Rago 1999) as is that of the percid Sander lucioperca in the shallow embayments of the Baltic Sea (Müller‐Karulis et al. 2013, Mustamäki et al. 2014). In South Africa, immature stages of the sciaenid Argyrosomus japonicus and the sparid Lithognathus lithoganthus, amongst others, are exploited as juveniles in estuaries, coupled with exploitation of adults in surf zones (Griffiths 1997, Lamberth & Turpie 2003). The heavy exploitation has led to major stock collapses in these species and current spawner stock biomass per recruit ratios at <5% of pristine levels. Adult stock abundance also is vulnerable to poor water quality and habitat deterioration. In North America, failed recruitments of alosines, anguillids, moronids and salmonids provide unfortunate examples of recruitment declines that are at least partly attributable to depleted adult abundances.
3.4.1.1 Stock structure, contingents and cohorts
Spawning populations of species using estuaries, particularly fishes that spawn within estuaries or their connected freshwaters, often have a diverse genetic makeup that is expressed in their particular spawning habits and behaviours (e.g. Secor 1999, Bottom et al. 2005a, 2005b, Secor and Kerr 2009, Levings 2016). Adult stock structure and variability are notably evident in anadromous fishes in which spawning adults return to their natal rivers and estuaries (Secor 2015). In the case of Pacific salmon species (Oncorhynchus spp.), there is diverse dependence on estuaries by fry, juveniles and smolts amongst the several species and races that reflect evolved spawning behaviours and estuarine dependence by fry and smolt stages prior to ocean migration (e.g. Simenstad et al. 1982, Thorpe 1994, Levings 2016). In the case of salmonids, the loss of spawning contingents, and overall losses of diversity in genotypic and phenotypic variability in pre‐smolt juveniles, combined with losses of estuarine nursery habitat, is a particular concern for recruitment success of O. tshawytscha in the Columbia Estuary (USA) (Bottom et al. 2005a, 2005b) and of salmonids in general in many Pacific coast estuaries in North America (Levings 2016, Quinn 2018).
Spawning contingents that display different life‐history patterns and variable dependence on estuaries and marine habitats are recognised for taxa other than salmonids, e.g. the moronid Morone saxatilis in Chesapeake Bay (Secor 1999, Secor & Piccoli 2007), Hudson River (Secor et al. 2001) and St Lawrence River (Morissette et al. 2016) populations. Similarly, research on the sciaenid Argyrosomus japonicus in South Africa indicates that this species has distinct marine and estuarine subpopulations, with the estuarine subpopulation showing critical declines in recruitment and abundance due to overfishing (Childs et al. 2015) with concomitant genetic diversity limitations and incidence of hybridisation with a related sciaenid species (Mirimin et al. 2014). In acipenserids, spawning seasons are highly variable (Bemis & Kynard 1997), and there is evidence of spring‐ and fall‐spawning races, for example, in Acipenser oxyrinchus living in the same estuaries and tidal rivers on the Atlantic Coast of North America and, similarly, dual‐spawning groups of acipenserid species in Europe and Asia (Balazik & Musick 2015). In another example of contingent behaviour, sub‐adults of some individuals of the amphidromous plecoglossid Plecoglossus altivelis ryukyuensis may repeatedly move between freshwater and estuarine habitats during their juvenile stage (Murase & Iguchi 2019).
In the Baltic Sea, the spring‐spawning clupeid Clupea harengus is represented by subpopulations or spawning stocks that are assessed and managed in fisheries as separate units (ICES 2018). Four C. harengus stocks are recognised for management: Baltic proper, Gulf of Riga, Gulf of Bothnia and Bay of Bothnia. Major differences between Gulf/Bay and Baltic proper stocks are related to their salinity tolerances and adaptations. A fifth spawning stock occurs in the Western Baltic, which is largely comprised of adults that feed and overwinter in the Danish Straits or outside the Baltic Sea (Dodson et al. 2019). Baltic C. harengus stocks have spring‐spawning and autumn‐spawning components that exhibit stock‐specific reproductive/recruitment behaviours, although the contribution by the autumn spawners is minor (Ojaveer 1988).
