Foraminiferal Assemblages As Bioindicators In The Western Caribbean: Albuquerque Cay (Colombia) Open Access

Coastal ecosystems are particularly vulnerable to anthropogenic disturbance. Overfishing, destruction of habitats and marine pollution are only a few examples of how human populations impact sustainability (Jackson et al., 2001; Halpern et al., 2019). For example, tropical reef environments have shown steady and often irreversible deterioration in recent decades. Besides local impacts clearly related to human activity, disease outbreaks and coral bleaching are major concerns (Brown, 1997; Hughes et al., 2003; Hoegh-Guldberg et al., 2007; Jackson et al., 2014). In the Caribbean Realm, numerous studies and governmental initiatives have promoted not only a better understanding of the reef ecosystems, but also their conservation in a close interaction with the local communities (e.g., McClanahan, 1999; Camargo et al., 2009; Bayraktarov et al., 2020).

Thanks to these efforts, and after decades of ecological studies and inter-institutional initiatives, the insular bodies of the San Andrés Archipelago (Colombian Caribbean) were declared by UNESCO as a part of the SeaFlower Biosphere Reserve (SFBR). As the archipelago represents one of the most extensive reef settings in the tropical Atlantic, this area now embraces the largest Marine Protected Area (MPA) in the Caribbean (Geister, 1992; Díaz et al., 2000; Geister & Díaz, 2007; Guarderas et al., 2008; Taylor et al., 2013; Ramirez, 2016). Because the isolation of some of these insular bodies challenges monitoring, their environmental vulnerability has not been well defined (Sánchez et al., 2005). For instance, Hurricane Iota in 2020 caused major environmental and economic impacts in the archipelago (Villegas, 2021; Gómez et al., 2022). In addition, recent environmental assessments in Roncador and Serrana, some of the non-populated cays in the SFBR, have clearly indicated a steady decline in the coralline and benthic assemblages in recent decades (Sánchez et al., 2019). Among the organisms those authors proposed as bioindicators were the benthic foraminifers.

Detailed studies of benthic foraminifera in the Caribbean remark on the high diversity of the reefal foraminiferal assemblages, like those reported in the Western Equatorial Pacific (Murray, 1991, 2006; Hallock, 1999). Therefore, substantial information is available on their distributions and the environmental controls (Brasier, 1975; Culver, 1990, Scott et al., 1990; Donnelly, 1993, Culver & Buzas, 1995; Gischler et al., 2003; Javaux & Scott, 2003; Oliver et al., 2014; among others). Though less intensively studied than other parts of the tropical Atlantic, assemblages have been reported for the southwestern Caribbean, especially the coastal areas of Colombia and Panama (e.g., Parada et al., 1985; Parada & Londoño de Hoyos, 1985; Parada & Pinto, 1986; Parada, 1996; Havach & Collins, 1997; Bernal et al., 2005, 2008; Velásquez et al., 2011; Gómez & Bernal, 2013). Fewer studies have been undertaken in the geographically extensive San Andrés Archipelago, notably in Serranilla and Serrana cays (Bock & Moor, 1971; Triffleman et al., 1991; Peebles et al., 1997), and more recently, Serrana and Roncador (Sanchez et al., 2019). In addition to analyzing the assemblages, Sánchez et al. (2019) applied the FoRAM Index (Foraminifera in Reef Assessment and Monitoring; FI) as a complement of other environmental assessments (coralline communities, echinoids) in both cays.

Since its introduction by Hallock et al. (2003) as a first-order estimate of water quality in western Atlantic and Caribbean reefal environments, the FI has been tested to assess water quality in reefal environments and settings worldwide (Schueth & Frank, 2008; Barbosa et al., 2009; Narayan & Pandolfi, 2010; Uthicke et al., 2010; Natsir & Subkhan, 2011; Carvajal-Chitty & Navarro, 2011; Stephenson et al., 2015; Pisapia et al., 2017; Humphreys et al., 2018). The first application of the FI in the Colombian Caribbean was the study of Velasquez et al. (2011) in three Marine Protected Areas (MPA). Their results were striking considering that the samples inside the MPA suggest a higher decline in comparison with the non-protected sites. Issues like an increase in non-regulated tourism, the proximity to densely populated areas, and the influence of the Magdalena River were considered as possible causes of that damage. The studies of Velasquez et al. (2011), Sanchez et al. (2019), and recently Rodríguez et al. (2023) demonstrated the potential use of the FI for biomonitoring the reef areas in the Southern Caribbean. However, to better interpret FI data, a broader knowledge of the distribution of the foraminiferal assemblages and the ecological constraints that might exist in the study areas is essential. Also, collection methodology is important, considering that not all samples from reefal settings are suitable to assess the FI (Hallock et al., 2012; Prazeres et al., 2020). Therefore, strengthening our baseline knowledge of the Colombian Caribbean and SFBR reefal ecosystems is also essential.

