Similar seed dispersal systems by local frugivorous birds in native and alien plant species in a coastal seawall forest

Ethics statement

Field studies were conducted with permission from the Management Bureau of Dafeng Milu National Nature Reserve.

Study site

The field work was conducted in the Dongchuan coastal seawall forest in Dafeng Milu National Nature Reserve, Jiangsu Province, China (32°59′N∼33°03′N, 120°47′E∼120°53′E). The forest was planted in the 1990s, and was mainly used for windbreak and dike consolidation. The region is characterized by a transition climate between subtropical and warm temperate. The mean annual precipitation is approximately 1,068 mm (concentrated in summer) and the mean annual temperature is 14.1 °C. The vegetation is dominated by Robinia pseudoacacia, with smaller areas of Metasequoia glyptostroboides and Populus canadensis. The width of the forest is about 20∼50 m. One side of the forest is an intertidal zone with the main vegetation types Spartina alterniflora, Suaeda glauca, Phragmites communis, and Imperata cylindrica, and the other side is farmland and pond culture. Annual typhoons blow over the trees, exposing many open forest gaps. Many bird species live in the forest, including the light-vented bulbul (Pycnonotus sinensis), common blackbird (Turdus merula), and black-billed magpie (Pica pica). From April to May and September to October, many passerines, mainly warblers and flycatchers, travel through the forest in order to complete their migration life cycle. In autumn, this area also has a small number of other fruiting, such as Lycium barbarum, Sapium sebiferum, Ligustrum lucidum, and Paederia scandens.

Field observation

Fruit characteristics of plant species

In autumn, P. americana and C. japonica are the most common fruiting species in the coastal seawall forest. Both fruit types are berry. When ripe, their colors change to black or purple-black. There was a significant difference in the fruit size between the species (t-test, t = 8.252, df = 38, P < 0.01) (Table 1). Cayratia japonica has bigger fruits than P. americana.

Phytolacca americana mainly grew in the forest gap habitat, while C. japonica was widely distributed in the understory along the seawall forest. The fruit ripening period of the two plant species basically overlapped, and both ripened between August and October. There was a significant difference in the height of fruit distribution between the two species (U-test, Z =  − 5.420, P < 0.05, n = 40). Fruit height of P. americana is higher than C. japonica.

Fruit feeding and seed dispersal by frugivorous birds

Ripe fruits were constantly eaten by frugivores, and no other animal groups were observed to eat P. americana or C. japonica fruits during the daytime. The relationship between the percentage of fully-ripe fruits and the percentage of fruits taken by birds showed a significant positive correlation for both plant species (simple linear regression, P. americana: β =0.863 ± 0.017, R²_adj = 0.978, P < 0.01; C. japonica: β =0.787 ± 0.034, R ² _adj = 0.898, P < 0.01) (Fig. 1). This suggested that the more ripen fruits, the greater proportion of missing fruits were eaten by birds or fallen on the ground.

A total of 11 bird species were observed feeding on the fruits. Vinous-throated parrotbill (Paradoxornis webbianus), blue-and-white flycatcher (Cyanoptila cyanomelana), and grey-backed thrush (Turdus hortulorum) were the three species that only fed on Phytolacca americana. Black-napped oriole (Oriolus chinensis) was the species that only fed on C. japonica. The rest of the bird species fed on both fruits of the two plants (Fig. 2).

The Pycnonotus sinensis and Pica pica had the highest feeding frequencies on P. americana and C. japonica, respectively. Pycnonotus sinensis and Sturnus cineraceus had the longest foraging times on Phytolacca americana and Cayratia japonica fruits, respectively. The Sturnus cineraceus was the species with the furthest first stopping distance after feeding on both plants (Table 2). However, there were no significant differences in bird feeding frequency (U-test, Z =  − 1.116, P > 0.05, n = 18), feeding duration (t-test, t = 1.960, df = 16, P > 0.05), or first stopping distance (t-test, t =1.401, df = 16, P > 0.05) between P. americana and C. japonica. However, the number of P. americana fruits consumed per visit was significantly larger than the number of C. japonica fruits consumed (t-test, t =  − 3.941, df = 16, P < 0.01) (Table 2).

Seed germination rate

The seed germination rate of intact fruits was lower than both hand-cleaned seeds and defecated seeds from both plant species. Phytolacca americana’s final seed germination rates across the three different groups were as follows: 79% for defecated seeds, 60% for hand-cleaned seeds, and 13% for intact fruits (Fig. 3). Additionally, C. japonica’s final seed germination rates were 87%, 71%, and 18%, respectively (Fig. 3). There was a significant difference of seed germination rates across the different groups between P. americana (ANOVA: F_2,15 = 13.283, P < 0.01) and C. japonica (ANOVA: F_2,15 = 4.591, P < 0.05). The seed germination of intact P. americana fruits was significantly lower compared to hand-cleaned seeds (LSD-test, P < 0.01) and defecated seeds (LSD-test, P < 0.01). The same result was shown in C. japonica. This showed that fruit pulp removal by local frugivorous birds’ digestion or other methods could promote seed germination success across both plant species.

Seedling microhabitats

We randomly found 300 P. americana seedlings and 280 C. japonica seedlings along the seawall forest in the study site. There were differences in microhabitat characteristics between the two plant species. Phytolacca americana seedlings preferred the forest gap habitat, and significantly more seedlings were found in the forest gap habitat than in the understory habitat (t-test, t = 5.684, df = 28, P < 0.01) (Fig. 4). However, C. japonica seedlings did not show an obvious preference when selecting a microhabitat (t-test, t = 0.092, df = 28, P > 0.05) (Fig. 4).

