Eur J Wildl Res (2013) 59:305–321 DOI 10.1007/s10344-013-0716-9

REVIEW

Density dependence in ducks: a review of the evidence Gunnar Gunnarsson & Johan Elmberg & Hannu Pöysä & Petri Nummi & Kjell Sjöberg & Lisa Dessborn & Céline Arzel

Received: 14 November 2012 / Revised: 21 February 2013 / Accepted: 19 March 2013 / Published online: 5 April 2013 # Springer-Verlag Berlin Heidelberg 2013

Abstract Density dependence (DD) is a central concept in population ecology and in the management of harvested populations. For example, DD underpins the idea of additive versus compensatory mortality and is a tenet in the paradigm of resource limitation and regulation. Yet the prevalence and importance of DD remains disputed in most organisms, including ducks, which are focal in game management, conservation and zoonotic diseases. Based on 154 data entries from 54 studies in the peer-reviewed literature, we here synthesize and evaluate the prevalence of DD in breeding ducks in relation to (1) species and guild (dabbling versus diving ducks), (2) stage in the breeding cycle (nesting, duckling, recruitment) or, alternatively, in terms of population dynamics, (3) study type (descriptive/nonmanipulative versus experimental), (4) continent (Europe versus North America), (5) spatial level (wetland, landscape, regional, continental) and (6) biome (tundra, boreal, nemoral, prairie, mediterranean). One conclusion from this Communicated by C. Gortázar G. Gunnarsson (*) : J. Elmberg : L. Dessborn Division of Natural Sciences, Kristianstad University, 291 88 Kristianstad, Sweden e-mail: [email protected] H. Pöysä Joensuu Game and Fisheries Research, Finnish Game and Fisheries Research Institute, Yliopistokatu 6, 80100 Joensuu, Finland P. Nummi Department of Forest Ecology, University of Helsinki, P.O. Box 27, 00014 Helsinki, Finland K. Sjöberg Department of Wildlife, Fish, and Environmental Studies, Swedish University of Agricultural Sciences, 901 83 Umeå, Sweden C. Arzel Section of Ecology, Department of Biology, Turku University, 20014 Turku, Finland

review is that it is difficult to find general patterns about the prevalence of DD unless data are broken down to subsets, for example, to stage or spatial level. With respect to stage, DD effects occur at all stages of the breeding cycle. During the nesting and duckling stages, the frequency of cases detecting versus not detecting DD is roughly the same. However, in cases referring to the recruitment stage, i.e. to survival of fledged ducks until 1 year old at the most, DD was the rule, suggesting that DD processes may operate mainly outside the breeding season. Further subdivision of data shows that spatial scale is important to the prevalence of DD in nesting ducks—rare on the wetland level and more common on higher spatial levels. In studies of population dynamics (i.e. based on time series data only), DD was more often found in diving than in dabbling ducks. This corroborates previous suggestions that dabbling ducks largely should be considered as r-selected species, in contrast to more K-selected diving ducks, which start to reproduce at an older age and often breed in more stable wetland environments where resources may be easier to track and populations thus often are closer to carrying capacity. However, the picture of DD in ducks is far from complete, and knowledge gaps for future studies to address include: (a) data from Russia, which holds a large part of the breeding ducks in the Northern hemisphere, (b) experimental studies on more species to separate density-dependent factors from other drivers of population change and to tease apart spatial and temporal interactions in the underlying processes, (c) time series analyses addressing population dynamics, especially from outside North America, (d) studies relating duck numbers to limiting resources, which arguably is the most relevant measure of density, (e) the timing of DD processes in relation to harvest and natural mortality. Keywords Anatidae . Density dependent . Duckling . Hunting . Limitation . Nesting . Population dynamics . Recruitment . Regulation . Waterfowl

306

Introduction Density dependence (hereafter abbreviated DD, as is densitydependent) is a theoretical and empirical cornerstone of population biology. The central idea of DD is that population growth rate is determined not only by exogenous and stochastic factors but that it may also depend on endogenous factors related to the density of the population itself. Increased density may have positive effects on growth rate as long as density per se is low, so called Allee effects (e.g. Stephens et al. 1999). However, DD is usually seen as being synonymous with negative feedback on vital rates with increasing population density, often expressed as decreasing birth rate and/or increasing mortality rate as carrying capacity is approached. In this sense, DD is a fundamental process not only within the paradigm and theoretical models of resourcelimited populations (Royama 1992; Cappuccino and Price 1995; McCallum 2000; Ranta et al. 2006) but also to infer and demonstrate interspecific competition (Dhondt 2012). DD processes are often invoked to understand the limitation and regulation of populations (Newton 1998). Such understanding has become increasingly important in a world with mounting challenges of persistence of natural populations subjected to, e.g., fragmentation, harvest, disease or climate change. Other reasons for the continued attention on DD in ecology have to do with the very nature of DD. Theoretical as well as empirical studies demonstrate that DD may be direct or indirect, intermittent, varied in time and space, delayed (i.e. mostly aspects of population dynamics) and not the least problematic to detect (e.g. Turchin 1990; Hanski et al. 1993; Åström et al. 1996; Zeng et al. 1998; Jonzén and Lundberg 1999; Paradis et al. 2002). Although DD is generally accepted by ecologists as a phenomenon operating in nature, some researchers question the concept of DD as such (e.g. White 2001; Berryman 2004), and there is also widespread semantic confusion in the literature (Herrando-Pérez et al. 2012b). A more moderate and widely accepted critique is that density per se should be seen as a mere proxy for the crowding effects likely to arise when population density gets high. Accordingly, it has been suggested that density should be related to the availability of the resources likely to be limiting, i.e. that it needs to be translated into a per capita measure based on e.g. abundance of food, nest sites, etc. (Elmberg et al. 2003; Lindström et al. 2005; Dhondt 2012). Another concern relates to recent works that emphasize the importance of habitat heterogeneity and individual quality to DD patterns and processes (e.g. Carrete et al. 2006; Dhondt 2012; Krüger et al. 2012). Many duck species are widespread and abundant, some also being well-known flagship species in conservation, hunting and ecological research. For these and other reasons (e.g. epidemiology), there is a large scientific literature on the population ecology of ducks. Quite a few of these papers

Eur J Wildl Res (2013) 59:305–321

address DD because many species are intensively hunted and frequently managed under the explicit assumption that DD processes result in harvest being compensatory to natural mortality rather than being additive (e.g. Anderson and Burnham 1976; Nichols et al. 1984; but see Pöysä et al. 2004; Sedinger and Herzog 2012; Pöysä et al. 2013). Consequently, flexible management strategies taking such processes into account have been developed for North American ducks (e.g. Johnson et al. 2002; Nichols et al. 2007). Even at first glance, some striking patterns emerge when the large body of research on DD in ducks is considered. Firstly, most studies focus on a specific event or stage in their annual cycle, i.e. very few have an annual, let alone lifetime, perspective. Secondly, studies differ in terms of which intrinsic and extrinsic factors are hypothesized to affect population growth rate. Thirdly, there is a huge variation in the spatial scale addressed, from single ponds to the continent level. To give an example, Pöysä (1987) found DD in feeding habitat use at a scale of <1 ha in the Eurasian teal (Anas crecca), while Murray et al. (2010) found support of DD in population dynamics at a continental scale (North America) in the closely related green-winged teal (A. (crecca) carolinensis). In addition, the methods used to study DD vary, from snapshot experimental studies to long-term time series analyses. To date, there has not been any extensive overview of DD in ducks. Considering the wide interest in ducks in population ecology, management, conservation and epidemiology, a compilation of patterns and a synthesis are overdue. In this paper, we review the scientific literature on intraspecific DD in ducks (family Anatidae, excluding swans and geese), addressing both population dynamics and the underlying processes. We only consider studies that investigated DD in breeding ducks (i.e. either as responses during different breeding stages or when breeding population size or duckling density are used as explanatory factors) simply because there are very few studies in other seasons (e.g. during migration and on winter sites) and also because regulative processes in breeding parameters are likely the most important ones in affecting population dynamics (e.g. Hoekman et al. 2002; Coluccy et al. 2008). Specifically, we explore the prevalence of negative DD with respect to species and guild (dabbling versus diving ducks), type of dependent variable, stage in the breeding cycle, study type, continent, spatial level and biome. Our goal is to describe, synthesize and evaluate patterns of DD and to pinpoint crucial knowledge gaps. We also discuss management implications, shortcomings and research needs derived from the results.

Selection of studies We searched the data bases Web of Science and Natural Sciences Collection (last accessed November 14, 2012) to

Eur J Wildl Res (2013) 59:305–321

find relevant peer-reviewed papers on DD in ducks (Anatidae, excluding swans and geese). We used search strings including combinations of the following words/ terms: density dependence, density-dependent, duck, waterfowl. In all papers found, we also searched the reference lists and considered older studies to the extent that we were able to retrieve them. During the review process of the manuscript, we were informed about one additional paper that was in press and which has been included (Lawrence et al. 2013). In a second step, we restricted our analysis to studies addressing DD in reproductive performance, survival or population dynamics in breeding ducks (i.e. we did not include studies on behavioural traits such as DD habitat selection). Accordingly, we included studies addressing DD during any part of the first year of life (see “stage” in the following text) as long as the dependent variable related to a measure from the breeding period. We further restricted our analysis to papers presenting empirical data on negative DD. In other words, theoretical studies and a few indicating positive DD due to Allee effects were left out. Our review was also restricted to intraspecific DD. We finally restricted the review to papers in which the authors themselves concluded whether DD occurred or not, which means that we may have missed studies addressing the topic without using the term DD. It should be noted though that in a few cases in which DD in population dynamics was addressed, we corrected obviously wrong interpretations by the authors or, if an interpretation was not given for a particular species, we concluded ourselves whether there was support for DD (final interpretations are shown in Table 1); these corrections/interpretations were based on quantitative parameter estimates presented in the original papers. In cases where authors published conclusions about DD based on the same data in several papers, we only included the latest and/or most comprehensive of these studies. This left us with 54 papers on which the present review is based.

