Area, isolation, disturbance and age effects on species richness of summer waterbirds in post-mining subsidence lakes, Anhui, China

1.   Introduction

Obligate avian brood parasitism is costly for hosts (Lyu and Liang, 2021) because hosts have to take care of the unrelated nestlings at the expense of losing their own nestlings and reproductive opportunities (Payne, 1977; Rothstein, 1990). Brood parasitism reduces the fitness of the host, and thus, hosts have evolved anti-parasitic strategies (Davies, 2000; Soler, 2014, Soler, 2017). The research on host anti-parasitic strategies has mainly focused on the egg stage of host reproduction; for example, many hosts have evolved the ability to recognize and reject foreign eggs from their nests (Brooke and Davies, 1988; Moskát et al., 2002; Lahti, 2006; Langmore et al., 2009; Liang et al., 2016; Liu et al., 2020, 2023a, 2023b; Yi et al., 2020). Egg rejection is an anti-parasitic strategy used by hosts after they realize that they have been parasitized. If parasitic eggs are successfully thrown out of the nest, this defensive action at the egg stage can save host's own reproduction; however, brood parasites usually take out eggs from the nest when laying eggs (Davies, 2000; Wang et al., 2020). In such cases, parasitism will still be a disadvantage for the host. Therefore, another important anti-parasitic strategy is to prevent parasitism in advance, termed nest defense (Feeney et al., 2012). In contrast to the aforementioned mechanism, host defensive strategy to prevent brood parasites from laying eggs in their nests can protect the entire reproduction attempt of the host. Hence, hosts might attack adult brood parasites near their nests to prevent them from approaching their nests (Davies and Welbergen, 2009; Welbergen and Davies, 2009; Hamao, 2011).

However, if recognition errors occur, each defense strategy might bring about potential costs to the host (Davies et al., 1996). This is because in response to egg recognition by hosts, brood parasites have evolved corresponding adaptations, including mimicking the shape and size of host eggs, the background colors of the eggs, and the colors and distribution of egg spots (Spottiswoode and Stevens, 2010). Brood parasites might also lay eggs with more visually covert colors (Langmore et al., 2009) and thicker shells (Antonov et al., 2009). This can cause hosts to misidentify and throw their own eggs or accidently break their own eggs by pecking (Zahavi, 1979; Rothstein and Robinson, 1998). Similarly, aggressive nest defense will also have costs. Attacks on brood parasites will take time and energy, which will shorten the time for foraging or feeding nestlings (Ueta, 1999). Moreover, host attacks can be used by brood parasites to find host nests (Banks and Martin, 2001) and might attract the attention of nest predators (McLean et al., 1986). The morphological mimicking of predatory birds, such as hawks, by some cuckoos can also lead to incorrect identification (Davies and Welbergen, 2008), as it is dangerous for hosts to attack these predators.

Considering these costs, when the risk of being parasitized changes, hosts should adjust their defense behaviours as an adaptation strategy (Lahti, 2006). To minimize the possibility of recognition errors, many host species have evolved the ability to recognize adult cuckoos (Davies and Welbergen, 2009; Trnka and Prokop, 2012), in addition to adjusting their nest defense behavior (Hamao, 2011) and egg rejection according to the risk of being parasitized (Zhang et al., 2021), as indicated by the host reaction to adult cuckoos (Davies et al., 1996; Brooke et al., 1998; Lahti, 2006) or cuckoo dummies (Zhang et al., 2022) close to the nest. That is, if the risk of being parasitized is reduced, the benefits of anti-parasitic behavior are reduced, and thus, hosts should reduce their anti-parasitic behavior accordingly. For example, in the East England, owing to reductions in cuckoo parasitism, the egg rejection rates by Eurasian Reed Warblers (Acrocephalus scirpaceus) dropped from 75% to 25% in 12 years (Brooke et al., 1998), and similar results were obtained in a survey lasting 30 years (Thorogood and Davies, 2013). Although Brooke et al. (1998) and Thorogood and Davies (2013) reported that a decline in egg rejection behavior is accompanied by a decreased parasitism rate, these changes required a long time (10 years or more) to be recorded, and there has been very little information on temporal changes in anti-parasitic behavior over a short term (but see Hamao, 2011; Zhang et al., 2021, 2022). Therefore, we tested whether host anti-parasitic defense is adjusted with changes in the risk of being parasitized in the short term, i.e., within a breeding season.

