Introduction

Parasite communities are defined by a set of species that live in a host population, the intra- and interspecific relationships between them, the parasite-host association and the environmental factors structuring that community (Bautista-Hernández et al. 2015). A descriptive study of parasite communities can be performed mainly at two different levels. One level is the component community, which includes all parasite populations in the analyzed host population. This level is related to evolutionary time scale processes, and its analysis provides a global view of the system (Poulin 2011). The other level is the infracommunity level, which includes all parasite populations in a single individual host. Comparisons can be made between individual hosts or “replicate communities.” This analysis provides a limited view in time since infracommunities are short-lived, and their maximum lifetime is the host's life. Therefore, they provide more precise information in relation to the individual host (Bush et al. 1997; Magurran and McGill 2011).

In bird hosts, parasite diversity has been associated with long life span, grade of immune response and dietary breadth, among others (Gutiérrez et al. 2019). Parasite richness has been associated with life history traits of the hosts, especially the migratory strategy (e.g., distance migrations, variety of habitats, geographic range size), immunocompetence, aggregations of host species in stopover sites and dietary preferences (e.g., generalist vs. specialist) (Thieltges et al. 2013; Hannon et al. 2016; Gutiérrez et al. 2019). Therefore, parasite communities provide reliable information for inferring patterns, interactions and ecological processes occurring in natural host-parasite systems.

Argentine Patagonia provides significant habitats to wintering migrant shorebirds, most of which breed in the Artic. This area is an important habitat for the two shorebird species chosen as avian model for this study, the Baird's sandpiper Calidris bairdii (Coues, 1861) and the White-rumped sandpiper Calidris fuscicollis Vieillot, 1819. Both species breed in the Arctic tundra (O'Brien et al. 2006). Calidris bairdii migrate using mainly the Mid-continental flyway and are observed using several diverse inland wetlands, such as river and lake shores, flooded grasslands and paddy fields. Calidris fuscicollis use mainly the Atlantic flyway route, being typically observed at the same latitude as C. bairdii but in large numbers over marine coastal areas in temperate regions of South America (O'Brien et al.2006). Despite these general patterns, individuals of both species often overlap their use of Patagonian sites feeding in mixed flocks.

Although helminth assemblages of migratory shorebirds are well studied in North America (Canaris and Kinsella 2001, 2007; Canaris et al. 2003; Didyk et al. 2007; Bondarenko and Kontrimavichus 2018), this topic has not been thoroughly considered in South America, and especially Argentina is a scarcely studied area, with few taxonomic records prior to the start of this research project (Table 1). To understand whether the parasitic fauna changes from non-breeding with respect to breeding grounds or whether birds carry parasites when they move along the route, it is first necessary to know the species involved in this dynamic.

Table 1 Helminth taxa reported worldwide for C. bairdii (BASA) and C. fuscicollis (WRSA) before the present research project
Full size table

The aim of this study is to describe and compare the helminth communities (i.e., their composition and structure) of C. bairdii and C. fuscicollis in their non-breeding sites in Patagonia, Argentina. We aim to determine whether they exhibit similarities and/or differences, considering that their hosts make similar distance migrations but differ in habitat preferences consistent with their migration routes. We expect that the specific richness and structure of their helminth communities give us some information about the habitat use and trophic behavior.

Materials and methods

Study area and sample processing

Birds either died accidentally or were found dead during different research projects performed in the austral summer (January) carried out from 1999 to 2009 in different inland and coastal marine sites in Patagonia, Argentina. Collection sites were located along two important migratory routes of these birds: the Mid-continental flyway and the Atlantic flyways. The former presents two different areas: the mountain range and the plateau, where areas with permanent or temporary bodies of fresh or brackish water can be observed with varied vegetation, predominantly halophytes and grass species (Blanco et al. 2012) (Fig. 1). In the Atlantic flyway, the marine coasts are characterized by the presence of cliffs and estuary ecosystems with long intertidal plains. There are also sandbanks areas, caves and rocky shelters. The intertidal plains allow the development of diverse benthic invertebrate communities. These ecosystems provide food for many bird species, including important populations of Nearctic migratory shorebirds (Blanco et al. 2012).