Variables considered From each study, we extracted the following information, which was included as variables in Table 1 and/or 2: (1) species (scientific name). Studies based on artificial nests of dabbling duck type were classified as Anas, and the ecologically very similar sibling species Eurasian teal and greenwinged teal were lumped as A. crecca/carolinensis; (2) dependent variable is the variable used to assess DD; (3) stage is a way of grouping the dependent variables into three broad categories according to biological/ecological relevance and sequence: “nesting” refers to studies addressing egg size, clutch size, nest survival (including brood/pair

307

ratios), “duckling” to studies about survival or brood size during the duckling stage and “recruitment” to studies on the survival of fledged ducks until 1 year old at the most. A fourth and alternative variable in this category is population dynamics, which means that the results are based on time series data; (4) explanatory variable is the independent variables used; (5) DD indicates whether negative DD was found or not (“yes”, “no” or “equivocal”); (6) study type is either “experimental”, which includes manipulation of either density, a crucial resource or a mortality factor, or “descriptive”, i.e., non-manipulative studies. The latter category includes two sub-types, namely, long-term time series analysis of inter-annual dependence of population size, and correlations between a dependent variable on the one hand and density on the other, usually based on a shorter time series or being snapshot studies; (7) continent is either North America (NA) or Europe (EU). One study carried out in the tropical Pacific (Seavy et al. 2009) was classified as “NA”; (8) spatial scale: admitting that there were a few borderline cases, we categorized studies by spatial extent as using either “wetland” (single water bodies or islands, usually much less than 10 km2), “landscape” (several wetlands, usually within an area of less than 100 km2), “regional” (usually not more than 100,000 km2) and “continental” (usually millions of square kilometers) as the level at which the dependent variable was evaluated; (9) biome corresponds to the following categories of biotic zones: “tundra”, “boreal” (coniferous dominated forest), “nemoral” (hardwood forest), “prairie” and “Mediterranean” (cf. Breckle 2002); (10) years denotes the number of seasons during which data were collected. Studies including more than one species and/or stage have produced separate entries for each species and stage treated. In this way, the 54 papers are used in 154 ‘cases’. We argue that this is biologically relevant, and conclusions in a certain study are usually statistically independent among species and stages. However, our dataset contains one more obvious source of pseudo-replication, namely, the fact that many of the papers based on time series of breeding numbers (e.g. from the North American waterfowl survey and Lake Mývatn in Iceland; see Table 1) are based on data from overlapping years and areas. The 154 cases used to evaluate patterns of DD in breeding ducks may at first glance seem as a good sample size enabling a thorough statistical meta-analysis. However, data are divided into many categories (see previous text discussion) and cover several stages of the breeding cycle, which disables contrasts due to too low actual sample sizes. Further, the dependent variables studied often differ even between studies addressing the same life stage, making it very problematic to calculate comparable measures of effect sizes needed for a proper meta-analysis (e.g. Harrison 2011). Also, as already mentioned, some data entries are to some

Dependent variable

Brood/pair ratio Duckling survival

Duckling survival Duckling survival Nest survival Nest survival Nest survival Egg size Brood/pair ratio Brood/pair ratio Brood/pair ratio Brood/pair ratio Brood/pair ratio Duckling survival Duckling survival Duckling survival Female ducklings/pair Nest survival Brood/pair ratio Brood/pair ratio

Brood/pair ratio Brood/pair ratio Brood/pair ratio Recruitment Recruitment Recruitment

Brood size Population growth rate Breeding population size Breeding population size Population growth rate Population growth rate Population growth rate Population growth rate Population growth rate

Species

A. platyrhynchos A. platyrhynchos

A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos

A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos

A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos A. platyrhynchos

Breeding Breeding Breeding Breeding Breeding Breeding Breeding Breeding Breeding

population population population population population population population population population

size size size size size size size size size

Breeding population size Winter population size Breeding population size Breeding population size Breeding population size Winter population size

Pair density Pair density Neighbour distance Pair density Nest density Pair density Pair density Pair density Pair density Pair density Pair density Hatchling density Pair density Pair density Pair density Pair density Breeding population size Winter population size

Pair density Brood density

Explaining variable

Equivocal Yes No Yes No No No Yes No

No No No Yes Yes Yesb

Yes No No Yes Yes Yes No No No No Yes No Yes No Yes Yes No Yes

Yes Yes

DD

NA NA EUR NA NA NA NA NA NA NA

Descriptive Descriptive Descriptive, Descriptive, Descriptive, Descriptive, Descriptive, Descriptive, Descriptive, Descriptive,

TSA TSA TSA TSA TSA TSA TSA TSA

NA NA NA NA NA

EUR NA EUR EUR EUR EUR EUR EUR EUR EUR EUR EUR EUR EUR EUR NA NA NA

EUR EUR

Region Landscape Wetland Region Continental Continental Continental Continental Continental Continental

Region Region Continental Region Continental

Landscape Landscape Wetland Region Landscape Landscape Landscape Landscape Landscape Landscape Wetland Region Wetland Landscape Region Region Region Region

Landscape Landscape

Prairie Prairie Tundra Prairie Multibiome Multibiome Multibiome Multibiome Multibiome Multibiome

Prairie Prairie Multibiome Prairie Multibiome

Nemoral Prairie Boreo–nemoral Nemoral Nemoral Boreo–nemoral Boreal Boreal Boreal Boreal Boreo–nemoral Nemoral Boreo–nemoral Boreal Boreal Prairie Prairie Prairie

Nemoral Boreal

Continent Spatial scale Biome

Descriptive Descriptive Descriptive Descriptive Descriptive

Experimental Experimental Descriptive Descriptive Descriptive Descriptive Descriptive Descriptive Descriptive Descriptive Descriptive Descriptive Descriptive Descriptive Descriptive Descriptive Descriptive Descriptive

Experimental Experimental

Study typea

Table 1 Studies addressing density dependence in breeding ducks (variables are explained in the main text)

26 13 27 26 41 47 31 45 51 40

26 26 18 26 30

2 2 1 10 7 7 16 16 16 12 12 16 13 12 10 8 26 26

2 2

Kaminski and Gluesing 1987 Hammond and Johnson 1984 Gardarsson et al. 2008 Vickery and Nudds 1984 Viljugrein et al. 2005 Jamieson and Brooks 2004 Zeng et al. 1998 Lillegård et al. 2008 Murray et al. 2010 Sæther et al. 2008

Kaminski and Gluesing 1987 Kaminski and Gluesing 1987 Pospahala et al. 1974 Kaminski and Gluesing 1987 Sheaffer 1998

Elmberg et al. 2005 Amundson and Arnold 2011 Andrén 1991 Hill 1984b Hill 1984a Pehrsson 1991 Pöysä 2001 Pöysä 2001 Pöysä 2001 Elmberg et al. 2003 Elmberg 2003 Hill 1984b Elmberg 2003 Elmberg et al. 2003 Gunnarsson et al. 2008 Coulton et al. 2011 Kaminski and Gluesing 1987 Kaminski and Gluesing 1987

Elmberg et al. 2005 Gunnarsson et al. 2006

Years Source

308 Eur J Wildl Res (2013) 59:305–321

A. clypeata

size Yes Yes No size Yes size No size No size Yes size No Yes Yes size No size No size No size No size Yes size No size Yes size No

Descriptive, Descriptive Descriptive Descriptive, Descriptive, Descriptive, Descriptive, Descriptive, Descriptive Descriptive Descriptive Descriptive, Descriptive, Descriptive, Descriptive, Descriptive, Descriptive, Descriptive,

Descriptive, Descriptive Descriptive Descriptive Descriptive Descriptive, Descriptive, Descriptive, Descriptive, Descriptive, Descriptive, Descriptive, TSA NA EUR EUR TSA EUR TSA NA TSA NA TSA NA TSA NA NA NA NA TSA NA TSA NA TSA NA TSA NA TSA NA TSA EUR TSA NA

TSA NA EUR EUR EUR EUR TSA EUR TSA EUR TSA EUR TSA NA TSA NA TSA NA TSA NA

Continental Region

Continental Continental Continental

Continental Landscape Landscape Wetland Region Continental Continental Continental Landscape Landscape Landscape Region Continental Continental Continental Continental Wetland Region

Continental Landscape Wetland Landscape Landscape Landscape Wetland Wetland Region Continental Continental Continental

Multibiome Prairie

Multibiome Multibiome Multibiome

Multibiome Boreal Boreal Tundra Prairie Multibiome Multibiome Multibiome Prairie Prairie Prairie Prairie Multibiome Multibiome Multibiome Multibiome Tundra Prairie

Multibiome Boreal Tundra Boreal Boreal Boreal Tundra Tundra Prairie Multibiome Multibiome Multibiome

Continent Spatial scale Biome

Descriptive, TSA NA Descriptive, TSA NA

Population growth rate Population growth rate Population growth rate

A. acuta A. acuta A. acuta A. acuta

Breeding population Pair density Pair density Breeding population Breeding population Breeding population Breeding population Breeding population Pair density Nest density Breeding population Breeding population Breeding population Breeding population Breeding population Breeding population Breeding population Breeding population

size No Yes size No Yes size No size No size No size No size No size Yes size No size Yes

Study typea

Population growth rate Breeding population size No Breeding population size Breeding population size Yes

Population growth rate Brood/pair ratio Duckling survival Population growth rate Breeding population size Population growth rate Population growth rate Population growth rate Nest survival Nest survival Brood size Breeding population size Population growth rate Population growth rate Population growth rate Population growth rate Population growth rate Breeding population size

A. americana A. crecca/carolinensis A. crecca/carolinensis A. crecca/carolinensis A. crecca/carolinensis A. crecca/carolinensis A. crecca/carolinensis A. crecca/carolinensis A. discors A. discors A. discors A. discors A. discors A. discors A. discors A. discors A. acuta A. acuta

