Tweed Foundation Annual Report 2020 final

The Tweed Foundation 2020 Annual Report



Contents Foreword 3 4 Ronald Campbell All Good Things Come to an End 5 9 Fry and Parr Where and How Many? 12 16 Salmon The Gala Smolt Trap 24 Smolt Tracking Project 31 Fish Counters 33 34 Review Tweed Salmon Catch Trends 37 39 Bird Predation What is Being Done? 41 43 Environment Water Temperatures – SRTMN 44 46 Trout Trapping Spawning Burns 47 48 Brown Trout Acoustic Tracking 51 Angling Catches 53 53 Brown Trout Spot Pattern Recognition 54 55 Grayling Angling Catches Education Trustees and Staff Accounts Aims and Objectives for 2020 Support Find us on Social Media Our Services Map of the Catchment Notes Published June 2021. © The Tweed Foundation. Front cover: Floy Tagged Salmon www.tweedfoundation.org.uk 1

2 www.tweedfoundation.org.uk

Foreword In a year when the Covid 19 pandemic significantly curtailed human movement, its effect inevitably impacted on the work of the Tweed Foundation and the delivery of its work programmes. Remarkably however the Tweed Foundation was able to continue with its smolt projects through the dedication of its staff and indeed it proved possible to increase the number of fish that were tagged. The results of this work are reported in the article on page 9. It is interesting to note that the percentage losses in the main river of 57% were similar to the previous year. A factor of the study was that it confirmed that a percentage of tagged fish were disappearing in the short section of the River Gala below the point where they were tagged. This raises the possibility of smolt tagged fish having a greater susceptibility to mortality than untagged fish and it is planned to carry out further studies in 2021 to investigate this. Following on from the 2019 programme an additional 1,000 smolts were tagged with P.I.T. tags at the Gala Smolt Trap where the biologists witnessed a huge smolt run with the number of smolts passing through the trap increasing by a factor of 3. This is fully reported in the article on page 11. The year also saw the return of the first of the salmon tagged as smolts in 2019 with 9 grilse being detected as they passed through the Gala fish pass. When the results from the P.I.T. tag reader installed in the fish pass for 2021 are known we will begin to have a picture of the likely return rate to the Gala Water. 2020 was a positive year in terms of the number of fish recorded by the counters on the Ettrick, Gala and Whiteadder with the Gala counter showing a huge increase reflecting the increased run of summer fish that were evident in the River. This is reported more fully in the article on page 16. The issue of changing trends on the River is examined by Dr Ronald Campbell in a very comprehensive paper on page 24 in which he looks at catches over the past 200 years and seeks to identify the long term trends. This is his swan song piece and is the culmination of his research on catch records over the past 30 years. It is a fascinating analysis and, at a time when there is huge concern about the reduction in salmon numbers, it does provide a measure of reassurance that the changes are, at least, in part due to a change in the cycle with the decrease in autumn catches being compensated in some measure by the increase in the summer catches. The Tweed has a huge resource of both netting and rod catch records which give the perspective essential to understand properly what is happening on the river at the present time. The pandemic unfortunately meant that it was not possible for the Tweed Foundation to participate in the National Electrofishing Programme Scotland (NEPS) but the biologists were able to continue with their regular monitoring of the catchment and their report “which is broadly encouraging” is found in the article on page 6. Other activities of the Tweed Foundation were restricted, but all in all it proved to be a busy and productive year. The year also saw the departure of Fay Hieatt who had worked tirelessly for the Foundation and River Tweed Commission over very many years initially as an Administrator and latterly as Director of the Foundation. She is to be thanked for her contribution to the development of the Foundation. The Foundation has since recruited Alison Gorrie as Executive Assistant and Jamie Stewart as Director. We welcome them both. Richard Onslow has also retired as a Trustee after many years service and we thank him for his support and encouragement. Lastly, I would like to record my thanks to the enormous commitment shown by the members of the Tweed Foundation Team during 2020 and for their delivery of a number of important projects including particularly the smolt projects. I would also thank our supporters and in particular the Fallago Environment Fund who provided a very generous sponsorship of the smolt tracking project and all our other sponsors and those with whom we work for their help and support over the year. Hugh Younger Chairman www.tweedfoundation.org.uk 33

Ronald Campbell: All Good Things Come to an End After thirty years of dedicated employment at The Tweed Foundation, attendance at more than 100 Commission meetings and six editions of the Tweed management plan, Dr Ronald Campbell, the senior biologist for The Tweed Foundation is to retire at the end of June 2021. Ronald has always had a link to fisheries, as his father, Niall Campbell, worked at the Freshwater Laboratory at Faskally (formerly known as The Brown Trout Research Laboratory), was a keen fisherman who wrote for Trout and Salmon magazine and was co-author of the book ‘Freshwater Fishes’. In a time when Ronald had a full covering of dark hair, samples were extracted by his father to make ‘The Ronald fly’ which was detailed in Trout and Salmon. After completing a B.Sc (Hons) in Zoology at Aberdeen in 1984 and Ph.D. in New Zealand, Ronald had relatively short postings of one to two years at the University College Cork, Institute of Terrestrial Ecology, Nature Conservancy Council and the University of the Mediterranean, Turkey. While out in Turkey, Ronald was encouraged to apply for the newly formed Biologist position at The Tweed Foundation. This was at the start of the fisheries trust movement that now covers most of Scotland, with the developing philosophy of evidence based fisheries management. Starting in 1990, Ronald begun work on the Tweed Fisheries Management plan which is used to coordinate the working activities of The Tweed Foundation and is one of his defining pieces of work. From 1992-1994, an ambitious radio-tracking project was undertaken to identify the spawning distribution of different stocks. On a number of occasions, Ronald took to the air in a small Cessna with the mobile tracking equipment to cover the catchment, enduring occasional bouts of high altitude sickness while trying to operate the detection equipment. Through the 1990’s, Ronald developed the scale reading programme for adult Salmon and Trout, with scale reading equipment first kept in his house and then a scale reading cubicle; contentment for Ronald was scale reading and listening to recordings of ‘Round The Horn’, much to the disapproval of younger members of staff. In this period, a number of assistants were given valuable experience that helped them to progress in their fish related careers. James Maclain, the first assistant, now works at the Natural Museum, London, as curator of fish. Callum Sinclair was director of Rivers and Fisheries Trusts Scotland and Matthew Gollack is now with the Zoological Society of London. One of Ronald’s mantras is “if you don’t know where you come from, you don’t know where you are going”, with the history of Salmon and Trout, in particular their catches, providing perspective on the current state of stocks today. Ronald has revelled in researching and documenting catch records, diaries, competition records, anecdotes, statistical accounts and any other reference to Tweed and its fisheries. Research from Trout diaries and competition records show how small parr sized Trout used to be fair game and were a significant part of catches, with increasing size limits reducing the size of catches. For Salmon, the insights gleaned from rod and net catches are significant, as they challenge the view that Salmon are in decline and instead show cyclical changes in run timing and sizes of fish that have happened before and are currently taking place. Ronald’s retirement will be a loss to The Tweed Foundation and the fisheries management sector, although his management plan and lengthy appendices mean that his influence will be long lasting. Although technically retired, Ronald will continue to be active in his historical research of fisheries management, with a number of books in the pipeline. His second home will probably be the library at Freshwater Laboratory. With his bee hives and productive garden for company, we wish Ronald all the best for the future. If you would like to pass on any messages to Ronald, his work email address [email protected] will remain active for the foreseeable future. 4 www.tweedfoundation.org.uk

Fry and Parr: Where and How Many? Rationale: The juvenile salmon and trout of the Tweed catchment have been systematically monitored by electric-fishing since 1988, providing a long series of records that can show any trends or sudden changes as well as natural variability. The 2020 electro-fishing monitoring programme for juvenile Salmon and Trout focussed on the Whiteadder, Eden, Leet and Eye Water as part of our triennial monitoring programme. The results confirm previous findings that Salmon spawning in the Eye Water is sporadic (most probably strays from neighbouring rivers), with no evidence for a self- sustaining population. Juvenile Salmon are largely absent from the Leet Water, in part due to its small size. Numbers of Salmon in the Eden Water dropped quite significantly from 2017, indicating a lack of spawning fish last year. Further details of the results (including Trout) can be found at https://arcg.is/1O8jn5 1. The Whiteadder is a unique catchment in the Tweed District, with the relatively recent removal or easing of obstructions to fish migration, diverse land use and geology and its proximity to the sea which moderates temperature variability in summer and winter. The results for juvenile monitoring, adult catches and stock structure are explained to some extent by these unique conditions, but there are still some unanswered questions that are explored in the following section. A Brief History of the Whiteadder The Whiteadder has a long history of Caulds (weirs) being built and maintained to divert water for mill operations, which has affected the upstream migration of Salmon and Trout. Historical records show that access has been intermittent over the last two hundred years for Salmon and Sea Trout, but by the end of the 1990s the river was almost fully open with the removal or easement of Edington and Newmills Caulds on the main stem of the Whiteadder and Cairns Mill and Kimmerghame Caulds on the Blackadder. The Whiteadder reservoir dam is the only remaining obstruction to Salmon migration. It was Cumledge Mill Cauld, as show in the picture here, washed away in the 1948 flood. completed in 1968 and blocked off a substantial area of spawning ground further upstream, with no provision for a fish pass. Electro-fishing monitoring documented the subsequent recovery, with just one out seven sites recording the presence of Salmon parr in 1988 compared to seven out of seven sites in 2000. Densities of parr increased from 1.3 per 100 m2 to 65 per 100m2, demonstrating that Salmon can quickly recolonise rivers that are opened up without human intervention through stocking; the Blackadder had not had salmon in it for at least 200 years, but within two years of Kimmerghame cauld being eased, salmon were spawning well upstream of it. The Whiteadder is well known as a ‘Spring’ river for Salmon, which means most of the fish return in the first part of the year (to the end of June). Like all Salmon, they wait until Autumn to spawn. The reason for their early return to the river is unknown, but it may well be through the natural selection of fish that enter the river early, which will give them a better opportunity to get past the series of caulds and waterfall at Elba when suitable flow conditions arise. 1 An explanation of the electro-fishing methodology can be found at https://arcg.is/14nHWi 5 www.tweedfoundation.org.uk