Fig. 1
figure 1

Geographical distribution and migration strategies of Calidris bairdii (BASA) and Calidris fuscicollis (WRSA) and study area in the nonbreeding range in Argentinean Patagonia. Dots show collection sites. Distribution data were consulted in BirdLife International (2021). Bird illustrations by Carl Christian Tofte

Full size image

All samples taken from birds were in compliance with all appropriate research permissions (permit numbers 19/04, 02/05, 10/6, 02/08, 48/08, 92/05 DF and FS Chubut and 06/05 DFS Santa Cruz province). Birds were dissected in the field, and viscera were fixed in 10% formalin or 96% ethanol.

Identification of gastrointestinal parasites

The viscera and body cavity were examined at the laboratory under a stereomicroscope. Collected helminths were recovered, counted and preserved in 70% or 96% ethanol. Nematodes and acanthocephalans were studied in temporary mounts using glycerin alcohol, lactophenol or eugenol. Digeneans and cestodes were stained using acetocarmine, cleared and mounted on permanent slides using natural Canada balsam. Also, scanning electron microscope and histological sections were made to assist in their identification. Specimens were identified to the lowest possible taxonomic level considering the material condition and the number of collected specimens. Helminths were identified according to Yamaguti (1963), McDonald (1981), Khalil et al. (1994), Gibson et al. (2002, 2005), Bray et al. (2008), Anderson et al. (2009), Gibbons (2010) and Amin (2013).

Data analysis

For component community analysis, prevalence (P = number of parasitized hosts/number of examined hosts), mean intensity (MI = number of parasites/number of parasitized hosts) and mean abundance (MA = number of parasites/number of examined hosts) were calculated for all parasites and for each parasite population following Bush et al. (1997). In addition, the following community descriptors were calculated: specific richness S = number of species in each component community; abundance N = number of helminth specimens; Shannon diversity index HB = −Σpi ln pi, where pi = ni/N proportion of individuals in the ith species; Berger-Parker's measure of relative dominance D = Nt/Nmax, where Nmax = number of individuals of the most abundant species and Nt = total abundance; and evenness E = H/Dmax, observed diversity (H) is the value calculated previously (Shannon's diversity index), and the maximum diversity (Dmax) is the natural logarithm of the observed richness.

Statistical differences for total prevalence between host species were tested using confidence intervals with the correction given in Agresti and Caffo (2000) for cases of extreme prevalence (e.g., > 95%). Similarity between component communities was analyzed through the Sørensen coefficient of similarity index (Magurran and McGill 2011) following the formula Css = 2c/Sa + Sb, where Sa = number of species present at component community a (specific richness of a), Sb = specific richness of component community b and c = species common to both component communities.

The infracommunity analyses were based on the following descriptors: specific richness S = number of species that exists in each infracommunity, Billouin diversity index \(\mathrm{HB}=\left(\mathrm{ln}N!- \sum \mathrm{ln}{n}_{i}!\right)/N\), which is applied for communities that have been fully surveyed; N is the total number of individuals in the sample, and ni is the number of individuals belonging to the ith species; Berger-Parker's measure of relative dominance D = Nt/Nmax, where Nmax = number of individuals of the most abundant species and Nt = total abundance; D shows the proportion of the most abundant species to the total number of parasites in each host; evenness E = HB/HBmax, where HBmax is \({\mathrm{HB}}_{\mathrm{max}}=\frac{1}{N}\mathrm{ln}\frac{N!}{{ \left[\kern-0.15em\left[ {N/S} \right]\kern-0.15em\right] !}^{S-r}+{\left( \left[\kern-0.15em\left[ {N/S} \right]\kern-0.15em\right] +1\right)!}^{r}}\) where \(r=N-S \left[\kern-0.15em\left[ {N/S} \right]\kern-0.15em\right] \) the maximum value of the diversity for the number of present species in each infracommunity. Statistical differences were tested between host species using confidence intervals under a 99% of significance level.

Similarity between infracommunities was assessed with Sørensen's coefficient of similarity index (Magurran and McGill 2011) following the formula \({C}_{N}=2jN/(aN+bN)\) where jN = sum of lowest recorded abundances of common species between infracommunities a and b, aN = abundance in the infracommunity a and bN = abundance in the infracommunity b. This index compares the number of shared species with the average number of species in a single infracommunity and indicates the proportion of shared species in each infracommunity (Magurran and McGill 2011).

