An updated checklist of the marine birds (Aves) of Golfo Dulce, southern Costa Rica

. Golfo Dulce is a fjord‑like embayment on the southern Pacific coast of Costa Rica, which encom‑ passes the marine protected area of Piedras Blancas National Park. The gulf is surrounded by mangroves, rocky shores, and sandy beaches that are home to migratory and resident bird species. Here, we provide a checklist of 73 marine birds, including records of six years of annual counts and incidental observations in Golfo Dulce. This paper represents an updated bird list, based on the 1998 Basic Information Units from Instituto Nacional de Biodiversidad. We added 16 species to the list, showing the importance of maintaining a monitoring program of marine birds in the Golfo Dulce area. We have also demonstrated the valuable contributions of citizen science for obtaining scientific information.


INTRODUCTION
Given its unique oceanographic and geomorphological conditions, Golfo Dulce in the southern Pacific coast of Costa Rica is considered one of four biological systems with a fjord-like structure in the tropics (Knauss 1978;Cortés 1992;Hebbeln et al. 1996).Along this tropical fjord, there are four coastal wetland habitats: mangroves, sandy beaches, sea grasses, and coral reefs (Cortés 1992;Weissenhofer et al. 2008;Samper-Villarreal and Silva-Benavides 2015;Samper-Villarreal et al. 2018).Aditionally, the gulf comprises rocky shores and small islets (Los Mogos and Isla Pelícano).
As explained by Barrantes and Chaves-Campos (2009), the complex topography and high tidal fluctuations of the Costa Rican Pacific coast have resulted in numerous habitats appropriate for marine birds.Despite the biological importance of Golfo Dulce for avian diversity, marine-bird studies or long-term monitoring in this environment are marginal, in comparison to studies regarding terrestrial birds in adjacent areas.
The first recorded list of marine birds in Golfo Dulce was built upon previous observations and was subsequently deposited on the former Basic Information Units (Unidades Básicas de Información; UBI) from Instituto Nacional de Biodiversidad in 1998.Although the UBI platform is no longer working, the marine birds list is available in Quesada-Alpízar et al. (2006).Origin of the data (dates and references) were not mentioned.Since then, no major updates have been made to the marine bird diversity status of Golfo Dulce.
Global changes, encompassing extreme climatic events, which are expected to impact marine and coastal zones (BIOMARCC-SINAC-GIZ 2013), make it essential to understand the distribution of birds in marine environments and coastal wetlands in a time frame.In addition to global changes, local anthropogenic threats to bird habitats in the Golfo Dulce region are ongoing.In the surrounding areas of Golfo Dulce (Osa and Golfito political divisions), tourism and real estate development have significantly expanded since the 2000s, especially in 2007-2009.Environmental impacts resulting from these activities have included damage to mangroves (Román and Angulo 2013).
Due to cause-effect association of microclimates and habitats, marine birds are indicators of a marine ecosystem's integrity (Rajpar et al. 2018).Therefore, fields like ecology and conservation biology that depend on taxonomic data require baseline information and continuous updating of species lists.Checklists can provide reliable information on changes in bird populations, phenology, and geographic and climate patterns at a regional scale (Droege et al. 1998), information that is required to inform future conservation and management practices.Hence, we provide here updated information on marine birds of the Golfo Dulce, which will be especially important at the current juncture of climate change.

STUDY AREA
Golfo Dulce (between 08°27′N-08°45′N and 083°07′W-083°30′W) is a 14 km wide and 50 km long gulf located on the southeast of Costa Rica's Pacific coast; the gulf has an area of 750 km 2 (Wolff et al. 1996) (Figure 1).This environment is part of the Tropical Eastern Pacific and comprises 21.5% of the marine biodiversity of the Pacific coast of Costa Rica (Morales-Ramírez 2011).Since the creation of Piedras Blancas National Park (PBNP) in 1991, 1200 ha are inside the marine protected area (SINAC 2015).
The inner basin of the Golfo Dulce (Figure 1) is 215 m deep and a shallow mouth 70 m deep; there is flux stratification within the water column of the gulf, in a pattern typical of fjord-like estuaries (Richards et al. 1971).The coral reefs in the gulf have a high topographic relief, low coral diversity, and live coral coverage, but a high dead coral coverage (Cortés 1992).The high diversity of fishes in Golfo Dulce (Cortés 1992;Bussing and López 1996;Guzmán-Mora and Molina-Ureña 2008;Palacios-Martínez 2015) may influence the occurrence of sea birds, herons, and gulls.