We tested this hypothesis using the Common Cuckoo (Cuculus canorus, thereafter cuckoo)–Isabelline Shrike (Lanius isabellinus) system, with the latter being a host of the former (del Hoyo et al., 2008; Ma et al., 2012; Yang et al., 2012). In Anxi area, Gansu Province, China, the parasitism rate of 21 Isabelline Shrike nests in May was 0, and 3 of 7 Isabelline Shrike nests in June were parasitized, and the parasitism rate was 42.9% (Ma et al., 2012). Isabelline Shrikes and cuckoos are both seasonal migratory species in China (Zheng, 2023). However, at our study site, Isabelline Shrikes begin to breed in April, whereas cuckoo begins to lay eggs in May. This means the risk of Isabelline Shrike parasitism changes seasonally; specifically, there is no risk in the early breeding season, whereas the risk is high in the late breeding season. Therefore, the difference in migration habits between the host and the cuckoo makes this system suitable for exploring the hypothesis that hosts will adjust their anti-parasitic strategies to cope with the risk of being parasitized over a short period. We predicted that Isabelline Shrikes will be more aggressive to cuckoos after they arrive. To verify this prediction, we experimentally investigated the response of Isabelline Shrikes to the cuckoo dummy. At the same time, we also compared their ability to recognize non-mimetic foreign eggs before and after the cuckoo arrives. Recent studies contended that bird egg rejection might change within a breeding season (Zhang et al., 2021), or that the mere presence of a cuckoo dummy with cuckoo calls in the breeding area could alter egg-ejection decisions in the host (Zhang et al., 2022). Consequently, we hypothesized that egg rejection by Isabelline Shrikes would be adjusted accordingly before and after the cuckoo arrives.

2.   Materials and methods

2.1   Study area

The field study was conducted from April to June 2020. The study site was the Fukang Desert Ecosystem Observation and Experiment Station (44°17ʹ–22ʹ N, 87°52ʹ–58ʹ E), Chinese Academy of Sciences, Xingjiang, northwestern China. The study area comprises a temperate continental desert climate, which is hot in summer and cold in winter, with an annual average temperature of 6.6 ℃ and annual precipitation of 164 mm (Yang et al., 2015). There are villages, deserts, and oasis agricultural farming areas around the shelter forest, and Cotton (Gossypium spp.) and Watermelon (Citrullus lanatus), among others, are mainly planted in the farming areas (Yang et al., 2015). Isabelline Shrikes mainly nest in the mixed shelter forest composed of Simon Poplar (Populus simonii), Chinese White Poplar (Populus tomentosa), Oleaster (Elaiagnus angustifolia), and desert shrub vegetation, with a nest height of approximately 1.5–3 m.

2.2   Egg recognition experiments before and after cuckoo arrival

The research on host egg recognition ability has been conducted mostly with artificial model eggs (Kilner, 2006), because if hosts cannot recognize the non-mimetic model eggs, it is more unlikely for them to recognize a real cuckoo egg. Cuckoo began to lay eggs in May, and thus, the Isabelline Shrike nests with the first egg laid in April were designated as “before cuckoo arrival,” whereas the nests with the first egg laid in May to June were designated as “after cuckoo arrival.” We placed one blue model egg in each Isabelline Shrike nest before and after cuckoo arrival and compared whether egg rejection of Isabelline Shrikes changed between these time points. The blue model eggs were made of clay and matched the size of host eggs, but the color was totally non-mimetic. After placing the model eggs, to reduce nest interference, we did not check the nests treated in each experiment until the 6th day. The responses of hosts were divided into rejection (the model egg was thrown out of the nest) or acceptance (the model egg was in the nest and was warm at the time of checking, considered the model egg was incubated by the host) (see also Yang et al., 2015).

