Seed dispersal by wintering ducks in a coastal wetland of eastern China

1.   Introduction

Dispersal is a key ecological process underlying species distribution and a pivotal determinant of plant distribution, demography, and genetic structure (Vekemans and Hardy, 2004; Caughlin et al., 2014; Silva et al., 2021). Waterbird-mediated zoochory enables wetland species to traverse the terrestrial matrix between discrete water bodies, thereby fulfilling a crucial ecological role for freshwater communities (Figuerola and Green, 2002; Green and Elmberg, 2014; Silva et al., 2021). Endozoochory is the primary way by which waterbirds disperse plant seeds, from which wetland plants benefit by enhanced dispersal and increased diversity (Hattermann et al., 2019; Silva et al., 2021; van Leeuwen et al., 2023). Waterbirds are effective seed dispersers due to their deliberate ingestion of large quantities of seeds as food, a portion of which are viably egested following multiple hours of retention in their digestive systems (van Leeuwen et al., 2012; Soons et al., 2016). They can move independently of the landscape structure during retention of seeds, either on small spatio-temporal scales between roosting and foraging areas or over long distances during seasonal migrations on larger spatio-temporal scales (Kleyheeg et al., 2017; Martín-Vélez et al., 2021b; van Leeuwen et al., 2023). The capacity of waterbirds to ingest numerous plant seeds not only boosts the quantity of seeds dispersed but also strengthens their role as seed disperser (Green et al., 2013; Lovas-Kiss et al., 2018; González-Varo et al., 2024). The digestive system of waterbirds can process these seeds, with some retaining their viability, enabling them to maintain their germination capacity after digestion (Wilkinson et al., 2017; van Leeuwen et al., 2023). In this process, among the seeds ingested by waterbirds such as ducks, some are digested within their system, while others remain undigested and are dispersed through their feces to new areas, thus facilitating the geographic spread of the seeds (Kleyheeg, 2015; Silva et al., 2021; Ando et al., 2023). In addition, endozoochory of waterbirds also contributes to seed dispersal in different habitats (Green et al., 2013; Lovas-Kiss et al., 2019). Waterbirds can spread seeds from wetland environments to terrestrial environments, or from coastal areas to inland areas, dispersing seeds to suitable habitats in a wider geographical range, thereby improving the viability and adaptability of plant populations (Kleyheeg et al., 2017; Almeida et al., 2022; Green et al., 2023).

As a major group within waterbird seed dispersal networks, ducks play a significant ecological role, not only in the diversity of seeds they disperse but also in ecological functions (Soons et al., 2016; Kleyheeg et al., 2019). Ducks usually spreads a variety of seeds, including aquatic plants and wetland marginal plants, such as the seeds of Nymphaeaceae, Typhaeaceae and Poaceae (Farmer et al., 2017; Lovas-Kiss et al., 2018; Almeida et al., 2022). Ducks have many advantages in seed dispersal due to their unique ecological habits and behavioral patterns. During foraging, they consume a greater quantity of plant seeds, and as migratory birds, they contribute more to the long-distance dispersal of seeds compared to resident species (Green et al., 2023; Urgyán et al., 2023; González-Varo et al., 2024). The wide distribution and high density of ducks in wetland ecosystems allow frequent access to different wetland habitats and act as a bridge connecting different plant communities in the seed dispersal network (Green, 1996; Amezaga et al., 2002). Moreover, wetland ecosystems often face water level fluctuations and seasonal changes, and ducks can flexibly adjust their activities and foraging behavior to cope with these changes, ensuring the continuity and effectiveness of seed dispersal (Guillemain et al., 2007; Ross and Adam, 2013; Soons et al., 2016).