Since 2015, as a part of a combined cooperation of Colombian governmental institutions (Comisión Colombiana del Océano, Dimar, Colombia BIO-Minciencias, among others), local authorities, several universities, and non-governmental organizations, scientific expeditions are being conducted in all the components of the San Andrés Archipelago (Expediciones Científicas SeaFlower). Located in the southwestern part of the SFBR, the isolated Albuquerque Cay was the expedition focus in 2018, and sediment samples were collected to survey foraminiferal assemblages.

The research reported here contributes to the knowledge of the composition of the reef-associated foraminifera of the SFBR and also provides data for FI analyses in this part of the Caribbean. Sediment samples were collected from shoreface and reef-lagoon sites around Albuquerque Cay, and foraminiferal assemblages in those samples were assessed. Previously collected data from Serranilla Cay (Triffleman et al., 1991) and unpublished data from Old Providence Island were also examined for comparative purposes. Using those assemblage data, FI values were calculated and compared with those reported in other cays of the archipelago and the Colombian Caribbean.

The San Andrés Archipelago comprises a group of islands, atolls, and carbonate banks in the western Colombian Caribbean. The islands are aligned in a general NNE direction, along a major submarine feature called the Nicaraguan Rise. In general, the insular bodies of the archipelago share a similar history. As Late Cretaceous volcanic basements subsided, they provided substrata for biogenic carbonates that accumulated since the late Paleogene (Geister & Díaz, 2007; Sánchez et al., 2015; Torrado et al., 2019). Modern oceanic circulation is influenced by the Caribbean Current and the Colombia-Panama Gyre, while the bathymetric configuration of the basement influences the distribution of the intermediate and deep-water masses (Geister & Díaz, 2007; CORALINA-INVEMAR, 2012).

Albuquerque Cay (12°10′N, 81°51′W) is located 35 km southwestern of San Andrés Island, the main insular body of the archipelago (Fig. 1). Including the pre-reefal terrace, this almost circular atoll has a diameter of 8 km and has two sandy cays: North Cay and South Cay. Located in the windward reefal terrace, both cays are protected from swell by a continuous windward reef barrier. In the inner section, the reef lagoon reaches depths around 1–5 m, while the leeward part has average depths of 9–15 m. Like other cays in the archipelago, the ecological reef zones are controlled by wave exposure on both the windward and leeward flanks. The inner lagoon harbors small reef patches (∼25%) of stony corals (Orbicella spp.), while the peripheral reef is composed of Acropora palmata among other stony corals, octocorals, seagrasses, and algae. In the shallower part of the reef lagoon and shoreface, the seafloor is covered by sand and pebbles, ramified octocorals and seagrasses, presenting also hardgrounds in the intertidal zone (Geister & Díaz, 1996, 2007; Gómez-López et al., 2012).

In September 2018, a total of 64 samples (∼25 g each) were collected in the intertidal zones (0.5–1.5 m) of both cays and along the reef lagoon (3–15 m; Fig. 1). The 48 intertidal samples were collected using a basic grab dredge, while the 18 reef-lagoon samples were provided through SCUBA diving by other researchers in the expedition ( Appendix A). Logistical issues prevented fixing the samples with ethanol for rose Bengal staining. Therefore, all the samples were dried following the collection each day. That method allowed an estimation of the symbiont-bearing forms that were alive when collected. Samples were sub-sampled (∼10 g) for granulometric analysis.