Fry and Parr: Where and How Many? Electro-fishing Results The Salmon Fry results for 2020 compared to previous results (summarised as average numbers before 2020) can be viewed in the map opposite. The result for each site is coloured coded to fit into one of six categories, ranging from ‘Zero’ to ‘Very High’, with averaged results before 2020 represented in the left hand circle and the 2020 result in the right hand one. The consistent geographical pattern of abundance shown by both the averaged results before 2020 and the 2020 results is quite clear, with the highest numbers of Salmon Fry in the middle and upper sections of the Whiteadder and Blackadder (mostly green circles). Lower numbers are found at the bottom of these tributaries (red and orange circles). The areas of high abundance indicate good levels of spawning and productive habitat for rearing Salmon Fry. The reason for the areas of low abundance is unclear; there is certainly plenty of suitable spawning and rearing habitat, indicating that another factor is affecting numbers. Temperature could be a factor as Salmon eggs require low temperatures to incubate and hatch properly. Published literature shows that at 10ºc eggs are under stress and 12ºc leads to egg mortality. As the Whiteadder and the Eye Water are the closest Tweed tributaries with Salmon to the sea, temperatures are moderated in winter time relative to the rest of the catchment. A pollution incident near Chirnside in 2020 that killed a number of fish in the local area may have affected some of the results and it is not known how far downstream this might have affected.  To start investigating why there are areas with low results, we walked some sections of the lower and middle Whiteadder in 2020 with an aerial drone to count redds and spawning fish, although this methods was found to be inconclusive. We have also installed two temperature loggers in these areas in October 2020 to see whether elevated winter temperature could affect egg viability. The loggers will show if there is an increased frequency of 10ºc or higher water temperatures relative to other parts of the catchment. 2020 Results Results between Chirnside and Cockburn Mill and upstream of Ellemford Bridge were typically in the very high category in 2020, which was similar to their averages Results around Abbey St Bathans, results dropped by several categories, suggesting reduced spawning in this area. Results were very poor for the Bothwell Water, again indicating a lack of spawning, which is surprising as flow records show that water levels were suitable for access by adult Salmon in the spawning period. Results for the Dye Water were mixed, with some increases and decreases suggesting localised, patchy spawning rather than good coverage in all areas. Results from the Blackadder in 2020, results were mostly consistent with results before 2020, with most sites not changing by more than one category. The primary area of Salmon Fry production continues to be from Charterhall up to Greenlawdean where ‘Medium’ to ‘Very High’ categories are found. The small numbers of Salmon Fry on the Langton Burn may well be migrants from further downstream as this watercourse is more suited to Trout spawning. A long term monitoring site which illustrates the recovery of Salmon in the local area is at Greenlawdean on the Blackadder, with the results shown below. In 1988 there were no Salmon Fry recorded and only a few in 1996 and 2000. In 2017, 53 Salmon per 100m2 were recorded, followed by an enormous increase in 2020 to 320 fry per 100m2. The increase could be attributed to greater levels of spawning the previous year or improvements in water quality from extensive riparian fencing further upstream. Electro-fishing results for Salmon Fry at Greenlawdean. 6 www.tweedfoundation.org.uk

Fry and Parr: Where and How Many? Electro-fishing Salmon 2020. www.tweedfoundation.org.uk 7

Fry and Parr: Where and How Many? Number of Adults With the opening up of the Whiteadder catchment to migratory fish, catches of adults would be expected to increase year on year to a point where fry and subsequent smolt production reaches capacity, which would then limit rod catches. To some degree this has occurred for the juvenile population, with the main stem of the Middle and Upper Whiteadder extremely productive for Salmon Fry, but this is not reflected in the Whiteadder rod catches shown below. From 2002 to 2008 it looked like this would be the case, with the expectation that rod catches would continue to be around 700 fish but after 2012, catches dropped to between 100 and 200 Salmon. While variability in smolt production and marine survival will influence the number of returning adult, we would still not expect these factors to influence the results so significantly and we do not see this level of variability or pattern in Tweed rod catches for Spring Salmon. 2020 Whiteadder rod catches. The reported annual rod catch should reflect the abundance of adult Salmon that return to the Whiteadder, particularly as it is a relatively small river that will have a relatively high exploitation rate (the chance of a fish being caught). Fly and lure fishing should be very efficient in catching Salmon as most pools in the lower and middle river can be covered and early running Salmon are also more likely to be caught than later running Autumn fish. Based on published figures and capture data from radio tracked fish, at least 20-30% of the population would be expected to be caught. We have no record of angling effort for the Whiteadder, but even if angling effort is relatively low, you would expect most fish will still be covered by fly or lure in the first month of returning to fresh water when we know they are most likely to be susceptible to capture. Under declaring of fish as a possibility does not explain these fluctuations as it would require a consistent decrease in declaration from most productive beats from 2013 onwards. The rod catch results for the Whiteadder continues to be an enigma with no obvious means of investigating these results further. Annual monitoring of smolt output could help investigations, but this would be a labour intensive, expensive exercise and would need another elevated catch to link the two together. If another bumper year becomes apparent, we will be on hand to collect scale samples which will at least provide some insight into the age class structure contributing to the increase. 8 www.tweedfoundation.org.uk

Salmon: The Gala Smolt Trap Rationale: While monitoring the fry and parr shows how they can vary over the years, the key figures are (1) how many smolts are produced from how many fry and parr and (2) how many adults return from how many smolts? In the third year of operation, the Gala smolt trap continues to provide new insights into the smolt migration patterns for a Tweed tributary, as well as generating important biometric data such as fish length, age structure and by PIT tagging a sample of fish, return rates back to the Gala Water. The primary aim of our trapping work is to estimate the annual production of Salmon and Trout smolts. The normal procedure for smolt traps that capture a proportion of the total run is to estimate the trap efficiency by dye marking1 samples of fish, returning them upstream, and recording how many are recaught in the trap at different flow levels. Trap efficiencies at different periods through the smolt run are then multiplied by the number caught in the trap to produce a population estimate. One of the assumptions in using this method is that marked fish that are returned upstream will behave in the same Gala smolt trap. way as unmarked fish, with their trapping efficiencies being the same. As we suspected some marked fish were delaying their migration back downstream, in 2020 we developed our dye marking system to make sure fish that are recaught in the trap could always be traced back to the date they were tagged. For each dye marking event, the number of recaptures is summarised in the Table below for the number of days after marking and the total recapture rate. Recaptures of Dye Marked Fish Panjet Date Number Number of days after dye marking Percentage Code marked 24-Apr-20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 27 28 29 32 33 35 36 37 40 recapture 2 27-Apr-20 50 3 02-May-20 50 61 1 1 5 1 1 1 1 4 1 1 1 50.00 4 06-May-20 50 5 08-May-20 42 11 14 2 532 111 1 1 45.10 6 11-May-20 50 7 13-May-20 50 2 1 13 1 11 7 3 1 1 44.00 8 15-May-20 50 9 18-May-20 50 8 5112 51 1131 1 71.43 10 20-May-20 50 11 22-May-20 49 8 31 1 1 32 12 44.00 12 27-May-20 50 13 53 4 21 11 5 1 1 21 1 58.00 3 43 4 1 52 1 1 48.00 2 652 1 11 2 5 50.00 10 4 3 1 321 13 12 62.00 21 1 1 11 2 55.10 21 2 1 1 1 3 58.00 10 2 1 3 2 42.00 The results, particularly for fish marked in late April to mid-May, clearly shows that there is a delay in fish migrating back downstream. For example, four fish tagged at the end of April took between 30 and 40 days to be recaptured. Ideally the majority of marked fish would take no more than a week to pass back downstream, allowing trap efficiency to be calculated for different flows. The results indicate that either the dye marking and trapping process induces a delay in migration (e.g. smolt are resting or feeding) or smolts are reluctant to enter the trap for a second time and find it difficult to pass over the cauld face in low water. The added delay of fish in the pool above the trap will increase the chance of predation, which will also affect the accuracy of estimates, with a resident Otter and several Goosanders regularly feeding there. Using dye marking for accurate smolt output estimation is therefore not going to produce accurate totals, particularly in years where flow levels are more variable through the smolt run. 1 A panjet is used to apply a small blue dot to the underside of a fish that will remain visible for around 6 weeks. A combination of different dot locations and in some cases two dots allow recaptures to be traced back to the date they were tagged. www.tweedfoundation.org.uk 9

Salmon: The Gala Smolt Trap Although we have not managed to produce an accurate smolt output estimate yet, a graph of cumulative totals provides an indication of run timing patterns (below). In all three years of sampling, the river level profile has been similar, with no significant rainfall throughout April or May. Without a rise in water to stimulate smolt movements, migration was delayed in 2018 due to below average temperatures in March and April (The Beast from the East). 2019 and 2020 are probably more representative of a normal smolt run, with numbers starting to pick up around the 21st-23rd April. Smolt trap cummulative totals graph. If the trap efficiency in 2020 was similar to 2018 and 2019, then the total of 15,960 smolts for 2020 suggests that smolt output was higher than the two previous years. As an indication of the potential productivity of the Gala Water, applying the highest calculated trap efficiency value in 2020 of 72%, then the estimated total smolt output would be 22,167 fish, equating to six smolts produced per 100 m2 of streambed. To overcome the issue of dye marked fish delaying their migration back downstream, an alternative approach is being tested for 2021, with 1,000 juvenile Salmon over 80 mm length caught by electro-fishing and P.I.T. tagged further upstream in late September 2020. With a tag detection antenna installed on the streambed several hundred metres downstream of the trap, the trap efficiency can then be calculated for P.I.T. tagged fish that migrate as smolts in 20211. As we don’t know how many PIT tagged fish will survive through the winter or migrate down into the main river before the main smolt run, the number of PIT tagged fish may need to be increased in future years. The results for 2021 will continue our progression towards a more accurate estimation of smolt output from the Gala Water. 1 Calculated by dividing the number of PIT tagged fish in the trap and released downstream by the total number of PIT tagged fish detected by the antenna. 10 www.tweedfoundation.org.uk

Salmon: The Gala Smolt Trap Having the smolt trap as well as the counter on the Gala means we can now count fish out and then count them back. As the fry and parr in the Gala are monitored by electric-fishing to estimate numbers and we can calculate the number of eggs deposited in the Gala, we are now in a position to quantify the main stages of the life-cycle of a stock of the Summer fish. Assuming that the ups and downs of the monitored Gala Summer stock will be typical of all the Summer stocks of the Tweed we will be in a unique position to be able to keep proprietors and anglers informed of the state of this major run, the factors affecting it and any suitable management actions. P.I.T. tagged fish returning to the Gala In 2019, 1,000 Salmon smolts were tagged with Passive Integrated Transponder (P.I.T.) tags at the trap to measure return rates to the Gala Water. Without a battery, these tags are the smallest type of detectable tag implants. With the installation of a detection antenna in the Gala fish pass, nine P.I.T. tagged Salmon returning as one sea winter fish were detected in September and October 2020. We will need detection data for 2021 (two sea winter) and 2022 (three sea winter) before we can produce an overall return rate for the 2019 cohort of smolts. We will also be able to relate these returning fish to their smolt sizes and date of tagging and, over time, build up a data set relating smolts to successfully returning fish. Over time we will collate information on return rates to the Gala Water. Improvements to the PIT reading equipment are planned for 2021, with an upgrade in the equipment to improve reliability and the antenna will be attached to the fish counter, allowing an estimate of fish length to be made along with a recorded video clip of the fish. We also plan to P.I.T. tag a sample of Sea Trout smolts to start estimating their return rates as well. The first P.I.T. tagged Salmon returning to the Gala Water after a year at sea. 11 www.tweedfoundation.org.uk