Recent studies have demonstrated that ducks disperse a high diversity and abundance of seeds from angiosperms with a non-fleshy fruit by endozoochory (Farmer et al., 2017; Lovas-Kiss et al., 2018; Green et al., 2022). However, the dynamic changes in seed dispersal by different duck species within the same habitat have not been fully explored. Significant differences exist among duck species in the dispersal of plant seeds, with these differences exhibiting dynamic changes over time (Guillemain et al., 2002; Ando et al., 2023). Even within the same duck species, different ecological periods may reveal varying potentials for seed dispersal (Soons et al., 2008; Lovas-Kiss et al., 2018). As seasons change, weather conditions fluctuate, and human activities intervene, the foraging behaviors and habitat preferences of ducks also change, leading to variations in the plant seed groups they disperse (Pernollet et al., 2015; van Leeuwen et al., 2023; Gross et al., 2020). Studies have shown that duck diet was similar across species when food was abundant (Guillemain et al., 2002; Gurd, 2008). As the number of individuals foraging increases later in the winter, ducks engage in spatial or behavioral niche differentiation to avoid competition, thereby increasing interspecific dietary differences (Guillemain et al., 2002; Ando et al., 2023).

In this study, we collected fecal samples from four common duck species in the coastal wetland of Dafeng, China and extracted the seeds contained within. We identified the seed species and statistically analyzed their quantities. The specific objective of this study is to determine the seed dispersal function of ducks and the differences and factors they exhibit over seasonal changes by quantifying the seed dispersal capabilities of different duck species and their characteristics at different times. We expected significant differences in the seed dispersal effects among different duck species, which reflect their known differences in morphology and foraging ecology (Kear, 2005a, 2005b; Almeida et al., 2022). Moreover, we predict that the seed dispersal effects of ducks will exhibit dynamic changes across different months and years, revealing the interaction between ducks and their habitats, as well as the impact of weather conditions change on duck diet and habitat (Kear, 2005a, 2005b; Silva et al., 2021).

2.   Materials and methods

2.1   Study area

This study was conducted in the eastern coastal area of Dafeng Milu National Nature Reserve (32°59′–33°03′N, 120°47′–120°53′E) in Jiangsu Province, eastern China (Fig. 1). The reserve, a coastal beach wetland, lies to the east of the Yellow Sea and to the south of Dongtai City. The climate of the reserve changes from subtropical to warm temperate resulting in drought conditions, reduced rainfall, and lower temperatures accompanied by frost. The area maintains a relative humidity of 80%. The annual precipitation is 1068 mm while 63% annual precipitation concentrates in four months from June to September. During winter, the reserve experiences the influence of the continental monsoon, resulting in drought conditions, reduced rainfall, and lower temperatures accompanied by frost. Conversely, in summer, the maritime monsoon brings higher temperatures and plentiful rainfall. The annual average temperature in the reserve is 14.1 ℃, with January temperatures averaging 0.8 ℃ and July temperatures peaking at 27 ℃. The reserve has many types of wetland habitats, such as rivers, tidal flats and ponds, which provide a suitable habitat for waterbirds.

Figure  1.  Distribution of four habitats and feces collection sites in the study area of Dafeng coastal wetland, Jiangsu Province, China.

DownLoad: Full-Size Img PowerPoint

2.2   Diversity of duck community in different habitats

To investigate the diversity of duck communities in Dafeng coastal wetland, we selected four typical habitat types (tidal flat, river, pond, farmland) for research. We set up four sample points within each habitat, totaling 16 sites in two wintering seasons (December to February 2022 and 2023) in the coastal wetland of Dafeng. The observation range of each sample point was a circular area with a radius of 500 m. Only the species and number of ducks that flew into the observation range were recorded. Two investigators conducted the survey at 8:00–11:00 and 14:00–17:00 each day for a total of 50 days during the study period. The habitat type of each sample point, the species and number of ducks were recorded. When the population density of ducks is low, the species and number are directly recorded, while photographs are taken to estimate the population size when the number is too high to count accurately.