Breeding population Pair density Breeding population Pair density Breeding population Breeding population Breeding population Breeding population Breeding population Breeding population Breeding population Breeding population

DD

Descriptive, TSA NA Descriptive, TSA NA Descriptive, TSA NA

Population growth rate Brood/pair ratio Brood/pair ratio Duckling survival Duckling survival Population growth rate Population growth rate Population growth rate Breeding population size Population growth rate Population growth rate Population growth rate

A. platyrhynchos A. penelope A. penelope A. penelope A. penelope A. penelope A. penelope A. penelope A. americana A. americana A. americana A. americana

Explaining variable

Breeding population size No Breeding population size No Breeding population size Yes

Dependent variable

Species

Table 1 (continued)

56 26

47 40 51

56 12 12 27 26 47 51 56 13 13 13 26 47 40 51 56 27 26

56 12 21 12 17 17 21 27 26 47 40 51

Lawrence et al. 2013 Vickery and Nudds 1984

Jamieson and Brooks 2004 Sæther et al. 2008 Murray et al. 2010

Lawrence et al. 2013 Elmberg et al. 2003 Elmberg et al. 2003 Gardarsson et al. 2008 Vickery and Nudds 1984 Jamieson and Brooks 2004 Murray et al. 2010 Lawrence et al. 2013 Weller 1979 Weller 1979 Hammond and Johnson 1984 Vickery and Nudds 1984 Jamieson and Brooks 2004 Sæther et al. 2008 Murray et al. 2010 Lawrence et al. 2013 Gardarsson et al. 2008 Vickery and Nudds 1984

Lawrence et al. 2013 Elmberg et al. 2003 Pöysä and Pesonen 2003 Elmberg et al. 2003 Pöysä and Pesonen 2003 Pöysä and Pesonen 2003 Pöysä and Pesonen 2003 Gardarsson et al. 2008 Vickery and Nudds 1984 Jamieson and Brooks 2004 Sæther et al. 2008 Murray et al. 2010

Years Source

Eur J Wildl Res (2013) 59:305–321 309

Anasc Anasf Anasf Anasc Anasf

Nest Nest Nest Nest Nest

Nest density Nest density Nest density Nest density Nest density Pair density Pair density Nest density Nest density Nest density Nest density Neighbour distance Neighbour distance Neighbour distance Nest density Nest density

survival survival survival survival survival

survival survival survival survival survival survival survival survival survival survival survival survival survival survival survival survival

Nest Nest Nest Nest Nest Nest Nest Nest Nest Nest Nest Nest Nest Nest Nest Nest

Anasc Anasc Anasc Anasc Anasc Anasc Anasc Anasc Anasc Anasc Anasc Anasc Anasc Anasf Anasf Nest density Nest density Neighbour distance Neighbour distance Neighbour distance

No No No No No

Yes Yes Yes Yes No Yes No Yes Yes Equivocald No Yes Equivocale No No No

No No Yes No No Yes Yes No No No Yes No

Breeding population size Yes Breeding population size Yes

size size size size size size size size size size size size

Recruitment Population growth rate

population population population population population population population population population population population population

A. rubripes A. laysanensis Anasc

Breeding Breeding Breeding Breeding Breeding Breeding Breeding Breeding Breeding Breeding Breeding Breeding

Population growth rate Population growth rate Population growth rate Population growth rate Population growth rate Brood size Breeding population size Population growth rate Population growth rate Population growth rate Population growth rate Population growth rate

DD

A. clypeata A. clypeata A. clypeata A. clypeata A. strepera A. strepera A. strepera A. strepera A. strepera A. strepera A. strepera A. strepera

Explaining variable

Dependent variable

Species

Table 1 (continued)

TSA TSA TSA TSA TSA TSA

TSA TSA TSA TSA TSA

NA NA NA NA EUR NA NA NA NA NA NA NA

Experimental Descriptive Descriptive Experimental Descriptive

Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental Experimental Descriptive Descriptive NA NA NA NA NA

EUR EUR EUR EUR EUR EUR EUR NA NA NA NA NA NA NA NA NA

Wetland Wetland Wetland Wetland Wetland

Landscape Landscape Landscape Landscape Region Landscape Landscape Landscape Landscape Landscape Wetland Landscape Landscape Wetland Wetland Landscape

Region Wetland

Continental Continental Continental Continental Wetland Landscape Region Continental Continental Continental Continental Continental

Mediterranean Mediterranean Mediterranean Mediterranean Mediterranean

Boreal Boreo–nemoral Nemoral Mediterranean Nemoral Boreo–nemoral Nemoral Prairie Boreal Prairie Mediterranean Prairie Prairie Mediterranean Mediterranean Mediterranean

Boreal Tropical island

Multibiome Multibiome Multibiome Multibiome Tundra Prairie Prairie Multibiome Multibiome Multibiome Multibiome Multibiome

Continent Spatial scale Biome

Descriptive NA Descriptive, TSA NA

Descriptive, Descriptive, Descriptive, Descriptive, Descriptive, Descriptive Descriptive, Descriptive, Descriptive, Descriptive, Descriptive, Descriptive,

Study typea

2 2 1 2 2

1 2 2 1 2 2 2 2 1 1 1 2 1 1 1 5

12 14

47 40 51 56 27 13 26 31 47 40 51 56

Ringelman et al. 2011 Ringelman et al. 2011 Ackerman et al. 2004 Ringelman et al. 2011 Ringelman et al. 2011

Elmberg et al. 2009 Gunnarsson and Elmberg 2008 Gunnarsson and Elmberg 2008 Elmberg et al. 2009 Padyšáková et al. 2010 Elmberg and Gunnarsson 2007 Elmberg and Gunnarsson 2007 Sugden and Beyersbergen 1986 Esler and Grand 1993 Larivière and Messier 1998 Ackerman et al. 2004 Sugden and Beyersbergen 1986 Larivière and Messier 1998 Ackerman et al. 2004 Ackerman et al. 2004 Ackerman et al. 2004

Zimpfer and Conroy 2006 Seavy et al. 2009

Jamieson and Brooks 2004 Sæther et al. 2008 Murray et al. 2010 Lawrence et al. 2013 Gardarsson et al. 2008 Hammond and Johnson 1984 Vickery and Nudds 1984 Zeng et al. 1998 Jamieson and Brooks 2004 Sæther et al. 2008 Murray et al. 2010 Lawrence et al. 2013

Years Source

310 Eur J Wildl Res (2013) 59:305–321

Dependent variable

Population growth rate Population growth rate Brood/pair ratio Fledglings per pair Brood/pair ratio Clutch size Brood size Clutch size Breeding population size Population growth rate Duckling survival Nest survival

Breeding population size Post-hatch survival Population growth rate Brood size Fledglings per pair Recruitment Breeding population size Population growth rate Population growth rate Population growth rate Population growth rate Duckling survival Fledglings per pair Population growth rate Breeding population size Breeding population size Population growth rate Population growth rate

Population growth rate Population growth rate Population growth rate Breeding population size Population growth rate

Species

H. histrionicus M. nigra B. clangula B. clangula B. clangula B. clangula B. clangula B. clangula B. clangula B. islandica B. islandica B. albeola

Ay. collaris A. fuligula A. fuligula A. americana A. americana A. americana A. americana A. americana A. americana A. americana A. americana A. valisineria A. valisineria A. valisineria A. valisineria A. valisineria A. valisineria A. valisineria

A. valisineria A. valisineria A. marila A. affinis A. affinis

Table 1 (continued)

Breeding Breeding Breeding Breeding Breeding

population population population population population

Breeding population Nest density Breeding population Breeding population Breeding population Breeding population Breeding population Breeding population Breeding population Breeding population Breeding population Clutch size Breeding population Breeding population Breeding population Breeding population Breeding population Breeding population

Breeding population Breeding population Pair density Pair density Pair density Breeding population Breeding population Pair density Pair density Breeding population Brood density Pair density

Explaining variable

size size size size size

Yes Yes Yes Yes Yes

size Yes Yes size Yes size No size Yes size Yes size Yes size Yes size No size Yes size No No size No size Equivocal size Yes size Yes size No size Yes

size Yes size Yes Yes Yes Yes size No size No No Yes size Yes Equivocal Equivocal

DD

Descriptive, Descriptive, Descriptive, Descriptive, Descriptive,

TSA TSA TSA TSA TSA

Descriptive, TSA Descriptive Descriptive, TSA Descriptive Descriptive Descriptive Descriptive, TSA Descriptive, TSA Descriptive, TSA Descriptive, TSA Descriptive, TSA Experimental Experimental Experimental Descriptive, TSA Descriptive, TSA Descriptive, TSA Descriptive, TSA

Descriptive, TSA Descriptive, TSA Experimental Experimental Descriptive Descriptive Descriptive Descriptive Descriptive Descriptive, TSA Descriptive Experimental

Study typea

NA NA EUR NA NA

NA EUR EUR NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA

EUR EUR EUR EUR EUR EUR EUR EUR EUR EUR NA NA

Continental Continental Wetland Region Continental

Region Wetland Wetland Landscape Continental Continental Region Continental Continental Continental Continental Landscape Landscape Landscape Region Continental Continental Continental

Wetland Wetland Landscape Landscape Wetland Landscape Landscape Landscape Landscape Wetland Landscape Landscape

Multibiome Multibiome Tundra Prairie Multibiome

Prairie Nemoral Tundra Prairie Multibiome Multibiome Prairie Multibiome Multibiome Multibiome Multibiome Prairie Prairie Prairie Prairie Multibiome Multibiome Multibiome

Tundra Tundra Boreal Boreal Boreo–nemoral Boreo–nemoral Boreo–nemoral Boreo–nemoral Boreo–nemoral Tundra Prairie Boreal

Continent Spatial scale Biome

51 56 25 26 51

26 19 25 13 50 50 26 47 40 51 56 8 8 8 26 41 40 47

25 25 12 12 8 21 21 7 7 25 3 2

Murray et al. 2010 Lawrence et al. 2013 Gardarsson and Einarsson 2004 Vickery and Nudds 1984 Murray et al. 2010

