggesting otter predation. Detection Event denotes each time the PIT tag was detected during the ~ca 10 hr period.	
Figure 6.5 Movement patterns of six individual brown trout during the 2015 Study Period demonstrating the different behaviours observed for individuals in the modified stream and how they were grouped into one of six categories: (1) Directional, (2) Exploratory Directional, (3) Drifter, (4) Resident Drifter, (5) Static and (6) Resident Static. Location of beaver dams are illustrated in relation to antenna locations with zero representing the Loch.	

Figure 6.6 Kaplan Meier plots for all four dams with data from 2015 and 2016 combined due to small dataset from 2016. The solid lines represent estimated proportion remaining in front of the dam, and the dashed lines represent 95% confidence intervals. In the legend, total refers to total number of fish tagged, repeats are the number of repeat passers, and censored is the number of fish that were never detected passing the dam.	

Figure 6.7 Covariate parameter estimates (b coefficients) for water temperature, rainfall, rainfall with 24 hr lag applied, fish mass, fish length and repeat passers. Open points represent parameter estimates and vertical lines represent ± 2*(SE). If the vertical line does not intersect with zero, that provides evidence that the effect of the covariate is significant.	

Figure 6.8 Passage prediction plots for: (1) fish that passed a dam for their first time and for fish that had already passed a dam, and (2) fish with fork length of 100 mm vs fish with fork lengths of 300 mm.  The solid and dashed lines represent estimated proportion remaining, and the grey areas represent 95% prediction intervals.

Figure 6.9 Mean swimming speed (m/min) between Dams 1 - 2, 2 - 3 and 3 - 4 for each movement category including ‘Directional’ [black bars], ‘Exploratory Directional’ [light grey bars], ‘Drifter’ [white bars] and ‘Resident Drifter’ [dark grey bars]. Error bars denote standard deviation.	

Figure 6.10 Mean [ SD] fork length of trout captured from the Loch, Modified stream and the control stream, for each movement category. Error bars denote standard deviation.	

Figure 7.1 Site of a study to investigate predator-prey dynamics in relation to grey heron (Ardea cinerea) presence relative to brown trout (Salmo trutta) habitat that has been either modified by beaver dam building activity or left unaffected (control). The map highlights the locations of trail cameras. Beaver Dam 1 (BD1), Beaver Dam 2 (BD2), Beaver Dam 3 (BD3), Beaver Dam 4 Below (BD4B), Beaver Dam 4 Above (BD4A), Upstream Control (USC1), Control (C1), Downstream Control 1 (DSC1), Downstream Control 2 (DSC2) and Downstream Control 3 (DSC3). The map also illustrates adjacent land use and habitat types, post-beaver modification in July 2016. The inset map provides a finer resolution view of the upstream beaver modified area and camera locations; the heat map represents relative water depths.	

Figure 7.2 Mean daily ambient air temperatures (C) (grey line) and total daily rainfall (mm) (black line) recorded during the study period (1 March 2016 – 31 March 2017). Data obtained from Met Office Lentran weather station approximate 6 km from the site.	

Figure 7.3 Camera trap (Acorn 6210MC, 940NM, Acorn LTD) deployed at the study site facing beaver dam 2 (BD2). The camera was securely attached to a fencing post to ensure fix-point photography with additional south facing solar panel backup installed to ensure continuous camera operation.	

Figure 7.4 Field of view was measured at beaver modified sites along the entire bank length visible in the camera frame (BD1A, orange line). In control sites, only the length of stream visible (DSC1, orange line) was measured as it was assumed herons could hunt over the entire width of the stream from either bank.	

Figure 7.5 Field of view of the 10 trail cameras installed at their respective locations, and surrounding habitat characteristics. Beaver modified sites comprised cameras located at Beaver Dam 1 (BD1), Beaver Dam 2 (BD2), Beaver Dam 3 (BD3), Beaver Dam 4 Below (BD4B), Beaver Dam 4 Above (BD4A). The control sites comprised Upstream Control (USC1), Control (C1), Downstream Control 1 (DSC1), Downstream Control 2 (DSC2) and Downstream Control 3 (DSC3).	

Figure 7.6 [A – G] Camera trap images illustrating piscivorous predation at the study site in Inverness-shire, Northern Scotland. A - grey heron capturing brown trout over spawning redds (BD4B); B - heron striking at prey in beaver pond, (BD4A); C - rare migrant black stork (DSC1); D - remains of predated brown trout on the banks of the control stream; E - brown trout eggs discarded on the banks of the control stream; F - otter on the control stream; G - American mink foraging on the control stream (C1).

Figure 7.7 Differences in heron abundance (number of heron images / riverbank metre [m-1]) for all camera locations within the modified and control streams between spring (black bars), summer (light grey bars), autumn (clear/white bars) and winter (dark grey). The bar plots illustrate the mean with error bars denoting standard deviation.	
Figure 7.8 Linear relationship between the abundance of heron (number of heron images / riverbank metre [m-1]) and brown trout (fish / river metre (m-1). Black dots and line indicate 1+ age groups of trout with YOY excluded. Grey shading indicates 95% confidence intervals.	

Figure 7.9 Mean heron abundance (number of heron images / riverbank metre [m-1]) in relation to three different river habitat types: Run, Glide and Pool. The box plots illustrate the median (horizontal line), interquartile range (boxes) and overall range up to 1.5 times the interquartile range (whiskers). All outliers are depicted (clear circles).	

Figure 7.10 Seasonal differences in heron abundance in the beaver modified and the control stream. The box plots illustrate the median (horizontal line), interquartile range (boxes) and overall range up to 1.5 times the interquartile range (whiskers). All outliers are depicted (clear circles).	

___________________________________________________________________________

Geographic location of data collection: Beauly, Inverness, Scotland, U.K.

Date of data collection 2014-09-01 to 2016-12-30

Associated Grants: The PhD was funded by the NERC Doctoral Training Partnership (SPITFIRE) based at the University of Southampton. Additional funding was provided by the Game and Wildlife Conservation Trust, Trout and Salmon Association and NatureScot.

Related publication:Robert J. Needham, Martin Gaywood, Angus Tree, Nick Sotherton, Dylan Roberts, Colin W. Bean, Paul S. Kemp. (2021). The response of a brown trout (Salmon trutta) population to reintroduced Eurasian beaver (Castor fiber) habitat modification. Canadian Journal Fisheries and Aquatic Sciences.


Dataset available under a CC BY 4.0 licence


Publisher: University of Southampton, U.K.


Date: June 2024