2.3   Dummy experiments before and after cuckoo arrival

We put dummies near Isabelline Shrike nests before (April) and after (June) the cuckoo arrival to compare whether Isabelline Shrike's nest defense would change between these two time points. During the incubation period, the female usually incubates the eggs, and the male is responsible for finding food for the female, so the female rarely leaves the nest (personal observation). Therefore, we will make the incubating females leave from their nests before the experiment begins, and then place dummies and cameras to observe. We used dummies of three locally distributed species with similar sizes as follows: 1) the cuckoo as the brood parasite; 2) the Eurasian Sparrowhawk (Accipiter nisus) as the potential nest predator; and 3) the Oriental Turtle Dove (Streptopelia orientalis) as a control without threat. These three dummies were used to test the responses of hosts in the same nest in a random order at an interval of 1 h apart from each other to allow for any carry-over effect to be minimized (e.g., Attisano et al., 2021). To quantify the attack intensity, we recorded the responses of hosts with a camera placed 3–5 m away from the nest and defined two variables reflecting the behavior: 1) the hosts do attack the dummy or do not attack the dummy but return to incubation when the dummy is presented within 10 min after returning; 2) number of host parents participating in the attack. We placed the dummy on a fixed pole 0.5–1 m away from the host's nest, keeping approximately the same height as the nest and facing the nest. The posture of the dummy remained upright, and the wings were folded. There were two dummies for each species to avoid pseudoreplication (e.g., Yu et al., 2020). When the dummy was placed next to the nest, the responses of hosts were recorded within 5 min after returning to the nest. The responses of hosts were classified as no attack (mobbing call, watching, or no response, see Appendix Video S1, S3 and S5) or attack (attack with contact or diving without contact with the dummy, see Appendix Video S2, S4 and S6). For each nest tested, the experiment was terminated once a host attacked the dummy. We predicted that attack of Isabelline Shrikes to the cuckoo dummy should be more after cuckoo arrival than before arrival, and that Isabelline Shrikes should attack the dove less than the cuckoo dummies, because the Oriental Turtle Dove does not pose threat to the nest. The Eurasian Sparrowhawk may also show a lower frequency of attack, as it may pose a threat to the adult Isabelline Shrikes. Dummy experiments were carried out under sunny days from 8:00 a.m. to 5:00 p.m. local time.

Supplementary video related to this article can be found at https://doi.org/10.1016/j.avrs.2023.100154

Due to the limited number of shrike nests available, dummy and egg experiments are conducted in the same nest. Previous studies have shown that presenting dummies and then model eggs to the same nest would lead to increased egg rejection in the host (Zhang et al., 2022). In this study, model eggs were first displayed and then the dummies were displayed, with no shrike parents being color-banded.

2.4   Statistical analysis

Statistical analysis was performed using SPSS v.20.0 for Windows (IBM, Armonk, NY, USA). Fisher's exact test was used to compare the egg rejection rates of Isabelline Shrikes to blue model eggs before and after cuckoo arrival, the attack rate with the same types of dummies, the rate of returning to nests for incubation, and the attack rate of Isabelline Shrikes on different types of dummies in the same period. As Levene's homogeneity of variance showed that the data were homogeneous and a one-sample Kolmogorov–Smirnov test showed that the data were not normally distributed, a Mann–Whitney U test was used to compare the number of pairs attacking dummies in each nest before and after cuckoo arrival. An exact binomial test was used to determine whether there was any difference in the sex of the host attacking the dummy when only one Isabelline Shrike attacked the dummy before and after cuckoo arrival. All statistical tests were two-tailed tests, and the average values were expressed as the mean ± SD. When the statistical result was P < 0.05, the difference was considered significant.

3.   Results

3.1   Egg rejection of Isabelline Shrikes before and after cuckoo arrival

There was no significant difference in the rejection rates of blue model eggs before (77.3%, 17 out of 22 nests) and after (63.6%, 14 out of 22 nests) cuckoo arrival for Isabelline Shrikes (P = 0.510, Fisher's exact test; Fig. 1).

Figure  1.  Comparison of blue model egg rejection rates before and after the arrival of Common Cuckoos by Isabelline Shrikes.

DownLoad: Full-Size Img PowerPoint

3.2   Isabelline Shrike attack rate on dummies before and after cuckoo arrival

The attack rate of Isabelline Shrikes on cuckoo dummies before cuckoo arrival (27.8%, 5 out of 18 nests) was significantly lower than that after cuckoo arrival (71.4%, 15 out of 21 nests; P = 0.010, Fisher's exact test). Further, the attack rate of Isabelline Shrikes on Eurasian Sparrowhawk dummies before cuckoo arrival (16.7%, 3 out of 18 nests) was significantly lower than that after cuckoo arrival (52.4%, 11 out of 21 nests; P = 0.043, Fisher's exact test). Finally, the attack rate of Isabelline Shrikes on Oriental Turtle Dove dummies before cuckoo arrival (33.3%, 6 out of 18 nests) was significantly lower than that after cuckoo arrival (76.2%, 16 out of 21; P = 0.011, Fisher's exact test).

3.3   Isabelline Shrike responses to different types of threats

There was no significant difference in Isabelline Shrike attack rates between cuckoo and Eurasian Sparrowhawk (P = 0.691, Fisher's exact test), Eurasian Sparrowhawk and Oriental Turtle Dove (P = 0.443, Fisher's exact test), or cuckoo and Oriental Turtle Dove (P > 0.999, Fisher's exact test) before cuckoo arrival. Similarly, there was no significant difference in the Isabelline Shrike attack rates between cuckoo and Eurasian Sparrowhawk (P = 0.341, Fisher's exact test), Eurasian Sparrowhawk and Oriental Turtle Dove (P = 0.197, Fisher's exact test), or cuckoo and Oriental Turtle Dove (P > 0.999, Fisher's exact test) after cuckoo arrival (Fig. 2).