The gulf is surrounded by four main estuaries forming expansive mangroves: those of the Esquinas and Coto Colorado rivers in the northern mainland and the Rincón and Tigre rivers draining from the Osa Peninsula.Six species of mangrove occur in Golfo Dulce: Rhizophora mangle L., Rhizophora racemosa G.Mey, Pelliciera rhizophorae Planch.& Triana, Avicennia germinans (L.) L., Laguncularia racemosa (L.) C.F.Gaertn., and Conocarpus erectus L. (SINAC 2015).Small patches of swamp forest dominated by the tree Pterocarpus officinalis Jacq.are also part of the coastal region.The Esquinas and Coto Colorado mangroves are bordered by these swamp forests, where P. officinalis is associated with Mora oleifera (Triana ex Hemsl.)Ducke, as described by Allen (1956).Steep rocky shores, rocky islets, and sandy beaches with gravel alternate along the coast, and this is especially evident in PBNP.These beaches transition into coastal forests where the spindrift, saltwater, and dry soil conditions influence the vegetation (Weissenhofer et al. 2008).

METHODS
We define marine birds as birds that spend at least a part of their life in a marine habitat, as defined by Schreiber and Burger (2001).Marine habitats include coastal areas, islands, estuaries, wetlands, and oceanic islands.Accordingly, we classify marine birds into three groups for: • Seabirds are colonial species that feed in salt waters and often migrate long distances from breeding grounds to wintering areas.In some cases, these birds may connect geographically different marine environments.This group includes Phaetontidae Brandt, 1840, Procellariidae Leach, 1820, Fregatidae Degland & Gerbe, 1867, Sulidae Reichenbach, 1849, Phalacrocoracidae Reichenbach, 1850, Pelecanidae Rafinesque, 1815, Haematopodidae Bonaparte, 1838, Stercorariidae J.E. Gray, 1871, and Laridae Rafinesque, 1815.• Shorebirds spend their lives on sandy beaches, mudflats, and shores of rivers and lakes, and they generally only interact with the edge of the marine environment.This group includes Recurvirostridae Bonaparte, 1854, Charadriidae Leach, 1820, and Scolopacidae Rafinesque, 1815.• Waterbirds inhabit and feed mostly in freshwater and generally interact with the marine environment only in the coastal zone.This group includes Ciconiidae J.E. Gray, 1840, Anhingidae Reichenbach, 1849, Ardeidae Leach, 1820, and Threskiornitidae Richmond, 1917.Because no species were collected and non-invasive methods were used to obtain data, scientific permits for collection were not required.Among the data reported here, some observations are incidental (observations in 2007 and 2014-2023), while all other observations derive from bird counts (in 2015-2017 and 2021-2023), following the area-search method.The selected area was a 24 km diameter circle, following the National Audubon Society's Bird Count guidelines (2022).The two methods used in this study are: (1) Area search (or route count).This method, usually repeated, attempts to identify, and count all the birds in a specified area within a specific time.It is most effective if repeated at the same time of day.It may involve an area search of the entire study site or of selected plots within the area (Ralph 1981;Ralph et al. 1993).The annual bird counts we undertook (Golfito Bird Counts, a citizen-science approach) were made in December, on the same day and at the same time (5:00-13:00 h) in each year.Teams of birders included a leader (a biologist or expert naturalist with a proven knowledge of bird diversity and tasked with maintaining data quality) and amateur bird observers.Specialists were chosen on the basis of their previous experience working with birds and their formal certifications as professionals in the field of biology or in nature guiding.One team of observers, including at least one specialist and four amateurs, surveyed each of these areas: 34 km (12.73 km 2 ) by boat (outboard motor) in the northern Golfo Dulce from Punta Gallardo to the Esquinas river estuary (Figure 1 shown as red lines).9 km (3 km 2 ) in the southern Golfo Dulce from the Coto Colorado river estuary to Playa Azul and Punta Piedra; one sector by boat (outboard motor) and one sector walking (Figure 1, shown as red lines) The search area included mangroves, river estuaries, sandy beaches, rocky shores, and an extension of 1 km from the land into the ocean.Birds were identified using field guides (Garrigues and Dean 2014) and from photographs.In the list, 44 records originate from the Golfito Bird Count census.
