Bird stomach contents analysis - final report: Goosander and Cormorant diet on four Scottish rivers 2019 to 2020

This study analysed the stomach contents of goosanders and cormorants collected from the Rivers Tweed, Dee, Nith and Spey during 2019 and 2020 in order to to assess whether there was evidence of substantial changes in the diets of these species of fish-eating birds since the 1990s.

1. Introduction

Throughout Europe there is often considerable concern about the potential impact of sawbill ducks Mergus spp. and the Great Cormorant (Phalacrocorax carbo, hereafter 'Cormorant') on stocks and catches of fish of commercial value (e.g. Marquiss & Carss 1994, Russell et al. 1996, Carss 2003, van Eerden et al. 2012). This also holds true for Scottish rivers, where Goosanders (Mergus merganser) and Cormorants, and Atlantic Salmon (Salmo salar) and Brown Trout (S. trutta) are the bird and fish species of most interest. This is more than purely biological, as there are world-renowned fisheries for these Salmonid species in Scotland and, hence, also considerable commercial interest. Angling and netting activities here are associated with 4,300 full-time jobs and contributed a total of £79.9 million to Scotland's economy in GVA, there are about 250 listed angling clubs across Scotland, and an estimated 125,000 river anglers (An Analysis of the Value of Wild Fisheries in Scotland (

Reviews in the past have concluded that there is generally a lack of unequivocal scientific evidence for the impact of piscivorous birds on fisheries (e.g. Carss in press), not necessarily because there is no impact, rather because impact is notoriously difficult to quantify (Carss et al. 2012). Furthermore, there are often limitations in the quality of available fish data, for example in terms of their abundance and behaviour particularly in deeper, wider sections of rivers, the vulnerability and responses of fish to predation in mainstems compared to tributaries (e.g. Marquiss et al. 1998, Carss et al. 1997, Wires et al. 2003). However, there are credible concerns that the potential for predation impact might well be exacerbated if a prey population is declining for other reasons. Such declines are certainly the case for Atlantic Salmon (hereafter 'Salmon') stocks in rivers in Scotland and elsewhere in Europe and Scandinavia (e.g. SalmonBusiness 2018, Ward & Hvidsten 2011).

Both Goosanders and Cormorants are so-called 'generalist' predators that appear to consume prey in relation to their abundance and availability in particular habitats, rather than being 'specialist' on particular prey species and/or specific foraging habitats. In certain places, at certain times of year, these birds undoubtedly prey upon juvenile Salmon (both parr and smolts) as reported in Marquiss et al.'s (1998) previous intensive research on these issues in Scottish freshwaters undertaken some 20 years ago. In the meantime, stocks of Salmon and European Eel (Anguilla Anguilla), both species eaten by these birds (Marquiss et al. 1998), have declined (e.g. Hindar et al. 2011 for Salmon, Correia et al. 2018 for Eel). The precise causes of these declines are unclear but include habitat loss, pollution, or overexploitation and changes in growth and survival, and in the distribution of predators or prey, often during the 'at sea' phase of the life- cycle and likely operating, to some degree at least, in a cumulative manner. There are thus likely to have been changes in the community structure of fishes in Scottish rivers in recent years and so it is timely to collect new comparative dietary information for two of the most abundant and widespread avian predators in these habitats.

Dietary analysis alone is insufficient to quantify the consumption of commercially important fish by birds. This would require site-specific data on (i) bird numbers, (ii) their daily food intake, and dietary composition. Further, demonstrating any impact based on such a knowledge of fish consumption by birds is complicated by the numerous other biotic and abiotic factors interacting to affect fish populations (e.g. Carss et al. 2012) but teasing out the role of predation is likely, in turn, to require considerable site-specific data on the fish themselves (e.g. behaviour, age-structure, population dynamics, abundance, potential compensatory responses).

The present research does not set out to quantify the impact of these birds, but to document the diet of Goosanders and Cormorants, and make comparisons with previous work. Much of this research was undertaken during the International Year of the Salmon's (IYS) 5-year initiative (Year of the All of it was carried out against the background of possible changes in fish abundance and/or community structure in Scottish rivers and heightened concern over the decline of the nation's wild Salmon and the environmental and man-made pressures faced by this iconic and economically important species.

Four Scottish Salmon rivers were included in this research: the Nith, Tweed, Aberdeenshire Dee, and the Spey. The overall aim of the work was to derive the proportion (by mass and by number) of each fish species in the diet of Goosanders and Cormorants, collated in a form directly comparable (and equivalent) "as contained in appendix 5.1 and 5.3, for Goosanders and Cormorants respectively, of [Marquiss et al. 1998] the 'Fish-eating Birds and Salmonids in Scotland' report published in November 1998 (ISBN 0 7480 7232 2)" (7.1 of contract requirements).

In the present study, a maximum of 36 Goosanders and 36 Cormorants with food in their stomachs were to be collected from each study river, with a minimum of 12 Goosanders and 12 Cormorants (with sufficient food in their stomachs) to be sampled during the smolt run period. Samples were collected under a scientific licence arranged between Marine Scotland, Scottish Natural Heritage (now NatureScot) and District Salmon Fishery Boards prior to the award of this contract. Arrangements were made to collect Goosander and Cormorant samples from the four study rivers in two periods:

(a) 1st March - 31 May 2019 - the smolt run period, and

(b) from 1st September 2019 – 29 February 2020

The latter period thus covered the autumn (Sep-Nov) and winter (Dec-Feb) and throughout the report information is generally presented separately for the smolt run period and the autumn/winter period.



Back to top