Figure  2.  Comparison of dummy attack on cuckoo, sparrowhawk and dove dummies before and after arrival of the Common Cuckoo by Isabelline Shrikes.

DownLoad: Full-Size Img PowerPoint

3.4   Comparisons of other Isabelline Shrike nest defense characteristics

The number of nests at which Isabelline Shrikes did not attack the dummies but returned to incubation when dummies were presented before cuckoo arrival (44.4%, 24 out of 54 nests) was significantly higher than that after cuckoo arrival (1.6%, 1 out of 61 nests; P < 0.001, Fisher's exact test). The number of Isabelline Shrike pairs attacking the dummies before cuckoo arrival (1.00 ± 0.00, n = 14) was significantly lower than that after cuckoo arrival (1.26 ± 0.44, n = 39; Z = −2.084, P = 0.037).

4.   Discussion

Our results showed that Isabelline Shrikes have high recognition abilities for non-mimetic model eggs, however, egg rejection of the Isabelline Shrike did not change between April and June within a breeding season, while its attack rate on cuckoo dummies became stronger in June.

To avoid the cost of misidentification, many host species have evolved the ability to adjust anti-parasitic strategies according to the risk of being parasitized (Lahti, 2006; Davies and Welbergen, 2009; Trnka and Prokop, 2012). Although some host awareness of brood parasites might be innate (Rothstein, 2001; Peer et al., 2011), other evidence suggests that in other systems, host awareness of brood parasites is acquired (Davies and Welbergen, 2009; Langmore et al., 2012). First, learning might be an important way to recognize adult brood parasites. For example, the Eurasian Reed Warbler increases its attack on the cuckoo around its nest after witnessing neighbors attacking cuckoos (Davies and Welbergen, 2009). Similarly, in an experiment in which dummies were displayed twice to the same nest, the attack intensity of Eurasian Reed Warblers on cuckoo dummies was higher the second time (Čapek et al., 2010). Second, among populations distributed in the same region as cuckoos, some host species are more aggressive to adult cuckoos than those distributed in other regions (Moskát et al., 2002). Different responses of the same host at different locations might be the result of genetic differences among populations, rather than differences in learning opportunities (Rothstein, 2001; Peer et al., 2011). However, a micro-geographic study of the gene-flowing Superb Fairywren (Malurus cyaneus) showed that it will react aggressively to cuckoo only after being exposed to this bird and will maintain an aggressive response even in years without the cuckoo. The micro-geographic variation in this response provides strong evidence for the learned recognition of brood parasites (Langmore et al., 2012). Third, in some species, the response of young and inexperienced individuals to brood parasites is weaker than that of older birds, and the possibility of accurately identifying parasitic eggs and nestlings is lower; thus, young individuals are at higher risk of being parasitized (Lotem et al., 1992; Amundsen et al., 2002). Our research results clearly show that Isabelline Shrikes will protect their nests more actively after the arrival of cuckoo. Therefore, a possible explanation for the change in Isabelline Shrike nest defense intensity before and after cuckoo arrival is that this species adjusts its attack behavior to cope with the increased risk of being parasitized. However, in controlled experiments, the attacks on Eurasian Sparrowhawk and Oriental Turtle Dove dummies were not different from those on Common Cuckoo dummies, therefore the attacks on cuckoo dummies may not be entirely due to the migration of cuckoos to breeding grounds, but also due to the decrease of the chance of further breeding success as the season progresses (Holmes et al., 1992).

Whether hosts have unique recognition and response systems upon encountering brood parasites has always gained attention (Trnka and Prokop, 2012; Ma et al., 2018). When hosts of brood parasites respond to different threats via anti-predation or anti-parasitism mechanisms, hosts can distinguish different types of threats but might not respond to harmless stimuli at all. For example, when hosts are in the egg-laying or incubation period (the cost is that the nest would be more prone to parasitism), the aggressiveness to the dummies of brood parasites is stronger than that in hosts in the nesting period (Trnka and Prokop, 2012). All our experiments were carried out at the egg stage of reproduction, and the results showed that Isabelline Shrikes adopt the same response to different threat levels when facing dummies of different species, which is similar to the results obtained in Oriental Reed Warblers (Acrocephalus orientalis) by Ma et al. (2018). This might be related to the competitive pressure of the breeding niche at the study site. Many species breed at one place, and the competition for the nest leads to Isabelline Shrikes strongly attacking harmless controls, such as the Oriental Turtle Dove. However, the attack on Eurasian Sparrowhawk might also originate from the strong local pressure of nest predation, which needs to be confirmed in future work. Both the pressure of nest predation and nest-site competition led to the strong nest defense behavior of the Isabelline Shrike, which indirectly increased the difficulty of the cuckoo to successfully parasite the Isabelline Shrike.