Principal components analysis (PCA) was used to detect patterns in infracommunities between host species and migratory flyways. The values of richness, abundance, diversity, dominance and evenness from each infracommunity were used as attributes. Birds with helminth abundance = 0 were removed from the data set. For the purpose of this analysis, sites were combined according to the migratory flyway, Mid-continental flyway: Laguna Leleque (45° 21ʹ S; 71° 6ʹ W), Estancia Quichaura (43° 25ʹ S; 70° 12ʹ W), Laguna del Toro (48° 14ʹ S; 67° 22ʹ W), Estancia María Cristina (44° 34ʹ S; 69° 35ʹ W), Sarmiento (45° 21ʹ S; 69° 2ʹ W) and Atlantic flyway: Bahía Bustamante (45° 5ʹ S; 66° 25ʹ W), Laguna del Ornitólogo (43° 8ʹ S; 65° 7ʹ W), Río Gallegos (51° 37ʹ S; 69° 13ʹ W), Bahía San Sebastián (53° 4ʹ S; 68° 14ʹ W), Golfo San José (42° 19ʹ S; 64° 19ʹ W) and Caleta Valdés (42° 31ʹ S; 63° 36ʹ W).

All statistical analyses were performed in R version 3.6.0 and Rstudio version 1.3.959 (R Core Team 2020) using the BiodiversityR (Kindt and Coe 2005), Vegan (Oksanen et al. 2020), dabestr (Ho et al. 2019) and factoextra (Kassambara and Mundt 2020) packages.

Results

A total of 44 C. bairdii and 50 C. fuscicollis were analyzed. In the Mid-continental flyway most collected specimens were C. bairdii (Nbairdii = 37, Nfuscicollis = 4), whereas in Atlantic flyway most were C. fuscicollis (Nbairdii = 7, Nfuscicollis = 46). Table 2 shows the helminth taxa found parasitizing each bird species and the values of their population parameters (i.e., P, MA, MI). Some of the species found during the course of this project have been taxonomically described in previous publications (Diaz et al. 2011; Capasso and Diaz 2016; Capasso et al. 2017, 2019, 2020). We found two species of the genus Nadejdolepis but for statistical analysis they were grouped as a single taxon due to difficulties in their differentiation under the magnifying glass.

Table 2 Summary of helminth component communities for sampled shorebirds in Patagonia, Argentina
Full size table

Component community

Of the 94 birds examined, 77 were parasitized (P for total parasitism = 82%). The total prevalence was higher in C. bairdii (P = 100%, 95% CI 100–100, N = 44) than in C. fuscicollis (P = 66%, 95% CI 52.9–79.1, N = 50). The most prevalent group in both host species was cestodes, followed by nematodes. Cestodes were also the most abundant in both communities (Table 2; Fig. 2).

Fig. 2
figure 2

Prevalence of main helminth taxa of Calidris bairdii (BASA) and Calidris fuscicollis (WRSA) from Argentinean Patagonia

Full size image

In C. bairdii hosts, 22 helminth species were represented by 2804 specimens, whereas in C. fuscicollis hosts, 17 helminth species were represented by 315 specimens. The similarity between both helminth component communities was CSS = 0.75. The S, N, HB, D and E for each component community are shown in Table 3.

Table 3 Ecological descriptors for helminth communities of shorebirds from Patagonia, Argentina
Full size table

Infracommunities

The analysis of infracommunities showed low values of diversity. The S, N, HB and D of C. bairdii infracommunities were significantly higher (mean = 4 species, 64 individuals, 0.65, 0.68 and 0.6, respectively) than C. fuscicollis (mean = 2 species, 10 individuals, 0.25, 0.82 and 0.43, respectively) (Fig. 3; Table 3). Most infracommunities in C. bairdii were composed of more than three species, reaching a maximum richness of ten parasite species, while in most C. fuscicolllis richness did not exceed two parasite species. No differences were found in evenness between infracommunities (Table 3). The values of CN for C. bairdii hosts were low, and only 5% of the comparisons showed a value > 0.5. Similarly, the values of CN for C. fuscicollis hosts were very low, ranging from 0 to 0.8, although in most cases no similarity was observed. Only 3% of the values were > 0.5.

