International importance of tidal flats in the Republic of Korea as shorebird stopover sites in the East Asian–Australasian flyway

,


INTRODUCTION
For long-distance migratory birds, stopover sites used for resting and refueling are important in linking breeding and nonbreeding sites, and they ultimately contribute to successful migration and conservation of biodiversity (Hutto 1998, Faaborg et al. 2010, Wang et al. 2022. A species' selection of a migratory route depends on the existence of suitable habitats at stopover sites (Martin et al. 2007, Murray andFuller 2015). East Asian tidal flats are recognized as essential stopover points in the migratory routes of millions of shorebirds in the East Asian-Australasian Flyway (EAAF; Barter 2006, Bamford et al. 2008. Coastal wetlands along the Yellow Sea area in the EAAF are particularly critical habitats for some migratory shorebird populations that are rapidly decreasing, especially Great Knot (Calidris tenuirostris) and Far Eastern Curlew (Numenius madagascariensis) populations (Nebel et al. 2008, MacKinnon et al. 2012, Murray and Fuller 2015. The Republic of Korea is located in the middle of the EAAF and contains 2500 km² of tidal flats. In particular, large tidal flats along the southwestern coast of the Korean Peninsula provide important habitats for migratory birds (Yoo andLee 1998, Koh andKhim 2014). These coastal wetlands serve as essential shorebird stopover sites where birds can stop to rest and refuel during spring and autumn while they migrate to their breeding and wintering sites (Moores et al. 2016). Currently, at least 20 coastal wetlands in southwestern Korea are recognized as internationally important for supporting more than 1% of the EAAF population of particular species or subspecies (Barter 2006, Moores 2006. To designate protected areas, Ramsar criteria are generally used to determine sites of international importance. Under the Ramsar sites criteria, waterbird-specific internationally important sites are designated based on criterion 5 (C5) and criterion 6 (C6); C5 specifies that wetlands must regularly support 20,000 or more waterbirds, and C6 states that wetlands must regularly support 1% of the individuals of a population from one waterbird species or subspecies (Ramsar Convention Secretariat 2018). However, shorebird populations could be underestimated at certain stopover sites if the site experiences rapid population turnover. For instance, Great Knots stay only 2.3 days during their northward migration (Ma et al. 2013), making it difficult to estimate exact population sizes. To remedy this issue, sites can be designated as internationally important stopover sites if the site supports more than 0.25% of the total estimated population of one or more species. This designation considers the movement of many individuals through stopover sites during migration (Bamford et al. 2008, Conklin et al. 2014).
In the Republic of Korea, since 2001, a census of migratory shorebirds that targets tidal flats has been conducted annually in the spring and autumn. However, most reports on important stopover sites and shorebird populations in the EAAF are based on data collected in or before 2000 (Barter 2002, 2006, Bamford et al. 2008, Conklin et al. 2014). In addition, the shorebird census results for the entire Korean Peninsula have not yet been analyzed. Although 24 internationally important sites for migratory shorebirds are recognized in the Republic of Korea (Bamford et al. 2008), few studies have examined the distribution of these migratory shorebirds. Hence, we investigated shorebird abundances using census data from 2014-2015 and 2019-2020 to identify internationally important stopover sites for shorebirds on Korean tidal flats. Since the shorebird census began in 2001, survey areas have been added and changed. In this study, we compared survey data from 2014-2015 (when the survey areas were fixed) with the survey data from 2019-2020 to evaluate population changes in the same areas over five years. Furthermore, we documented habitat changes at these sites to provide fundamental data for the conservation of shorebirds and the management of tidal flats that provide resting and refueling sites for migratory birds.

Survey methods
We surveyed 18 coastal areas historically known as major habitats for migratory shorebirds in the Republic of Korea (Fig. 1). These areas were designated by the Korean Ministry of Environment as regular survey sites to monitor the migratory status of shorebirds using the Yellow Sea area as a stopover site (NIBR 2020). The survey was conducted twice in autumn, when shorebirds migrate to non-breeding sites (southward migration), and twice in spring, when shorebirds migrate to breeding sites (northward migration).
We selected survey periods that coincided with spring tides, which facilitated counting shorebird populations because high-spring daytime tides concentrate birds within high-tide roost sites. Sites were surveyed in September (September 9-15 in 2014; September 9-15 in 2019) and October (October 8-14 in 2014; October 9-15 in 2019) during autumn, and in April (April 16-22 in 2015; April 17-23 in 2020) and May (May 1-10 in 2015; May 3-9 in 2020) during the spring season. The shorebird census was simultaneously performed in all survey areas to avoid counting the same individuals at multiple sites. The fieldwork was performed by 60 experts working in groups of 2. One to two hours before high tide, the survey began with a point count in and around roost sites as well as a line census along the coastline. At high tide, shorebirds foraging on tidal flats are forced to move to adjacent supratidal sites (roosts), such as fish farms, salt ponds, or wetlands, where they rest until the tide recedes (Choi et al. 2019). We identified species using high-resolution telescopes (20-60×) and binoculars. The survey was performed by groups of two experts using standardized survey methods, including consideration of habitat status (threat factors, habitat changes, etc.) and mapping (resting and feeding sites).