(2) Incidental observations.This method is particularly important in determining the presence of rare and unusual species not normally observed, not captured by nets or traps, or missed during regular censuses or counts (Ralph et al. 1993).This is an unstandardized method.The incidental observations included in our list are data obtained by us while conducting birdwatching tours or fieldwork on other taxa in the study area from 2007 to 2017 for LS and from 2014 to 2023 for ARB.All these records are based on observations of birds seen and identified on-site.We categorized our incidental observations as either from the inner or outer basins of the Golfo Dulce using an imaginary line (Figure .1, shown in blue) based on Cortés (1990).The incidental observations covered the whole of the study area (750 km 2 ).Twenty-five records in the list originate from our incidental observations.Data obtained from bird counts were deposited in the Costa Rican Ornithologist's Union (Unión de Ornitólogos de Costa Rica) database, and the Golfito Bird Count results have been documented in the eBird platform since 2021.We also reviewed the literature to find and verify historical records and compare our results to those obtained from UBI records.We used Google Scholar and accessed repositories in the Universidad Nacional, Universidad de Costa Rica, and Universidad Estatal a Distancia to find research articles, field-course reports, theses, and dissertations using five keywords (both in English and Spanish): "Golfo Dulce", "Golfo Dulce birds", "Golfo Dulce marine birds", "Costa Rica marine birds", and "Golfito birds".The literature found the following references: Araya Ajoy (2005), Young and Zook (2006), Barrantes and Chaves-Campos (2009), Morales-Ramírez (2011), Bessesen (2015).A field guide (Garrigues and Dean 2014) was also reviewed to verify records of rare species, but only data specifying the Golfo Dulce were included in our list.The UBI bird list (Quesada-Alpízar et al. 2006) was cited only by Morales-Ramírez (2011).
In the present list, three records originate from this literature review (Garrigues andDean 2014, Young andZook 2016) and confirm the mentioned species in the UBI that were not observed in this study.We used eBird, a citizen science platform, which has become integrated with existing regional platforms, to increasing the data available (Sullivan et al. 2014).All records in eBird  filtered by Osa Peninsula and Golfito were included in our data.Single , photographic, and audio records were considered, since rare or unusual observations on this platform are reviewed by experts.One record originates from our review of eBird.
The checklist is arranged following Sandoval and Sánchez (2020), and the taxonomy is according to the American Ornithological Society (Chesser et al. 2023).Species status resident, migratory, or breeding migrant in Costa Rica follows the information by Stiles and Skutch (1989) and Garrigues and Dean (2014).Abundance codes follow the guidelines developed by the American Birding Association (2023).For this purpose, we estimated abundance averaging the number of individuals observed on the area search or route count methods, and these values were used to classify species as rare, uncommon, and common.Incidental observations were used to classify species as accidental or rare.
• Common.Found in moderate to large numbers (>10 individuals on average) and easily found in appropriate habitat at the right time of year.• Uncommon.Found in small numbers (2-10 individuals on average), and usually, but not always, found with some effort in appropriate habitat at the right time of year.• Rare.Occurs annually in very small numbers (<2 individuals on average).Not to be expected on any given day but may be found with extended effort over the course of the appropriate season.• Accidental.Highly unpredictable (<3 reports under incidental observations).Not typically found most years, with fewer than 10-15 historical records.Also included are vagrants out of their usual range.

RESULTS
Our results indicate that the marine avifauna of Golfo Dulce is composed of 73 species in 16 families.Twenty species are resident and 50 are Nearctic migrants.Two species have resident and migratory populations, and one is a breeding migrant (Table 1).In this list four species are classified as globally Near Threatened by the International Union for Conservation of Nature (IUCN 2023).
We document 32 species from the Golfo Dulce, of which we classify 28 as rare.Sixteen species are newly recorded from the area, in relation to the UBI list; these are marked as "newly recorded from the area".
