(Article reproduced from TDSFB Annual Report 2003 / 04)
Introduction
The autumn run is the mainstay of the Tay salmon fishery. The autumn run provides the largest catches, the biggest rents and ultimately provides most of the revenue for the Board to conduct its activities throughout the entire district. But something is happening to the autumn run. In the last number of years catches have fallen from what they were, though 2004 was a considerable improvement. Why have recent autumn runs generally not been as good as they were?
Autumn salmon
There is little doubt that autumn salmon represent a particular strain of fish which inherit from their parents the tendency to return to freshwater late in the year, though the precise date of return may perhaps be subject to some modest variation according to environmental influences the fish encounter at sea. This genetic trait is maintained because, at spawning, the autumn salmon are segregated from the spring salmon and summer grilse. The springers and summer grilse spawn in highland tributaries like the Ericht, Tilt, Lyon and Dochart but the true autumn salmon largely spawn in the mainstem of the Tay, the Earn, the Eden, the Dighty Burn and in lower tributaries of the Tay such as the Isla and even the lower Tummel.
Autumn salmon spawn later than spring salmon; December / January as opposed to early November. To some extent their distribution may reflect winter water temperatures. Cold winter weather prevents fish ascending steep highland tributaries, and so autumn fish tend to be associated with lowland and relatively warm rivers, though not exclusively so. Winter temperatures in the main stem of the Tay are on average warmer than tributaries like the Tilt or Shee on account of the time it takes for the vast waters of Loch Tay to cool.
Looking at Britain as a whole, true autumn salmon are almost unknown in the northern Highlands but start to appear in the more lowland rivers of Aberdeenshire, the lower part of the Esks, the Tweed, Ayrshire and Solway rivers and the north east of England. Many rivers on the west coast of England and Wales also have late runs, culminating in Cornwall where the ultra-mild rivers Camel and Fowey fish on to the 15th of December with a full expectation of catching silver fish on the last day.
Since autumn salmon spawn lower down the Tay system they do not migrate rapidly upstream as do fish in May or July and so are more obliging to anglers in the lower reaches of the Tay. Autumn fish have also spent longer feeding in the sea so they are larger than earlier fish. Whereas a grilse in June may only weigh 3 pounds, by October grilse may average 8 to 10 pounds, with some male specimens up to 12 pounds or more. Most of the autumn fish now caught on the Tay are in fact large grilse, though this is often not appreciated by anglers. While the majority of these grilse can be males, the autumn two sea-winter salmon which are mainly female, usually are in the high teens of pounds but the small proportion of males can weigh over 20 pounds.
When do the fish enter the river?
What might loosely be called autumn salmon start to appear around August and fresh run fish can continue to enter the river right to the end of the season and beyond. Some fish destined to spawn in the lower Tay can enter the river even as late as the start of the season on the 15th of January. However, the strength of runs in different months is not always the same and indeed may vary considerably over time.
Changing autumn runs over time
It is well known that over the decades, autumn runs have been subject to great change, not only on the Tay, but throughout Scotland.
The end of the 19th century and the first few decades of the 20th century was generally a good time for autumn fishing in Scotland but as the 20th century wore on, the abundance of autumn fish reduced markedly and indeed spring fish increased.
By the late 1940s autumn fishing had reached its low point and was very poor, perhaps even poorer than spring runs are now. This autumn decline is well known for those rivers, like the Tweed, where detailed study of long term catch records has been carried out. Elsewhere we have to rely on comments from contemporary writers. For example, Richard Waddington writing in his first edition of Salmon Fishing (1947) stated that “autumn fishing proper, for fish that have run into the river during September and October, is virtually non-existent in the rivers of the north-east of Scotland today. In the Dee and the Spey even the odd autumn fish is a rarity now. In the Don there is still a very small run, a mere shadow of its former self……”. In 1956 G.P.R. Balfour-Kinnear wrote in Trout and Salmon, “the Spey does not have any appreciable autumn run, and the autumn runs on the Tay and the Dee are also negligible, so that these rivers are not, in my opinion, worth fishing after the summer. It is possible that those disapproving of autumn fishing have such rivers in mind, not having fished a proper autumn water.”
From the 1950s autumn runs started to increase again and by the late ‘60s had become quite numerous. However, there was some difference in what constituted autumn fish. The autumn fish on the Tay or the Spey were very different from the Tweed.