Data Analysis
For each survey area, we compared our pooled data to previous survey data published by the National Institute of Environmental Research (NIER), the National Institute of Biological Resources (NIBR), and the Shorebird Network Korea (NIER 2003, 2005, NIBR 2007, 2020, 2014. To estimate shorebird population size, we mainly used data from Hamblin (2016), but also referenced Bamford et al. (2008) for species missing from B. D. Hansen, R. A. Fuller, D. Watkins, D. I. Rogers, R. S. Clemens, M. Newman, E. J. Woehler, and D. R. Weller (2016, unpublished manuscript). In particular, population estimates of the critically endangered Spoon-billed Sandpiper (Calidris pygmaea) reflect the results of recent studies that estimated a population size of 800 (BirdLife International 2017, Bradfer-Lawrence et al. 2021, Zöckler et al. 2021, Mundkur and Langendoen 2022. We defined internationally important stopover sites as sites that supported more than 0.25% of the total estimated population of one or more species and also identified the number of sites of international importance for each species. We compared population changes between 2014-2015 and 2019-2020 for species for which over 5% of the population was accounted for during southward and northward migration. For each study site, we  compared population changes based on peak counts during migration and compared the number of internationally important species (observed in populations that accounted for > 0.25% of the total estimated population). Finally, we counted the number of shorebirds arriving at each study site and categorized species according to whether 1%, 5%, or 20% of the total estimated population was observed.

Shorebird migration patterns
In this study, we observed 120,729 individuals of 42 species in the autumn of 2014; 264,634 individuals of 38 species in the spring of 2015; 120,225 individuals of 40 species in the autumn of 2019; and 228,472 individuals of 44 species in the spring of 2020. These data reflect the sum of peak counts from all sites. The observed population size in the spring was consistently about double that observed in the autumn season (Fig. 2).
The dominant species and population sizes of shorebirds migrating to Korea differed between spring and autumn. For the dominant species Dunlin (Calidris alpina), approximately 90,000-100,000 individuals were observed during the northward migration (spring), but only 20,000-30,000 individuals were observed during the southward migration (autumn). For the

Distribution of internationally important species
Across the entire survey period, 19 species were observed in populations that accounted for more than 0.25% of the total estimated population (staging threshold; internationally important species). This included 16 species in the autumn of 2014, 17 species in the spring of 2015, 18 species in the autumn of 2019, and 19 species in the spring of 2020 (Table 1). Among these species, the Common Greenshank, Terek Sandpiper, Far Eastern Curlew, Grey Plover, Dunlin, Eurasian Whimbrel (Numenius phaeopus), and Eurasian Curlew were observed in nearly all survey areas. For 11 species, 5% of their total EAAF population was observed, and for three species, 20% of their total EAAF population was observed; this includes three species (Bartailed Godwit, Kentish Plover, and Eurasian Oystercatcher) in 2014-2015, and four species (Far Eastern Curlew, Kentish Plover, Terek Sandpiper, and Eurasian Oystercatcher) in 2019-2020 (Table 2). Several globally threatened species were also observed during their migration. For one, the Spoon-billed Sandpiper (

