Portal de Periódicos da CAPES

Could falling female sex ratios among first‐winter northwest European duck populations contribute to skewed adult sex ratios and overall population declines?

2018; Wiley; Volume: 160; Issue: 4 Linguagem: Inglês

10.1111/ibi.12649

1474-919X

Anthony D. Fox, Thomas K. Cristensen,

Fish Ecology and Management Studies

For the first time, we present new evidence to show declines in the female sex ratio among first-winter as well as adults among four common duck species based on wings provided voluntarily by Danish hunters since 2000. We consider the potential causes are unrelated to changes in methods of assigning wings to age/sex classes or to changes in methods of hunting, but speculate whether they could result from primary or secondary sex ratios (embryo or hatching) or pre-hunting season differential survival. Whatever the explanation, because of their quarry status and declining abundance, we urgently need investigations into primary, secondary and tertiary (i.e. adult) sex ratios of these species and improved demographic monitoring to better understand the causes of these changes and promote potential mitigation management measures. Sustained increases in the abundance of many huntable duck species in Europe since the Second World War (e.g. Eltringham & Atkinson-Willes 1961, Owen et al. 1986, Nagy et al. 2014) were thought to result from enacted hunting legislation that limited their previously unrestricted harvest in response to concerns at that time (e.g. Berry 1939, Owen et al. 1986). Implementation of domestic and international wetland protection mechanisms (such as the Ramsar Convention and EU Birds and Habitats Directives) has also limited the rate of habitat loss and degradation, latterly through the establishment of cohesive flyway networks of EU Natura 2000 sites. In North America, it took complex and expensive administrative structures, extensive monitoring procedures and associated continental-scale research programmes under the North American Waterbird Management Plan (US Fish & Wildlife Service/Canadian Wildlife Service 1986) to deliver the maintenance of abundant waterfowl populations and perpetuate their supporting habitats, achieved through managed subsistence and recreational harvests (subject to continuous review) consistent with their conservation. By comparison, in Europe by the late 1990s, it appeared a similar achievement had been attained, as most common huntable duck species showed favourable conservation status (in the sense that most were increasing in the face of existing harvest rates) with comparatively little investment in research, monitoring or coordinated international management actions, beyond site safeguards and the status quo. Unfortunately, between-year flyway-level changes in abundance of many common duck species listed on the EU Bird Directive's Annex II list of huntable species suggest the populations of some species are stabilizing and beginning to decline (Nagy et al. 2014, Ramão 2015). Hunting of Annex II species can occur under national legislation as long as this 'is compatible with the maintenance of the populations of these species at a satisfactory level …' and '…with the principles of wise use'. For any closed population, the annual rate of change reflects the balance between annual reproductive success and mortality. To be able to manage common duck populations in a way compatible with wise use and maintaining population size, we need to identify and understand the processes that drive an increasing or stable population into decline through some annual assessment of these two parameters. Regrettably, we lack extensive duck ringing programmes in Europe to provide data, especially on survival rates. However, long-term data on the age and sex composition of common duck species from wing samples, voluntarily contributed by Danish hunters, are available to assess potential changes in annual breeding success among birds killed in Denmark in autumn and early winter (Christensen & Fox 2014). Proportions of first-winter specimens among duck wings voluntarily submitted by Danish hunters show apparent long-term declines in annual reproductive output in four common duck species (Eurasian Wigeon Mareca penelope hereafter Wigeon, Northern Shoveler Spatula clypeata hereafter Shoveler, Northern Pintail Anas acuta hereafter Pintail and Common Goldeneye Bucephala clangula hereafter Goldeneye) between 1982 and 2009 (Christensen & Fox 2014). In the case of the Wigeon, the annual production of young significantly contributed to the rate of change in flyway population indices by annual mid-winter counts (Fox et al. 2016a), although it was not possible to demonstrate the contribution from climate change to declines in productivity (Pöysä & Väänänen 2018), as seemed the case previously (Mitchell et al. 2008). Christensen