Writing in Trout and Salmon in October 1976, Tony George commented that autumn runs had been increasing for the previous fifteen years but stated that the “true autumn run of the Tay takes place from mid July to the latter half of August, and fresh sea-liced fish run only in comparatively small numbers thereafter.” Also writing in the late 1970s John Ashley-Cooper, commented that on the Tay the “autumn run, or perhaps the late summer run would be a truer designation, has increased beyond all recognition during the past twenty years. At any time from mid-August onwards the lower river starts to stock up with fish…..By mid September their number runs into thousands, and this increase continues up to the end of the season…..Many of these fish…are in varying stages of redness; but bright silver fish are plentiful in August and the first half of September, and an odd one, though not many, can still be found in October. Rod-fishing ends after October 15th, none too soon, as most of the fish appear to be near spawning by then, and the Tay seems to lack a true autumn run of fresh fish in October and November, like that of the Tweed." He further commented that the “Tweed is almost the only river in Scotland which still has the genuine autumn run, so much renowned in our grandfathers’ days. Fresh silver fish enter this river in numbers during October and November…… Moreover the autumn fish, approaching extinction as they were in the 1940s, have been steadily increasing in numbers over the past twenty years”.
Not much can be gleaned about changes in autumn runs
from the Tay fishing reports in the Trout
and Salmon magazine
in the 1960s and 70s but it is notable that even then
the commentators seemed to bemoan most seasons as being
some sort of disaster. Tony George in his October 1976
article made an interesting observation: “The
tone of much that is written on our salmon fisheries
in recent years is of an almost unrelieved gloom....
One would never imagine from many writers that big
catches are made anymore. How many readers are aware,
for example, of the magnitude of the autumn catches
made in the Tay in some seasons?” He then went
on to describe some very substantial Tay catches he
had personally witnessed and noted that on Islamouth
it was felt necessary to have a limit of 20 fish per
day or 120 per week during the period 1st August to
15th October.
Catch data - nets
To corroborate the claims of angling writers, the obvious thing is to analyse any catch data which exist. The first data analysed are from the nets which used to operate in the Tay estuary.
Figure 1 shows the catches of salmon and grilse combined caught in Tay district nets during the months of July and August from 1952 to 1996 as reported to the Scottish Office. Throughout this period, July and August always had the highest catches of any month in the year, but in the late 1950s and through the 1960s catches in both months were of similar size. Bearing in mind the season ended on 20th August, then August must have had the biggest runs during this period. However, catches in July and August were relatively low in the early 1950s but in both months they increased markedly from then on. August catches reached a peak around 1970, but July continued to increase for several years. After about 1970 August catches commenced a long term decline and July surpassed it as the best month.
So, as described by the commentators, the huge increase in August net catches during the 1950s and 60s indicate a strengthening “autumn” run, but why did they fall after 1970? For that we must now look at rod catches.
Figure 1. Net and coble catches of salmon and grilse combined in the Tay
District in July and August, 1952 – 1996. Data from Fisheries Research Services
Catch data - rods
It was fortunate that in 2004, Morton Heddell-Cowie, a researcher employed by the Tweed Foundation, committed the extensive historic rod catch data of the former Tay Salmon Fisheries Company to computer. Not a pleasant task but one for which we are grateful! One of the longest and most informative data sets is from Stobhall and the autumn catches are shown in Figure 2.
Figure 2. Catches of salmon (including grilse) from the Stobhall beat,
1934 to 2004, for the months of August, September and October.
Figure 2 exemplifies the pattern described by angling writers of initially low autumn catches which then picked up in the 1950s and peaked in the mid to late 1980s. However, this differs from the net catch where August fell after about 1970. Indeed the peak in the Stobhall August catch took place 10 years and September nearly 20 years later than the peak of the August net catch. Since, according to Government statistics, there was no diminution in netting effort over this period, one possible explanation is that from about 1970 to the 1980s the peak of the “autumn” run entering the Tay shifted later, from perhaps August to September – i.e. from before to after the end of the netting season. Such a shift in the timing of the run would have reduced exploitation by nets and given a boost to angling in the Tay and may have been responsible for the era of great angling catches in the 1970s and 1980s.
However, close inspection of the Stobhall catch reveals further differences between individual months. For example, while August catches have fallen since the early 1980s, October catches have tended to remain much more constant over the last 30 years. This is in spite of the fact that netting, which only ever affected August, reduced in the 1990s and ceased completely after 1996, and should have made more August fish available to anglers. Is this further indication of a latening of the run?
One way of testing this is to compare the sizes of individual fish caught in the different months. As they have spent longer at sea, the fish which enter the river in late autumn are bigger than those entering in the early autumn. Thus, if in the 1970s most of the fish caught in late autumn were old coloured fish, then their weights should be similar to those of fresh fish caught in some previous month. On the other hand if the late autumn is characterised by fresh run fish, then their average weights ought to be bigger than fish caught in previous months.