3.4.1.2 Maternal effects
Variability in age structure, growth and condition of adults contributes to variability in reproductive output and potentially to success of recruitment of estuary‐dependent fishes. Quantity and quality of eggs and sperm may depend on the condition, size and age of adults. As such, the quality and potential of eggs and larvae to survive depend on maternal (or paternal) investments. In a review of marine fishes, maternal effects were documented for many taxa (Marshall 2016). However, Ottersen et al. (2013) analysed recruitment success of 38 stocks of marine fishes, including some estuary‐dependent or ‐associated stocks (e.g. the Baltic Sea clupeid Clupea harengus membras, Baltic Sea gadid Gadus morhua and European pleuronectid Pleuronectes platessa) and found little evidence of a significant effect of maternal age on recruitment success in those stocks although the authors acknowledged that their data were sparse and difficult to interpret. In a meta‐analysis of effects of maternal age and size on marine fish stocks and implications for recruitment, Venturelli et al. (2009) reported that populations represented by older, larger individuals have a higher maximum reproductive rate than populations of equivalent size represented by younger, smaller individuals, concluding that expanded age structure is important for sustainability in many exploited fish stocks. The relationship between female size and egg number for Pacific salmons can vary with latitude (Fleming and Gross 1990). In the salmonid Salmo trutta, the anadromous traits displayed in some females provide an adaptive advantage and greater fitness during early ontogeny (Goodwin et al. 2016).
Based on laboratory harvesting experiments conducted on the short‐lived, estuarine atherinid Menidia menidia, Conover & Munch (2002) and Conover & Baumann (2009) found strong inferential evidence suggesting that under heavy, age‐ or size‐selective exploitation, a population's reproductive potential could evolve rapidly. For M. menidia experiencing age‐ or size‐selective harvesting of larger, faster‐growing individuals, progeny matured at younger ages and smaller sizes than in unexploited populations. There had been a genetic shift towards production of slower‐growing, earlier‐maturing M. menidia in only four generations. Size‐selective fishing is common in coastal and estuarine fisheries, and it is probable that heavily exploited species in estuaries are more prone to selective harvesting than offshore species. For example, size‐selective fishing has been shown to favour fast‐growing individuals that mature at smaller sizes and younger ages in the anadromous osmerid Osmerus eperlanus from the Gulf of Riga, Baltic Sea (Arula et al. 2017) and Osmerus mordax in the northeastern USA (Chase et al. 2019).
The pleuronectid Pleuronectes platessa has experienced declines in size at maturity and shifts in growth rate and age structure in the southern North Sea in the past century that are attributed to changing fishing mortality rates. Rijnsdorp (1993) inferred that these changes were in part genetic, but not independent of concurrent changes in productivity of juvenile coastal and estuarine habitats. He also found that, while growth dynamics changed, the reproductive investment by females did not (i.e. relative ovary weights of females remained constant), although there was a shift in recent decades towards larger but fewer eggs in ovaries (Rijnsdorp 1993). In laboratory experiments on P. platessa, Kennedy et al. (2007) concluded that larger females produce bigger eggs and larvae with greater yolk volumes that grew faster than larvae from eggs of smaller females. Maternal effects on spawning also were demonstrated in laboratory experiments on the estuarine pleuronectid Pseudopleuronectes americanus (Buckley et al. 1991). In a protracted spawning season, large, early‐spawning females produced bigger eggs than those produced by small, late‐spawning females, implying that conservation of large females in P. americanus will maintain large egg sizes, overall fecundity and presumably increased survival potential of early‐life stages. Overall, there is compelling evidence that selective forces (primarily fishing) can modify age structure and precipitate changes in reproductive and recruitment potential of estuary‐dependent fishes such as P. platessa and P. americanus.
In the anadromous moronid Morone saxatilis from the east coast of North America, effects of changes in age structure of the adult stock and its effects on reproductive success and