Fig. 3
figure 3

Ecological parameter distribution in Calidris bairdii (BASA) and Calidris fuscicollis (WRSA) infracommunities. Dots show the underlying distribution of observations in each group

Full size image

The first two axes of the principal component analysis, which sorted the infracommunities according to host species, explained 93% of the association between the variables (Table 4). The first component (66%) was positively associated with richness, diversity, evenness and abundance and negatively with dominance. In Fig. 4a, a gradient is observed from higher richness and diversity values (and to a lesser extent abundance) in the positive quadrant of axis 1 represented mostly by C. bairdii infracommunities to high dominance values in the negative quadrant of axis 1, represented mostly by C. fusicicollis. Only C. bairdii infracommunities are located at the upper right quadrant of axis 1, showing the highest values of abundance, richness and diversity. The second component (27%) was positively associated with abundance and negatively with evenness. This axis showed high abundance variability for C. bairdii and low for C. fuscicollis and high values of dominance in C. fuscicollis and low in C. bairdii. In Fig. 4b, the principal component analysis could not identify clear aggregation or progressive change in the structure of infracommunities between flyways (environmental inland-marine gradient).

Table 4 Eigenvalues and factor loadings of the first two axes of principal component analysis
Full size table
Fig. 4
figure 4

a Ordination of infracommunities produced by principal component analysis (PCA) based on community variables. Numbers correspond to hosts. Hosts 1 to 44 belong to Calidris bairdii (BASA) species (circles and ellipse in red) and from 45 to 77 to Calidris fuscicollis (WRSA) species (triangles and ellipse in blue). b Ordination of infracommunities located within the Mid-continental flyway (circles in red) and Atlantic flyway (triangles in blue). c Correlation circle showing the variables (vectors) in relation to component 1 and 2

Full size image

Discussion

In the present study, 26 helminth taxa were found, included in 14 families and 20 genera. Of these taxa ten were cestodes, ten digeneans, four nematodes and two acanthocephalans. Cestodes and nematodes were the predominant groups. Nine taxa were determined to species level, 16 to genus level and 1 to family level. Two species were found to be new to science: Parorchis sp. nov. and Maritrema pichi (see Capasso et al. 2019). Two species were found as immature stage: Profilicollis sp. and Microsomacanthus sp. The remaining taxa were found as adults.

During stopovers, hosts having low degree of specific overlap will be characterized by a low parasite richness and parasite species with high host specificity. In contrast, in areas with a high overlap of host species, hosts will have a greater number of parasite species, most of which are generalists (Edwards and Bush 1989). Most helminth genera and some species found in C. bairdii and C. fuscicollis were also found in other shorebird species in the northern hemisphere, e.g., Nadejdolepis, Aploparaksis brachyphallos, Trichocephaloidis, Kowalewskiella cingulifera, Microsomacanthus, Plagiorchis, Levinseniella, Strongyloides, Echinuria, Tetrameres, Eucoleus, Arhythmorhynchus comptus (Canaris and Kinsella 2000, 2007; Kinsella et al. 2007; Canaris et al. 2012), suggesting that they have low host specificity. This could be related to the opportunities for parasite transmission at stopover sites. During migrations, these birds do not spread out randomly over wide areas but congregate at a few sites where the availability of prey species and roosting places are predictable. The overlap of host species would be reflected in a high number of generalist parasite species.

At component community level, it was observed that parasite richness and diversity were higher in C. bairdii than in C. fuscicollis. The evenness and dominance indices indicated that helminth species were uniformly represented in each component community. As migratory behavior is a main characteristic of these birds, and throughout their migration they frequent different types of environments, they are exposed to a variety of parasites. Commonly, birds with more diverse feeding habits (generalists) will be positively associated with high helminth species richness because they consume more potential intermediate hosts (Thieltges et al. 2013; Gutiérrez et al. 2017). Considering that most helminth parasites have indirect life cycles, the variety of food resources during stopovers may increase the risk of parasite transmission. This was most evident in C. bairdii. The overall similarity between the two component communities was high, and they shared most of the helminth species. However, the overall similarity between infracommunities in both host species was very low. These results reflect a high number of helminth species that parasitize few host specimens and are found in low abundances. This phenomenon destabilizes the community structure and gives a low predictability to infracommunities, suggesting that parasite infracommunities in both host species are heterogeneous and unpredictable. Furthermore, infracommunities are unsaturated, which means that their richness does not coincide with that observed at the component community level, generally reaching half the value of the component community.