Vickery and Nudds 1984 Mihelsons et al. 1985 Gardarsson and Einarsson 2004 Hammond and Johnson 1984 Péron et al. 2012 Péron et al. 2012 Vickery and Nudds 1984 Jamieson and Brooks 2004 Sæther et al. 2008 Murray et al. 2010 Lawrence et al. 2013 Anderson et al. 1997 Anderson et al. 1997 Anderson et al. 1997 Vickery and Nudds 1984 Viljugrein et al. 2005 Sæther et al. 2008 Jamieson and Brooks 2004

Gardarsson and Einarsson 2004 Gardarsson and Einarsson 2004 Pöysä and Pöysä 2002 Pöysä and Pöysä 2002 Söderholm 2004 Fredga and Dow 1983 Fredga and Dow 1983 Eriksson 1979 Eriksson 1979 Gardarsson and Einarsson 2004 Savard et al. 1991 Gauthier and Smith 1987

Years Source

Eur J Wildl Res (2013) 59:305–321 311

Nest survival Duckling survival

T. tadorna T. tadorna

f

e

d

c

b

No No No Equivocal No Yes Yes No Yes Yes Yes No

DD

Breeding population size Yes Breeding population size Yes

Breeding population size Breeding population size Nest density Breeding population size Pair density Pair density Nest density Breeding population size Duckling density Breeding pop. density Breeding pop. density Breeding population size

Explaining variable

Natural nests of several Anas species (Anas spp.)

Dependent on density treatment

Density effect in late season, but not in early

Studies using artificial duck nests, usually resembling those of A. platyrhynchos

“Weak support for DD”, i.e. p<0.1

TSA stands for time series analysis

Population growth rate Population growth rate Clutch size Clutch size Brood size Fledglings per pair Fledglings per pair Recruitment Population growth rate Recruitment Population growth rate Breeding population size

A. affinis/marila A. affinis/marila S. mollissima S. mollissima S. mollissima S. mollissima S. mollissima S. mollissima S. mollissima O. leucocephala O. leucocephala O. jamaicensis

a

Dependent variable

Species

Table 1 (continued)

Descriptive Descriptive

Descriptive, Descriptive, Descriptive Descriptive Descriptive Descriptive Descriptive Descriptive Descriptive, Descriptive Descriptive, Descriptive,

Study typea

EUR EUR

TSA NA TSA NA EUR EUR EUR EUR EUR EUR TSA EUR EUR TSA EUR TSA NA Landscape Landscape

Continental Continental Region Wetland Landscape Landscape Region Wetland Landscape Region Region Region Nemoral Nemoral

Multibiome Multibiome Nemoral Nemoral Boreo–nemoral Boreo–nemoral Nemoral Nemoral Boreo–nemoral Mediterranean Mediterranean Prairie

Continent Spatial scale Biome

13 13

47 56 8 49 34 34 8 49 31 20 20 26

Patterson et al. 1983 Patterson et al. 1983

Jamieson and Brooks 2004 Lawrence et al. 2013 Swennen 1991 Coulson 1999, 2010 Hario and Rintala 2006 Hario and Rintala 2006 Swennen 1991 Coulson 2010 Hario and Rintala 2006 Almaraz and Amat 2004 Almaraz and Amat 2004 Vickery and Nudds 1984

Years Source

312 Eur J Wildl Res (2013) 59:305–321

Eur J Wildl Res (2013) 59:305–321

313

extent pseudo-replicates (evident in, e.g., several studies based on time series data from North American aerial surveys). Therefore, we decided not to do a statistical metaanalysis but instead base patterns and interpretations on actual numbers.

Density dependence in different stages In the following section, we present the results and patterns from our compilation, sorted by the stages in the breeding cycle (i.e. nesting, duckling and recruitment) or, alternatively, in terms of population dynamics (i.e. time series analyses). We use the term ‘case’ instead of ‘study’ because some studies provided more than one entry in the analyses (see Table 1). Nesting stage Among 33 entries pertaining to nesting ducks, 15 found support for DD and 15 did not (Table 2). This pattern of very similar frequencies remained evident when comparing

North American versus European cases, as well as when dabbling ducks and diving ducks were contrasted (note though that only four of 30 cases with a clear result concern diving ducks). However, when entries were grouped by spatial scale (Table 2), there was a strong difference between the ‘wetland’ level, where DD was not found at all, and the ‘landscape’ level, where DD was found three times as often as a lack of it. Nest predation is the main process studied during the nesting stage and, moreover, the most likely causative agent to explain variation in nest success in general. It is well known that many predator species targeting duck nests are mobile and have large home ranges (Sargeant et al. 1993; Phillips et al. 2003; Elmberg and Pöysä 2011). This could explain why DD processes have rarely been documented on the wetland level and more likely to manifest at higher spatial levels, where functional and especially numerical responses of predators occur (cf. Schmidt and Whelan 1999). However, a cautionary note about nonindependence of data is warranted; seven of the nine cases at the wetland level refer to the same study system and research group, i.e. nest survival of wild and artificial nests

Table 2 Pair-wise comparisons of the frequency of density dependence in cases within three different stages of the breeding cycle as well as for population dynamics (factors and variable names are explained in the main text) Factor

Variable name

Sum Dabbling ducks Diving ducks Study type Experiment Descriptive Continent Europe North America Spatial level Wetland Landscape Region Continental Biome Tundra Boreal Boreo–nemoral Nemoral Guild

Prairie Mediterranean Tropical Multibiome

Nesting

Duckling

Totala

Recruitment

Population dynamics

DD No DD Equiv.b DD No DD Equiv.b DD No DD Equiv.b DD No DD Equiv.b DD

No DD

15 14 1 8 7 9 6 0 13 2 0 0 2 3 4

15 12 3 6 9 6 9 9 4 2 0 0 0 3 3

3 2 1 3

5 1 0 0

0 9 0 0

2

1 2 3

1

N/A not applicable a

The sum of nesting, duckling and recruitment columns

b

Equivocal results

16 11 5 4 12 13 3 4 8 3 1 0 6 4 3

20 15 5 3 17 11 9 1 14 4 1 1 7 2 1

3 1 2

2 0 0 1

8 0 0 1

2

3 3 1 2

1

10 4 6 1 9 5 5 0 5 3 2 0 2 1 3

1 0 1 0 1 1 0 1 0 0 0 0 0 0 1

2 0 0 2

0 0 0 0

1 1 1

1 1

1

41 29 12 13 28 27 14 6 25 7 3 0 12 5 11

36 27 9 8 28 18 18 11 20 4 1 1 7 3 7

7 3 4 3 4 1 6 2 5 0 0 0 1 0 1

34 16 18 N/A N/A 10 24 8 2 8 17 8 0 1 0

36 30 6 N/A N/A 4 32 3 0 6 26 3 1 0 0

9 1 0 3

8 9 0 1

5 0 0 0

6 1 1 17

6 0 0 26

314

in the Suisun Marsh in central California (Ackerman et al. 2004; Ringelman et al. 2011; neither study reporting any DD from the 1-ha level and up). These two studies also explain the striking pattern of no DD for the Mediterranean biome (Table 2). For the other biomes, cases with and without DD are roughly equally common, but note that all five prairie cases found support for DD (Table 2). Previous research covering various taxonomic and functional groups of birds show that nest failure is one of the strongest determinants of breeding success and hence fitness (Martin 1988). In the light of this, DD nest predation is often hypothesized to operate, and the overall pattern found in the present review is thus rather unexpected. Many of the duck studies in this category are based on well-replicated robust experimental designs, but as is evident from Table 2, this approach does not provide a different picture than descriptive studies. Ringelman et al. (2011) pointed out that DD nest predation may be difficult to detect if the range of observable nest density is constrained due to a floor/ceiling effect, but available data do not permit a comprehensive evaluation of how densities used in experiments relate to those occurring in nature. However, three European nest predation studies arguably used nest densities ranging from normal to close to unnaturally high with respect to the biome wherein they were conducted (Elmberg and Gunnarsson 2007; Gunnarsson and Elmberg 2008; Elmberg et al. 2009). The importance of nest density relative to other factors affecting nest survival was assessed by Elmberg et al. (2009), who found that density was three times as important to survival of artificial duck nests as was biome (Mediterranean versus boreal) and 1.5 times as important as phenology (early versus late nests).

Eur J Wildl Res (2013) 59:305–321

cases concern the mallard Anas platyrhynchos, for which there are also 14 cases of descriptive studies of DD (Table 1). In the latter, DD was detected in much fewer cases than a lack of DD (four versus ten). The mallard duckling study system thus shows that experiments give different results than descriptive studies, a fact calling for attention in the future. Considering continents, DD appears to be less prevalent in NA than in Europe, but this is probably a pattern confounded by the fact that almost all experimental studies have been carried out in Europe. In contrast to the nesting stage, the duckling stage cases carried out at small geographic scales show a higher proportion of DD. At first, this seems logical as ducklings are not as mobile as adults or the main predators, which both readily move between wetlands and operate at a landscape level. In other words, in the case of ducklings, DD is likely to lead to reduced survival on the wetland level or to an exodus from the natal pond (Dzus and Clark 1997), both of which will be recorded as a reduction in the number of remaining ducklings. However, it should be noted that four of the six cases carried out at the wetland level are the experimental studies mentioned earlier in this section, possibly confounding this conclusion. It is also possible that bottom-up (food supply) regulatory processes (e.g. Sinclair and Krebs 2002) prevail on this scale, making DD easier to detect. At larger scales, cases are few, but a majority of them does not show any support for DD. There is little effect of biome on the prevalence of DD. However, for the prairie biome, there is an interesting difference when nesting stage and duckling stage cases are contrasted; DD is more prevalent in the former, whereas the opposite is true for the latter.