Each sample was washed with distilled water using the 63 and 150-μm fraction sieves. The retained material was dried, and around 100–300 specimens per sample were counted to obtain statistically relevant populations (Fatela & Taborda, 2002; Forcino, 2012). We collected the foraminifera of the 150-μm fraction in all the samples, and if possible, we also considered the 63-μm fraction until we had ∼300 specimens in material with good recovery. Therefore, since we considered the 63 and 150-μm fraction sieves, the total counts combined the specimens of both fractions. We are aware that in nearshore material most of the specimens are larger than 150 μm, but we wanted to explore the content of the >63-μm fraction, since focusing only in coarser and medium grain fractions could cause underrepresentation of smaller heterotrophic taxa (e.g., Hallock, 2012). Foraminiferal taxonomy followed local and regional references (e.g., Parada & Pinto, 1986; Triffleman et al., 1991; Bolli et al., 1994; Havach & Collins, 1997; Peebles et al., 1997). To make FI estimations, the assemblages were also considered in the functional group categories defined by Hallock et al. (2003): symbiont-bearing, stress-tolerant, and other small heterotrophic. As noted in previous studies, FI values range between 1–10, defining unfavorable conditions for coral growth (FI <2), likely reefs undergoing environmental change (FI 4–6), and environmental conditions supporting reef growth or recovery (FI >6).

Based on recent observations on use of the FI (Hallock et al., 2012; Prazeres et al., 2020), FI values were only calculated for the reef-lagoon samples. Similar limitations were considered for data from previous studies in the archipelago (Triffleman et al., 1991; Sanchez et al., 2019) and unpublished data from Old Providence Island (Expedición SeaFlower 2019). In addition, an estimate of fragmented foraminiferal tests in each sample was made. Diversity (Shannon-Wiener; H) and evenness (E) calculations were conducted using the PAST software (Hammer et al., 2001). To test similarities between the foraminiferal assemblages from both cays, a permutational multivariate analysis of variance (PERMANOVA) was also conducted, using a Bray-Curtis dissimilarity distance. However, since the picking method differs from previous foraminiferal surveys in the Caribbean, our H and E estimations will be considered as exploratory.

We followed the sedimentological and geomorphologic indications defined for Albuquerque Cay by Martínez-Clavijo et al. (2021). Some of the reef samples of our study (samples labeled as EST) were also assessed by those authors. In addition, we also explored the sedimentological criteria reported by Triffleman et al. (1992) in Serranilla (northern SFBR). To distinguish three common species found in Caribbean reef environments, Archaias angulatus, Amphistegina gibbosa, and Asterigerina carinata, we have abbreviated their genus names as Ar., Am., and As., respectively.

In general, most of the analyzed sediments can be defined as medium-coarse grained biogenic sands, poorly sorted, with a CaCO₃ content >80%, following the report of Martínez-Clavijo et al. (2021;  Appendix A). The coarser material and the least sorting were found in the intertidal samples, while some of the samples along the reef lagoon were defined as muddy biogenic sands (samples CAC-8; EST-22). In addition to benthic foraminiferal tests, the biogenic elements included fragments of mollusks (the dominant components in the samples), corals, echinoid spines, ostracod valves (moderately diverse), and fish remains (otoliths and teeth). Planktic foraminifers (mainly Globigerinoides and Trilobatus) were also observed in low proportions.

The foraminiferal assemblages of Albuquerque are typical of hypersaline and marine lagoon settings, with a dominance (>91% of total assemblages) of porcelaneous and hyaline tests ( Appendices A-C). The proportion of porcelaneous tests distinguished samples collected at the North Cay from most of the samples in the South Cay and the reef lagoon samples (Fig. 2). A total of 27 genera (nine symbiont-bearing genera) and 46 species were identified in Albuquerque (Figs. 3–4). Archaias angulatus dominated (>40% of total assemblages) in the intertidal samples of both cays (Fig. 5) and was abundant in lagoonal samples, essentially co-dominating with Amphistegina. Heterostegina and Asterigerina were more common in the lagoon material (>5% of total assemblages). Though in lower proportions (2–6% of total assemblages), Sorites was also common in the intertidal samples from North and South cay pie charts data (Fig. 5).