Distribution of internationally important stopover sites
Because they contained more than 0.25% of the population of one or more shorebird species, all surveyed areas were classified as internationally important stopover sites (Table 3). Among these sites, Yubu Island was regarded by far as the most important area; this site supported 13 species at internationally important levels and approximately 90,000 individuals in 2014-2015, and it supported 15 species at internationally important levels and approximately 55,000 individuals in 2019-2020. The second and third highest counts in 2014-2015 were recorded at Namyang Bay, which supported 42,794 individuals and Yeongjong Island, which supported 40,713 individuals. In 2019-2020, the second and third highest counts were recorded at Asan Bay, which supported 24,014 individuals and Biin Bay, which supported 20,702 individuals.
In general, when compared to records from the early 2000s, the number of observed internationally important species decreased, with these numbers dropping sharply in Saemangeum (including Mangyeunggang and Dongjingang estuaries;  Lee et al. 2018), and the number of shorebirds observed in the Mangyeunggang and Dongjingang estuaries markedly declined (Fig. 3). In addition, between 2014-2015 and 2019-2020, the population size of shorebirds at Yeongjong Island and Namyang Bay also declined, likely due to the reclamation of coastal areas. In contrast, the number of species and individuals was either maintained or slightly increased at Suncheon Bay and Gomso Bay, which have been protected wetland areas.  Among the study sites, 14 sites in 2014-2015 and 17 sites in 2019-2020 harbored more than 1% of several shorebirds' migrating populations (Table 4). Moreover, two sites supported more than 5% of one or more shorebird species' populations in 2014-2015, while three sites did so in 2019-2020. In particular, Yubu Island proved to be the most important stopover site in the Republic of Korea because it supported more than 20% of the Eurasian Oystercatcher population both in 2014-2015 and 2019-2020 and did the same for the Kentish Plover in 2019-2020 (Table 4). Yubu Island also supported more than 5% of the population of Bar-tailed Godwits (NT) during their northward migration in 2014-2015 and similarly supported more than 5% of the population of Far Eastern Curlews (EN) during their northward migration in 2019-2020. The other sites that supported more than 5% of certain species' populations were Mangyeunggang Estuary (> 5% of the Kentish Plover population in 2014-2015), Biin Bay (> 5% of the Eurasian Oystercatcher population in 2019-2020), and Ganghwa Island (> 5% of the Far Eastern Curlew population in 2019-2020).

