Run timing changes: evidence from netting catches on Tweed and Spey

Despite the onset of social media the river reports in the Trout & Salmon magazine are still an important source of information and there are usually some revealing insights into thinking across the country. I note for example that this month the Tweed correspondent is still hanging on to some hope that the back-end run will re-appear this year. History suggests that he will be disappointed.

As evidence to support that assertion I am going to present below a series of slides showing the catches from the Sandstell netting station, which was operated at the mouth of the Tweed and the Raik nets on the Spey. The Raik nets were mainly net & cobble nets which operated in the lower few miles of the Spey mainstem. But before presenting those catches here is a graph showing the Spey rod catch split into spring salmon (to end of May), summer salmon and grilse from 1952 to 2017 (Data from Marine Scotland Science and Spey Fishery Board (since 2002)).

Trends in Spey rod catch 1952 to 2017. The most stable component of the stock has been the summer salmon. The spring catch passes all three NASCO tests with 2016 and 2017 well up the list of recent catches. It was only the wet September of 2017 that saved the autumn component from failing all three tests. This failure of tests 2 & 3 show that the reduction in the autumn catch is not a recent phenomenon. The reduction in the autumn catch is mainly as a result of the decline in the grilse catch. There is a difference of almost 3,000 between the recent five year average grilse catch and the peak of the early 1990s.

The collapse of the spring catch in 1981 was matched with an upsurge in the grilse catch. The grilse catch exceeded the spring salmon catch from 1982 until 2009. It is as if they can’t be in the same room at the same time. [/caption]

Below are the decadal catches from the Tweed and Spey netting stations with the catches split into salmon and grilse and expressed as a percentage of the total by month. The whole series will be presented one after the other with additional comment at the end. The Tweed catches are shown in the left graph and the Spey on the right. The sequence starts in the 1850s and ends in the 1980s.

There are some minor differences between the two rivers but the overall pattern is similar over a period of 140 years. The most noticeable change is the dominance of spring catches in the 1920 and 1930s. There was evidence of an increasing spring catch from 1900 and this lasted until the 1950s. This increase in the spring catch was offset with a decline in the grilse catch. This shows that when changes in run timing occur they tend to be long lasting and, based on this analysis, occur over large geographical areas. Tony George in his seminal PhD on cyclical changes in run timing in Scotland considered that the rivers from the Findhorn to the Tweed followed the same trends concurrently and this analysis supports that.

When I was making these graphs it occurred to me that whilst the Spey Raik net catches in the Spey in the 1970s showed low spring catches the spring rod catches in the 1970s were great (see below and first graph).

Until I had completed this analysis I was working under the impression that it was the nets that provided the best sample of the run and that the rod catch was biased for a variety of reasons. The netting effort data shows that in 1972 there was a 25% reduction in the number of crews operating in the spring months and this maintained until the 80s. This 25% reduction in effort is likely to have had a lesser impact on catches as no doubt the least profitable crew/station would have been cut first. What I take from this is that if you want to understand the characteristics of the run at any particular time you have to utilise all the data available.

Salmon run timing changes occur in a predictable cyclical way (or at least it has until now).
When changes occur they can last for decades.
The main change is a swing in the relative proportions of spring salmon and grilse, the two are never present in abundance at the same time.
These changes occur over large geographical areas at similar times, although there are always differences from river to river.
We have been losing the grilse run for a number of years; there will be year to year variation, but based on historical evidence there is little prospect of an upturn in grilse catches in the the immediate future.
The stage is set now, will the spring run deliver?

Thanks to Dr. Ronald Campbell, Tweed Foundation, for access to the Tweed nets data.

There are 7 comments for this article
  1. Dave Wilkinson at 12:00 pm

    Thanks for this full analysis but I have some problems, due to aged eyes, i reading the detail on your graphs. I left a message with your switch asking for large copies.

    Though retired, I was a low-grade statistician using business applications. I would thus have some doubts re error as follows:-

    Using old catch stats is a real issue since kelts were killed as late as c 1850 I think.
    Seasonal effort is impossible to obtain, I assume. Knowing a few old net & cobble fishermen, then the marginal stations would likely have turned to sea-trout in low-flow years or not fished.
    Whereas the Tweed and Spey would make up a considerable portion of the Scottish net and rod catch, the overall total for Scotland would be better maybe ? E.G. at low river flows then any coastal netting stations in the Moray Firth ( for Spey ) and hang-netting off the Tyne ( N’ld for Tweed ) may have biased the seasonal mix of fish according to the gear used. Catches in the latter fishery were huge in the 19C – I have them in front of me: 129,100 salmon recorded at North Shields ( R. TYne nets + sea -nets in 1873.