Common smaller porcelaneous taxa included Quinqueloculina, Triloculina, Pyrgo, and Spiroloculina, while smaller hyaline taxa were Rotorbinella rosea and Cymbaloporetta squammosa. Among the porcelaneous taxa, Quinqueloculina was the most common, with slightly higher proportions in the North Cay (∼8% of total assemblages). Of the smaller rotaliids, R. rosea was the most abundant with >20% of total assemblages in both the South Cay and the reef lagoon samples. Proportions of C. squammosa were higher in the intertidal samples from both cays. Low proportions (<1% of total assemblages) of Eponides repandus and Cribroelphidium spp. were also present. Agglutinated forms mostly correspond to Textularia oviedoina and Clavulina angularis (<3% of total assemblages), somewhat more common in the lagoon samples (Fig. 5).

Higher diversities (H>2.0) were found in the reef lagoon samples and the western intertidal set of samples of North Cay. In contrast, H<1.5 was recorded for samples from the east flank of both cays and the northeastern part of the atoll (samples CT1-3). The highest diversity estimations (H = 2.4) are found in samples CAC-1 and CAC-2 in the reef lagoon (Fig. 6). Evenness values indicate a contrast between the lower values in North Cay (E<0.3) and those from the South Cay (E = 0.3–0.4). In general, higher evenness values are observed in all the lagoon samples (E>0.4; Fig. 6). A PERMANOVA analysis suggested a significative difference between the three main sample groups (lagoon, North Cay, South Cay). In PERMANOVA tests, any separation between groups is not significant if more than ∼5% of the permuted F-statistics have values greater than that of the observed statistic (Fig. 7; Anderson, 2001). Besides the general overview provided by the PERMANOVA analyses, the composition of some of the analyzed material suggests a closer relationship between the foraminiferal assemblages from the reef lagoon and most of the intertidal samples along the South Cay. In contrast, the intertidal samples from the North Cay indicate a more heterogeneous composition. In most of the analyzed samples, especially in the intertidal material, fragmented specimens were predominantly Archaias. angulatus.

An overview of the samples from Albuquerque Cay indicates a fairly typical Caribbean foraminiferal assemblage. Because different taxa are adapted to microenvironments, some differences can be seen in local foraminiferal assemblages, despite mixing of sediments by waves and currents. Such differences can be seen in the ternary diagram, the pie charts, and the PERMANOVA analysis (Figs. 2, 5, 7). For example, the proportions of common taxa such as Ar. angulatus, R. rosea, and Am. gibbosa indicate that the North and South cays are influenced by somewhat different environmental factors, and assemblages in both cays differ from the reef lagoon.

Archaias angulatus, the most abundant species identified in samples from Albuquerque Cay, is common throughout Caribbean and western Atlantic platforms settings. This species thrives in shallow, relatively low-energy environments, often attached to seagrasses (Culver & Buzas, 1982; Hallock et al., 1986; Martin, 1986; Wilson, 2006). On the isolated, current-swept banks in the SFBR such as Serranilla bank, the Archaias-Asterigerina assemblage was found in more protected sites and in inverse proportion to the R. rosea dominated assemblages characteristic of higher exposure (Triffleman et al., 1991). Thanks to its relatively large size and shell robustness, Ar. angulatus has a high preservation potential (Martin, 1986), and its discoidal shape favors re-mobilization during high-energy events (e.g., storms). These characteristics allow a higher frequency of their shells in shallow and back-reef foraminiferal thanatocoenoses (Li & Jones, 1997; Li et al., 1998; Wilson, 2006). Among the damaged and fragmented tests recorded in the Albuquerque Cay samples, most belonged to Ar. angulatus, with evidence of abrasion, bioerosion, and broken margins (Fig. 4). A similar observation was made in Serranilla Bank sediments, where the tests of Ar. angulatus tests indicated moderate damage (Trifflemann et al., 1991, 1992). Thus, Ar. angulatus, at >50%, is likely over-represented in the thanatocoenoses of the North Cay, resulting from storm-driven, postmortem transport. In comparison, in samples from the reef lagoon and the South Cay, Ar. angulatus tests made up 30–40% of the total assemblages.