Salmon: Smolt Tracking Project In 2019 The Tweed Foundation initiated a study to investigate the survival of Salmon smolts as they migrate down the river. Due to the potential variability in factors that can influence smolt survival, in particular water levels in Spring, the study is expected to continue over a four year period, with the final year potentially investigating whether shooting for scaring or lethal control can improve smolt survival. The overall aims of the study are to :- 1) Estimate salmon smolt survival in the main stem of the river (Middle and Lower Tweed). 2) Identify any areas of the river where there are high losses. 3) Test alternative management strategies to reduce losses (if high losses are detected). As the 2019 tracking work was a trial study with a relatively small size (60 fish), we were extremely cautious in extrapolating the results to the wider population of Salmon smolts and assigning losses to particular predators. The 2020 study built on the work carried in 2019, with a larger sample size of 240 tagged fish and a much more extensive monitoring network, with 22 tracking receivers deployed, compared to nine in 2019. All of the fish were tagged using Vemco V5 (5 mm) acoustic tags that produce a coded 180 khz sonic ‘ping’ on average every 30 seconds, uniquely identifying each fish. To increase the recovery time between tagging and migration and investigate whether recovery time influences smolt survival, we planned to catch 80 fish as presmolts in March by electro-fishing further upstream. This would allow fish to recover for at least four weeks before migrating past the first tracking receiver in the lower Gala Water. The Tweed smolt tracking project complements other related work carried out by The Tweed Foundation, including dietary analysis of Goosanders and Cormorants to update data collected in the 1990s and regular bird counts to monitor numbers in the river. Smolt Survival to the Sea (Aim 1) One hundred and fifty-five tagged fish were detected at Galafoot and tracked down the river. With 66 fish detected at Berwick, this produces a figure of 57.4% smolt loss (42.6% survival), compared to a percentage loss of 59% in 2019. Out of the 155 detected fish, only 15 had been caught electro-fishing as pre-smolts in late March. Unfortunately such a small sample size of pre-smolts prevents a meaningful comparison of survival between fish caught electro-fishing and fish caught at the trap. In the report for 2019 (https://arcg.is/1KzynD), consideration was given to how representative the sample of tagged fish was of the natural population. With 39 tagged fish being detected in the main river in 2019 and 155 detected in 2020, this only represents a tiny percentage of the total smolt population. Given the expensive nature of tracking projects, mainly due to the cost of tags, this is an inherent limitation that is unlikely to change. We must therefore remain cautious in extrapolating the survival rate of tagged fish to the wider population. The 2019 report also considers the potential influence  of Goosanders and Cormorants on the survival of tagged smolts and the wider population.  Due to Coronavirus restrictions in the 2020 study, the standard bird counting method of walking sections of river from Galafoot to Berwick on the same day could not be carried out, but a survey using vehicles was possible and was considered to be reasonably accurate, with only small amounts of the river not checked.  The number of Goosanders counted on April 17th 2020 was half the number counted in 2019 and Cormorant numbers were similar. During the main part of the smolt run on the 7th May, numbers of Goosanders and Cormorants were similar between years. The possible influence of predation by large Brown Trout contributing to losses is also a consideration, but we currently have no means of estimating their population size and consumption rates of smolts. Ad hoc samples can be collected, either through the occasional large Trout that is killed by an angler in Spring or through stomach flushing, but we still have no effective means of carrying out representative sampling. We are looking to start collecting some catch data from sample Salmon fisheries regarding the frequency and location of large Trout captures and we are continuing to consider possibilities for investigating this question further. The effect of fish carrying the tag (known as tag burden), their recovery from surgery and the potential for tags to be shed from fish on the survival percentage continues to be another consideration.  In order to conclude whether reported tag losses are representative of the natural population, peer review studies on this subject are required. Tagging studies investigating these questions are planned for 2021 and 2022 in partnership with Marine Scotland Science. 12 www.tweedfoundation.org.uk

Salmon: Smolt Tracking Project Distribution of Smolt Losses (Aim 2) Results can also be plotted as percentage survival against distance down the river. The graph clearly illustrates that losses are higher in the Middle Tweed than Lower Tweed, which is a similar geographical pattern to the 2019 results. When looked at a finer scale, the demarcation between higher and lower losses does not precisely match the boundary of the two sections of river (confluence with the Teviot just below Kelso cauld), with loss rates noticeably lower beyond Sprouston in 2019 and Upper North Wark in 2020. The rate of loss is very similar between years is very similar from Galafoot to Makerstoun, with a loss rate of 1.19 smolt per kilometre in 2019 compared to 1.26 in 2020. There is a higher rate of loss between Makerstoun and Sprouston in 2020 compared to 2019, but the 2019 results may be influenced by the relatively small size and lower number of receivers. Smolt tracking graph distance vs survival v4a. The similar geographical distribution of losses in 2019 and 2020 suggests that there are consistent factors influencing the results.  In 2019, losses of tagged fish were much lower below Sprouston on the lower river and it was suspected that the transition from low to high survival was at the confluence of the Teviot and Tweed where the input of extra smolts and increased water volume decreased the probability of predation.  Without receivers in the immediate area of the Tweed- Teviot confluence, this theory was conjectural.  However, the 2020 results show that the loss rate found in the Middle Tweed continued down to Upper North Wark, suggesting other factors are having a greater influence on results. Bird counts continue to show that numbers of Goosanders and Cormorants are higher in the Middle Tweed and the upper parts of the Lower Tweed, despite these areas having the highest number of beats that have licences for predation mitigation (scaring and removal).  What is particularly interesting is that in the unusual absence of any predation management due to COVID restrictions, numbers of Goosanders were much lower in the 2020 April count compared to 2019 and were similar in May.  This leads to the question of whether current bird predation management is effective in disrupting predation on the smolt run and leads to a measurable improvement on smolt survival.  Studies carried out www.tweedfoundation.org.uk 13

Salmon: Smolt Tracking Project by The Dee Rivers Trust in 2019 found no detectable difference in survival of tagged smolts between sections of river with bird management compared to control sections without any management (http://www.riverdee.org.uk/f/articles/ Smolt_tracking_2018_River-Dee-Trust.pdf). Loss Rates compared to Other Rivers The Tweed in-river percentage losses in 2019 and 2020 are similar to those in the River Deveron where smolt losses of 60% and 57.5% were recorded in 2016 and 2017 respectively. Recorded in-river losses in the Dee where the same tags were being used between 2016 and 2019 are lower, ranging from 21 to 31 percent. There is no obvious explanation for the discrepancy in the results between rivers. Neither the Deveron or Dee show the Tweed pattern of higher losses in the upper part of the study section and lower losses in the lower part of the river (excluding the estuary or harbour area). This may well be related to individual river characteristics (e.g. availability of sheltering habitat for smolts) or the geographical distribution of fish eating birds, which would require further investigation. Smolt Survival in the Lower Gala Water While the original purpose of the study was to investigate smolt losses in the main river, as this is where predation is thought to be highest, a question mark was raised over the loss of six smolts on the lower Gala Water in 2019 between the smolt trap and Galafoot (a distance of 3.4 km). With such a small sample size, we were careful not to draw any conclusions, although bird predation as a factor on its own would be unlikely as only four Goosanders were counted in this section in April and May. Also of note for the 2019 results, the percentage loss was zero when tagged smolts migrated quickly through the lower section of the Gala in high water and 35% in low water with a slower migration period of 4 days (table below). In 2020, similar percentage losses were again detected in low water conditions (table below), with 32% losses recorded for fish that were caught and tagged electro-fishing. Most significantly, for a sample size of 209 fish caught at the smolt trap, 28% of fish disappeared in this section. Again, under low water conditions, fish caught electro-fishing or at the trap took between five and seven days to migrate out of this section. Mobile tracking equipment was tested in this section to see if tags could be detected but unfortunately noisy, broken water, which is common in this area, prevented the equipment from working efficiently. Survival of smolts under different flow conditions and capture methods Year Capture Water level Average Number Number Difference Percentage method transit detected detected Percentage loss 2019 High water time at Galafoot 0 2019 Electro-fishing Low water (days hour or 6 loss (%) per km (% 2020 Electro-fishing Low water : minutes) tagged at 26 7 per km) 2020 Electro-fishing Low water Skinworks 11 69 Trap 0d 2:14 15 00 4d 6:12 cauld 140 35 10.3 7d 18:02 26 32 9.4 5d 08:58 17 28 8.2 22 209 Using the percentage loss figure per km, the Lower Gala figures are between eight and ten times higher than the Middle Tweed. Predation as the main factor causing the high Gala Water losses can be considered using basic calculations of daily food intake and numbers of Goosanders or Otters in this section*. As some context, 15,960 Salmon and 2,277 Trout were captured at the Gala smolt trap in 2020 and this represents only a proportion of the Gala smolt run (between 40 and 70 %). For predators to be consuming around 30% of the smolt run in low water would require a combination of high consumption rate estimates and a low smolt output estimate based on a high trap efficiency. 14 www.tweedfoundation.org.uk

Salmon: Smolt Tracking Project The possibility of tagged fish having a greater susceptibility to mortality compared to untagged fish, either through predation or disease, cannot be discounted at present. There is no instance of such high losses in other studies that we know of for what appears to be a healthy section of water without any issues that could cause high levels of mortality. The anomalous results for the Gala Water certainly strengthens the case for research that investigates the possible effect of tags or tagging on smolt survival. Plans are in place for studies in Scotland (including Tweed) that will compare the survival rates for fish with different sizes of tags to the natural population. The closest we can get to a tagged fish that will act naturally is to use P.I.T. tags which are much smaller than the acoustic tags used for this project and as they do not have a battery, can be implanted into fish in October the previous year. 1,000 fish were tagged in October 2020, acting a control group to compare survival to fish with larger tags in 2021. A P.I.T. tag (top) compared to a V5 acoustic tag. Conclusions from the Study • The consistency of overall percentage loss in 2019 and 2020 and the geographical distribution of losses suggest that the factors influencing survival are similar in both years. Low flow conditions in both years were probably a contributing factor, as other tracking studies and smolt monitoring confirm that rises in river level promote migration and higher average flows lead to higher survival. Tracking with Sea Trout smolts from the Yarrow in 2010 and 2011 showed this too, with a higher survival in the wetter spring of 2011 than in the drought of 2010. The premise is that a higher volume of water (in some cases coloured) reduces the risk of predation and faster water facilitates movement down to the estuary. • With two years of data now collected as a baseline, particularly for the Gala Water, we now have the opportunity to investigate the possible effect of tags on smolt survival, with a project being carried out in 2021 and 2022. If survival percentages of tagged fish do reflect losses of the natural population, then this will be a major piece of evidence for informing the debate on how to improve survival of smolts. With the study on tag effect being carried out with a focus on the Gala Water, further baseline data will also be collected. • With the evidence pointing to higher losses of smolts in the Middle Tweed, the natural end point of the study would be to carry out a comparison of smolt survival in this area between sections of river with bird management (scaring and shooting) versus control sections without any management, which would complement the work carried out by The Dee Rivers Trust. *From general observation at the trap, a resident otter is known to live in proximity to the Skinworks cauld. A paper by by Nolet and Kruuk (1994) reports from other research that Otters can eat between 119 and 150 g of fish per kilo of their body weight per day.  Taking the average weight of a tagged smolt as 24.7g and an otter size of 10 kg, a maximum intake of 1500g / 24.7 = 60.7 smolts per day can be potentially consumed. Multiplied by 61 days = 3,702 smolts in April and May.  Again, this will probably be an overestimate as resident fish including other fish species such as eels may also be consumed and the home range of the resident otter may extend beyond the monitored section. To provide some context to the possibility of goosanders causing these losses, 4 birds x 61 days x 10 smolts per day (probably an overestimate) = 2,440 smolts for the period of April and May.  This figure may well be an overestimate as birds are unlikely to exclusively feed in this section of river and resident fish will also be consumed. www.tweedfoundation.org.uk 15

