Scottish Marine and Freshwater Science Vol 6 No 12: The demography of a phenotypically mixed Atlantic salmon (Salmo salar) population as discerned for an eastern Scottish river

Mortality Rates, Management Actions and Decadal Trends

The magnitude and pattern of PEA escapement ( i.e. the proportion of pre-estuary returning adults who potentially survive to spawn) differed appreciably between early2 SW' fish and the two later-running sub-stocks. For early2 SW' fish PEA survival increased more or less smoothly from only 40% in 1980 to nearly 90% in the late 2000's (Fig. 5.b). By contrast both late-run 2 SW' fish and grilse showed about the same 60% escapement in the late 2000's as they showed in 1980 (Fig. 5.e,h respectively). In both cases the smoothed escapement dropped a little during the late 1980's and then showed a rise between 1990 and 2005, before dropping back to a value close to that in 1982.

The key driver of the multi-decadal change in PEA escapement for early-run 2 SW' salmon was a steady increase in the proportion of fish surviving the estuary nets (Fig. 5.a), leading to a smoothed estuarine net-fishery survival which rose from 40% in 1980 to nearly 90% in 2008. Early MSW survival from the estuarine net-fishery thus increased appreciably over the decade 1980 to 1990, prior to the removal of the Morphie Dyke nets in 1991, and actually at a similar rate to the increase in net-survival between 1991 and 2000. These changes may well have resulted from economic pressures altering effort by the net-fishery. By the times of the two management-agreements that delayed the opening-dates to the netting-season (2001, 2005), survival from the nets was already some 90% for early2 SW', and, although net-survival continued to rise after 2001, it perforce increased more slowly (being already so close to 100%). Over the same period early2 SW survival from the rod-fishery, which was over 90% in 1982, initially decreased to less than 80% in the late 1980's, before rising steadily to virtually 100% following the introduction of catch-and-release in 2005.

For grilse, the smoothed estuarine net-fishery survival changed much less, from roughly 75% to about 80% over the same three decades of the study, while the rod fishery survival, which was over 90% in 1982 dropped a little, to just above 80% by 1990, where after it rose almost to its 1982 value around 2003 before falling back to close on 80% by 2009 (Fig.5.g).

Figure 5. Decadal trends in the density independent parts of the life-cycle of three North Esk sub-stocks, derived using grilse-error corrected rod-capture data and the default allometric marine-mortality model ( DAM). Frames a), d) and g) show the component processes of the overall pre-estuary to spawner survival shown in frames b), e), and h). For these frames (a,b, d,e, g,h) 'year' is the year the adults returned (and succumbed to/ survived the fishery, and spawned or not). Frames c), f) and i) show the smolt to pre-estuary survival: here (frames c,f,i) 'year' means the year of common smolting; note the slightly different scales for survival, used to show detail, across frames c,f,i. All frames show yearly values (points) and a smoothed trend (line) obtained by loess smoothing using R routine loess with default parameters. Frames a), d) and g, show the percentage of the pre-estuary numbers surviving the estuarine (net) fishery (circles with dashed line) and the percentage of those survivors who also survive the rod fishery (squares with solid line). Vertical dashed lines show dates of management interventions: 1991: Morphy Dyke nets moved to rods, 2000: net season delayed until April 1 ^st, 2001: MSW rod-caught fish released, 2005: net season delayed until May 1 ^st

The estuarine net-fishery survival of late-run 2 SW' fish showed a similar pattern to grilse, which run the river at much the same time. Survival increased from only 60% in 1981 to nearly 75% by the mid-1990s, followed by a plateau to the early 2000s, with some suggestion of a small decrease thereafter (to 2009) (Fig.5.d). The rod survival for the late2 SW' sub-stock also showed similar changes to the grilse sub-stock, decreasing from over 90% in 1981 to c. 75% in the early 1990s, rising thereafter to apparently plateau at some 95% around 2005 (Fig.5.d).

At-sea survivals showed two roughly decadal-period fluctuations during the study (Fig.5.c,f,i). As each of the three values is exponentially related to the smolt-year-classes' estimated overall survivals (equation 8), which is the same for all sub-stocks, the pattern of variation was necessarily the same for all sub-stocks; namely wide variability around a common, but quite flat, overall trend. The smoothed trends showed survival in 2005 being almost identical to that in the early 1980s, with a small drop centred on 1985 and a rather larger drop centred on 2000. However, the survival drop in the early 1980s (Fig.5) was heavily influenced by a single, very high, estimated survival for the 1980 smolting cohort. With the exception of the1980 estimate, the pattern of survival would be quite consistent with a trend of declining oceanic fishery mortality, and thus increased survival, between 1980 and 1990. Due to its 1980 outlier, this particular data-set was thus equivocal about whether the magnitude of any changes in marine survival were co-incident with the regulation of the oceanic fisheries in West Greenland (1980) or Faroes (1990) (see Discussion, Marine Fisheries, below). However, it is important to note that in this analysis the marine mortalities were assumed independent of sea-age class (except in so far as sea-age is correlated to body-size or period at sea), whereas the marine fisheries were known to have affected only those sea-age classes which were exposed to the fisheries concerned, (primarily the MSW in the case of West Greenland). Hence the absence of an observable effect in a parameter which subsumes at-sea predation mortality with non-estuarine coastal fisheries mortality and distant-water fisheries mortality does not constitute strong evidence in either direction.

Comparisons Across Data-sets and Models

The estimates of the demographic process parameters for early2 SW' and grilse were rather insensitive to grilse-error correction, as were estimates of estuary-net survival for late2 SW'. The major differences were in both the (large) magnitude and temporal variation of the rod-survival estimates for late-running 2 SW' fish. The like results to Table 2 and Fig.5 for 'uncorrected' data are given in Supplementary Material part C.