Rotorbinella rosea is more common (>20% of total assemblages) in the reef lagoon and the subtidal samples from South Cay (Fig. 5). This species (previously identified as Discorbis rosea, e.g., Triffleman et al., 1991) is a frequent to dominant element in the Caribbean sediments (Culver & Buzas, 1982; Wilson & Ramsook, 2007). Like Ar. angulatus, several studies have observed that R. rosea shells endure abrasion under dynamic environmental conditions (Brasier, 1975; Rose & Lidz, 1977), and its more compact shape enhances preservation among the hyaline assemblages (Wetmore, 1987). The abundances of R. rosea in Albuquerque share similarities with those described in Serranilla, where R. rosea was more abundant in the samples related to the platform margins, decreasing in the western interior and the northwest sector of the bank (Trifflemann et al., 1991). However, the recent geological and ecological contexts of Albuquerque Cay contrast with those from Serranilla and the proximal banks in the northern Nicaraguan Rise (Rosalind, Diriangen, and Bawihka banks), which are subject to extremely strong and seasonal reversal of currents and are characterized by bioerosional sponge-algal benthic communities (Trifflemann et al., 1992; Peebles et al., 1997). The much weaker oceanic currents that influence Albuqueque Cay are related to the Colombia-Panama Gyre (CORALINA-INVEMAR, 2012).

Geomorphological studies at Albuquerque, indicated that the North Cay is completely embedded in the windward lagoon terrace, while the South Cay has a major interaction with the 5–10 m deep lagoon basin and the fore-reef terrace (Geister & Díaz, 2007; Martínez-Clavijo et al., 2021). That location and exposure of the South Cay might explain the similarity in the abundances of R. rosea, and other common intertidal genera (Cymbaloporetta, Asterigerina) of this cay, with the assemblages from the lagoon samples.

After Ar. angulatus and R. rosea, the most common species in the lagoon samples of Albuquerque was Am. gibbosa (Fig. 5). In comparison with its lower abundance in the intertidal samples (7% and 12% of total assemblages in North and South cays, respectively), percentages of Am. gibbosa exceeded 20% of the total assemblages in most of the lagoon samples, reflecting ecological preferences for somewhat deeper conditions, as previously discussed by several authors (e.g., Hallock, 1981, 1999; Hohenegger, 1995, 2004). According to many studies based on living and dead assemblages, Amphistegina spp. occur abundantly in tropical and sub-tropical coral-reef-associated areas worldwide (Brasier, 1975; Havach & Collins, 1997; Hohenegger et al., 1999, Langer & Hottinger, 2000, many others). Considering their abundance in the reef settings and the light dependence of their diatom endosymbionts (Leutenegger, 1984), this genus is a relevant component in the substrate stability and the growth of reefal structures, while in ecological assessments they can be used as a first-order indicator of water quality (e.g., Hallock, 1981, 2012, Langer & Hottinger, 2000; Prazeres et al., 2012). Therefore, the common presence of Am. gibbosa in the reef-lagoon samples from Albuquerque is not only expected for this kind of environment, but it is also a good indicator that water quality remains favorable for them to thrive.

In other areas of the SFBR, Am. gibbosa has been also reported in reef-lagoon settings, specifically in coarse to fine-grained carbonate sands in Serranilla (Am. gibbosa, >20 m water depth; Triffleman et al., 1991), Roncador/Serrana (<5% of total assemblages; Sanchez et al., 2019), and recently in reef-lagoon material of Old Providence (10–20% of total assemblages; Expedición SeaFlower 2019). Moreover, in areas such as the platform of Belize, Am. gibbosa and As. carinata (also a common element in the reef lagoon of Albuquerque) have been also reported (Gischler et al., 2003), while in areas such as Bocas del Toro (Caribbean Panamá), Am. gibbosa is associated with typical reef assemblages (inner to middle shelf, 6–82 m water depth; Havach & Collins, 1997). In the Colombian Caribbean (National Natural Park Corales del Rosario y San Bernando, Cartagena Bay), Amphistegina is common in both the reef lagoon and the fore-reef settings (Parada & Londoño de Hoyos, 1983; Parada & Pinto, 1986; Bernal et al., 2005; Velasquez et al., 2011; Rodriguez et al., 2023).