for the management of a stock is that enough fish should escape caiwesnaaSnnaddlmsftuoolclynkt:chaFeniirsbhneuCcrosoeuurnynteatdere.rasFsisfhorcothuentneerxatrgeernuenraotniotnheanEdtttrhiciks, g information on run timing, size distribution and numbers for Rationale: The most basic need for the management of a stock is that enough fish should escape all the pressures on them to spawn and fully their nursery areas for the next generation and this best known if the fish each species and stock can be counted. Fish counter are run on the Ettrick, Gala and Whiteadder, providing information on run timing, size distribution and numbers for Salmon and Trout. Ettrick The Salmon total for the Ettrick counter was 2,841 fish, a counwetlceomr ewincareasse 2fro,m8t4he1resuflitssohf 2,01a8 anwd 2e01l9cthoatmweere the lowest on record. Relative to the results before the new 8 andcou2nt0er 1wa9s intsthalleadt(Figwuree1r),ethet2h020etoltaol wwasehigshter othann results in 2003, 2007 and 2008. It remains to be seen whether the poor results of 2018 and 2019 were a ‘blip’ in the results; efore the new counter was installed,if totals similar to 2020 are collected in subsequent years, this will add evidence to the poor results of 2018 and 2019 being results in 2003, 2007 and 2008. Itattributable to the floods of 2015/2016 affecting recruitment and subsequent returns of adult fish as one or two sea winter poor results of 2018 and 2019 werefish. While Salmon Fry numbers in the Ettrick appeared to be unaffected by the floods based on electro-fishing results s similar to 2020 are collected incollected in 2016, we cannot exclude the possibility that parr were affected. If there was a drop in recruitment, then it will evidence to the poor results of 2018then be interesting to see if there are recurring poor returns of Spring Salmon to the Ettrick every five years (2023 and 2024), as the floods of 2015/2016 affectingthe typical age of an Ettrick fish is two years in the river, two at sea, returning in the fifth year. This pattern was seen after the urns of adult fish as one or two seasevere floods of 1977, with depressed Spring catches detected every five years until 1997. With the installation of the Ettrick umbers in the Ettrick appeared to becounter, we will have more objective data this time to assess whether there is a reoccurrence of this phenomenon. on electro-fishing results collected inWhen numbers of adults are converted to estimated eggs deposition, the totals for 2020 was 12,763,902 eggs, more ssibility that parr were affected. Ifthan double the deposition estimated in 2018 and 2019. It will be very interesting to see whether a corresponding , see if thereEttrick Fish Counter. are recurring poor then it will then be interesting toincrease in fry numbers is detected in 2021 when electro- fishing monitoring sites are revisited. aTronudt nu2m0be2rs4(s)h,owansintFhigueret2y) cponitcinauel tao gbeebeolofw aavneraEgetftorrfiischk ttrick every five years (2023Consistent with results from 2018 and 2019, under 40 cm (Brown Trout) and over 40 cm (Sea Trout and a small proportion of large Brown Trout). The reduced number o at sea, returning in the fifth year. This pattern was seen after theof Trout under 40 cm can to some degree be explained by a higher threshold filter in the new counter to reduce turbulence events that cause false counts. However, the Trout total for fish over 40 cm in 2020 (unaffected by the filter) essed Spring catches detected every five years until 1997. With thewas 916 fish compared to an average of 1,717 before 2017. There is no clear explanation for the reduced numbers, and they are very different from the above average numbers detected by the Gala counter. It is conceivable that recruitment , we will have more objective data this time to assess whether thereof Trout, was affected by the floods of 2015/2016, with the 2020 cohort being the last age class that could be affected. enon. onverted to estimated eggs deposition, the totals for 2020 was uble the deposition estimated in 2018 and 2019. It will be very esponding increase in fry numbers is detected in 2021 when electro- ted. 16 www.tweedfoundation.org.uk

Salmon: Fish Counters Figure 1: Fish counter Ettrick graph Salmon. Figure 2: Fish counter Ettrick graph Trout. www.tweedfoundation.org.uk 17

Salmon: Fish Counters Gala The Gala Water fish counter was upgraded in August, with the installation of a new stainless-steel light tunnel which houses the scanner, lights and camera. The main improvement of the new system is a nitrogen filled space immediately in front of the camera which significantly improves the quality of video clips, greatly increasing identification rates. The counter location was also moved from the bottom of the fish pass to the top of it, reducing the number of fish that repeatedly pass up and down through the scanner. The 2020 estimate total was 2,478 Salmon (Figure 3), which was higher than any other recorded total for the Gala Water. Relative to other totals, this was 414 fish The new light tunnel which fish swim through, with camera housing. more than the second highest total in 2019. As would be expected, estimated total egg deposition (13,653,344) and eggs deposited per 100m2 (3,746) are also records. Trout totals in the graph (Figure 4) are split into fish under 40 cm that are almost all Brown Trout and fish over 40 cm which will be mainly Sea Trout, but will include some larger Brownies as well. It is important to make this split as numbers of each form of Trout reflect different life history strategies that are in part influenced by different factors. The 2020 Trout total for fish under 40 cm was 948, the highest on record and well above the average of 428. There is a suggestion of an upward trend in numbers for this size range, but without data for 2016 and 2018, no firm conclusions can be made. 1,475 Trout over 40 cm were estimated, slightly above the average of 1,267. A number of floy tagged Salmon were spotted passing through the Gala counter. Unfortunately the tag code to identify where the fish had been caught and released could not be read from the video clip. For 2021, PIT tags have been added to the floy tags used at Middle Tweed beats, which will be registered by the detection loop added to the counter. This will identify some caught and released salmon as Gala fish and will allow an estimate to be made of the contribution of the Gala Water to the catches of the Middle Tweed. 18 www.tweedfoundation.org.uk

Salmon: Fish Counters Figure 3: Fish counter Gala graph Salmon. Figure 4: Fish counter Gala graph Trout. www.tweedfoundation.org.uk 19

Salmon: Fish Counters Whiteadder The Whiteadder counter continues to provide information on run timing, sizes of fish and a minimum annual total, with an unknown proportion bypassing the fish counter by ascending the cauld face. Bearing this in mind, 576 Salmon were counted (Figure 5) relative to the average of 634. 95 Trout under 40 cm were counted (average 199) and 380 over 40 cm (average 569) (Figure 6). The camera which records video clips of fish for species identification was out of operation for 2020 which meant that the totals for Salmon and Trout were estimated using data from previous years to predict species. A complete overhaul of the fish counter is planned for 2021, with a new light tunnel, lights and camera system that will improve the quality of recorded videos Fish counter Whiteadder. and associated identification rates. We are very grateful to Ahlstrom-Munksjo, the owners of the Chirnside Mill, for funding this upgrade. 20 www.tweedfoundation.org.uk

Salmon: Fish Counters Figure 5: Fish counter Whiteadder graph Salmon. Figure 6: Fish counter Whiteadder graph Trout. www.tweedfoundation.org.uk 21

Salmon: Fish Counters Egg Deposition Rates While annual totals are useful for monitoring changes in abundance, using length distributions to estimate egg deposition provides a more informative assessment tool as it is total egg deposition rather than simply numbers that will affect fry recruitment in the following year. Significant changes in run timing and sizes of adult Salmon are evident in rod and net catches, but each method of fish capture is susceptible to biases in the sizes of fish that are caught 1. The Tweed fish counters provide us with the opportunity to study the effect of changes in abundance, size and egg deposition on recruitment without these biases. A combination of radio tracking in the early 1990’s, catch records, genetic data and scale readings provide evidence that Ettrick is predominantly a Spring stock and the Gala is Summer stock. Analysis of catch records shows that three sea winter Spring Salmon are becoming more common but more significant changes are taking place in the Summer stock, with greater increases in two sea winter fish and an upward trend in catches. To see whether this tallies with the fish counter data, the lengths of Salmon split up into size intervals for each counter (the first two graphs on the following page) show the annual variability in sizes of fish based on the underlying proportions of one, two and three sea winter fish and the variability in lengths for each of these cohorts. The 2020 size distribution for the Ettrick is within the range of proportions found in previous years but the Gala is distinctly different, with a sizeable increase in numbers of fish over 80 cm. As the relationship between fish length and number of eggs (fecundity) is curved upward rather than a straight line relationship, an increase in numbers of fish over 80 cm has an exaggerated (positive) effect on egg deposition which is shown when estimated eggs per female fish are plotted for the Ettrick and Gala (graph over page). 5,510 eggs per female fish was estimated in 2020 for the Gala Water relative to an average of 3,483 eggs before then; the increase in eggs per fish reflecting the marked increase in the average length of fish from 63 cm (2008 to 2019) to 74 cm in 2020. The effect of many more fish in the 80 cm plus size category is that the 2020 estimated egg deposition is double the estimate for the next highest total in 2008 of 6,621,376 eggs. The increase for the Ettrick is less marked, but the 2020 estimate is still the second highest on record. When the estimated number of eggs per female is then converted to estimated egg deposition per 100m2 for the streambed area upstream of each counter, the estimated egg deposition is 1,042 eggs per 100 m2 for the Ettrick catchment and 3,076 egg per 100 m2 for the Gala (graph over page). While the figures of 250 and 500 eggs per 100 m2 are reference points rather than targets specifically calculated for the Ettrick Water, a figure of 800 eggs per 100 m2 has been calculated for the North Esk (considered a productive river). The egg deposition target used for the national assessment carried out by Marine Scotland Science is 274 eggs per 100 m2. It therefore appears likely that egg deposition for both catchments exceed the minimum requirement to maximise smolt output. A photograph showing the contrast in the numbers of eggs a small (left) and large (right) female fish can carry. 1 Presented on youtube by Dr Campbell (https://www.youtube.com/watch?v=MumWKZNmMuQ&t=48s) 22 www.tweedfoundation.org.uk

Salmon: Fish Counters Fish counter Gala lengths graph. Fish counter Ettrick lengths graph. Fish counters – Egg deposition graph Ettrick vs Gala.eggs per female fish. Fish counters – Egg deposition graph Ettrick vs Gala eggs per 100m2. The Wider Context In the context of recent changes in run timing, the increased numbers and sizes of early Summer Salmon in 2020 correlate with the increased number of fish detected by the Gala fish counter. The less significant increase in egg deposition for the Ettrick could be because more of this stock is from earlier running Spring fish that enter the river from February to early May and have not undergone the changes detected for Summer Salmon. Although less signficant, the increase in numbers of Ettrick Salmon and the corresponding egg deposition in 2020 relative to 2018 and 2019 could, in part, be attributable to the increased summer run in 2020. With the rod catches showing a downward trend in Spring fish in February, March and April, a decrease in abundance of Springers in this period (assuming most upstream of the Whiteadder return to the Ettrick) could be disguised by an increase in early Summer fish, which also spawn in this catchment. Unfortunately genetics does not currently allow us to differentiate Spring and Salmon stocks and it is quite possible that there is interbreeding of these stock components, making spawning target estimation even more difficult. For both catchments, it will be very interesting to see whether the increased egg deposition will lead to increased fry numbers in 2021. With our developing ability to estimate smolt output for the Gala Water, we will be able to investigate the effect of high egg deposition on recruitment and subsequent numbers of adult returning to the Gala. With at least ten years of data, we will then be able to start building a stock-recruitment graph that will be able to give us a more informed assessment of how many adults are needed to maximise production, rather than relying on egg deposition targets from other rivers. www.tweedfoundation.org.uk 23