and Fox (2014) found no correlations between North Atlantic Oscillation (NAO) indices or summer temperatures in the breeding areas and the annual ratio of first-winter birds to adults for any species in the study, suggesting it was not macro-environmental meteorological conditions that were consistently affecting breeding success. Using identical methods (Christensen & Fox 2014), we are now in a position to extend the time series from Danish hunter-derived wings to 2017. This confirms continued significant declines in all four species and shows that the Eurasian Teal Anas crecca (hereafter Teal) now also shows a statistically significant decline in the proportions of young among samples of submitted wings (Fig. 1). Tufted Duck Aythya fuligula showed no significant relationship between year and age ratio in the wing sample, but showed recent increases in adult female sex ratio, (Fig. 2). Pintail, Shoveler, Teal and Wigeon all showed long-term declines in the proportion of females among adult birds sampled in the wing samples, although only Teal and Wigeon attained statistical significance, whereas Tufted Duck significantly increased and Goldeneye fluctuated without a clear trend (Fig. 2). A male-biased sex ratio among common dabbling and freshwater diving ducks has long been recognized (e.g. in North America: Leopold 1933, Mayr 1939, Bellrose et al. 1961). This is thought to arise from differential predation resulting from the vulnerability of nesting and brood rearing females (the males taking no part in either activities, exposing females to heavier predation rate, e.g. Sargeant et al. 1984). Support for this hypothesis has been generally forthcoming from the contention that hatching sex ratio for most common duck species approaches parity (Clutton-Brock 1986 and see the major review in Blums & Mednis 1996). For this reason, there has also been the assumption that the sex ratio amongst first-winter common duck species has approached unity (e.g. Mitchell et al. 2008), as first-winter birds are not likely to have been exposed to the differential predation, hunting or other sources of mortality risk of older age classes. Earlier analysis of the 1982–2009 Danish wing survey suggested stable or non-significant increases in female sex ratios among first-winter wings (T.K. Christensen unpubl. data). However, the addition of data from more recent years now shows significant long-term declines in Wigeon and Teal, whereas Pintail (albeit non-significant), Shoveler, Tufted Duck and Goldeneye all showed statistically significant quadratic model fits, which imply an increase in the proportions of females among first-winter birds until c. 2000 followed by a subsequent decline (Fig. 3). These results suggest that for Shoveler, Teal, Wigeon and Tufted Duck the female sex ratio among harvested first-winter birds in Denmark is now significantly less than 50%. Assuming this ratio reflects that in the population as a whole, this implies that the sex ratio of ducks entering their second year is already less than parity. Furthermore, because of the progressive change that we observe in the time series presented here, this factor is increasingly contributing to the skewed adult sex ratio of these duck species and potentially to the recent declines observed in these populations. There could, of course, be multiple interacting potential explanations for these patterns. The management of the Danish Wing Survey scheme passed from Ib Clausager to T.K.C. in 2004 and there has been some turnover (but also considerable overlap) in staff identifying the wings to species, age and sex classes. However, we are confident that there have been no changes in techniques or adoption of new knowledge than could contribute to the observed declines in females among duck wings. This is especially so for the long-term declines in Teal and Wigeon, but also the more recent declines in other species. Although there are long established claims that hunters select for males (e.g. McIlhenny 1940), there is no evidence for this given that the sex ratio among the wing sample apparently reflects ratios in the field (e.g. Fox et al. 2016b). There is no doubt that first-winter birds are over-represented among the hunted wing samples and, among Wigeon at least, that this bias progressively declines through the hunting season (e.g. Fox et al. 2016b). Hence, while we fully accept that changes in hunting techniques or methods of attracting ducks to shoot could potentially affect proportions of young ducks harvested by wildfowlers, it is difficult to see how this may disproportionately contribute to elevating the hunting mortality of juvenile males that predominate increasingly in the wing samples. It seems more feasible to accept that changes in the relative vulnerability of the sexes to hunting could contribute to higher