Previous studies in helminth communities of several Scolopacidae species carried out in North America showed similar results to the present study (Canaris and Kinsella 2000, 2007; Kinsella et al. 2007; Canaris et al. 2012). Helminth component communities from North America tend to have a single dominant helminth species, associated with many unpredictable species, high species richness and high diversity. As in the present study, authors indicated that those communities are the result of a continuous recruitment of a few helminth species in low abundances (Canaris and Kinsella 2007). These features seem to be generally true for several shorebirds whose annual life cycle includes occupation of both inland and marine habitats. Considering that some avian hosts can switch seasonally between freshwater and saline habitats, increasing the diversity of habitats frequented, it is assumed that the parasite richness also increases (Gutiérrez et al. 2017). The results of the PCA indicated that the structures of the infracommunities are not strictly in relation with the preponderance of continental or marine environments in their migratory flyways. As evidence of this, in the present work we found some parasite species with terrestrial or freshwater life cycles in birds captured in marine environments (e.g., A. brachyphallos, Echinocotyle sp., Trichocephaloidis spp., Eucoleus sp.) and parasite species with marine life cycles in birds caught inland (e.g., Parorchis sp. nov., A. comptus). This demonstrates a mixed habitat use.

Regarding the higher prevalence, richness and diversity of helminths in C. bairdii, it is to be expected that they would make more stopovers in more diverse foraging sites than C. fuscicollis during southbound migration and in their non-breeding grounds in Patagonia. We outlined before that bird species with a broad diet will ingest prey that act as intermediate hosts for a wide range of parasites and parasite richness will increase (Thieltges et al. 2013; Gutiérrez et al. 2017). Accordingly, this might be the case for C. bairdii.

Based on the present results, some new questions have arisen: Are the lower parasite prevalence, richness and abundance of C. fuscicollis related to a narrower diet during their stay in Patagonia or are those features related to the use of less diverse habitats than C. bairdii? Are these parasitological parameters related to physiological characteristics of the hosts, for example, reducing the impact of parasites activating immune responses? The hypothesis that high helminth richness communities reflect a generalist trophic pattern is supported by the high diversity of prey items observed in C. bairdii stomach contents (Capasso 2019). Therefore, it can be suggested that the broad food spectrum facilitates an increase in parasite incorporation pathways (Anderson and Sukhdeo 2011). Given the higher values in the ecological parameters (richness, prevalence and abundance), as well as the high intensities of infection by parasites with complex life cycles such as cestodes and digeneans, it is expected that C. bairdii incorporates a greater variety of prey than C. fuscicollis.

According to Morand (2015), environmental diversity would increase encounters with a greater variety of parasitic species. One plausible explanation for the patterns observed in these two bird species may be related to the migratory strategies: birds that fly over the continent may have more opportunities to stop over to refuel, while birds that migrate following the seashore may find fewer suitable stopovers (Piersma 2003). Also, present parasitological results are consistent with the recent confirmation that C. fuscicollis conducts extensive non-stop flights from the northern to southern hemisphere (Anderson et al. 2019), explaining the lower helminth prevalence found (i.e., less chance of encountering parasites along the route).

However, it should be considered that the structure of the helminth communities also reflects the ability of host species to tolerate different parasite loads. So far, no studies have been carried out to determine the immunological condition of these two bird species that can help to understand whether the parasite load is related to factors intrinsic to each host bird species and/or to each bird specimen. Future studies combining parasitological and immunity indices will be necessary to determine whether immune inversion differs between host species and between populations that frequent different types of environments.

Conclusions

The present study represents a valuable contribution to the knowledge of the helminth diversity of migratory shorebirds migrating between Arctic and sub-Antarctic regions. Ecological results suggest that the characteristics of the helminth communities of these two migratory shorebird species in Argentine Patagonia depend on the differential use of environments, considering that the migration distance performed by birds is similar and they are frequenting both marine and freshwater environments. In this way, helminth communities’ structure is directly related to the opportunities for the transmission of parasites, in both the non-breeding areas and stopovers during migration. It can be inferred that helminth communities of studied shorebirds are structured based on similarities and differences. The similarities would be determined by the fact that (1) both host species share, partially, breeding and non-breeding grounds in the Nearctic, (2) cover the same migration distances and (3) have a generalist trophic habit. Differences could be attributed to: (1) diversity of frequented environments, (2) number of stopovers and (3) prey item variability and availability during stopovers in the sub-Antarctic.