Table 1.List of the marine birds found in Golfo Dulce, southern Costa Rica region between 2007 and 2023, their status, classification, abundance, habitat (main habitat used by the species), and source of record.Status: R = Resident, M = Migratory, BM = Breeding migrant.2A Observations.COSTA RICA -Puntarenas • Golfo Dulce; 08°43′11″N, 083°20′18″W; 14.V.2018; A. Rojas-Barrantes obs.; first observed in the estuary of Esquinas river, solitary.The species was also observed solitary or in small groups in sandy beaches of the northern region of Golfo Dulce and Tigre river.
Identification.Large shorebird with black-and-white pattern.Very long orange legs and pointy bill are distinctive.

Recurvirostra americana Gmelin, 1789
American Avocet Newly recorded from the area Figure 2B Observations.COSTA RICA -Puntarenas • Golfo Dulce; 08°40′43″N, 083°19′32″W; 18.IX.2018; A. Rojas-Barrantes obs; this provides the first record for the southern Pacific coast of Costa Rica.This species is a casual migrant in the country and has been reported in Palo Verde (northern Pacific coast), Tortuguero (Caribbean slope), Lake Arenal, and Chomes (central Pacific coast) (Garrigues and Dean 2014).R. americana forages in saltmarshes and ponds, but we observed a single individual foraging at low tide at Playa Caña Blanca (Figure 1), a sandy beach in northern Golfo Dulce.Later, in December 2022, R. americana was observed again foraging along with waterfowl in rice fields near the mouth of the Coto Colorado river, reported on eBird database https://ebird.org/checklist/S123757261 with photographic record.
Identification.Large shorebird with black-and-white pattern in wings and back.Long upturned bill.Long, gray-blue legs.No other shorebird has these characteristics.
Identification.Large and straight red to orange bill, and black head are distinctive.Identification.Set apart from the similar and very common P. squatarola by the lack of black axillar patch saw it at flight.Additionally, on nonbreeding plumage shows a clear whitish supercilium line.
Identification.Two distinctive black bands on the chest set them apart from similar plovers.
Identification.Set apart by the barred patter in the tail, that is narrower than in very similar L. griseus.
Identification.Yellow legs and base of the bill set them apart from other similar plovers on breeding plumage.
Identification.Set apart from similar P. tricolor by the white forehead in non-breeding plumage.Breeding individuals have a distinctive red coloration on the neck.
Identification.Set apart from similar phalaropes by the distinctive red coloration in breeding plumage.Set apart from similar nonbreeding P. lobatus by the dark coloration on tail feathers and non-streaked back.
Identification.Set apart from similar jaegers by the round shape of the extended tail feathers.
Identification.Set apart from similar jaegers by a slender body shape, smaller head with thinner bill, and two pointed central tail feathers.
Identification.Set apart from similar gulls when showing black band on primaries when flying.
Identification.The slender black bill is distinctive and set them apart from similar gulls.Also, small black patch behind the eyes can be used to recognize this species.

Identification.
A distinctive black bird with a white cap.

Phaethon aethereus Linnaeus, 1758
Red-billed Tropicbird Figure 3E Observations.COSTA RICA -Puntarenas • Golfo Dulce; 08°39′17″N, 083°18′20″W ; 20.IX.2018; A. Rojas-Barrantes obs.; inner basin, recorded one time.Observed individual landed on a boat, holding a prey (medium size fish).Bird seemed either exhausted or sick, as it did not show signs of distress or attempted to fly away when approached by humans.Later, on December 2023 P. aethereus was observed in flight, reported on eBird database https://ebird.org/checklist/S156151255Identification.Body with white feathers with black-barred back and base of upper wing.Black eye stripe and yellowish bill.
Identification.Big size and lightly curved bill are distinctive.Only bird with these characteristics in the study area.It might be confused with Jabiru mycteria Lichtenstein, 1819 however, there is no geographic overlap.
Identification.Differs from other boobies by the dark color of the feet and color of the bill (yellow).Head is complete white in adults.
Identification.Differs from other boobies by their blue legs.