    Then there is the possible bias from more recent large sea-fisheries at Greenland and the Faroes. Out-of-control Danish bats in the early stages of the former and winter long-lining in the latter.

    It occurred to me that work in the farmed salmon filed may assist as they had to reduce grilsing losses. It was found that increasing photo-period in sea-cages delayed grilsing.
    A certain Icelander (OV) was convinced that the 4 D C isotherm i the sub-Arctic was the area salmon preferred. I assume this because its prey was there. As global warming may be interfering with the photperiod in the 4C area is this a possible factor in grilsing? If so,one might expect a random pattern of run-timing of Ss sub-groups to become apparent rather than a switch to stable cycles over a number of years.

    Then the Girnock Burn research concluded, I think, that early-running grilse were of the same or similar genetic make-up. Maybe spawning in similar areas. If true the it seems to me that environmental factors at sea may be the issue for Ss and it’s prey fish too.

    Om smolts, I have the old stats for Ss farming in Iceland, maybe this is abandoned now for economic reasons. Probably competition from farmed salmon being the most likely. As you say, it works on the Eastern Ranga for sportsfishing. I was interested in what you say about trials here failing. But Icelandic smolts and post-smolts likely have a much smaller journey. to sea-feeding areas than most fish from Scottish rivers. Perhaps a trial on the North Coast of Scotland away form salmon farms. Many thanks & good luck with the sea-trout, please give my regards to Andy Walkern. Dave

  2. Euan at 10:34 pm

    Given all the pressures that exist nowadays Brian might it be a good idea to increase our smolt output utilising private money in smolt release ponds which have been proved to work elsewhere? If this would not work on the Spey I look forward to the explanation why. E

    • Brian Shaw Author at 9:49 am

      Thanks for the comment Euan. Whilst no one would argue against increasing natural smolt production I would take issue with your point that smolt stocking, whether from release ponds, or not, would work on the Spey. There have been some viable fisheries established using smolt stocking from release ponds, notably the Ranga’s in Iceland but all the evidence suggests that it doesn’t work in Scotland. The Tay DSFB produced a useful summary of smolt stocking in Scotland, see here, and Marine Scotland Science produced this publication in 2015 . Since 2015 catches on the Lochy, where smolt stocking has been a prominent feature of their strategy, have declined even further. Salmon catches on the Wester Ross River Carron increased during a period of intensive stocking, some of which were at the smolt stage, but in recent years the catches have declined (see page 8 in this report.

      In my view the evidence for successful smolt stocking in Scotland is wafer thin. But your specific point was about smolt release ponds, which may or may not be the key part. I am not aware of much recent work using smolt release ponds, with the exception of the Wester Ross Carron where they have been used. The scientific consensus is that stocking is harmful to wild fish hence why any application to stock smolts in the Spey is unlikely to be granted. However experimentation is an essential part of progress and in the right place it would be good to see a proper trial undertaken in Scotland. I have had similar discussions with Ian Gordon, there may be locations in Scotland where such work could be done but it would need to be in a place where there was little risk to existing wild stocks.

      I attend Spey Fishery Board meetings and although it was generally before my time, whenever smolt stocking was discussed it was never approved as a concept, and in my time I have seen little evidence of support from within the board for stocking at this life stage My point is that this is not only the biologists view.

  3. nick warren at 9:14 am

    excellent and informative thansk very much for taking the time to do this !

    • Brian Shaw Author at 1:21 pm

      Thanks Nick. This is such an important factor contributing to the current low status of stocks that it needs some discussion. I intend to publish more on the same theme soon.

      Regards Brian

  4. David Turtle at 10:45 pm

    A fascinating summary and an excellent bit of research, Brian. Thank you.

    It certainly provides hope for us Spring rods but will I still be alive to see a return to the catches of the past?


    • Brian Shaw Author at 9:14 am

      Thanks David, given all the pressures that exist nowadays, compared to 100 years ago for example, it is likely that when the runs change they will not reach the highs of previous cycles. A repeat of the huge spring catches to the rods of the 1950s is unlikely, unfortunately. It might still be quite good though. Regards Brian

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.