hunting mortality. Mallard Anas platyrhynchos and female American Black Duck Anas rubripes in poor condition at capture were more likely to be shot and recovered in the same autumn, especially among male Mallard (Hepp et al. 1986, Conroy et al. 1989, Dufour et al. 1993). This suggests a condition bias in the harvest, which could lead to an over-representation of males among the wing surveys if they were in poorer body condition. Mallard shot over decoys also tended to have lower body mass and condition indices than those shot by other means or trapped (Greenwood et al. 1986, Reinecke & Shaiffer 1988), suggesting a condition–bias interaction with hunting method, although we have no evidence to suggest that hunters supplying wings have progressively changed their hunting techniques. Nevertheless, if male ducks of all age classes were increasingly in poorer condition compared with females by virtue of density-dependent access to food, stress, oxidative status, contaminant loads or some other mechanism, we might expect to see the increases in the representation of males in hunting bags that we have witnessed. However, most studies suggest the winter behavioural dominance of males over females (e.g. Hepp & Hair 1984, Brodsky et al. 1988), which may ultimately affect access to food reserves (Alexander 1987, Choudhury & Black 1991), suggesting density-dependent access to optimal food resources would favour males, a factor thought to contribute to a skewed adult male sex bias. Nor can we find convincing arguments why male ducks should be more sensitive to other environmental factors that might enhance their representation in the hunting bag over females. Rather, we prefer the hypothesis that the sex ratios of commoner duck species in the shot wing sample are generally representative of that in the population, accepting some age bias, as confirmed by Fox et al. (2015) for Wigeon. If this is the case, we need to look for a mechanism to explain long-term decreases in the proportion of females among first-winter Teal and Wigeon cohorts and shorter-term declines among Shoveler, Tufted Duck and Goldeneye. Such trends are likely to contribute to the increasingly skewed sex ratio in subsequent age classes, which we see in the adult wing samples. Donald (2007) found that along a gradient of IUCN conservation status categories, as threat level to avian population survival increased, so adult sex ratios became more male-biased, concluding that this resulted from increasing female mortality (attributed to general 'stress') rather than production of more males or their improved survival. Although this could be the case here, it is difficult to see a stress mechanism that could selectively elevate early first-year mortality of female ducks over males. Despite reviews (e.g. Charnov 1982, Clutton-Brock 1986) suggesting the extreme rarity of deviations in avian primary sex ratios from parity and an absence of evidence for their adaptive significance (Komdeur & Pen 2002), several studies since the 1980s have suggested that offspring sex in birds can be manipulated actively by females at laying. Offspring sex is influenced by maternal sex hormones (Pike & Petrie 2003), so potentially embryonic sex expression could be manipulated in this way. For instance, Seychelles Warblers Acrocephalus sechellensis show biased hatching sex ratios in the absence of nest-helpers, producing 77% males in low-quality territories compared with 13% in high-quality territories (Komdeur et al. 1997). Interannual variation in sex ratio also mirrored that of the adult sex ratio in the preceding year in the Saltmarsh Sparrow Ammodramus caudacutus (Benvenuti et al. 2018). Captive female Blue-faced Parrotfinches Erythrura trichroa of similar body condition produced unbiased sex ratio offspring on high-quality diets but more males under poor feeding conditions (Pryke & Rollins 2012). Because nutritionally stressed fostered sons were more likely to survive than were daughters in this species, this implies females could potentially adjust offspring ratios in relation to the apparent quality of the environment to support differential sex survival rates. Komdeur and Pen (2002) and Navara (2018) reviewed the potential drivers of maternal avian sex allocation, broadly examining sexual dimorphism, dispersal and philopatry, social environment (especially the presence/absence of helpers), male quality/attractiveness, female quality/condition, food availability, seasonal effects and variation across the laying order. Studies from the poultry industry also show late-onset incubation reduces gosling quality, survival and female sex ratio at hatching, suggesting another potential mechanism to affect secondary sex ratios at hatch (e.g. Amantai et al. 2018). Experimental evidence suggests that, whereas testosterone and oestradiol have no