Duckling stage Recruitment stage This is the stage of the breeding cycle with the highest number of cases addressing DD. Yet the numbers of cases with or without reported DD are nearly tied, and this is true also when dabbling and diving ducks are considered separately (Table 2). When studies are broken down further into categories, however, there appears to be some differences in the prevalence of DD. For example, there are more instances of DD in experimental than in descriptive cases. Experimental studies have the advantage of being able to control for many confounding factors that otherwise may create ‘noise’ in the data, and in this sense they can be a better method to identify DD patterns (e.g. Harrison and Cappuccino 1995). A general critique against experiments is that they are sometimes carried out in environments that differ much from natural settings (cf. Ackerman et al. 2004), but concerning DD in ducklings (seven cases; see Table 2), all were carried out in the wild and under natural conditions. Four of the seven

In contrast to the nesting and duckling stages, there were clearly more detected cases of DD (ten) than non-DD cases (one) for the recruitment stage (Table 2). This trend seems to be quite robust, considering the different factors with decent sample sizes in Table 2, i.e. cases were evenly distributed among continents and duck guilds. This finding underlines the importance of studying multiple stages during the breeding period in order to be able to judge whether DD occurs or not (cf. sequential DD; Åström et al. 1996). Moreover, a lack of DD patterns may in fact be an effect of regulatory mechanisms; negative DD occurring during one stage may diminish the probability of detecting the process in later stages. For example, in Elmberg et al. (2005), DD was not evident for almost fledged ducklings, whereas such effects were indeed observed in an earlier stage (the number of hatched broods). Similarly, when increasing the density of breeding common goldeneye Bucephala clangula pairs by

Eur J Wildl Res (2013) 59:305–321

providing nest boxes, Pöysä and Pöysä (2002) found that while the number of breeding pairs increased, the per capita brood and fledgling production was reduced. It is interesting to note that all cases but one showing DD are from descriptive studies, whereas such studies do not provide a similar pattern for the nesting and duckling stages. Based on this pattern, it seems as if DD processes operate mainly in fledged ducks outside the breeding season. When seeking an explanation for the very high proportion of recruitment cases showing DD, it should also be noted that all but one were carried out above the wetland level, that is, mainly on landscape and regional levels. In contrast, many studies carried out during the nesting and duckling stages were done at the wetland level.

315

one case (Seavy et al. 2009) at the smallest scale come from a single lake (i.e. Lake Mývatn; Pöysä and Pesonen 2003; Gardarsson and Einarsson 2004; Gardarsson et al. 2008). The same problem is true for biome (DD more frequent on tundra wetlands). Finally, the occurrence of delayed DD, i.e. a lag of two or more years in the dependence of per capita growth rate (e.g. Turchin 1990; Holyoak 1994), in duck population time series has received relatively little attention. Jamieson and Brooks (2004) found some support for delayed DD in two diving ducks, the redhead (Aythya americana) and canvasback (A. valisineria), but not in the other eight species studied. Other studies explicitly considering delayed DD has not found any support for it (Zeng et al. 1998; Viljugrein et al. 2005; see also Murray et al. 2010; Lawrence et al. 2013).

Population dynamics Our data set comprises 70 cases in which DD was studied in population time series. About half of the cases indicated DD, whereas the other half did not. However, when the two duck guilds were considered separately, a clear pattern emerged. DD was more frequent in diving ducks than in dabbling ducks (Table 2). This supports suggestions in earlier studies that diving ducks are more K-selected and dabbling ducks more r-selected (Patterson 1979; Bailey 1981; Vickery and Nudds 1984; see more in the following text). However, recent analyses of North American data suggest that distinctions based on differences in life history characteristics may not reflect patterns of population dynamics (Sæther et al. 2008; Murray et al. 2010; Lawrence et al. 2013). DD seemed to be more frequent in European (c. 71 %) than in North American (c. 43 %) duck population time series. However, compared to those in North America, relatively few studies have addressed DD in time series of European ducks. It is also important to note that most of the studies from North America are based on data from the annual Breeding Waterfowl Population and Habitat Survey (US Fish and Wildlife Service 2005), which means that results from different studies are not independent. Even so, it is interesting to note that studies in which the same data base has been used have reached rather divergent conclusions about the prevalence and strength of DD in population dynamics (e.g. Viljugrein et al. 2005 versus Lillegård et al. 2008, Sæther et al. 2008 versus Murray et al. 2010; see also Lawrence et al. 2013). A shortcoming in the European data is that time series are from three study areas only: Lake Mývatn in Iceland, south-eastern Finland, and the Iberian peninsula (Pöysä and Pesonen 2003; Almaraz and Amat 2004; Gardarsson and Einarsson 2004; Gardarsson et al. 2008). Considering spatial scale, there seems to be more DD cases at wetland level than at other scales. However, all but

Is density dependence important in breeding ducks? One obvious conclusion from this review is that it is hard to see general patterns in the prevalence of DD, unless data are broken down to subsets, e.g. to stage or spatial level. In case of the former, DD effects obviously occur at all stages of the breeding cycle. During the nesting and the duckling stages, the proportion of cases detecting versus not detecting is roughly the same. However, in cases referring to the recruitment stage, i.e. to survival of fledged ducks until 1 year old at the most, DD was the rule. Another more general pattern concerns the cases addressing population dynamics, in which DD was more frequently demonstrated for diving ducks than for dabbling ducks. This result is in line with the general opinion expressed by, for example, Patterson (1979) and Bailey (1981), namely, that ducks can be categorized either as r- or K-selected species (or species representing a “slow–fast” continuum). For such a distinction, the annual reproductive capacity is considered as being of central importance (Sæther and Engen 2002; Sæther et al. 2002). Among ducks, the potential annual reproduction rate may not vary so much, although mean age of first breeding differs a bit between dabbling and diving ducks (Snow and Perrins 1998). A different prevalence of DD in dabbling versus diving ducks is most likely linked to the stability of their breeding habitats (Southwood 1977). The K-selected diving ducks often breed in relatively stable habitats where populations may approach the carrying capacity of the environment, resulting in DD feedback. The more r-selected dabbling ducks, on the other hand, have the ability to quickly colonize more variable, intermittent and newly created breeding habitats, where carrying capacity is very variable and hard to track (Patterson 1979; Bailey 1981; Vickery and Nudds 1984). We would like to stress though that this type of habitat stability-based explanation

316

for the difference between dabbling and diving ducks may apply primarily to breeding habitats that are highly variable between years, for example, the North American prairies, Mediterranean wetlands and parts of Australia. Nichols (1991) argued that most conclusions about population dynamics in North American ducks are based on mallard data and that it is not clear how relevant these are to other species. Since then (1991), DD has been studied, and detected, for several other species (Table 1). DD processes are indeed important for understanding population dynamics, for example, because harvest decisions are frequently based on the assumption that hunting mortality is either compensatory or additive to natural mortality. In the present and much larger review, the mallard is also the most frequent study organism, which is to say that some generalizations can be made about DD in this species and also that these do not necessarily apply to other species. Interestingly, the mallard is often regarded as an r-selected species and, by definition, therefore not considered to be regulated strongly by DD effects. Nevertheless, we show that DD effects have been demonstrated in all stages of mallard life: nesting– duckling–recruitment–population dynamics. Although DD is widely accepted as important in affecting population growth rate, the prevalence of DD effects may vary among species with different life histories and/or among populations in different ecological contexts (Carrete et al. 2006). This is further complicated by the fact that underlying mechanisms to DD may differ between populations and that even the same population subjected to DD may be so due to several causes (Carrete et al. 2006). Moreover, some researchers dispute DD as such; in the scientific literature, there are alternative suggestions or interpretations to the patterns emerging in the present review (e.g. White 2001; Berryman et al. 2002; Ziebarth et al. 2010). However, the purpose of our review was not to value the different articles, even if some, based on present knowledge and with consideration to criticism from authors who challenge DD effects, indeed could be questioned. As a matter of fact, in some cases in which studies have been cited as having detected DD, it turns out that quite a few of the original studies actually do not address DD at all or have come to other conclusions than when cited (e.g. Batt et al. 1992). In these cases, we have relied on the original work. Another problem which confuses the overall picture and makes generalizations about DD in ducks hard is the fact that methods differ a lot among studies. As a case in point, some papers addressing the same question with data from the same system have reached opposite conclusions. For example, although using largely overlapping data (Breeding Waterfowl Population and Habitat Survey, US Fish and Wildlife Service 2005; see also Smith 1995), to study duck population dynamics, Viljugrein et al. (2005) found support for DD population dynamics in prairie-nesting mallards,

Eur J Wildl Res (2013) 59:305–321

whereas Lillegård et al. (2008) did not. The latter authors suggest that, in addition to differences in the composition of data sets, one important cause for diverging conclusions may be that Viljugrein et al. (2005) detrended the data by the annual variation in the number of ponds, a procedure that may lead to an overestimation of the strength of DD. In general, differences in the methods used to model population dynamics of North American ducks probably explain the quite different pictures given in studies about the importance of DD in population dynamics (see Table 1). Another source of confusion is the uncertainty of the population abundance estimates for North American ducks (Lillegård et al. 2008), a problem that has been recognized to be influential also in other global collections of population time series (e.g. Knape and de Valpine 2012 and references therein). One could argue that a potential bias in our synthesis is that studies finding negative results (i.e. non-significant) of DD are underrepresented in the literature due to withheld reporting and editorial rejection of such manuscripts (Csada et al. 1996). However, there are also arguments against this suspicion. Koricheva (2003) found that ecological studies with non-significant results are published as frequently as those with significant result, although in journals with different impact factors. Since papers in all types of peerreviewed journals, regardless of impact factor, were used in our review, we do not think that under-reporting seriously biases our conclusions. We do suspect though that descriptive studies without clear trends are underreported.