Review: Tweed Salmon Catch Trends This article is based on a more detailed document which will be posted on the new Tweed website when this is set up. Over the last 200 years or so, the nature of the Tweed’s salmon fisheries has changed several times. Sport fishing for salmon can be said to have started in its modern form here with the arrival of the railways in the 1850s and anglers then expected Autumn salmon fishing, mainly after the end of the netting season on the 14th September. It was with some concern therefore that in the early 20th century it was realised that Grilse numbers were falling to levels not seen before: “In Scotland, certain rivers which used to hold abundance of Grilse do not appear now to do so. In the Tweed quite recently, as I am informed, the absence of Grilse has been evident to all” 1. This was not just a Tweed phenomenon, J. Arthur Hutton, the pioneer scale-reader, writing of the R. Wye could say “In 1912, I drew attention to the falling off in the numbers of Grilse and 1913 seems to indicate that this tendency is still going on. Mr. Calderwood informs me of a similar falling off in practically every river in Scotland in 1913. So far no explanation is forthcoming as to the cause” 2. The effects of this change and others on angling can be followed in the annual reports on Tweed salmon fishing by the Rev. Dr. William MacCullum in The Scotsman: 1921: BEST SPRING AND WORST AUTUMN FISHING ON RECORD – One regrets the failure of the Tweed as an Autumn river……. Since 1859 Tweed spring fishing has begun on February 1, but did not become a distinct feature of the river until 1911 ….. who can say why the lively little fish appear in the river so early in the year, and who can tell what they are? Are they delayed grilse? If not, the disappearance of grilse is another of the puzzling problems connected with the salmon. …… Have we killed out the grilse race or have they changed their habits? Are they remaining longer in the sea and running up the river in the following year as spring salmon?..... 1922: UNFAMILIAR GRILSE – The netsmen reported a run of grilse this year. There was a time when these little fish of the same hatch as the small springers of the coming season were more numerous than the salmon in the Tweed, but they had almost disappeared in recent years, and seemed to be staying in the sea until they were of the status of salmon. Why have they been pushing up the river in the grilse stage this year? It has been suggested that the scarcity of herrings and the absence of the normal crustacean fauna may have to do with the re-appearance of grilse in fresh water, but who knows.” The gloom about Grilse was forgotten, though, in the report for 1935: A WONDERFUL YEAR – a hundred years hence men may be talking about the Tweed in 1935 as an example of the good old days. It certainly has been a wonderful year …… Some say there has never been such a spring season on the Tweed, but “vast shoals of salmon” were in the river in the opening months exactly a hundred years ago. …. But it is almost certain that no living man can look back on bigger bags than those of last spring. Lady Joan Joicey landed 40 fish at one outing, returning a dozen kelts and retaining 28 clean, clear, salmon. On the same day, February 15, the Tweed and Teviot yielded more than 200 springers. Rod catches continued to be mainly Spring (though with heavy Autumn Salmon also a feature) until 1968 when the proportion of the annual total caught before the 1st September first dropped below 50% as Autumn catches, particularly of Grilse, increased. It then stayed consistently below the 50% level from 1973 onwards, though good catches in February and March continued into the 1980s. Change came again in 2016 when for the first time since the 1970s the proportion of the annual catch taken on the river before the 1st September went back above 50%. The RTC’s rod catch records for the Tweed only go back to 1947 and are only reasonably complete from the late 1960s, so for a long data series, records from individual beats have to be used. The catches from a number of beats in the Coldstream area can be put together to give a data series from 1865 to 2020. However, as different years have different number of beats contributing data, these results have to be given as percentages rather than numbers (Figure 1). The features noted in The Scotsman reports can be seen on this graph, the decline in the Autumn catches around the start of the 20th century; the unexpected Grilse run reported on in 1922 fits with the variability around that time, and then the rise of Spring catches into the 1930s. Although there are two “Late Phases” apparent on this graph, when most fish were caught after the 1st of September, they were not the same. In the first, in the latter 19th century, the fish were (roughly) half and half Autumn Grilse and Salmon; in the second, at the end of the 20th and the beginning of the 21st century, the fish were mainly Autumn Grilse. During the “Early Phase” of the middle 20th century, when most fish were caught before the 1st of September, they were 1 Calderwood, W.L., 1909: Decline of Grilse, its Significance. Appendices to the Twenty-seventh Annual Report of the Fishery Board for Scotland HMSO Edinburgh. 2 Hutton, J.A., 1914: Results of the Wye scale reading 1909-13. Salmon & Trout Magazine No. 7. 24 www.tweedfoundation.org.uk

Review: Tweed Salmon Catch Trends Figure 1: Rod catches in the Coldstream area 1865-2020 as percentages of the annual total taken before and after the 1st September. Figure 2: Summer (June-August) and Autumn (September-November) catches and trends on the River Tweed 1980-2019. 25 www.tweedfoundation.org.uk

Review: Tweed Salmon Catch Trends mainly Spring Salmon. [see Section B of the online document for details of the catches at different times] At present, it looks as if another “Early Phase” is starting (Figure 2), but it does not therefore follow that the fish will be mainly Spring Salmon again: in fact, it looks as if they will be mainly Summer Salmon. If this is the case, then wet Summers will be needed to give good catches. Figure 2 clearly shows the issue for the fishery: while numbers in the Autumn have decreased by thousands, numbers have (so far) only increased by hundreds in the Summer. The increase for Autumn 2020 comes almost entirely from the September catch, a month of mixed Summer and Autumn fish. Trends in Sizes and Sea-Ages of Fish It is not only in numbers and in run-timing that salmon can change, their size (reflecting time spent at sea) also varies and can also explain much of the other changes. The more dramatic changes in sizes in recent years are shown in the following diagrams which have been produced from the catch records of a lower Tweed beat for recent decades and parts of decades, 1985-1994; 1995-2004; 2005-2014 and 2015-2020 (six years only). The diagrams for other periods can be found on the website document. The proportions of the fish of each sea-age [1 Sea-winter (1SW), 2 Sea-winter (2SW), 3 Sea-winter (3SW)] for each period are given in the tables below each graph. When most fish were killed in the 1980s and 1990s, many more scale samples were taken and so reflected the population in the river. As fewer fish were killed, the number of scale samples decreased and so they do not now necessarily reflect the population. They can still, however, suggest explanations for changes in the population so, for example, in some months there are larger fish than in the past and at the same time the scale samples for those months have more 3SW fish in them, suggesting that the increase in larger fish is due to more older fish rather than to the fish just getting bigger. In May and June, Figure 3 shows that larger salmon have become more common in these months in recent years: fish over 12lbs were 7.8% for 1985-1994; 3.4% for 1995-2004; 8.9% for 2005-14 and 15.0% for 2015-2020 and they have also become more numerous. The proportion of 3SW fish in the scale samples also increased from 2005 to be almost 30% which matches with the increase in size. Figure 3: Sizes of fish caught in May and June, months with upward catch trends and that are almost all Salmon with only a very few Grilse. 26 www.tweedfoundation.org.uk

Review: Tweed Salmon Catch Trends Scale-readings: 1SW 2SW 3SW n= 1.41% 98.59% 0 71 May & June 1.11% 95.56% 3.33% 90 1985-1994 1.52% 80.30% 18.18% 66 1995-2004 0 70.83% 29.17% 24 2005-2014 2015-2020 For July and August, Figure 4 shows that larger salmon have also become more common in these months in recent years: fish over 12lbs were 11.6% for 1985-1994; 7.0% for 1995-2004; 12.3% for 2005-14 and 17.2% for 2015-2020. Numbers have also considerably increased for these months. The commonest sizes have also changed: for 1985-1994 and 1995- 2004 these were 4 to 6lbs, after 2005-2014 they were 6 to 12lbs. This happened because the type of fish being caught changed from mainly Grilse in the 1980s and 1990s to mainly Summer Salmon in the 2000s: the scale samples show that while 70-80% of the readings were Grilse before the 2000s, 50-60% are now Salmon. Figure 4: Sizes of fish caught in July and August, months with upward catch trends, and that have had significant changes in the sizes of fish caught. Scale-readings: 1SW 2SW 3SW n= 70.67% 29.33% 75 July & August 77.68% 22.32% 2.59% 112 1985-1994 43.01% 54.4% 2.27% 193 1995-2004 40.91% 56.82% 44 2005-2014 2015-2020 (The graphs for other times of year are given in the online document) www.tweedfoundation.org.uk 27

Review: Tweed Salmon Catch Trends The present situation is, therefore, that while the total number of fish caught in the Tweed has dropped considerably over the last 7 years, presumably reflecting fewer fish returning from the sea, size and age analyses show that this is not a random “collapse” it is a structured change, a sign that the fish are adapting to changed conditions in their environment. In this case it means more fish are staying longer at sea and so more return as Salmon rather than Grilse and earlier in the season than later. While the drop in Autumn catches has decreased overall totals, catches in Summer are increasing, 2020 showing a remarkable rise in June and July – and not only are there more fish in Summer, they are now mainly larger Summer Salmon instead of being smaller Summer Grilse. This means a great increase in the number of eggs being spawned by this type of salmon just as the number being produced by Autumn fish decreases – what this means for Tweed salmon will become apparent in future seasons. The shift from Grilse to Salmon has implications for the number of fish returning to the Tweed. Salmon spend longer at sea than Grilse do, and so have more chances of dying there, which means they have a lower return rate. The effect of the proportion of Grilse on fish numbers can be seen in Figure 5 which shows the numbers of fish caught at the Sandstell netting station, which was the best in the Tweed estuary, from the middle of the 18th century up to the end of its individual records in the 1970s. The really large catches were when the Grilse proportion was high and conversely during the Grilse “depression” 3 (as it has been called) of the 1920s-1930s the annual catches averaged less than a thousand fish a year. As the Grilse proportion started to increase after the “depression” the numbers also increased. Figure 5: The numbers of fish caught at the Sandstell netting station from the 1740s to the 1970s and the proportion of Grilse in the catches. This illustrates an important point – not all decreases in salmon numbers are due to human impacts on their rivers, there are large-scale natural changes and trends that can have significant effects on their numbers as well. The key point is to be able to distinguish man-made impacts which can actually be repaired, from natural impacts which can only be mitigated at best. 3 Menzies, W.J.M. & G.C.J. Smart, 1966: Salmon Runs in Scotland. The Salmon Net II. 28 www.tweedfoundation.org.uk