effect, artificially elevated levels of the principal stress hormone, corticosterone, in laying female quails resulted in a significantly female-biased sex ratio at laying (Pike & Petrie 2006). However, in general, we remain very ignorant about the potential maternal mechanisms that could affect avian sex allocation and secondary sex ratios. Specifically, we suffer from the little recent attention specifically paid to whether Anatidae females show an ability to manipulate the sex ratio of their offspring and under what circumstances they may do so. Differential sex allocation in ducks may also occur passively, for instance, because of environmental effects, for example through the intervention of endocrine-disrupting pollutants (Colborn et al. 1993). There is now a rich literature linking sex determination in a range of aquatic organisms to a range of contaminants, including common industrial chemicals, fire-retardants and pesticides (e.g. Windsor et al. 2018). Studies show that organochlorine pesticide levels within gull eggs affect the hatching sex ratio (Fry & Toone 1981) and aquatic birds using heavily mercury-polluted water have also demonstrated offspring sex-ratio effects (Bouland et al. 2012), although in both cases contamination induced more female embryos, rather than fewer as in the present case. Quite why the currently falling female ratios among first-winter duck wings submitted by Danish hunters should be returning to relative low levels in the 1980s among Pintail, Shoveler, Tufted Duck and Goldeneye is also puzzling and perhaps linked to changes in levels of stress hormones or endocrine-modifying contaminants during this time series. We tend to think of the boreal forests (where many of these duck species commonly breed) as relatively pristine, stable environments. However, climate change is having considerable effects on the release of organic matter into lakes (e.g. Monteith et al. 2007) and is predicted to increase in the future (e.g. Larsen et al. 2011). Elevated levels of dissolved organic matter are known to enhance bioavailability of ionic mercury to aquatic organisms (e.g. Ravichandran 2004). Climate change is also affecting the frequency of northern forest fires, against which fire-retardants are increasingly used in Fennoscandia (C. Arzel pers. comm.). However, we lack environment data to confirm any such relationships and in the face of such rampant speculation, it is clear that we urgently need some focused investigations to look at embryo, hatching and fledging sex ratios among common duck species and to account for associated environmental factors that may be responsible for any observed deviations from parity. Recent confirmation for the decline in the proportions of females in the Common Pochard Aythya ferina (hereafter Pochard) population in the field compared with the 1980s (Brides et al. 2017) offers some independent support for the patterns provided by Danish hunter wings. Unfortunately, numbers of Pochard wings submitted to the Danish wing survey in 2009–2017 have been too few to generate reliable estimates since those of Christensen and Fox (2014). Nevertheless, if declines in first-winter and adult female sex ratios are a general feature of our common huntable duck species, we need to confirm that this is part of a longer-term trend. Assessing independent historical sources of data on long-term changes in the annual sex and age ratios of commoner duck species (for example private citizen science projects and the EURING duck ringing database) represents a clear future research priority. We also need to instigate a rigorous mechanism for systematically collecting sex- and (where possible) age-ratio date in the field in a representative fashion to follow such changes in the immediate future. Whatever is driving these trends, it is becoming increasingly evident that our knowledge of the demography and population dynamics of our common and huntable duck species remains lamentably poor, adding to the urgency of implementing greatly improved monitoring and conservation research for these populations (Elmberg et al. 2006), especially within some kind of adaptive management framework (Holopainen et al. 2018). We are deeply grateful to the very many contributors who voluntarily contribute wings to the annual duck survey in Denmark and to Ib Clausager for organizing the scheme and coordinating collection, starting in 1982. We extend our considerable thanks to Celine Arcel for her ideas, discussions and literature sources, as well as the participants of the PEDS5 conference held at Cumbrae, Scotland, in April 2018 for their contribution by stimulating thinking about causes of changes in sex ratios of common duck species. We also thank an anonymous referee for comments on an earlier draft of the manuscript. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.