DISCUSSION
This study confirmed the international importance of tidal flats along the southwestern coast of the Republic of Korea as shorebird stopover sites, given the internationally important numbers of waterbirds supported by all 18 of these sites. In particular, Yubu Island contained the highest species diversity. In addition, these sites supported internationally important proportions of the populations of 15 shorebird species, including the Spoon-billed Sandpiper, which is designated as critically endangered, as well as two endangered and four near-threatened species. For 19 species, more than 0.25% of the total estimated migrating population was accounted for. In addition, several individuals of threatened species, including the critically  endangered Spoon-billed Sandpiper and the endangered Far Eastern Curlew, Great Knot, and Spotted Greenshank were observed. This study also documented six near-threatened species, specifically the Eurasian Curlew, Bar-tailed Godwit, Eurasian Oystercatcher, Black-tailed Godwit, Grey-tailed Tattler (Tringa brevipes), and Red-necked Stint. All surveyed areas were classified as internationally important stopover sites because they each contained more than 0.25% of one or more shorebird species' populations. However, the size of shorebird populations declined between 2014-2015 and 2019-2020. Barter (2002) estimated that approximately two million shorebirds (about 40% of EAAF populations) used stopover sites in the Yellow Sea area, which includes the Republic of Korea, North Korea, and China, during their northward migration, and about one million shorebirds used this area during their southward migration. In particular, more than 90% of Great Knot, Bar-tailed Godwit, Grey Plover, Kentish Plover, Eurasian Curlew, and Far Eastern Curlew individuals presumably used the Yellow Sea areas in the spring during their northward migration to breeding sites (Barter 2002). In this study, more than 20% of the populations of Eurasian Curlew, Far Eastern Curlew, Kentish Plover, Terek Sandpiper, and Eurasian Oystercatcher, as well as more than 5% of 11 other species' populations, were observed during the northward and southward migratory period. Generally, shorebirds tend to congregate in high-quality habitats in which birds can forage on macrobenthos in tidal flats, adding to evidence that habitat and food conditions likely determine shorebird populations among stopover sites (Choi et al. 2015, Wang et al. 2022). In addition, roosting sites are as important as food resources for supporting shorebird populations during stopovers (Burton et al. 1996, Rogers et al. 2006, Peter and Otis 2007. Rogers et al. (2006) found that the energetic cost of roosting in the Great Knot is closely related to roosting site availability and management of roosting sites is important in preserving existing habitats. Among the study sites, Yubu Island was identified to be an optimal stopover site, harboring the highest number of shorebirds observed in the Republic of Korea and offering many roosting and feeding sites in various environments. Given the high abundance of migratory shorebirds, these results indicate that the Republic of Korea's tidal flats are important stopover sites. Especially in the Yellow Sea area, the importance of Korean tidal flats is emphasized because it can be difficult for shorebirds to find high-quality alternative migratory sites when the original habitat is lost or degraded (Murray et al. 2014, Zhang et al. 2018, Wang et al. 2022).
In the present study, the number of observed shorebirds was higher during the northward migration than during the southward migration. This disparity has been noted before (e.g., Choi et al. 2016) and has several possible explanations; first, the path of northward migration frequently differs from that of southward migration (Newton 2007, Gill et al. 2009, Minton et al. 2010, Lindström et al. 2011. For example, the Ruddy Turnstone (Arenaria interpres) and Bar-tailed Godwit used southward migration routes that differed from those of their northward migration along the EAAF (Minton et al. 2011, Battley et al. 2012. Second, populations of a species migrating southward are more spread out over time because failed breeders, juveniles, and successful breeders migrate through stopover sites at different times (Tomkovich 1997, Choi et al. 2011. Therefore, the number of birds observed on southward migration routes would be lower even if the number of birds migrating northward and southward was the same.
In the EAAF, many shorebird populations have declined over the past several decades because of habitat destruction within stopover sites (Piersma et al. 2016, Studds et al. 2017). Since the early 2000s, counts of shorebirds in the Yellow Sea area have generally declined, with population decreases ranging from as little as 15% to as much as 90% (Wang et al. 2022). In the Republic of Korea, between 2014-2015 and 2019-2020, the observed population of Kentish Plovers was reduced by almost half, and populations of Eurasian Curlews, Dunlins, and Bar-tailed Godwits also declined by 20%. In particular, the Great Knot, which previously had the largest population size among shorebirds stopping over in the Republic of Korea, i.e., approximately 130,000 individuals in 2005 (NIER 2005), experienced a sharp decrease in population size because of the substantial habitat loss resulting from the construction of the Saemangeum dike (Moores et al. 2008, Garnett et al. 2011. Consequently, the Great Knot was uplisted from "least concern" to "vulnerable" in 2010 and further uplisted to "endangered" in 2015 after its global population decreased from 380,000 to 292,000-295,000 individuals (Wetlands International 2015). More recently, our results showed that only a small number of Great Knot individuals (about 30,000) were confirmed to use tidal flats along the southwestern coast of Korea during their northward migration. Great Knots primarily used Yubu Island, which supported 14,000-15,000 individuals during both years. In 2014-2015, this species also relied on Namyang Bay, which supported 13,100 individuals, and Gomso Bay, which supported 4361 individuals, but the number of Great Knots observed at these sites declined to 4108 and 9700, respectively, in 2019-2020.