Similar fruit characteristics between the two plant species

Bird-mediated seed dispersal is important for the establishment of many fleshy-fruited plant species (Wenny, 2001). Many alien plants with fleshy fruits are able to establish effective dispersal interactions with local birds based on having attractive fruit characteristics that are similar to those of native plants (Gosper, Stansbury & Vivian-Smith, 2005). Our study showed that the alien P. americana species had similar characteristics (e.g., fruit type, shape, color when ripe, and ripening period) to the native C. japonica species (Table 1). This similarity suggests that the alien species P. americana has great potential to have its seeds dispersed by local frugivorous birds that have already established a positive relationship with the native plant species C. japonica.

Previous studies have found that small fruits and seeds are more effectively removed and dispersed than larger ones (Rey et al., 1997), and birds prefer visiting trees with larger crop sizes (Sallabanks, 1993; Korine, Kalko & Herre, 2000). A study in a coastal belt forest in South Africa found that fleshy-fruited invasion plant species effectively dispersed by birds were small-seeded, open-habitat species with longer fruiting lengths compared to native plant species (Bitani et al., 2020). If the former conclusion is true, and based on the fact that P. americana has smaller fruit size but more seeds than C. japonica (McDonnell et al., 1984; Hansen & Goertzen, 2006), P. americana could use these traits as an advantage over C. japonica in order to improve its dispersion efficiency.

The forest gap microhabitat formed by light disturbance was more suitable for the growth of alien plant species (Willamson & Fitter, 1996; Ridenour & Callaway, 2001). We found that alien plant species seedlings tended to occupy the forest gap microhabitat, which was more suitable for seedling growth. However, the native plant seedlings showed no preference in microhabitat selection. This suggested that alien plant species seedlings had developmental potential for successful population regeneration in the coastal seawall forest. These results were consistent with the findings of Burnham & Lee (2010) and Martins et al. (2004).

Similar seed dispersal systems

A fruiting species can be consumed by a variety of potential bird dispersers, and a potential bird disperser can remove a variety of fruit species (Dugger et al., 2018; Schneiberg et al., 2020). In our study, seven out of 11 local frugivorous birds consumed fruits from both plant species, suggesting that a stable relationship between the alien plant and local birds had already been established. Three bird species were found to only consume the alien species, showing that it had a wider popularity with local bird species (Whelan & Willson, 1994; Levey & Martínez del Rio, 2001). Ten out of 11 and eight out of 11 bird species who ingested the whole fruit may be potential seed dispersers of the alien species and the native species in study site, respectively. Four bird species were pecking birds since pecking may have damaged the seeds (Table 2).

Pycnonotus sinensis is a small and common frugivorous bird species that we studied during our field observation. Because of the lack of insects, ripening fruits were their main food resource in the coastal seawall forest in autumn. We typically observed three to five Pycnonotus sinensis individuals foraging on Phytolacca americana trees with large crop sizes and purple-black ripe fruits. Ultimately, Pycnonotus sinensis interacted the most with the alien species and was its main seed disperser.

Overall, we did not find an obvious pattern of preference across the local birds’ selection and utilization of fruits from these two plants, which reflected the degree of similarity between the introduced plant species P. americana and the native species C. japonica. We found some different feeding behaviors between the two plant species, but according to the number of consumed fruits, P. americana showed more easily used by birds. We found that the frugivores feeding on the fruits were mostly resident birds. Migratory birds contributed less to seed dispersal according to the lower feed frequency and shorter feed duration of the three migratory species, Oriolus chinensis, Turdus hortulorum, and Cyanoptila cyanomelana. Therefore, we concluded that resident birds were the main contributors to P. americana and C. japonica seed dispersal. This may be due to the fact that birds migrating through the seawall forest in autumn were mostly insectivorous flycatchers and warblers (Liu et al., 2020).

Effect of birds on seed germination rate

Bird ingestion may increase the seed germination rates of different plant species, and many fruit plant species, including invasive species, have been shown to benefit from ingestion by frugivorous birds (Combs et al., 2011; Carrion-Tacuri et al., 2012; Czarnecka, Orlowski & Karg, 2012; Thabethe et al., 2015).

Our results showed that the seed germination rates from intact fruits of both plant species were lower than from ingested and manually de-pulped seeds (Fig. 4), which was consistent with the results of Barnea, Yom-Tov & Friedman (1990), Yagihashi, Hayashida & Miyamoto (1999) and Traveset (1998). This suggested that local frugivorous bird ingestion similarly and positively affected the germination success of alien and native plant species in the study site. The local birds were legitimate seed dispersers/seed predators or pulp peckers of the fleshy-fruited alien plant. The higher seed germination rate of ingested and manually de-pulped seeds may be attributed to the removal of pulp or appropriate seed coat abrasion, but more trials are necessary to prove this conclusion. The final cumulative germination percentage of the native plant species was similar to that of the alien plant species across three different groups. This suggested that the seed dispersal pattern in P. americana was similar to that of C. japonica.

Many alien plants have interfered with the dynamic between frugivores and native plants (Farwig & Berens, 2012). Alien species with abundant and nutritious fruits may prevent birds from visiting native plants. This reduces seed dispersal success and recruitment (Hansen & Müller, 2009; Mokotjomela, Musil & Esler, 2013). In our study, the alien plant P. americana had more abundant fruits and attracted more visiting frugivores than the native plant C. japonica. The mutualistic interaction between the alien plant and native frugivores is likely to lead to an invasion with a negative impact on the local ecological environment. Appropriate measures should be taken to control a potential invasion of P. americana in the study site.