Implications DD is of central importance and directly linked to the concepts of sustainable harvesting and compensatory hunting mortality (Anderson and Burnham 1976; Nichols et al. 1984; Boyce et al. 1999; Pöysä et al. 2004; Sedinger and Herzog 2012; Pöysä et al. 2013). In that context, our review provides good news because we show that DD has been documented in all stages of the breeding cycle of ducks. However, more crucial than the occurrence of DD per se is the exact timing of DD processes in relation to the timing of harvest (e.g. Kokko et al. 1998; Kokko 2001). Possible occurrence of sequential DD, as has been experimentally demonstrated for the mallard (Elmberg et al. 2005), complicates these temporal aspects further (Åström et al. 1996; Ratikainen et al. 2008). Most duck hunting in Europe and North America occurs in autumn, after the young have fledged and before the onset of winter, when lower temperatures may lead to increased mortality risk through higher metabolic requirements combined with lower food availability. Unfortunately, little is still known about the timing of DD events after the breeding season (i.e. during the recruitment stage), and it is thus not

Eur J Wildl Res (2013) 59:305–321

possible to assess the prevalence of compensatory mortality during that period (i.e. post-harvest DD mortality). If compensation does not occur during the autumn/winter period or if it is only partial, hunting mortality will be additive and result in lower breeding population size. In that case, either the survival of breeding individuals or reproductive output in the next breeding season should be DD in order to achieve compensation (i.e. compensatory natality; e.g. Boyce et al. 1999). Even though the cases reporting DD were not more frequent than those not finding DD during the nesting and duckling stages, it did occur in some systems, indicating that possibilities for compensation during the breeding season exist. Is this reflected in the outcome of harvest management models? The only system in which this question can be evaluated is the Adaptive Harvest Management programme for North American waterfowl, in which four alternative population models for a given stock are considered, the alternatives being combinations of two survival hypotheses (additive versus compensatory hunting mortality) and two reproductive hypotheses (strongly versus weakly DD) (e.g. Nichols et al. 2007). Current assessment of the alternative models for midcontinent mallards, for example, indicate strong support for weakly DD reproduction and more support for additive than compensatory hunting mortality (US Fish and Wildlife Service 2011). All in all, mallard is the best studied species in terms of the occurrence of DD (see Table 1). Still, even for this species, we do not yet have a single study system for which there is enough information about the occurrence of DD at all the critical life stages and in population dynamics. This means that, in general, management and conservation of duck populations worldwide is currently carried out under considerable uncertainty. Habitat loss or deterioration, possibly resulting in concentration of individuals to remaining suitable habitat patches, may accelerate DD effects on vital rates and also affect population dynamics. Indeed there are indications that the strength of DD in ducks may vary both spatially and temporally, depending on changes in habitat availability and variability (cf. Grøtan et al. 2009). For example, geographical gradients in the strength of DD in population dynamics of some North American duck species have been found to be driven by geographical gradients in the availability and variability of wetlands (Sæther et al. 2008). On the other hand, Murray et al. (2010) found that the degree of DD in population dynamics of North American ducks has weakened from 1955–1979 to 1980–2005, i.e. in conjunction with a putative decline in wetland availability. Overall, there are differences among duck species in their flexibility to respond to temporal changes in habitat quality (Johnson and Grier 1988), and those differences may also be reflected by changes in DD processes due to redistribution of breeding individuals among habitats.

317

Recent models based on different climate change scenarios predict severe drought and loss of wetlands in the North American prairies, a key breeding area for ducks (Poiani and Johnson 1991; Sorenson et al. 1998; Johnson et al. 2005). Changes in land use due to agricultural development also cause loss of breeding habitats in the prairie areas (Bethke and Nudds 1995). Moreover, although much less dramatic, declines have been reported in the number and size of lakes and wetlands also from boreal areas where wetland water levels are more stable between years (Roach et al. 2011 and references therein). Predicted changes in habitat quality and availability, potentially affecting the strength of DD in different population processes of ducks, may thus have complex and unpredictable consequences for the management and conservation of duck populations. Our review of DD in vital rates and population dynamics in ducks also has implications for the ever-lasting debate about whether populations exhibit DD regulation or not (e.g. Turchin 1995; Turchin 1999; White 2001; Berryman 2004; Ratikainen et al. 2008; Ziebarth et al. 2010 and references therein). A large part of the confusion in this debate stems from the fact that the exploration of DD has traditionally focused on population time series (e.g. Sibly et al. 2005; Brook and Bradshaw 2006; Ziebarth et al. 2010; Knape and de Valpine 2012), whereas the actual mechanisms underlying population dynamics and regulation have received less attention. Still, the main goal should be to understand population dynamics and regulation. To that end, in addition to analysing population time series, it is needed to recognize potential regulatory mechanisms and to study them directly (Murdoch 1994; Turchin 1999). In this respect, our review provides a rare example of a system for which critical DD mechanisms have been identified and the occurrence and prevalence of DD at critical stages of the life cycle have been assessed, extending to the assessment of DD population regulation. Even though there still are serious gaps in the knowledge about the prevalence of DD in population processes of ducks, the present synthesis should help in building more realistic population models for ducks and other species. Successful management and conservation need information about both the occurrence and type of population regulation and the mechanisms underlying the observed dynamics (e.g. Pöysä and Pesonen 2003; Reynolds and Freckleton 2005; Sibly et al. 2005; Herrando-Pérez et al. 2012a).

Desiderata for future studies There are some inevitable and nevertheless problematic biases in the research to date on DD in ducks. The geographic–biotic focus on temperate parts of Europe and North America is natural as these are main breeding areas for many species and because ducks are a widely managed

318

and harvested resource in these countries. However, there are no studies from Russia despite this country holding a large share of the ducks breeding in the Northern Hemisphere. There is also a strong bias to hunted species, which is also natural but at the same time unfortunate in the sense that DD effects may be relevant to study in other species, some of which are declining and of conservation concern (Delany and Scott 2006). Future studies should also address some other biases and knowledge gaps not evident from Table 2. Firstly, time series analyses are strongly concentrated to North America (56 of 70 entries) and often rely heavily on data from areas with large between-year variations in habitat availability and suitability. Hence, a more representative and comprehensive picture of DD in population dynamics would be obtained if the time series approach is applied to data from Europe and other areas. Secondly, all experimental work explicitely or implicitly concern very few species, and mostly the mallard. Obviously, this needs to be complemented with additional species. Thirdly, further experimental studies are needed to infer causality from the DD patterns revealed in descriptive studies. Especially valuable are studies combining descriptive and experimental data and/or using different spatial levels simultaneously (cf. Ackerman et al. 2004; Elmberg et al. 2009). Already in 1989, Wiens discussed the scale dependence of DD, specifically the processes responsible for making DD ‘arise’ or ‘disappear’ from one level to the next. However, scaling up experimentation beyond the landscape level will always be problematic but can, in some cases, be achieved through changes in management and harvesting policies applicable to larger areas. Perhaps the most problematic shortcoming when the literature on DD in breeding ducks is seen as a whole is the notorious lack of studies in which per capita fitness measures are related to availability of limiting resources. As was pointed out in the “Introduction”, in most cases, density per se can merely be a proxy for the agents actually causing a reduction in fitness as population density increases. Even studies that are exceptions to this lack (e.g. Elmberg et al. 2003) have shortcomings, for example, in the sense that they address only one of several possible limiting factors. Finally—although outside the scope of the present review sensu stricta—DD needs to be studied throughout the annual cycle of focal organisms (Dhondt 2012), that is, to a larger extent also outside the breeding season. Fall and winter staging sites as well as wintering areas can be limiting when it comes to food availability (e.g. DuBowy 1988; but see Bethke 1991). In this context, not only habitat and food availability are central; DD predation is a largely neglected topic as are DD effects of parasitism and social interactions. The ultimate duck predator—man—also needs more attention, that is, the possible role of harvest acting as a DD process.

Eur J Wildl Res (2013) 59:305–321

Conclusions Subdividing studies and data by the variables considered in this review produces some intriguing patterns in the proportion of cases reporting DD. In contrast to the nesting and the duckling stages, which have roughly equal proportions of cases in favour/against DD, the latter is more frequently reported for the recruitment stage. This suggests that DD processes are more likely to occur for fledged ducks outside the breeding season, a possibility that needs careful consideration before judging if populations are subjected to regulatory mechanisms or not. In studies based on time series data (population dynamics), it is clear that DD is more frequently reported for diving ducks than for dabbling ducks. This may be due to diving ducks generally being closer to their carrying capacity, hence experiencing conditions more typical of K-selected species, in contrast to dabbling ducks that more often breed in more variable and ephemeral wetlands (conditions typical for more r-selected species). Despite the large body of research synthesized here, much remains to be understood about the timing of DD in the annual cycle. Management and conservation of duck populations worldwide are thus currently carried out under considerable uncertainty. Moreover, successful management and conservation need information on both the occurrence and type of population regulation and the mechanisms underlying the observed dynamics, and the present synthesis should help in building more realistic models. Although a fair number of studies have considered DD in breeding ducks, there are still some major knowledge gaps that need to be considered in future studies. These include (a) studies from Russia, which holds a considerable part of the duck populations in the Northern Hemisphere, (b) more experimental studies on more species carried out during other stages than nesting are needed to distinguish DD processes from density-independent processes, (c) more time series analyses on European ducks and, finally, (d) studies relating duck numbers to limiting resources, which likely is the most relevant measure of density. Acknowledgments We are grateful to Todd Arnold, Tom Nudds, Pablo Almaraz and three anonymous reviewers for commenting on this manuscript. This study was supported by grants from the Swedish Environmental Protection Agency (V-220-08, V-205-09) and the Kone Foundation, Finland.