Between the two survey periods, populations decreased in certain areas (Geumgang Estuary, Namyang Bay, and Mangyeunggang Estuary) and increased in others (Asan Bay, Biin Bay, and Geumgang Estuary). In addition, although population sizes decreased overall, the number of species present at internationally important levels in each survey area increased.  Wang et al. (2022) suggest that these populations have dramatically decreased in China (-69.74% and -74.98%, respectively). The growth of Far Eastern Curlew and Terek Sandpiper populations in Korea is the likely product of habitat destruction because habitat loss increases the likelihood of populations moving to remaining habitats (Yang et al. 2011, Wang et al. 2022. It is thought that habitat destruction due to large-scale development along the western coast of Korea influenced shorebird populations, which used other remaining areas as stopover sites, and also changed species composition. Therefore, habitat loss within EAAF stopover sites might cause shorebirds to relocate to other sites, especially Korean tidal flats. We identified 18 sites as internationally important stopover sites for shorebirds of the EAAF. Habitats at these sites are of great importance to shorebirds and deserve international conservation effort, but only seven of these areas are designated as protected areas in the Republic of Korea. In addition, habitat quality in the region was recognized as internationally important in spite of continued deterioration (Barter 2006, Bamford et al. 2008, MacKinnon et al. 2012. The areas without legal protection experienced slight or sharp decreases in the migrating populations of internationally important species. In particular, Saemangeum (Mangyeunggang and Dongjingang estuaries), Yubu Island, Yeongjong Island, Namyang Bay, and Nakdonggang Estuary showed remarkable decreases in the number of visiting shorebird species and their population sizes. These areas have suffered significant habitat loss due to tidal flat reclamation, coastal development (airports, factories, ports, etc.), and environmental pollution. The rate of habitat loss outside protected areas was globally highest (-1.83% year) in the Yellow Sea area, which includes the Republic of Korea, North Korea, and China (Murray and Fuller 2015). In particular, Saemangeum, which was once a very important site supporting hundreds of thousands of shorebirds, now stands as evidence of the functional correlation between reclamation and biodiversity decline. After the completion of the Saemangeum seawall in 2006, the population of shorebirds in Saemangeum dramatically decreased (Moores et al. 2016), and shorebird populations in the adjacent Geumgang Estuary zone (including Yubu Island, Biin Bay, Geumgang Estuary, and Gomso Bay) slightly increased. However, this population increase in adjacent regions did match the population decrease in Saemangeum, and the local population has not yet recovered. When stopover sites are destroyed or become unusable, adjacent habitats become relatively more important for congregating birds . Therefore, it is necessary to precautionarily expand the legal protection of wetland environments to conserve biodiversity and ecosystem services.
The Republic of Korea once had vast tidal flats, but large portions have been lost due to development projects (Hyun et al. 2013, Lee et al. 2018). Nonetheless, this study revealed that many shorebirds still use stopover sites in Korea's southwestern tidal flats, including Ganghwa Island, Yeongjong Island, the Songdo tidal flats, Namyang Bay, Saemangeum, the Yeonggwang coast, and especially Yubu Island, which represented the most important stopover site in the Republic of Korea. In addition, some endangered species, such as the Spoon-billed Sandpiper, Spotted Greenshanks, Great Knots, and Far Eastern Curlews, were also observed in Korea's tidal flats in internationally important numbers (NIBR 2007, 2020, NIER 2003, 2005. To preserve populations of migratory shorebirds, it is necessary to conserve an extensive number of existing key stopover sites along the EAAF (Wang et al. 2022). In the Republic of Korea, the protected area system has been highly effective in preventing habitat destruction, even though the rate of habitat loss in designated areas has been relatively high (Murray and Fuller 2015). Therefore, it is necessary to legally protect important areas and adjust the spatial scope for protection and management. Namyang Bay is known as a major habitat for shorebirds, and despite continued reclamation projects after the dike construction was completed, the area still meets the requirements of an internationally important stopover site. Similarly, although the tidal flats and coastal topography of Yeongjong Island have undergone many changes due to the construction of the Incheon International Airport, the northernmost and southernmost parts of this area, where the tidal flats and waterways are well developed, are still used by shorebirds as resting and foraging sites. Lastly, the areas of Saemangeum (Mangyeunggang Estuary, Geumgang Estuary, Yubu Island, and Biin Bay) are geographically close to one another and located within the area where shorebirds relocate with the tide. These sites are of highest priority for conservation because they harbored the largest number of observed shorebirds among the surveyed sites. In addition, diverse species, including internationally endangered species, used this land as a stopover site. Therefore, it is necessary to expand the scope of the Seocheon Tidal Flat Wetland Protected Area (protected area specially designated by the Ministry of Oceans and Fisheries of Korea that includes Yubu Island and Biin Bay) to ensure the protection and efficient management of shorebird habitat.
In 2021, the Getbol Korean Tidal Flats were designated as World Heritage Sites based on containing "the most important and significant natural habitats for in-situ conservation of biological diversity, including those containing threatened species of outstanding universal value from the point of view of science or conservation" (UNESCO World Heritage Centre 2021). The Korean government has fulfilled the obligation to protect these areas by law in their entirety as Wetland Protected Areas (WPAs) under the Wetlands Conservation Act (WCA) and other laws and regulations (e.g., the Conservation and Management of Marine Ecosystems Act) that have been applied to restrict damaging activities. In addition, the Korean government is also carrying out various protection policies by designating WPAs in areas other than World Heritage areas. These programs include regular investigation of the ecological status of wetlands, as well as the pollution status and land use status in sites around protected areas, and implementing policies such as prohibiting building installations, the collection of natural objects, and the capture of animals and plants. These efforts will further strengthen the outstanding universal value inherent in the Yellow Sea and the EAAF and support the biodiversity of communities in and around the wetlands.

Author Contributions:
All authors contributed to the study's conception and design. Material preparation, data collection, and analysis were performed by Ju- Hyun Lee, In-Cheol Kim, Jong-Ju Son, Jae-Ung Jang, and Si-Wan Lee led by Ha-Cheol Sung. The first draft of the manuscript was written by Ju-Hyun Lee and In-Cheol Kim and