Review: Tweed Salmon Catch Trends Similar trends to those seen on the Tweed have been found to the south and reported on in the Environment Agency’s Salmon Stocks and Fisheries in England and Wales 2018 4. This shows [Fig 17, page 42] how catches on a number of rivers in England have changed from being more Grilse than Salmon to being more Salmon than Grilse, and that this process started in 2012. This trend is general and is being followed in all the regions of England and Wales [Fig 18, page 43] where MSW salmon are increasing both in numbers and in proportions of catches. The graph of total catches [Fig 19, page 44] shows how the majority of fish caught in England and Wales are now Multi Sea Winter Salmon after the recent steep decline in Grilse. This is a key graph as it confirms that the current fall in fish numbers is not a random collapse, it is structured: it is Grilse that are declining, while salmon numbers are stable, or increasing. All these trends are similar to what is being seen in the Tweed analyses. Unfortunately, no similar data is available for anywhere in Scotland for comparisons to be made with the north. (The EA diagrams can also be seen in the website document). The obvious question, of course, is what drives such large-scale changes? The answer seems to be Sea-surface temperatures (SSTs). When these are colder than average in the north Atlantic, Grilse predominate and when they are warmer, Salmon do. A long data series of Grilse to Salmon proportions can be worked out for the Sandstell and other estuary netting stations as shown in Figure 6. Figure 6: The average Grilse to Salmon ratios per decade for Tweed estuary nets 1740-2004. The different phases apparent on this graph are: c1740 – c1800: more Salmon than Grilse c1800 – c1870: more Grilse than Salmon c1870 – c 1910; about equal proportions c1910– c 1970: more Salmon than Grilse c1970 – to end of series, more Grilse than Salmon These can be linked with warmer and colder periods in sea-surface temperatures. The 1860-1900 period of “equal” Salmon and Grilse interestingly links with a period when sea-surface temperatures were very variable, warmer and cooler 4 https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/919734/SalmonReport-2018- 29 assessment_final.pdf www.tweedfoundation.org.uk

Review: Tweed Salmon Catch Trends for a few years at a time. It looks as if this “confused” the fish, the environment not giving a strong lead as to whether the Grilse or the Salmon route led to better breeding (see the website document for more details). However, other rivers in Scotland seem to have had more Grilse at this time. The conclusion of this review, inevitably, has to be that there are no fixed baselines for Tweed salmon, not in their numbers, nor sizes, run-timing, sea-age or any other characteristic, only change is constant. This is nicely illustrated by another Scotsman report, of the 1st of February, 1922: “Times change and the habits of the fish with them. In 1784, thirteen years after the Tweed was subject to special legislation, those engaged in the administration of the Acts of Parliament passed a resolution that ….. “the salmon taken in the Tweed since the year 1776 are greatly short of the usual number” but in the end of the century the Earl of Home wrote:- “The greatest number of salmon ever caught by a rod in one day was on April 9, 1795 by Lord Home….the number caught thirty-six, the weight of the fish ran from 6lb to 36lb”. And just as the fish must adapt to changing conditions, so anglers must adapt to changing fish. Salmon Fishing Museum The River Tweed Salmon Fishing Museum in Kelso Town Hall tells the story of rod and line salmon fishing techniques which were developed on the Tweed during the middle of the 18th century and which influenced the economic and cultural development of the Eastern Borders. The Tweed Foundation’s Dr Ronald Campbell contributed to the museum’s displays of more than 2,000 objects relating to the history and heritage of salmon fishing on The Tweed. Ranging from maps, models, books, illustrations, film and photographs through to rods, reels, fishing flies and a replica 19th-century fishing bothy – complete with a tweed-clad boatman! In pride of place is the newly-commissioned carving of a 69¾ lb salmon caught on the Tweed by the Earl of Home in circa 1735 – the sport’s largest rod caught British salmon for which there is credible evidence. As well as highlighting the fascinating history of rod and line salmon fishing, it is hoped that the museum will also help to raise awareness and appreciation of the river, its heritage and fragility, and of the need for all of us to help to play our part in its conservation. Created by a team of volunteer salmon fishing experts and enthusiasts from the region, the project took over three years to plan and develop with financial support from private donations, the Fallago Environment Fund and Scottish Borders Council. Entry is free. www.salmonfishingmuseum.com 30 www.tweedfoundation.org.uk

Bird Predation: What is Being Done? Rationale: There are significant populations of fish-eating birds on Tweed and, as Smolts are the ‘end product’ of up to three years of freshwater life, any loss to predation is irrecoverable and results in fewer salmon even having a chance of returning from the sea. While progress continues to be made in investigating the survival of Salmon smolts through the Tweed tracking study, work on other aspects of bird predation relating to dietary analysis, population estimation and reviewing the licence procedures has been slow. The stomach contents of 36 Goosanders and 36 Cormorants were submitted for analysis as part of a study on four rivers in Scotland at the end of 2019 but unfortunately COVID restrictions have limited the ability of the contractor to complete the work on time. A preliminary report for samples taken during the smolt run has been produced, including comments provided by The Tweed Foundation, but this cannot be circulated at present. Longer term, we are interested in the possibility of EDNA sampling which is currently being trialled by The Atlantic Salmon Trust as part of the Moray Firth Tracking Project. This would involve using faecal samples of both bird species, with genetic analysis being used to estimate dietary composition, potentially a quicker method of sampling, but we will have to wait and see whether abundance of different species can be accurately measured. While local counts of Goosanders and Cormorants are important for making an assessment of economic damage to the fishery and to monitor the numbers, it is also important to have an estimate or index of the national population. Tweed and other fishery boards use rigorous counting methods but unfortunately this data is not currently used for estimating the national population. To address this issue, Marine Scotland Science is currently analysing data from a number of rivers in Scotland and we hope that data collected by fishery boards will be used in future. Tweed Bird Counts Despite COVID restrictions from March through to June, we still managed to carry out our regular bird counts between Ettrick Mouth and Berwick in January, April, May and October, although the Spring counts were carried out by vehicle rather than foot, meaning a few birds will have been missed. The graphs on the following page indicate the annual variability in numbers of Goosanders and Cormorants on the river. Counts of Goosanders in January and October show much greater fluctuations than for Spring time. For example, nearly seven times the numbers of Goosanders were counted in October 2010 compared to October 2020. Given this level of variability, it is difficult to detect any conclusive trends, but certainly since 1994 there is no evidence that the Goosander population is increasing in any of the months we carry out counts. It is interesting to note that all of the Goosander counts were below average in 2020, particularly for January and October when bird numbers are normally at their highest. For the October count, numbers in 2019 were a third of the long term average (101 vs 299). We do not know why Tweed numbers were less this year; as bird numbers depend on the balance of immigration and emigration plus recruitment from breeding summer breeding, it is difficult to determine the root cause for these low results. Several other large east coast rivers have confirmed their Goosander counts were similar to previous years. Apart from a few exceptions, Cormorant numbers are lower than Goosanders in all months of the year, although it is important to remember that being twice the size of a Goosander, they can eat twice the quantity of fish. Occasionally there are spikes in numbers, perhaps due to poor feeding conditions at sea pushing them inland or an influx of birds from the continent. In response to a build-up of Cormorants at a roost on the Middle Tweed, regular counts are now carried out at all of the main roosts. Fishery officers now have night vision binoculars that assist with counting at dusk or night time. The chart on the following page shows the results for monitoring at the Middle Tweed roost, with cyclical patterns of high numbers through winter and a drop in numbers from January through to September due to most birds leaving to either breed or feed in the marine environment. For the four winter periods that have been sampled so far, the highest numbers were recorded over the winter period in 2017 and 2018, followed by lower numbers in 2019 and 2020, which correlates with the recent changes in main river counts. www.tweedfoundation.org.uk 31

Bird Predation: What is Being Done? Bird count results for January. Bird count results for April. Bird count results for May. Bird count results for October. Bird predation Cormorant Roost chart. The application of a bird deterent laser is detailed in the 2019 annual report. 32 www.tweedfoundation.org.uk

Environment: Water Temperatures – SRTMN Rationale: With forecasts of changes in rainfall, flow patterns and temperature, it is important to try and forecast impacts on fish habitats so that any step to mitigate these changes can be identified. The Tweed Foundation’s progress and participation The shade provided by mature trees in the lower Leader Water reduces summer temperatures. within the Scottish River Temperature Monitoring Network (SRTMN) was disrupted considerably as a result of the COVID-19 restrictions during 2020. The project, which is led by Marine Science Scotland (MSS) predicts river temperatures across the whole of Scotland and allows Fisheries Trusts to identify areas within river catchments where high water temperatures are already, or are becoming, an issue for trout and salmon. Local Fisheries Trusts can then explore the potential to reduce temperatures by planting trees to create shade within these areas. The Tweed Foundation, along with several other Scottish Fisheries Trusts, collects data from a number of water temperature loggers spread throughout their catchments. This data feeds into a model that was created by MSS which can make the temperature predictions. In January and February 2020, several field visits were made to sections of river identified by the model as being vulnerable to rising temperatures. Following on from the field surveys the Tweed Foundation in partnership with the Tweed Forum as lead on the Riverwoods project will contact land owners/managers in a bid to create more riparian woodland. The main focus will be to bring about targeted riparian woodland to improve the fishery and enhance biodiversity. Delivery of the Riverwoods project in Scotland will be overseen by a working group consisting of Tweed Forum, Fisheries Management Scotland, Woodland Trust, Scottish Wildlife Trust, Scottish Land and Estates and Scotland, the Big Picture. In addition to the disruption described above, the Spring Covid-19 restrictions resulted in the Tweed Foundation being unable to download data from many of the temperature loggers during the Spring temperature data download period. Graph 1 – Water temperature readings (at 15m intervals) for the River Tweed at Makerstoun during Spring, Summer and Autumn 2020. Note that temperatures peaked at 22.5ºC during late June. Trout and Salmon begin to experience stress at temperatures over 20ºC. www.tweedfoundation.org.uk 33

Trout: Trapping Spawning Burns Rationale: How do Brown Trout relate to Sea Trout? Trapping spawning burns shows what fish are spawning in them and whether these are all Sea Trout; all Brown Trout or various combinations. Identifying trout at traps will show if they return to the same burns repeatedly. The numbers of juvenile trout upstream of traps can be related to the numbers of spawning adults caught in them. The temporary trout traps in the headwaters of the Upper Tweed and Gala Water, which are used to monitor trout spawning populations, were not deployed during 2020. This gave the opportunity to try something different at new locations during the 2020 spawning run. Instead of using traps to sample adult trout, electro-fishing was used in late October and early November to intercept trout as they ran upstream to spawn, with the Leithen Water and a burn that flows into the lower Gala Water being sampled. Electrofishing was chosen as suitable sampling method as it has been successfully used to catch mature trout on other rivers. The sample of trout from the burn in the lower Gala Water was relatively small (22 trout), although it should be noted that as a result of the sampling method and frequency only a modest proportion of the spawning population would likely have been caught. A larger sample from multiple years is required to carry out a more thorough analysis of the data and the burn will be visited again in 2021. A much larger sample (64 trout) was collected from the Leithen Water, although again this will only have been a small proportion of the breeding stock. Given the sampling method small Brown Trout are also more likely to evade capture than larger trout and may be under represented in the results. The relationship between Brown Trout and Sea Trout from the Leithen Water is similar to that seen in other burns downstream of Peebles that have been trapped and greatly resembled the results from the nearby Glensax Burn. As with these other burns the males of the spawning population are predominately Brown Trout, with only a small percentage of Sea Trout. In contrast the females are predominately Sea Trout, with a much smaller percentage of Brown Trout. Chart 1 shows the length frequency for the Leithen Water population whilst Chart 2 shows the Glensax Burn population. Due to the different growth patterns Brown Trout (predominately under 450mm) can clearly be distinguished from Sea Trout (predominately over 450mm) within these length frequency charts. A typical female trout (Sea Trout) from the Leithen Water spawning population. 34 www.tweedfoundation.org.uk