The abundance classification of ducks is based on the following: when the proportion of individuals of a particular species within the total duck population exceeds 10% (P > 10%), this species is the dominant species; when 1% < P < 10%, this species is common; when P < 1%, it is a rare species (Zhang et al., 2017). We also used Shannon-Wiener diversity index, Pielou evenness index, and Simpson dominance index to compare the diversity of duck communities in different habitats (Sutherland et al., 2004; Li et al., 2022).

2.3   Plant seeds recovered from duck feces

The collection of duck feces was conducted to investigate the dispersal of plant seeds by ducks in two wintering seasons (December to February 2022 and 2023), with fixed-point observation at sites where single-species group or a group with over 95% conspecific individuals were located to ensure accurate correspondence between the collected feces and ducks. Fresh feces were collected within 1 m × 1 m quadrats set up at duck resting sites, supplemented by the shape, length, color and other characteristics of the feces to distinguish them. The samples were sealed in bags and labeled with the quadrat number, bird species name, and fecal count. Post-collection, each fecal sample was rinsed with tap water through a 250-μm sieve to isolate plant seeds. The identification of intact seeds primarily relies on their morphological integrity. The intact seeds should maintain their original shape without any obvious damage or missing parts; seed coat should be intact, without cracks or damage (Wang and Hou, 2023). Seeds that are broken but whose remaining portion appears larger than half the size of the intact seed from a visual perspective can be counted as broken seeds, whereas smaller seed fragments are not included in this count. Although broken seeds will not be viable and are not dispersed, broken seeds provide valuable information about the effects of the duck digestive tract on seeds.

To identify the species of plant seeds, we first classified the seeds according to the appearance of the seeds, and DNA identification was performed on the seeds that could not be identified from the appearance. We used an improved 2 × CTAB method to extract total DNA from plant seeds. The DNA was fragmented into 500–800 bp pieces. The NEBNext Ultra II series reagents were used for DNA end repair, adaptor ligation, and selection of DNA fragments of suitable size. After PCR amplification and magnetic bead cleaning, second-generation genome sequencing was performed. The GetOrganelle software was used to reassemble the plastid, and the NCBI database was used for Blast nucleotide comparison to find species with high similarity.

To investigate the seed dispersal function of ducks, we conducted rarefaction analysis using R package iNEXT to examine the relationship between the number of fecal samples from four duck species and the cumulative number of recorded plant taxonomic groups. This method allowed us to assess the diversity of plant seeds in duck feces and compare the seed dispersal potential among different duck species (Soons et al., 2016; Martín-Vélez et al., 2021a).

3.   Results

3.1   Diversity of duck community in different habitats

During the study period, a total of 13 species of ducks were recorded in coastal wetland of Dafeng (Appendix Table S1). Among them, the population size of Spot-billed Duck (Anas zonorhyncha) and Mallard (A. platyrhynchos) was the largest. The number of species and individuals both decreased significantly in February (Fig. 2). In addition, the abundance of ducks in the second winter was higher than that in the first winter. The overall Shannon-Wiener diversity index of the experimental area was 1.144, Pielou evenness index was 0.477, and Simpson dominance index was 0.401. The diversity and abundance of ducks vary across different habitats. Among these habitats, tidal flat supports the greatest diversity and abundance of ducks (Shannon-Wiener diversity, tidal flat: 1.304; river: 0.344; Pielou evenness index, tidal flat: 0.307, river: 0.470; Simpson dominance index: tidal flat: 0.636, river: 0.182). There were few species and quantity of ducks in farmland, only five Spot-billed Ducks were recorded, and no duck species were found in pond.

Figure  2.  Interannual and monthly variation of duck species and individual numbers in the coastal wetland of Dafeng.

