Peter Gough's definitive reply

Address:                                                           Our Ref:

Your Ref:

Mr Mike Ashwin

By e-mail                                                          Date: 14.12.20

Dear Mike,

Thank you for your letters dated 2nd and 10th December:

  • Letter dated 02.12.20 and headed: “Corrections, explanations and recommendations to the NRW published Salmon stock performance in Wales 2019 Report”
  • Letter dated 10.12.20 and headed: “Response to NRW correspondence of 30.11.20 and 03.12.20 ref NWATFCC information requests and recommendations on Welsh Stock reporting and Byelaw Impact”

Please treat this letter as a response to both.

Your letter of the 2nd December

Paragraph 2:

NRW’s Report on Salmon Stock Performance in Wales 2019 was circulated to all Local Fisheries Groups in Wales and to the strategic Wales Fisheries Forum but was received by them without significant comment (including from your colleagues in CPWF and the Angling Trust who are members of these bodies).

Section 1:

The terms Conservation Limit (CL), Management Objective (MO) and Management Target (MT) are defined in the report, as is the nature and purpose of the formal statistical compliance procedure and the link between the compliance outcome for a stock (i.e. its ‘risk status’) and the management response as informed by the Decision Structure (DS). Nowhere in the text is it stated that (in your words) “average stocks must be above the MT to achieve MO”. We think you mean ‘CL’ here and not ‘MT’?

Technically, a river that is achieving its MO will be classed as ‘not at risk’. We think it unlikely that there are “many” examples of rivers classified as ‘not at risk’ where very few egg estimates from the latest 10-year period meet the CL? However, we take the point that a river which shows a marked upward trend in egg numbers could be

classified as ‘probably not at risk’ (projected 5-years into the future) with few or perhaps any egg estimates from the last 10 years meeting the CL (hence your reference to the Wye). In those circumstances, managers would need to form a judgement as to whether measures to protect the stock should be maintained or relaxed (or even strengthened). In the case of the Wye, byelaws were already in place requiring mandatory C&R and these were maintained by NRW to support a still vulnerable stock.

Section 2:

In line with the DS, voluntary C&R to protect vulnerable salmon (and sea trout) stocks had been promoted on rivers in Wales for several years before NRW produced its Technical Case setting out the rationale for the byelaw proposals. The case for mandatory C&R was, in large part, necessary to address the minority of fishermen who refused to adopt C&R measures voluntarily, but who had been given ample opportunity to do so. You will be fully aware of the debate on this aired at the Local Inquiry; the independent judgement on this and other matters made by the Inspector; and the decision of Welsh Government which followed. That debate included the different legislative requirements pertaining to Wales but not England (underlying the different response to the DS in the two jurisdictions).

The DS will be examined as part of the wider review of stock assessment processes identified in the NASCO IP and, as part of that examination, it may be appropriate to better tailor the DS to the different legislative demands of England and Wales.

However, for now, the DS in its current form remains as relevant to Wales as England – not least to inform the protection and management of sea trout as well as salmon stocks.

Section 3:

The ranking system shown in Table 2 is not new – it was used in the Technical Case. It does not reflect any formal change in “policy” by NRW, rather it is just a means of bringing together various statistics to illustrate the vulnerability of stocks,

You suggest that additional ‘performance tests’ should be included in the ranking system, but it is not clear to us what you mean by:

  • “Individual rivers actual annual Formal Compliance to CL (5 or 10 year % values)
  • A rivers actual average 5-year attainment to CL target (its MAT%)”

The trend in the latest 10-year series of (log10) egg estimates is an important component of the statistical compliance process because it captures the general ‘direction of travel’ for the stock, and by so doing incorporates an additional level of

precaution in the assessment (i.e. we would be more concerned by a stock in decline than one showing an improving trend).