Accordingly, some host species have non-aggressive responses to brood parasites, although these behaviors will also change during the breeding cycle. For example, the Whitehead (Mohoua albicill), which is unique to New Zealand, responds to dummies of the obligate brood parasite Pacific Long-tailed Cuckoo (Urodynamis taitensis) in a unique way during the egg laying period; specifically, it will quietly return to its nest (McLean, 1987). Although our results also show that some Isabelline Shrike individuals returned to the nest to incubate eggs (Appendix Video S1, S3 and S5) during the dummy test, unlike Whitehead, these birds returned to the nest after observing dummies for a period of time, instead of secretly returning to the nest.

The ability of Isabelline Shrike to recognize foreign eggs was not significantly different before and after cuckoo arrival, which is different from a study conducted by Zhang et al. (2021) on Daurian Redstart (Phoenicurus auroreus). They found that this bird has a significantly increased egg rejection rate after the arrival of cuckoo. However, this might be because the Daurian Redstarts studied had egg polymorphism (Yang et al., 2016), and there was a significant difference in the egg rejection rate between the two types of egg color. The presence of cuckoos around the nest further stimulates the recognition of foreign egg. These results are a reflection of the difference between different color types of Daurian Redstarts. In contrast, genetic basis of the egg recognition ability of Laniidae birds might also be the cause of this difference (Rothstein, 2001; Peer et al., 2011). Red-backed Shrike (Lanius collurio) rejects 93% of cuckoo eggs (Lovászi and Moskát, 2004) and 100% of non-mimicking model eggs (Moskát and Fuisz, 1999), even in decades without parasitism (Lovászi and Moskát, 2004). If the egg-recognition ability of Red-backed Shrike is phenotypically plastic, the recognition ability should decline rapidly in the years without parasitism (i.e., Brooke et al., 1998; Thorogood and Davies, 2013). However, research has shown that some host populations of Laniidae birds that are not parasitized can preserve the egg recognition-related genes for a long period (Peer et al., 2011). The most extreme example is that Loggerhead Shrike (Lanius ludovicianus) in North America maintained an almost 100% rejection rate of foreign eggs for a long period without parasitism (Rothstein, 2001). The vestige behavior of Loggerhead Shrike might reflect its contact with certain brood parasites in ancient times or it might be inherited from the same species in the eastern hemisphere through an earlier speciation process (Rothstein, 2001).

Generally, this study emphasizes the importance of the lag in breeding time caused by the difference in migration habits between hosts and brood parasites. If cuckoos were to migrate to the breeding site earlier, more host nests could be used. Conversely, if Isabelline Shrikes begin to breed earlier, it can avoid being parasitized. The rejection rates of nonmimetic model eggs for Isabelline Shrikes did not change before and after cuckoo arrival. This may be because the egg recognition ability of the Isabelline Shrike is regulated by genetic basis, or it may simply be because the shrike performs nest sanitation of foreign eggs. This has yet to be confirmed, but the over 70% egg rejection rate of the Isabelline Shrike must make it difficult for the cuckoo to successfully parasite. The attack of Isabelline Shrikes on dummies including cuckoos after the arrival of cuckoos was significantly stronger than that before the arrival of cuckoos. Although this may only be due to the general attack on nest invaders caused by nest predation and nest site competition pressure, cuckoos will be strongly attacked by Isabelline Shrikes when parasitizing Isabelline Shrikes (Appendix Video S7), making the success of the parasitism more difficult.

Ethical statement

The experiments reported here comply with the current laws of China. Fieldwork was carried out without specific permit. Experimental procedures were in agreement with the Animal Research Ethics Committee of Hainan Provincial Education Centre for Ecology and Environment, Hainan Normal University (permit no. HNECEE-2016-004).

CRediT authorship contribution statement

Bo Zhou: Writing – original draft, Methodology, Investigation, Formal analysis. Wei Liang: Writing – review & editing, Validation, Supervision, Resources, Funding acquisition, Conceptualization.

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.

Acknowledgments

We would like to thank the editor and two anonymous reviewers for their constructive comments on our manuscript, and we thank Ming Ma and the Fukuang Desert Ecosystem Observation and Experiment Station, Chinese Academy of Sciences, Xingjiang, for their kind help and support with fieldwork.