DownLoad: Full-Size Img PowerPoint

3.2   Plant seeds recovered from duck feces

During the study period, due to limited sampling effort, we collected fecal samples from four out of 13 duck species. A total of 2659 feces of Spot-billed Duck, Gadwall (Mareca strepera), Mallard, and Eurasian Teal (Anas crecca) were collected in tidal flat (Appendix Fig. S1). A total of 228 plant seeds were isolated from feces, 7.48 % of the feces contained at least one plant seed, 1.24 % of the feces contained intact seeds, and 6.24 % of the feces contained only broken seeds. The proportion of broken seeds in Spot-billed Ducks are higher than those in the other three duck species, and the seeds isolated from the feces of the other three species are relatively large in the proportion of intact seeds (Table 1). Spot-billed Ducks consistently exhibited a higher number of seeds per feces across all months compared to the other three duck species (Fig. 3). Over the study period, the number of seeds per feces showed a certain degree of seasonal variation. Towards the end of the second winter, there was a general decline in the number of seeds per feces.

Table  1.  Number of duck feces and plant seeds in the coastal wetland of Dafeng.

	Spot-billed Duck	Gadwall	Mallard	Eurasian Teal
Number of feces	1475	545	492	147
Number of feces with intact seeds	7	12	8	6
Number of feces with only broken seeds	161	5	0	0
Number of seeds	197	17	8	6
Number of intact seeds	7	12	8	6
Number of broken seeds	190	5	0	0
Proportion of broken seeds (%)	96.45	29.41	0	0

 | Show Table

DownLoad: CSV

Figure  3.  Number of intact seeds per feces in four duck species across different months during the study period.

DownLoad: Full-Size Img PowerPoint

By analyzing the appearance identification of seeds and DNA identification of 19 seed samples, we successfully identified 219 seeds, representing 15 distinct plant species across 7 families and 13 genera (Fig. 4). Of these, identification precision reached the species level for nine plants and genus level for six others. Among the fecal seeds of ducks, those of Bolboschoenus planiculmis were most abundant, with seeds from other plant species being less prevalent (Appendix Table S2).

Figure  4.  Fifteen species of plant seeds collected from the feces of ducks. (a) Bolboschoenus planiculmis; (b) Rumex dentatus; (c) Humulus scandens; (d) Triadica sebifera; (e) Polygonum sp.; (f) Cyperus sp. 1; (g) C. sp. 2; (h) C. sp. 3; (i) Echinochloa sp.; (j) Salix matsudana; (k) Populus sp.; (l) Lolium multiflorum; (m) Oryza sativa; (n) Digitaria sanguinalis; (o) Medicago sativa.

DownLoad: Full-Size Img PowerPoint

The rarefaction curves revealed that Gadwalls had the highest seed diversity, followed by Spot-billed Duck, Eurasian Teal, and Mallard (Fig. 5). Despite the smaller sample size for the Gadwall, a large number of plant taxonomic groups were detected, indicating a high potential for seed dispersal by Gadwalls. As the number of samples increased, the diversity of plant seeds in duck feces also increased, with varying rates among different duck species. The slope of the curve reflected the cumulative rate of plant seed types at the same number of fecal samples.

Figure  5.  Rarefaction analyses showing the accumulated number of plant taxa recorded in feces of Spot-billed Duck, Gadwall, Mallard and Eurasian Teal, in relation to the number of fecal samples. Error bars represent 95% confidence intervals.

DownLoad: Full-Size Img PowerPoint

4.   Discussion

In this study, we found that the majority of duck species involved in seed dispersal are similar across Eurasia, suggesting that certain duck species may play a significant role in seed dispersal globally (Lovas-Kiss et al., 2019; Silva et al., 2021; Almeida et al., 2022). Duck species in China exhibit efficient endozoochorous seed dispersal capabilities. This finding resonates with European research outcomes, indicating that these duck species can effectively disperse plant seeds across different geographical regions (Kleyheeg, 2015; Almeida et al., 2022; Green et al., 2023). Additionally, we observed similarities among many plant taxonomic groups. For instance, seeds from wetland plant families such as Polygonaceae, Cyperaceae, and Poaceae can be dispersed by ducks, suggesting that these plants may have adapted strategies for seed dispersal via duck (Martín-Vélez et al., 2021a; Silva et al., 2021; Almeida et al., 2022). This long-distance seed dispersal pattern is of significant importance for the genetic diversity and ecological adaptation of plant populations (Viana et al., 2013; Green and Elmberg, 2014; Lovas-Kiss et al., 2023; González-Varo et al., 2024).