DISCUSSION
The 1998 UBI list includes 58 bird species in 11 families.From this list, we could not confirm five species by either our own observations or from historical records, the literature, or citizen-science platforms; these five species are: Ixobrychus exilis Gmelin, 1789 , Calidris alpina,Linnaeus, 1758, Calidris canutus,Linnaeus, 1758 Calidris melanotos Vieillot, 1819, and Tigrisoma lineatum Boddaert, 1783.We suggest that the record of T. lineatum, a species which does not occur on the Pacific slope, may be a misidentification of T. fasciatum, which is rare and a recent colonizer of the interior area of the Osa Peninsula (Garrigues and Dean 2014), or perhaps of the more common T. mexicanum.Two species in the UBI list, Agamia agami and Chroicocephalus philadelphia were not confirmed by our observations, but there are literature records.Garrigues and Dean (2014) mentioned both species as rare in the Golfo Dulce coastal area, and eBird has recent observations in nearby area La Gamba for A. agami.For C. philadelphia, Garrigues and Dean (2014) mentioned May 1979 as the observation date.At the time of publication of Garrigues and Dean's (2014) field guide, the last observation of C. philadelphia was in January 2013 in Playa Potrero, Guanacaste, northern Pacific of Costa Rica, and is described as a casual winter visitor.The most recent observation of C. philadelphia in Costa Rica reported on eBird (eBird list S99276953) was in Tamarindo, Guanacaste in 2019.
The family with the most species observed was Scolopacidae (shorebirds).This is expected due to the suitability of coastal habitats as a feeding ground in Golfo Dulce.Shorebirds congregated in mangrove ecosystems at river estuaries, with higher congregations at the mouths of the Esquinas, Coto Colorado, and Tigre rivers, and at Playa Azul, a long sandy beach near Golfito Bay (Figure 1).Four Near Threatened species, Egretta rufescens, Thalasseus elegans, Calidris pusilla, and Charadrius nivosus, were all observed using these congregation sites.The ability of a site to support large numbers of shorebirds is largely determined by the extent of intertidal sand and mudflats and by the density, availability, and seasonal predictability of their invertebrate prey; sites with greater densities of invertebrates typically support higher shorebird densities (Galbraith et al. 2002).
Golfo Dulce shows a relatively healthy environment (Morales-Ramírez 2011), and it does not show high rates of pollution from pesticides or hydrocarbons (Acuña-González et al. 2004;Sponberg 2004;Sponberg et al. 2011).Nonetheless, as suggested by Morales-Ramírez (2011), efforts to restore Golfito Bay are needed.We note the importance of addressing the effects of coastal developments for tourism without sustainable planning, especially in Golfito Bay, where development has impacted bird habitats, such as mangroves.According to Román and Angulo (2013), tourism and fishing are the activities that most directly impact marine-coastal resources in the region; however, there are no historical records to show how these activities have affected marine avian diversity in Golfo Dulce.
The conservation of marine birds have received little attention in Costa Rica (Barrantes and Chaves-Campos 2009).Pollution from pesticides, sewage, and solid waste, as well as habitat destruction, are threats to marine avifauna in Costa Rica.With the increase in recent pollution events from hydrocarbon (from fishing vessels) and small-scale palm-oil spills (from tankers for oil exportation) in Golfito Bay (pers.obs.), coupled with the inadequate management of solid waste, we recommend the frequent monitoring of this type of sea pollution be undertaken.Inadequate environmental guidelines in the port for the palm oil and fishery industries could be responsible for increased threats to marine birds in Golfo Dulce from these activities.
Climate change is a major threat to marine avifauna in Golfo Dulce due to rising sea levels, warming and acidification of water, and more frequent extreme weather events.Projections have inferred that PBNP is highly vulnerable to future climate change, with outcomes that include changes in temperature and precipitation, ocean swells, the frequency of extreme climatic events, and the El Niño Southern Oscillation (BIOMARCC-SINAC-GIZ 2013).
Sea warming will impact the abundance and quality of the prey for waterbirds and change the biota of the intertidal habitats in which these birds forage, with changes expected to local and even national patterns of bird distributions (Kendall et al. 2004).The rise in the sea level may change the complexity of the coastline, thereby fragmenting habitats, and the reduction in foraging area may be a major contributor to the decline of shorebird populations (Convertino et al. 2012).As mentioned earlier, most species in this study are shorebirds.A rising sea level would make intertidal areas to fully subtidal and result in reduced foraging habitats for shorebirds (Galbraith et al. 2002), and such a change would be greatest where the intertidal zone is not able to shift inland either by topography or by seawalls (Galbraith et al. 2002).Such a situation may occur in the northwestern Golfo Dulce, where forests act as a topographic wall.Because of the minor anthropogenic development in this area of the gulf, the coastal wetland habitats are especially important as feeding grounds for Nearctic migratory species.