Trout: Trapping Spawning Burns The percentage of male and female Brown Trout and Sea Trout within the Leithen Water spawning population Male Brown Trout Sea Trout Female 47% 14% 14% 25% Chart 1 – Length frequency chart for the Leithen Water trout spawning population. Chart 2 – Length frequency chart for the Glensax Burn trout spawning population. 35 www.tweedfoundation.org.uk

Trout: Trapping Spawning Burns As documented in previous reports, this relationship is quite different to that seen in burns at Tweedsmuir near the headwaters of the Tweed. Spawning within these burns is dominated by long lived and late maturing (and therefore large) Brown Trout. As a result, the area between Peebles and Tweedsmuir is a priority for trapping going forward. Chart 3 – Length frequency chart for the trout spawning population from a Tweedsmuir burn. Almost all of the trout in the population are Brown Trout. 36 www.tweedfoundation.org.uk

Brown Trout: Acoustic Tracking Rationale: Do Brown Trout stay in much the same areas all the time or do they move around and if so, is it for long or for short distances? This is an important consideration for the trout fishery as it needs to be known where trout that the anglers in any particular part for the catchment actually come from – are they local fish or from further away? The fish counters on the Ettrick and Gala show Brown Trout running upstream in spawning time which suggests at least some large scale movement. As a result of the uncertainty regarding COVID-19 restrictions in the first half of the year the decision was taken not to purchase acoustic tags for Brown Trout tracking during 2020. As such, no new trout were tagged. However, up to eight tags from Brown Trout tagged during 2019 were potentially still active during 2020, if the fish had survived the 2019/2020 winter. Tracking Brown Trout from the Gala Angling Association Water Acoustic tagging within the Gala AA water takes place during the trout fishing season with Brown Trout over 40cm caught by volunteer anglers being tagged. The tagged trout are tracked to the rough area of their spawning grounds during their migration at the end of the season using tracking receivers spaced out along the length of the Tweed. The purpose of the tagging is to identify if any areas of spawning are of particular importance to the Brown Trout caught within the Association’s water. Six Brown Trout from the Gala Angling Association stretch of the Tweed were tagged during May and June 2019. Of those six trout, four returned from their spawning grounds in November and December 2019 and were potentially alive within the Gala AA water during the 2020 trout fishing season. Of these four trout, two were detected making spawning migrations in autumn 2020. The first of these two trout (tag #4970, 42cm when it was tagged in June 2019) spawned between Stobo and Tweedsmuir in Autumn 2019. During its 2020 spawning migration it appeared to be making a repeat migration before disappearing somewhere between Innerleithen and Peebles. It is possible that the trout spawned in a different location in 2020. However, no further tag detections have been recorded which suggests it was taken by a predator. The second trout (tag #4969, 41cm when it was tagged in June 2019) has been “leading us on a merry dance”. In Autumn 2019 the trout undertook a migration away from where it was tagged, only to return the best part of two months later. Movements with this timing and duration usually indicate a spawning migration. In this instance what made the migration unusual was that it was downstream instead of upstream. In Autumn 2020 we were expecting the same downstream migration once again. However, this time the trout moved upstream and appeared to spawn somewhere between Stobo and Tweedsmuir, before again returning to where it was tagged 56 days later. The migration in 2020 looks more likely to have been the spawning migration. We will never know for sure why the trout moved downstream during 2019 only to return after a few months. Our best guess is that the trout was exploring for a better territory whilst other trout were away spawning, before deciding it was better off where it was. A 42cm Brown Trout (tag #4970) which was caught and tagged within the Gala Angling Association stretch of the River Tweed on 4th June 2019. It successfully spawned in autumn 2019 and returned to the Gala AA water. It started the same migration in autumn 2020 but disappeared from detection before reaching its spawning grounds. www.tweedfoundation.org.uk 37

Brown Trout: Acoustic Tracking Tracking Brown Trout Migration Post Spawning in a Tweed Headwater Burn For a number of years a sample of Brown Trout kelts (17 in total) dropping downstream from a spawning burn at Tweedsmuir have been trapped and acoustic tagged. As with the Gala AA tagging, the tagged trout are picked up by tracking receivers spaced out along the length of the Tweed. The burns that flow into the Tweed at Tweedsmuir have been identified as being important spawning grounds for large Brown Trout. The purpose of the tagging is to record the downstream distribution of these trout and assess their importance to the catches of different Angling Associations on the main stem of the Tweed. Four Brown Trout kelts between 43cm and 50cm were tagged post spawning in November 2019. Of the four trout only one (tag #4973) returned to Tweedsmuir to spawn in 2020. After spawning in 2019 it had moved downstream fairly quickly to the Tweed near Galashiels, where it had spent much of 2020, before returning to Tweedsmuir at spawning time. After spawning it once again moved downstream to the Tweed near Galashiels. Whilst one return in four may seem quite low, over the four years we have been acoustic tagging at Tweedsmuir the survival from one year to the next is roughly 50:50 (although the sample size is still relatively small). As such, from only four trout tagged in 2019, one return is well within the natural variation that would be expected from a sample size of only four fish. A female Brown Trout (tag #4973) which was caught and tagged on 8th November 2019 whilst leaving a spawning burn at Tweedsmuir. After being tagged she migrated downstream to the Tweed near Galashiels where she stayed until early autumn 2020. In early autumn 2020 she migrated upstream and returned to the same burn to spawn in October. She quickly returned to the Tweed near Galashiels post spawning. 38 www.tweedfoundation.org.uk

Brown Trout: Angling Catches Rationale: Monitoring the catches and fishing efforts of a sample of anglers is the only way in which a reasonable indication of the state of the Brown Trout stocks and fishing can be produced at present. The recording of the amount of effort and the methods that produced catches also means that they can be compared over the years. The sizes and ages of the trout being caught are also indicators of the state of stocks. Without a sound and reliable knowledge of the past nature and state of the Brown Trout stocks of the Tweed it is impossible to properly evaluate their present state. In total 33 catch log books were returned for the 2020 Brown Trout fishing season, covering 298 fishing trips and 842 hours of angling effort across 13 Angling Association waters. This is down on the return rate for previous years, most likely as a result of limitations put on ticket sales by some Angling Associations due to the Covid-19 restrictions which were in place when angling opened up in late May. As a result angling effort was most likely lower than would normally be the case in some areas. River Tweed The catch rates for Brown Trout over 25cm (10”) from the Upper, Middle and Lower Tweed in late May were amongst the highest since the current recording scheme began in 2006, whilst the catch rates for June were the highest on record. The May and June catches likely benefitted from the river being rested during April and most of May, and from reduced angling pressure in some areas. However, the catch rates for Brown Trout between 25cm (10”) and 35cm (14”) were elevated well above the ten year average. Catches of bigger and smaller trout remained relatively close to the ten year average. This indicates a higher abundance of 25-35cm Brown Trout than would normally be the case and the likelihood that the catches in the first half of the trout season would have been above average, even if the 2020 season had proceeded as normal. In contrast to those in May and June, the catches in July were much more variable between different areas, and were generally around or below average during August and September. Different catch trends are often observed between the first and second half of the season as the catch rate of Brown Trout over 30cm (12”) tends to drop off markedly after the end of June. Catches of Brown Trout between 25cm and 35cm were well above average in many Angling Association waters during the 2020 trout fishing season. www.tweedfoundation.org.uk 39

Brown Trout: Angling Catches Tributaries Unlike the Tweed, which saw the same trend across all areas, catch rates of Brown Trout over 25cm (10”) were much more variable between different Tweed tributaries during the 2020 fishing season. Catch rates were good on the Whiteadder, again as a result of catches of trout between 25cm - 35cm (10” – 14”) that were well above the 10 year average. In stark contrast the catch rate for the Teviot was well below average. The catch rates for the Ettrick/Yarrow and Leader Water were slightly above average. The catches of undersize trout (<25cm (10”)), which dominate the catches of many Tweed tributaries, also varied between areas with above average catch rates on the Leader, average catch rates on the Whiteadder and below average catch rates on the Teviot and Ettrick/Yarrow. The release rate of “takeable” size Brown Trout across the whole Tweed catchment remains at a very high level. Of the 550 Brown Trout over 25cm (10”) that were recorded within the catch log books none were killed. Catch log book data indicates that catch and release is practiced by most Tweed Brown Trout anglers. Picture courtesy of Scott Hamilton. 40 www.tweedfoundation.org.uk

Brown Trout: Spot Pattern Recognition The Brown Trout spot pattern recognition programme continued in 2020. As the programme mostly records adult sized Brown Trout over 12” (30cm), and as the highest catches of trout of this size are in the first half of the season, the COVID-19 lock-down in April and May had a significant impact on the 2020 recording. In essence, it shortened the main working period from three months to one. Despite this 17 participating anglers were still able to send in 191 pictures of Brown Trout over 12” from their Association waters. The spot pattern on the left gill cover was recorded for each trout and stored in a database, allowing individual trout to be identified and providing information on movements and recaptures. In general, the recaptures and movements during 2020 mirrored those seen in previous seasons, but to a lesser degree due to the shortened season. These have been reported on in detail in previous annual reports. Given the high rate of recaptures within some Angling Association waters (as reported previously) one of the recommendations from the spot pattern recognition programme is catch & release for Brown Trout over 12” (30cm). As it turns out catch and release for trout over 12” is, for the most part, already in place within the Tweed district. Many Angling Associations already protect most Brown Trout over 12” through the size limits recommended by The Tweed Foundation whilst a move towards a culture of catch & release amongst Tweed Brown Trout anglers can be clearly seen within Brown Trout catch records. With catch and release largely being practiced the onus is now to promote best practice. With four years worth of large numbers of pictures having been sent in by anglers from the Peeblesshire Trout Fishing Association and Gala Angling Association we now have a large sample for two different Angling Associations that can be used to look for signs of damage caused by a previous angling capture. In almost all of the angling recaptures that have been photographed by the volunteers the trout have shown no obvious signs of damage. Whilst some forms of damage are subtle and may go undetected if a fish is recaptured, this still suggests good practice amongst the anglers submitting pictures. However, in some of the pictures that have been sent in there are signs of damage caused by previous captures by anglers who have not participated in the programe (this is known because the fish had not previously been photographed). The most obvious sign of damage caused by a previous capture is healed damage to the maxilla (see the pictures at the bottom of the page and the top of page 43) which feedback from anglers would suggest is likely to have been caused by the removal of a barbed fishing hook. As the maxilla is much tougher than most of the surrounding tissues barbed hooks are more likely to be difficult to get out when fish are caught in this place. Between 5.4% and 6.6% of photographs sent in by anglers from the Peeblesshire TFA and Gala AA respectively had damage to the maxilla (Table 1). It is assumed that trout are equally likely to be hooked on both sides of the mouth and as the pictures only show one side of the mouth the actual rate of damage is likely to be double the rate recorded within the photographs. A picture of a Brown Trout showing the position of the maxilla. 41 www.tweedfoundation.org.uk