Figure 3 shows a comparison of the average weights of grilse caught at Stobhall in the months of August and October, 1950 to 2003. In this instance grilse were identified from salmon on the basis of a more detailed analysis of weight frequency distributions for each month which are not presented here for the sake of brevity.
Figure 3. Average size of grilse caught at Stobhall in August and October, 1950 - 2003
Figure 3 shows that for both August and October the size of grilse increased after the 1950s peaking in the early 1970s. Since then the size of grilse has declined. However, the gap between the size of grilse in August and October was relatively large in the late 50s and 1960s but in the early 70s and late 80s the difference was relatively small, but in the 1990s became much wider. This suggests that in the early 70s and in the late 80s a greater proportion of the fish caught in October may have been in the river for a longer time compared to the 50s /60s or the 1990s. But would this have to mean that the 50s and 60s was also a good autumn period? Well, no actually. In the 1950s and 60s although such grilse as were caught on Stobhall in October may have been genuine late runners, the catches were actually very small (Figure 4), so there appears to have been few of them. But October grilse catches in recent years have been much higher, so this analysis of catch data suggests that there really has been more genuine late running fish in recent years.
Figure 4. Catches of grilse from Stobhall in August and October, 1950 – 2003
Recent evidence for even later runs
While a few decades ago really late run fresh fish seem to have been scarce on the Tay, they now appear to be much more common. There are several types of more direct observational evidence which indicate this is the case.
A feature of recent Januarys, including 2004, has been that numbers of unspent fish have been caught when the new season opened. A big proportion of these fish were still bright silver, some even sea-liced. Many of the kelts are also in very good order suggesting they were not long in the river before spawning. In recent seasons fish have still been actively spawning on many beats in the main stem on January 15th. Although the Board has not systematically collected data on this, a number of observers of spawning in the main stem of the Tay right up to Kenmore report that spawning largely takes place in December and January nowadays as opposed to November in the 1980s. This is all suggestive of a significant run of spawners taking place even during the early winter.
Even more conclusive evidence of a late autumn run was obtained in 2003. As a means of obtaining broodstock for the hatchery, the Board requested that some ghillies angle for salmon in the main stem of the Tay after the fishing season had ended. This proved surprisingly successful. On the first attempt in early November some six or seven ghillies landed 86 fish on the Ballathie beat in one day! Many of the fish caught on that occasion were coloured and the females generally were near full maturity and stripped easily before the end of November. However, on a subsequent outing after a spate in late November the catch rate ran at between 5 and 10 fish per ghillie for the day, but on this occasion few of the fish were coloured or anywhere near full maturity. Most were silvery, ranging from the slightly tarnished to mint fresh and weeks away from spawning. The same thing happened on a couple of outings in mid December when it was found to be practically impossible to catch any coloured fish. One ghillie landed ten fish in sixteen casts and all were quite silver, one was even sea-liced. Some of the less silver of the silvery fish were taken to the hatchery and did not yield eggs until well through January.
Reports of increased numbers of very late running fish also come from other rivers. On the North Esk, for example, the Freshwater Fisheries Laboratory has had a fish counter on the lower river since 1981. In the 1980s September was a good month and November practically inconsequential, but since then, September counts have declined appreciably while November is now one of the more important months of the year. It looks as if the whole of the “autumn” run has slipped a full month later. In January 2004 one or two slivery “baggots” (late fish yet to spawn) were caught on the Helmsdale, the first time such fish have ever been caught.
Conclusion – have the runs changed?
Over the last 50 years autumn runs have seen remarkable change. Initially autumn running fish were not at all common but increased greatly in the 1960s. By 1970 large runs of “late summer” fish entered the but the peak seems to have been before the end of netting on 20th August. Thereafter the run seems to have gradually become ever later, the peak slipping into September in the 1970s and 80s and since then perhaps even later, with a significant run now taking place even after 15th October. The pattern of change in the size and timing of the peak runs is suggested as something like Figure 5.
Figure 5. Suggested representation of the timing and size of the Tay “autumn” run
The interpretation of these changes by anglers who experienced them first hand can be different. Some just perceive a dramatic fall, from the early autumn peak they recall not so long ago. Some assume that prior to the peak they remember that it must always have been like that, but in fact it never was. The only certainty we can conclude about salmon runs is that indeed they do change, and doubtless they will change again. As to why they change, that is a subject in itself
|