Given that much of the data of interest for this study comes from observations made in the Esquinas river, we emphasize the need to strengthen the research and monitoring program of the corresponding conservation area for PBNP (Osa Conservation Area).This river and its estuary, which serve as the natural limit of the protected area, consists of flood-prone areas vulnerable to climate change scenarios, and is also being affected by anthropogenic impacts such as the use of illegal fishing arts and water pollution by pesticides from upstream agricultural plots (SINAC 2015).At the same time, improving the control and protection program in Esquinas river and its habitats would contribute to the conservation of the ecological integrity of the important feeding grounds for marine birds.
Long-term monitoring, such as annual bird counts and counts in selected plots, is a fundamental tool in understanding how ecosystems change over time.Therefore, obtaining ecological data is critical to understanding the resilience, stability, and changes in an ecosystem (Chiappa-Carrara et al. 2017).This highlights the relevance of projects such as the Golfito Bird Counts, from which most of the data for this study was obtained.The Golfito Bird Count, a citizen science approach, is an annual monitoring project, supported by the community participation.It allows obtaining information of interest for monitoring environmental changes in the marine-coastal ecosystems of Golfito, and analyzing the long-term population dynamics of birds at a local level.This project also promotes that scientific knowledge must be accessible and beneficial for the community.
As citizen-science projects are rapidly increasing in aims of achieving the 2030 Agenda for Sustainable Development Goals (de Sherbinin et al. 2021), it is imperative that citizen science follows standards and is curated by experts so that the data obtained are reliable over time.According to Si and Ding (2011), Sullivan et al. (2014) and Randler (2021) one example of successful citizen science project is applied in ornithology, and one of the earliest to do it, because birdwatchers spend a large amount of time collecting data.Basic and low-involvement projects like Christmas Bird Counts that can be based on a single time point in data collection, can conveniently contribute to the generation of knowledge, while social dimensions are also addressed (Randler 2021).
As suggested by Sullivan et al. (2014), citizen science is better accomplished by expanding the range of activities beyond data collection, including community engagement, data curation, data synthesis and analysis, pattern visualization, and delivery of results to a broad community of possible stakeholders.In this context, the Golfito Bird Counts, followed a well-established methodology and sampling protocol.At the same time, it has expanded its scope of action, reaching a community commitment and democratization of science that ensures the continuity of this joint effort.
It is remarkable the modern use of global citizen science platforms like eBird.The eBird project is generating biodiversity data which are increasing for different uses (Sullivan et al. 2014), in our case, it allowed us not only to confirm records, but to grow the regional data input, which can be harnessed for monitoring the distribution of resident and migratory birds for future research in the area.
We recommend continuing the Golfito Bird Counts through the involvement of citizens and with strategic alliances, including Piedras Blancas National Park.Moreover, specific marine monitoring would be particularly beneficial to study seasonality and relative abundance of bird species over time.

Figure 1 .
Figure 1.Golfo Dulce, on the southern Pacific coast of Costa Rica.Bird counts routes (BCR) are shown in red.The division of the gulf into inner and outer basin is shown as a blue line.
Differs from other boobies by their red legs.Differs from other boobies by the combination of white head and grey legs and bill.Barrantes obs.; first observed in Golfito Bay, commonly seen solitary or in small groups in shallow water near the shore and river estuaries.Newly recorded from the area Identification.The shape of the needle-like bill is distinctive.Show sexual dimorphism, where females are different than males in having a brownish upper part of the body.COSTA RICA -Puntarenas • Golfo Dulce; 08°38′12″N, 083°10′38″W 15.XII.2015; A. Rojas-Barrantes.obs.; first observed in Golfito Bay, near Playa Cacao.Commonly seen solitary or foraging with other birds near the shore and river estuaries.Newly recorded from the area Identification.The shape of the bill is distinctive.Juveniles are pale.The long bill is distinctive.Dark colors can set them apart from similar herons like T. mexicanum.