Among the seeds obtained from duck feces, only the seeds of O. sativa were found in the feces of two duck species, while the seeds of other plant species were found in the feces of only one duck species (Appendix Table S2). The plant species consumed by the four duck species exhibited differences in the seeds collected from their feces, possibly reflecting distinct dietary preferences associated with their respective ecological niches (Ando et al., 2023). The presence of O. sativa seeds in the feces of two duck species may be attributed to its widespread distribution in the habitat and the higher feeding preference of ducks for this plant (Pernollet et al., 2015). The number of seeds we recorded is lower compared to other studies (Figuerola et al., 2002; Urgyán et al., 2023; Jiménez-Martín et al., 2024), which might be due to these ducks primarily feeding in farmlands at night, thus ingesting fewer seeds from wetland plants (Guillemain et al., 2010; Kleyheeg et al., 2017). Moreover, due to the difficulty in accurately counting seed fragments and the digestive effects of ducks on seeds, the actual number of seeds dispersed by ducks could far exceed our observed data. This also increases the likelihood that most of the plant seeds observed were found in the feces of only a single duck. In addition, we also speculate that this phenomenon may be related to the differences in feeding behavior and digestive capacity among duck species, suggesting that different duck species have differences in seed dispersal groups and possess different dispersal functions (Soons et al., 2016; Silva et al., 2021; van Leeuwen et al., 2023). Our study found that the number of B. planiculmis seeds dispersed by Spot-billed Ducks fluctuates between different years and months, showing a downward trend particularly in February (Appendix Table S3). This phenomenon may be associated with fluctuations in water levels and changes in food supply, and the seasonal changes in food supply could also lead to variations in the number of plant seeds dispersed by ducks at different times (Guillemain et al., 2002; Urgyán et al., 2023). When the fruiting period overlaps with the presence of migratory waterfowl, the probability of seed dispersal increases, which is due to the seasonal changes in bird diets (Urgyán et al., 2023; Jiménez-Martín et al., 2023). Changes in water levels can affect the habitat and food resources of ducks, thereby influencing their seed dispersal (Guillemain et al., 2002). Studies have shown that mallards and teals have a significant overlap in their ecological niches during early winter, and when food resources are depleted, they adopt alternative strategies, resulting in switches in feeding depth/seed size (Guillemain et al., 2002).

Our rarefaction analyses indicate that Gadwalls and Eurasian Teals exhibit higher diversity in seed dispersal. The slopes of the rarefaction curves for Gadwalls and Eurasian Teals are steeper than those for Spot-billed Ducks and Mallards, suggesting that, at the same sample size, the feces of Gadwalls and Eurasian Teals contain a greater variety of plant seeds. This result is consistent with the findings of Almeida et al. (2022) in the comprehensive database on the ingestion of plant seeds by European waterfowl, where it was shown that both Gadwalls and Eurasian Teals had a higher number of seed species per sampled fecal unit than Mallards. Additionally, Soons et al. (2016) also found that Eurasian Teals disperse seeds with higher diversity compared to the Mallards in the Camargue. This could be due to the more extensive diet of M. strepera and A. crecca, which consume a wider range of plant species, thereby carrying a richer diversity of plant seeds in their feces (Figuerola et al., 2002; Sebastián-González et al., 2020). This broad diet may be a result of their habitat providing diverse plant resources or due to these duck species having greater ecological adaptability and flexibility in food selection (Guillemain et al., 2002). Studies have indicated that Eurasian Teal exhibits considerable dietary diversity, whereas the larger Spot-billed Duck and Mallard may compete with other duck species for more nutritious plant rhizomes, forcing other species to consume a greater variety of plants to obtain sufficient nutrition (Ando et al., 2023). Furthermore, Spot-billed Ducks and Mallards may have a higher degree of seed digestion in their digestive systems. For instance, seeds in the digestive tract of Mallards may remain for a longer duration, leading to increased fragmentation and a reduction in the number of seeds detected (García-Álvarez et al., 2015; Green et al., 2023).