References Ackerman JT, Blackmer AL, Eadie JM (2004) Is predation on waterfowl nests density dependent?—tests at three spatial scales. Oikos 107:128–140 Almaraz P, Amat JA (2004) Multi-annual spatial and numeric dynamics of the white-headed duck Oxyura leucocephala in southern

Eur J Wildl Res (2013) 59:305–321 Europe: seasonality, density dependence and climatic variability. J Anim Ecol 73:1013–1023 Amundson CL, Arnold TW (2011) The role of predator removal, density-dependence, and environmental factors on mallard duckling survival in North Dakota. J Wildl Manage 75:1330–1339 Anderson DR, Burnham KP (1976) Population ecology of the mallard. VI. The effect of exploitation on survival. United States Department of the Interior, Fish and Wildlife Service Resource Publication 128 Anderson MG, Emery RB, Arnold TW (1997) Reproductive success and female survival affect local population density of canvasbacks. J Wildl Manage 61(4):1174–1191 Andrén H (1991) Predation: an overrated factor for over-dispersion of birds' nests? Anim Behav 41:1063–1069 Bailey RO (1981) A theoretical approach to problems in waterfowl management. Trans N Am Wildl Nat Resour Conf 46:58–71 Batt BDJ, Afton AD, Anderson MG, Ankney CD, Johnson DH, Kadlec JA, Krapu GL (1992) Ecology and management of breeding waterfowl. University of Minnesota Press, Minneapolis Berryman AA, Lima Arce M, Hawkins BA (2002) Population regulation, emergent properties, and a requiem for density dependence. Oikos 99(3):600–606 Berryman AA (2004) Limiting factors and population regulation. Oikos 105(3):667–670 Bethke RW (1991) Seasonality and interspecific competition in waterfowl guilds—a comment. Ecology 72(3):1155–1158 Bethke RW, Nudds TD (1995) Effects of climate change and land use on duck abundance in Canadian prairie–parklands. Ecol Appl 5(3):588–600 Boyce MS, Sinclair ARE, White GC (1999) Seasonal compensation of predation and harvesting. Oikos 87:419–426 Breckle S-W (2002) Walter's vegetation of the earth—the ecological systems of the geo-biosphere, 4th edn. Springer, Heidelberg Brook BW, Bradshaw JA (2006) Strength of evidence for density dependence in abundance time series of 1198 species. Ecology 87(6):1445–1451 Cappuccino N, Price PW (1995) Population dynamics: New approaches and synthesis. Academic Press, San Diego Carrete M, Sánchez-Zapata JA, Tella JL, Gil-Sánchez JM, Moleón M (2006) Components of breeding performance in two competing species: habitat heterogeneity, individual quality and densitydependence. Oikos 112(3):680–690 Coluccy JM, Yerkes T, Simpson R, Simpson JW, Armstrong L, Davis J (2008) Population dynamics of breeding mallards in the Great Lakes states. J Wildl Manage 72:1181–1187 Coulson JC (1999) Variation in clutch size of the Common Eider: a study based on 41 breeding seasons on Coquet Island, Northumberland, England. Waterbirds 22(2):225–238 Coulson JC (2010) A long-term study of the population dynamics of Common Eiders Somateria mollissima: why do several parameters fluctuate markedly? Bird Study 57(1):1–18 Coulton DW, Clark RG, Howerter DW, Anderson MG, Wassenaar LI (2011) Costs and benefits of natal dispersal in yearling mallards Anas platyrhynchos. J Avian Biol 42:123–133 Csada RD, James PC, Espie RHM (1996) The “file drawer problem” of non-significant results: does it apply to biological research? Oikos 76(3):591–593 Delany S, Scott DA (2006) Waterbird population estimates, 4th edn. Wetlands International, Wageningen Dhondt AA (2012) Interspecific competition in birds. Oxford University Press, Oxford DuBowy PJ (1988) Waterfowl communities and seasonal environments: temporal variability in interspecific competition. Ecology 69:1439–1453

319 Dzus EH, Clark RG (1997) Overland travel, food abundance, and wetland use by mallards: relationships with offspring survival. Wilson Bull 109:504–515 Elmberg J (2003) Density-dependent breeding success in mallards Anas platyrhynchos on a eutrophic lake. Wildl Biol 9:67–73 Elmberg J, Nummi P, Pöysä H, Sjöberg K (2003) Breeding success of sympatric dabbling ducks in relation to population density and food resources. Oikos 100:333–341 Elmberg J, Gunnarsson G, Pöysä H, Sjöberg K, Nummi P (2005) Within-season sequential density dependence regulates breeding success in mallards Anas platyrhynchos. Oikos 108:582– 590 Elmberg J, Gunnarsson G (2007) Manipulated density of adult mallards affects nest survival differently in different landscapes. Can J Zool 85:589–595 Elmberg J, Folkesson K, Guillemain M, Gunnarsson G (2009) Putting density dependence in perspective: nest density, nesting phenology, and biome, all matter to survival of simulated mallard Anas platyrhynchos nests. J Avian Biol 40:317–326 Elmberg J, Pöysä H (2011) Is the risk of nest predation heterospecifically density-dependent in precocial species belonging to different nesting guilds? Can J Zool 89:1164–1171 Eriksson MOG (1979) Aspects of the breeding biology of the goldeneye Bucephala clangula. Holarct Ecol 2(3):186–194 Esler D, Grand JB (1993) Factors influencing depredation of artificial duck nests. J Wildl Manage 57:244–248 Fredga S, Dow H (1983) Annual variation in the reproductive performance of goldeneyes. Wildfowl 34:120–126 Gardarsson A, Einarsson A (2004) Resource limitation of diving ducks at Mývatn: food limits production. Aquat Ecol 38:285–295 Gardarsson A, Petersen A, Einarsson Á (2008) Population limitation in dabbling ducks at Mývatn, Iceland. Acta Zool Acad Sci Hung 54(suppl 1):131–143 Gauthier G, Smith JNM (1987) Territorial behaviour, nest-site availability, and breeding density in buffleheads. J Anim Ecol 56(1):171–184 Grøtan V, Saether B-E, Engen S, van Balen JH, Perdeck AC, Visser ME (2009) Spatial and temporal variation in the relative contribution of density dependence, climate variation and migration to fluctuations in the size of great tit populations. J Anim Ecol 78:447–459 Gunnarsson G, Elmberg J, Sjöberg K, Pöysä H, Nummi P (2006) Experimental evidence for density-dependent survival in mallard (Anas platyrhynchos) ducklings. Oecologia 149:203–213 Gunnarsson G, Elmberg J (2008) Density-dependent nest predation— an experiment with simulated mallard nests in contrasting landscapes. Ibis 150:259–269 Gunnarsson G, Elmberg J, Dessborn L, Jonzén N, Pöysä H, Valkama J (2008) Survival estimates, mortality patterns, and population growth of Fennoscandian mallards Anas platyrhynchos. Ann Zool Fenn 45:483–495 Hammond MC, Johnson DH (1984) Effects of weather on breeding ducks in North Dakota. US Fish and Wildlife Service, Fish and Wildlife Technical Report 1 Hanski I, Woiwod I, Perry J (1993) Density dependence, population persistence, and largely futile arguments. Oecologia 95:595–598 Hario M, Rintala J (2006) Fledgling production and population trends in Finnish common eiders (Somateria mollissima mollissima)— evidence for density dependence. Can J Zool 84:1038–1046 Harrison F (2011) Getting started with meta-analysis. Methods Ecol Evol 2:1–10 Harrison S, Cappuccino N (1995) Using density manipulation experiments to study population regulation. In: Cappuccino N, Price PW (eds) Populations dynamics. Academic Press, San Diego, pp 131–147

320 Herrando-Pérez S, Delean S, Brook BW, Bradshaw CJA (2012a) Decoupling of component and ensemble density feedbacks in birds and mammals. Ecology 93(7):1728–1740 Herrando-Pérez S, Delean S, Brook BW, Bradshaw CJA (2012b) Density dependence: an ecological Tower of Babel. Oecologia 170(3):585–603 Hill DA (1984a) Clutch predation in relation to nest density in mallard and tufted duck. Wildfowl 35:151–156 Hill DA (1984b) Population regulation in the mallard (Anas platyrhynchos). J Anim Ecol 53:191–202 Hoekman ST, Mills LS, Howerter DW, Devries JH, Ball IJ (2002) Sensitivity analyses of the life cycle of midcontinent mallards. J Wildl Manage 66(3):883–900 Holyoak M (1994) Identifying delayed density dependence in timeseries data. Oikos 70:296–304 Jamieson LE, Brooks SP (2004) Density dependence in North American ducks. Anim Biodivers Conserv 27(1):113–128 Johnson DH, Grier JW (1988) Determinants of breeding distribution of ducks. Wildl Monogr 100:1–37 Johnson FA, Kendall WL, Dubovsky JA (2002) Conditions and limitations on learning in the adaptive management of mallard harvests. Wildl Soc Bull 30(1):176–185 Johnson WC, Millett BV, Gilmanov T, Voldseth RA, Guntenspergen GR, Naugle DE (2005) Vulnerability of northern prairie wetlands to climate change. Bioscience 55:863–872 Jonzén N, Lundberg P (1999) Temporally structured densitydependence and population management. Ann Zool Fenn 36:39–44 Kaminski RM, Gluesing EA (1987) Density- and habitat-related recruitment in mallards. J Wildl Manage 51:141–148 Knape J, de Valpine P (2012) Are patterns of density dependence in the global population dynamics database driven by uncertainty about population abundance? Ecol Lett 15:17–23 Kokko H, Pöysä H, Lindström J, Ranta E (1998) Assessing the impact of spring hunting on waterfowl populations. Ann Zool Fenn 35:195–204 Kokko H (2001) Optimal and suboptimal use of compensatory responses to harvesting: timing of hunting as an example. Wildl Biol 7(3):141–150 Koricheva J (2003) Non-significant results in ecology: a burden or a blessing in disguise? Oikos 102:397–401 Krüger O, Chakarov N, Nielsen JT, Looft V, Grünkorn T, Struwe-Juhl B, Møller AP (2012) Population regulation by habitat heterogeneity or individual adjustment? J Anim Ecol 81:330–340 Larivière S, Messier F (1998) Effect of density and nearest neighbours on simulated waterfowl nests: can predators recognize highdensity nesting patches? Oikos 83:12–20 Lawrence JD, Gramacy RB, Thomas L, Buckland ST (2013) The importance of prior choice in model selection: a density dependence example. Meth Ecol Evol 4:25–33 Lillegård M, Engen S, Sæther B-E, Grøtan V, Drever MC (2008) Estimation of population parameters from aerial counts of North American mallards: a cautionary tale. Ecol Appl 18(1):197–207 Lindström Å, Enemar A, Andersson G, von Proschwitz T, Nyholm NEI (2005) Density-dependent reproductive output in relation to a drastically varying food supply: getting the density measure right. Oikos 110:155–163 Martin TE (1988) Processes organizing open-nesting bird assemblages: competition or nest predation? Evol Ecol 2:37–50 McCallum H (2000) Population parameters: Estimation for ecological models. Blackwell Science, Oxford Mihelsons HA, Mednis AA, Blums PN (1985) Regulatory mechanisms of numbers in breeding populations of migratory ducks. Proc Int Ornithol Congr Mosc 18:797–802 Murdoch WW (1994) Population regulation in theory and practice. Ecology 75(2):271–287