Brown Trout: Spot Pattern Recognition A picture of a Brown Trout from one of the Tweed Foundation fish traps with damage to the Maxilla caused by an angling capture. Table 1 – The percentage of Brown Trout that had damage to the Maxilla in pictures sent in by anglers from the Peeblesshire TFA and Gala AA Association Percent Damage Percent Damage Peeblesshire Left side only Both sides (assumed) Gala 5.4% 10.8% 6.6% 13.2% Whilst the Brown Trout photographs that have been sent in show that a high percentage of anglers are practicing catch & release to a standard that has no obvious long term impact on the trout, there is still room for improvement. The Foundation encourages all angler to visit the Keep Fish Wet website www.keepfishwet.org/ which provides a wealth of information on best practice catch and release. The Tweed Foundation would like to thank all of the anglers who sent in pictures during the 2020 Brown Trout fishing season. 42 www.tweedfoundation.org.uk

Grayling: Angling Catches The Earlston Angling Association‘s catch & release Grayling (January) and Inter-club (August) Competitions provide our main source of Grayling catch data each year. Unfortunately water levels in January and COVID-19 restrictions in August resulted in both competitions being cancelled and the loss of the associated catch data. This left the Grayling catch from the Brown Trout catch log books as the main source of Grayling catch data for 2020. Although less detailed, it is still possible to compare the 2020 Grayling catch with previous years. Reports from anglers were of very good Grayling fishing during 2020 and the angling catch data supported their accounts. Of the four areas where there is enough Grayling catch data to carry out analysis the Upper Tweed, Lower Tweed and River Teviot all recorded their highest catch rate of Grayling over 10” (25cm) since the current recording scheme began in 2006. The Middle Tweed recorded its second highest catch rate. Whilst no Grayling size data is recorded within the Brown Trout catch log books, reports from anglers indicated that catches were dominated by Grayling in the 10” to 13” (25-33cm) size category. This size class is largely made up of one year old Grayling and suggests good survival of the 2019 year class. “Strong” and “weak” year classes are frequently seen within Grayling populations, most likely as a result of natural variations in temperatures and flows during important stages of development, and have been found a number of times on the Tweed. Anglers on the River Teviot reported some very good catches of Grayling during the 2020 fishing season. Catches of Grayling under 10”, which are predominately young-of-the-year Grayling fry, were much more variable between different areas, but the Lower Tweed and River Teviot recorded their highest catch rate for Grayling of this size/age. www.tweedfoundation.org.uk 43

Education: Events Snapshot 2020 January The 2020 events started with a visit to the Salmon and Trout Yorkshire Group. A presentation was given on the Tweed’s Grayling and the spot pattern recognition work being carried out on Trout as part of the Tweed Trout and Grayling Initiative. February In February, the Tweed Foundation visited the MCaorlcdh stream and District Angling Association’s AGM Mwahrcehrseawatnheufiprsdt sachteoowl viasist ogf itvheeyneaor nwhtehnethwe TowrekedoFfouthndeatTiown veiseitedd Priorsford Primar STcrhoooultataPneedblGesrfaorytlhineigr WInoriltdiaoftWivoerk. day. Several classes learnt about the work carried out the Tweed Foundation before getting the chance to try their hand at invertebrate identification. The Tweed Foundation also gave an evening talk to the Peebles Wildlife Group which was well attended by local residents. March March saw the first school visit of the year when the Tweed Foundation visited Priorsford Primary School at Peebles for their World of Work day. Several classes learnt about the work carried out by the Tweed Foundation before getting the chance to try their hand at invertebrate identification. The Tweed Foundation also gave an evening talk to the Peebles Wildlife Group which was well attended by local residents. 44 www.tweedfoundation.org.uk

Education: Events Snapshot 2020 September/October Throughout September and October, the Tweed Foundation visited three schools – Newlands Primary in Romannobridge, Kingsland Primary in Peebles and Holy Trinity Church of England First School in Berwick. Each school had two sessions, learning about the invertebrates and fish living in the Tweed. With the exception of one school (for which live samples were taken to the school), all of the sessions were carried out on the river bank. During the invertebrate session, classes learnt about the life cycles and importance of invertebrates in our rivers. They got to sample for invertebrates themselves, carrying out kick sampling before using biological keys to identify what they found. During the second session, the classes learnt about the different fish species living in the Tweed system. By observing live samples and answering questions, the children learnt about the life-cycles of the fish and how to identify each species before learning about the importance of salmon to the Tweed. 2020 Summary At the start of 2020, the Tweed Foundation had planned to run several school projects; host Tweed Start days; partake in shows and events, and carry out work with community groups. Unfortunately, COVID-19 restrictions limited the educational work that could be carried out and much of what was planned was unable to go ahead. However, with the situation around COVID looking to be improving, we hope to be able to work with schools and community groups throughout the catchment in 2021, when restrictions allow. www.tweedfoundation.org.uk 45

Trustees and Staff Stanhope, Upper Tweed As at May 2021 The Tweed Foundation Foundation Staff Drygrange Steading, Melrose, Roxburghshire, TD6 9DJ Jamie Stewart (Director) since October 2020 Tel: (01896) 848271 Fax: (01896) 848277 Alison Gorrie (Executive Assistant) Email: [email protected] Dr Ronald N B Campbell (Policy & Strategy Biologist) Web: www.tweedfoundation.org.uk James H Hunt (Monitoring & Data Biologist) Company No. SC366380 Kenneth A Galt (Trout & Grayling Biologist) Registered Charity No. SC011055 Patrick Barbour (Education Officer) Barry Wright (Scale Reading Biologist) Trustees Principal Bankers Hugh P Younger (Chairman) Douglas J Dobie Clydesdale Bank Lord Joicey 9 High Street, Galashiels, TD1 1RY Richard Onslow (Retired 2020) John P H S Scott Independent Examiners The Duchess of Sutherland W Allan Virtue Rennie Welch Sheena A West Chartered Accountants & Registered Auditors Douglas H Younger Academy House, Shedden Park Road, Kelso, Roxburghshire, TD5 7AL Fund Managers Cazenove Capital Management Limited 18 Charlotte Square, Edinburgh, EH2 4DF Acknowledgements The Tweed Foundation is very grateful to Tweed fishery proprietors, the River Tweed Commission, Foundation Benefactors, private donors, Friends of the Foundation, the Scottish Government, Marine Scotland, FishPal, the local Angling Clubs and Associations, and many others for their financial assistance with our studies. Without this support we would not be able to achieve the very substantial amount of studies undertaken each year on behalf of the River. Thank you 46 www.tweedfoundation.org.uk

Accounts The Tweed Foundation Limited (A charitable company limited by guarantee) StatemTheenTtwoefeFdinFoaunncdiaatlioAnctivities FoSrtathteemyeeanrteonfdFeidna3n1cDiaelcAemctbiveirti2e0s20 for the year ended 31st December 2020 Unrestricted Restricted 2020 2019 Total Total funds funds funds funds £ Notes £ £ £ 203,332 164,337 Income and endowments from 2 163,257 40,075 Donations and legacies 37,793 78,282 4 10,020 - Charitable activities Charitable activities 11,893 25,900 23,017 23,399 Governance and office costs 4,800 - 10,020 - Investment income 278,962 266,018 Other income 3 23,017 - 5 4,800 - 9,975 Total 212,987 65,975 51,147 132,473 Expenditure on 6 5,634 384 6,018 Raising funds 65,190 7 43,387 51,433 33,606 Charitable activities 8,046 900 125,200 Research and conservation - 292,391 Scientific staff 124,300 62,749 42,470 Governance and office costs 62,749 10,107 38,260 16,097 Depreciation, etc 28,153 793,562 Total 228,882 54,778 283,660 Net gains on investments 15,642 - 15,642 809,659 NET INCOME/(EXPENDITURE) (253) 10,944 11,197 Reconciliation of funds 746,692 62,967 809,659 Total funds brought forward Total funds carried forward 746,439 74,164 820,603 Extract from the 2020 Accounts The financial information set out on this page has been extracted from The Tweed Foundation Limited’s accounts, on which Rennie Welch LLP, the independent examiners, reported without qualification. Copies of the accounts are available upon request. www.tweedfoundation.org.uk 47

Aims and Objectives for 2020 ENVIRONMENT: AIM FOR 2020 ACHIEVEMENT FOR 2020 Water temperatures To assist Marine Scotland Science, Achieved. Monitoring of the network installed in 2016 FISH STOCK implement the national network continued and the data shared with MSS, which has STRUCTURE: programme of temperature produced some ‘heat’ maps and identified areas where Genetics monitoring (STRMNS). tree planting may be most beneficial to help reduce water temperatures for juvenile fish. JUVENILE Tissue samples from juvenile salmon Partly achieved. Discussions began with some landowners STOCKS: have been provided to MSS for their in early 2020 regarding planting trees to reduce water Fry Indexing and research on identification of Spring temperatures in some key areas. However, the introduction Electro-fishing Salmon stocks genetically with the of COVID restriction has limited progress. objective of being able to map where JUVENILE Tweed’s Spring Salmon spawned. Achieved – A map of early and late running stocks has been STOCKS: produced by MSS. Gala Smolt Trap To continue the fry index surveys in the District. In particular, to Partly achieved. In 2020 the Fry Index surveys were on JUVENILE understand the physical conditions the Rivers Whiteadder, Eye, Eden, Leet and on the Ettrick STOCKS: that determine the boundaries and Yarrow Waters. The Marine Scotland Science national Salmon Smolt between trout and salmon areas and electro-fishing programme was cancelled due to COVID Monitoring Studies that affect productivity. To take part but will hopefully be repeated in 2021. These sites will help in Marine Science Scotland’s National to inform Marine Scotland’s Conservation Limits model by PREDATION: Electric-fishing programme (NEPS). with juvenile data: previously only adult fish catches have Juveniles been considered. and Adults To monitor the Salmon and Trout of the Gala Water over their whole life Achieved: the dedicated smolt trap on the Gala Water was cycle in fresh water. operated in 2020 with the purpose of estimating the total run of Salmon smolts every year and collecting biometric To add value to the current data. 1,000 Salmon smolts (1,000) were again P.I.T. tagged in monitoring system for fry and parr the Spring and another 1,000 fish were tagged in Autumn stocks. To allow closer monitoring further upstream to assist with smolt output estimates in of the factors limiting successful 2021. 9 PIT tagged fish returned in 2020 as Grilse. smolt migration and to be able to establish the percentage that Achieved: Following the pilot study in Spring 2019, 240 complete their migration, and assess smolts were acoustically tagged in 2020, with a calculated where most smolt losses occur in loss of salmon smolts in the main river of 57.4%. the Tweed system. Achieved. Bird counts were carried in January, April, May To continue monitoring predators in and October. All counts recorded below average numbers the catchment both through main of Goosanders but Cormorants were close to the long stem counts and through index term average. monitoring at roosts and at strategic points throughout the District which will allow a measure of whole catchment distribution to be made. To analyse the frequency of damage types on young fish in relation to run timing and sizes using the existing fish traps. 48 www.tweedfoundation.org.uk