The activity range of wintering ducks is not confined to tidal flat but extends to surrounding habitats in Dafeng. This movement facilitates seed dispersal between different habitat types, thereby influencing the composition and diversity of plant communities (Kleyheeg et al., 2017; Ando et al., 2023; Barta et al., 2023). During the wintering period, the internal movement of ducks within the study area may affect seed distribution patterns, such as the concentration in specific feeding areas potentially leading to increased seed densities in certain regions (Pernollet et al., 2015; Kleyheeg et al., 2017). Furthermore, wintering ducks typically rest by day and feed at night. Their nocturnal habitats may differ from their daytime locations, which could have a certain impact on seed dispersal (Guillemain et al., 2010; Kleyheeg et al., 2017). The spring migration following the wintering season provides an important pathway for long-distance seed dispersal (Green et al., 2023). Located on the East Asian-Australasian Flyway, Dafeng supports a large number of ducks that stopover, rest, and winter here, significantly impacting regional seed dispersal dynamics (Sun et al., 2023). Although migratory waterbirds are considered to have “flyways,” the movement patterns of many species vary considerably, with individuals frequently switching between biogeographic regions that were previously thought to belong to different flyways (Guillemain et al., 2017). Despite migration occurring only during a small fraction of the annual cycle, with the remainder of the time dominated by daily feeding activities and returning to roosting sites, these routine movement patterns still provide extreme dispersal distances exceeding the maxima expected with most abiotic dispersal mechanisms (Green et al., 2023; Lovas-Kiss et al., 2023; Urgyán et al., 2023). Studies have predicted through the data from ring recoveries that migratory ducks can facilitate seed dispersal on a large scale during the migration period, with over 3.5% of seeds predicted to be dispersed more than 100 km and up to 1600 km (Viana et al., 2013).

Although considerable progress has been made in the study of waterbird zoochory, the current research focus is largely concentrated in Europe (Green et al., 2023). Our study reveals that there are interspecific differences in the seed dispersal function of ducks in the coastal wetlands of eastern China, with these differences exhibiting seasonal variability. The extensive information on endozoochory by teal and mallards in Europe is highly relevant to our findings, as our research suggests the existence of similar dispersal patterns in China. Nevertheless, our study has its limitations, with a relatively small number of duck species analyzed, which may not fully represent the seed dispersal functions and relationships among ducks. We did not identify the species of the broken seeds, and future research needs to compare the proportion of broken to intact seeds between plant and bird species to obtain more comprehensive information. Daytime surveys cannot capture the nocturnal feeding behavior of ducks, which will affect the integrity of seed dispersal data, and it is recommended that future studies use GPS tracking to observe where waterbirds feed and to more comprehensively gather seed dispersal information. Additionally, the study of other waterbird species at the same location can further explore the seed dispersal function of waterbirds. Furthermore, our research is confined to a single location. Future studies should observe the seed dispersal by waterbirds across more locations and seasons, incorporating a broader temporal and spatial perspective to continuously enrich the understanding of seed dispersal by waterbirds.

CRediT authorship contribution statement

Xin Luo: Investigation, Data curation. Ning Li: Writing – original draft, Supervision. Wei Tai: Investigation. Yao Cai: Data curation. Zheng Wang: Supervision.

Ethics statement

Our study was carried out in agreement with the current law of China on the protection of wildlife.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.avrs.2024.100209.