Eur J Wildl Res (2013) 59:305–321 Murray DL, Anderson MG, Steury TD (2010) Temporal shift in density dependence among North American breeding duck populations. Ecology 91(2):571–581 Newton I (1998) Population limitation in birds. Academic Press, San Diego Nichols JD, Conroy MJ, Anderson DR, Burnham KP (1984) Compensatory mortality in waterfowl populations: a review of the evidence and implications for research and management. Trans N Am Wildl Nat Resour Conf 49:535–554 Nichols JD (1991) Responses of North American duck populations to exploitation. In: Perrins CM, Lebreton J-D, Hirons GJM (eds) Bird population studies: Relevance to conservation and management. Oxford University Press, Oxford, pp 498–525 Nichols JD, Runge MC, Johnson FA, Williams BK (2007) Adaptive harvest management of North American waterfowl populations: a brief history and future prospects. J Ornithol 148(Suppl 2):S343– S349 Padyšáková E, Šálek M, Poledník L, Sedláček F, Albrecht T (2010) Predation on simulated duck nests in relation to nest density and landscape structure. Wildl Res 37(7):597–603 Paradis E, Baillie SR, Sutherland WJ, Gregory RD (2002) Exploring density-dependent relationships in demographic parameters in populations of birds at a large spatial scale. Oikos 97:293–307 Patterson IJ, Makepeace M, Williams M (1983) Limitation of local population size in the shelduck. Ardea 71:105–116 Patterson JH (1979) Can ducks be managed by regulation? Experiences in Canada. Trans N Am Wildl Nat Resour Conf 44:130–139 Pehrsson O (1991) Egg and clutch size in the mallard as related to food quality. Can J Zool 69:156–162 Péron G, Nicolai CA, Koons DN (2012) Demographic response to perturbations: the role of compensatory density dependence in a North American duck under variable harvest regulations and changing habitat. J Anim Ecol 81(5):960–969 Phillips ML, Clark WR, Sovada MA, Horn DJ, Koford RR, Greenwood RJ (2003) Predator selection of prairie landscape features and its relation to duck nest success. J Wildl Manage 67(1):104– 114 Poiani KA, Johnson WC (1991) Global warming and prairie wetlands. Bioscience 41:611–618 Pospahala RS, Anderson DR, Henny CJ (1974) Population ecology of the mallard. II. Breeding habitat conditions, size of the breeding populations, and production indices. US Fish Wildl Serv Wildl Res Rep 115:1–73 Pöysä H (1987) Costs and benefits of group foraging in the teal (Anas crecca). Behaviour 103:123–140 Pöysä H (2001) Dynamics of habitat distribution in breeding mallards: assessing the applicability of current habitat selection models. Oikos 94:365–373 Pöysä H, Pöysä S (2002) Nest-site limitation and density-dependence of reproductive output in the common goldeneye Bucephala clangula: implications for the management of cavity-nesting birds. J Appl Ecol 39:502–510 Pöysä H, Pesonen M (2003) Density dependence, regulation and openclosed populations: insights from the wigeon, Anas penelope. Oikos 102:358–366 Pöysä H, Elmberg J, Gunnarsson G, Nummi P, Sjöberg K (2004) Ecological basis of sustainable harvesting: is the prevailing paradigm of compensatory mortality still valid? Oikos 104(3):612– 615 Pöysä H, Dessborn L, Elmberg J, Gunnarsson G, Nummi P, Sjöberg K, Suhonen S, Söderquist P (2013) Harvest mortality in North American mallards: a reply to Sedinger and Herzog. J Wildl Manage, doi:10.1002/jwmg.532 Ranta E, Lundberg P, Kaitala V (2006) Ecology of populations. Cambridge University Press, Cambridge Ratikainen II, Gill JA, Gunnarsson TG, Sutherland WJ, Kokko H (2008) When density dependence is not instantaneous: theoretical

Eur J Wildl Res (2013) 59:305–321 developments and management implications. Ecol Lett 11:184– 198 Reynolds JD, Freckleton RP (2005) Population dynamics: growing to extremes. Science 309:567–568 Ringelman KM, Eadie JM, Ackerman JT (2011) Density-dependent nest predation in waterfowl: the relative importance of nest density versus nest dispersion. Oecologia 169(3):695–702 Roach J, Griffith B, Verbyla D, Jones J (2011) Mechanisms influencing chances in lake area in Alaskan boreal forests. Global Change Biol 17:2567–2583 Royama T (1992) Analytical population dynamics. Chapman & Hall, London Sæther B-E, Engen S (2002) Pattern of variation in avian population growth rates. Phil Trans R Soc B 357:1185–1195 Sæther B-E, Engen S, Matthysen E (2002) Demographic characteristics and population dynamical patterns of solitary birds. Science 295:2070–2073 Sæther B-E, Lillegård M, Grøtan V, Drever MC, Engen S, Nudds TD, Podruzny KM (2008) Geographical gradients in the population dynamics of North American prairie ducks. J Anim Ecol 77:869– 882 Sargeant AB, Greenwood RJ, Sovada MA, Shaffer TL (1993) Distribution and abundance of predators that affect duck production— prairie pothole region. US Fish and Wildlife Service Resource Publication 194 Savard J-PL, Smith GEJ, Smith JNM (1991) Duckling mortality in Barrow's goldeneye and bufflehead broods. Auk 108(3):568–577 Schmidt KA, Whelan CJ (1999) Nest predation on woodland songbirds: when is nest predation density dependent? Oikos 87:65–74 Seavy NE, Reynolds MH, Link WA, Hatfield JS (2009) Postcatastrophe population dynamics and density dependence of an endemic island ducks. J Wildl Manag 73(3):414–418 Sedinger JS, Herzog MP (2012) Harvest and dynamics of duck populations. J Wildl Manage 76(6):1108–1116 Sheaffer SE (1998) Recruitment models for mallards in eastern North America. Auk 115:988–997 Sibly RM, Barker D, Denham MC, Hone J, Pagel M (2005) On the regulation of populations of mammals, birds, fish, and insects. Science 309:607–610 Sinclair ARE, Krebs CJ (2002) Complex numerical responses to topdown and bottom-up processes in vertebrate populations. Phil Trans R Soc B 357:1221–1231 Smith GW (1995) A critical review of the aerial and ground surveys of breeding waterfowl in North America. Biol Sci Rep 5:1–252 Snow DW, Perrins CM (1998) The birds of the western Palearctic, vol 1. Oxford University Press, Oxford

321 Sorenson LG, Goldberg R, Root TL, Anderson MG (1998) Potential effects of global warming on waterfowl populations breeding in the Northern Great Plains. Clim Chang 40:343–369 Southwood TRE (1977) Habitat, the templet for ecological strategies? J Anim Ecol 46(2):336–365 Stephens PA, Sutherland WJ, Freckleton RP (1999) What is the Allee effect? Oikos 87(1):185–190 Sugden LG, Beyersbergen GW (1986) Effect of density and concealment on American crow predation of simulated duck nests. J Wildl Manage 50:9–14 Swennen C (1991) Development and population dynamics of common eider Somateria mollissima colonies in the Netherlands. Dan Rev Game Biol 16(1):63–74 Söderholm S (2004) Density dependent breeding success of goldeneye Bucephala clangula at Angarnsjöängen. Ornis Svec 7:169–179 Turchin P (1990) Rarity of density-dependence or population regulation with lags? Nature 344:660–663 Turchin P (1995) Population regulation: Old arguments and a new synthesis. In: Cappuccino N, Price PW (eds) Population dynamics: New approaches and synthesis. Academic Press, San Diego, pp 19–40 Turchin P (1999) Population regulation: a synthetic view. Oikos 84(1):153–159 U.S. Fish and Wildlife Service (2005) Waterfowl population status, 2005. Department of the Interior, Washington U.S. Fish and Wildlife Service (2011) Adaptive harvest management: 2011 hunting season. Department of Interior, Washington Weller MW (1979) Density and habitat relationships of blue-winged teal nesting in northwestern Iowa. J Wildl Manage 43:367–374 White TCR (2001) Opposing paradigms: regulation or limitation of populations? Oikos 93:148–152 Vickery WL, Nudds TD (1984) Detection of density-dependent effects in annual duck censuses. Ecology 65:96–104 Wiens JA (1989) Spatial scaling in ecology. Funct Ecol 3(4):385–397 Viljugrein H, Stenseth NC, Smith GW, Steinbakk GH (2005) Density dependence in North American ducks. Ecology 86(1):245–254 Zeng Z, Nowierski RM, Taper ML, Dennis B, Kemp WP (1998) Complex population dynamics in the real world: modeling the influence of time-varying parameters and time lags. Ecology 79(6):2193–2209 Ziebarth NL, Abbott KC, Ives AR (2010) Weak population regulation in ecological time series. Ecol Lett 13:21–31 Zimpfer NL, Conroy MJ (2006) Models of production rates in American black duck populations. J Wildl Manage 70(4):947–954 Åström M, Lundberg P, Lundberg S (1996) Population dynamics with sequential density-dependence. Oikos 75:174–181