The compliance graphs for the Wye, Severn, Mawddach, Conwy and Dee are shown below:

Sorry Graphs are missing

The report doesn’t provide egg deposition estimates as a percentage of the CL, but this information is given in the more comprehensive annual assessment report for E&W published by Cefas, the Environment Agency and NRW and referred to in the text of the Wales report (with link).

Section 4:

The ~25-year time-series of egg deposition estimates – given for all principal rivers in Appendix I – are provided mainly to give greater historic context (i.e. beyond the usual figures of the latest 10-year series of egg estimates). The log scale adjustment (note the numbers on the scale are arithmetic) is appropriate to illustrate patterns and trends when levels of egg deposition are highly variable within and between rivers. We don’t intend to reproduce these graphs on the usual arithmetic scale, but If you want to see the associated raw data then we can provide them. Your suggested changes to the MT line can be incorporated in next year’s report (if these figures are repeated then).

Regarding the review of stock assessment procedures as identified in the NASCO IP and rod catch raising factors – both these matters have been referred to on numerous occasions. They are covered in the wider assessment reports and in the NASCO IP itself and both documents have been brought to the attention of LFGs and WFF (including discussion on these matters). They have also been addressed repeatedly in correspondence with yourself (often copied to your associates).

Your letter of 10th December

Note to avoid repetition we’ve only responded to issues which have not been addressed in previous correspondence.

First bullet point:

The Wye is the only river in Wales where owners’ rod catch returns and not statutory licence returns are routinely used to derive run, spawner and egg estimates. This is because an effective system of collecting owners’ returns has been in place for the entire Wye system for many decades, and throughout this time this has provided the most reliable and complete source of catch data.

Second bullet point:

We have responded to this matter in our letter of the 3rd December 2020.

Third bullet point:

The average trap efficiency at Chester Weir (since the programme began in full in 1992) has been ~20% for salmon, ~25% for whitling sea trout and ~33% for older sea

trout. Clearly, in some years the efficiency will have been above and in some below these averages; however, the statement in the Dee Angler Report “that, on average, 20-30% of the run is trapped at Chester” is not unreasonable.

As you indicate, rod exploitation rates used on the Wye and Usk to derive run figures from rod catch track the extant rates estimated for the Dee (i.e. use the rolling 5-year average rate). Modelling developments (discussed in recent correspondence) may change that approach in due course, but until those developments are in place, there are no plans to alter existing methods on any river. This is also the case with other modifications to stock assessment procedures which may arise from the current review (expected completion by 2022 – as stated in the NASCO IP).

Regarding the ratios of tagged to untagged fish among the catches of logbook and other anglers on the Dee: the figures you produce for the latter assume that all recaptures are fully reported by non-logbook anglers, which we feel is highly unlikely. Indeed, we can’t estimate this reporting rate, and, outside the logbook system, we are less sure overall about the validity of the ratio of tagged to untagged fish provided by the wider angling population. There are signs in the last few years (the years you focus on) that differences in the ratio of tagged to untagged fish reported by logbook anglers and the wider angling population have been growing. The fact that fewer fish have been around to tag and recapture doesn’t help in that respect. Similarly, some anglers may also be less willing to report tags if they disagree with the recent byelaw changes. We can only work to improve on that situation, including by seeking to introduce fishery-independent methods to evaluate the ratio of tagged to untagged fish (as outlined in previous correspondence on these matters).

Fourth bullet point:

Rivers are ranked in Table 2 of the Wales assessment report as described in the text of that report (see last bullet point of Section 2). The P-values arise from the statistical compliance procedure; i.e. they are derived from linear regression methods and relate to the probability of an upward trend in (log10) egg deposition estimates for the last 10- years.

I hope this addresses your concerns, although I recognise that there may remain some differences of opinion between us.

Yours sincerely


Principal Advisor Fisheries

Cc        Simon Toms, Environment Agency Ian Davidson, NRW

Ffôn/Tel            03000 653501


Croesewir gohebiaeth yn y Gymraeg a’r Saesneg Correspondence welcomed in Welsh and English

Leave a Reply




You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>