Other symbiont-bearing taxa in Albuquerque, such As. carinata or Heterostegina (1–4% of total assemblages), are also present in the reef lagoon settings of the western Caribbean (e.g., Parada & Londoño de Hoyos, 1983; Parada & Pinto, 1986; Triffleman et al., 1991; Gischler et al., 2003; Gischler & Möder, 2009; Sánchez et al., 2019; Rodriguez et al., 2023). In contrast, symbiont-bearing taxa such as Sorites, Laevipeneroplis, and Peneroplis were more common in the intertidal material from both cays (1–6% of total assemblages), suggesting their preference by more dynamic energetic environments, as proposed by some authors (e.g., Peneroplis; Culver, 1990). A similar trend was observed with C. squammosa, a common species found in both cays of Albuquerque (6% of total assemblages). Its encrusting preferences and pseudopelagic adaptations allow this species to thrive in the intertidal zone (Brasier, 1975; Steinker & Reiner, 1981). Finally, the abundance of Quinqueloculina (4–8% of total assemblages) and other miliolids in both cays and the reef lagoon was expected for Albuquerque. In general, miliolids such as Quinqueloculina and Triloculina are abundant in carbonate platforms (e.g., Gischler et al., 2003), usually related to low-energy environments (e.g., Lidz & Rose, 1989; Havach & Collins, 1997).

We calculated FI values for the reef-lagoon samples in Albuquerque, comparing them with estimations from previous surveys (published FI data and FI values based on published foraminiferal counts) from Serrana, Roncador, and Serranilla banks (Triffleman et al., 1991; Sánchez et al., 2019;  Appendix B) and from preliminary data we collected from Old Providence Island (northern SFBR). The results are summarized in Figure 8, and some considerations are described below.

While re-mobilization of the material might cause over-representation in the counts of the most common symbiont-bearing taxa of Albuquerque (Archaias, Amphistegina, Asterigerina, Heterostegina, Laevipeneroplis, Sorites), the low abundance (<1.5% of total assemblages) of the stress-tolerant taxa among the foraminiferal assemblages support our FI values that indicate environmental conditions suitable for reef accretion (e.g., Hallock et al., 2003). In general, Clavulina was the most common stress-tolerant genus, while Cribroelphidium and Elphidium were very scarce in the samples. These samples were collected in areas where patch reefs occur (e.g., Geister & Díaz, 2006).

To test the implementation of the FI in other areas of the SFBR, we also considered unpublished data from Old Providence (19 samples; 1.5–11.8 m water depth). A key aspect with this island is its major environmental heterogeneity (mangrove areas, small riverine affluents, rocky cliffs), human presence (around 5000 inhabitants) and therefore, a higher impact of the reef settings (e.g., marine pollution). And while the foraminiferal assemblages from Old Providence resembled those from Albuquerque, with common Archaias, Amphistegina, Asterigerina, Rotorbinella, Sorites, Quinqueloculina, among others, some differences were noted, consistent with the greater environmental heterogeneity (total counts in  Appendix A). For instance, Planogypsina and Borelis were found, as were stress-tolerant Clavulina and Elphidium, and even Ammonia in sample PS12. Thus, the range of FI values for Old Providence (FI = 3–8.4) reflects that variability at the non-reefal end of the spectrum. An additional consideration for future studies is that Old Providence was deeply affected by Hurricane Iota in November 2020 (Villegas, 2021; Gómez et al., 2022).

As noted above, the FI was calculated for the first time in the SFBR by Sánchez et al. (2019) in the Serrana and Roncador cays. The foraminiferal survey (four samples in Roncador, four samples in Serrana) was conducted in 2015 and 2016, respectively. In general, the foraminiferal assemblages in both cays were dominated by small heterotrophic taxa (Discorbis, Nonion, Rotorbinella) and miliolids (Miliolinella, Triloculina), while the symbiont-bearing taxa (Amphistegina, Laevipeneroplis) comprised ∼10% of the total assemblages. Moreover, stress-tolerant genera included Bolivina, Clavulina, and Elphidium (Sánchez et al., 2019). Therefore, the foraminiferal assemblages and the FI estimations indicate a higher environmental stress in Roncador and Serrana (FI = 2–4; Fig. 8). Those lower FI values were also consistent with other evidence such as the widespread decline in the coral cover (e.g., today <21% at Serrana shallow reefs and 35% in 1995), urchin assemblages, and bottom complexity (rugosity).

An additional comparison utilized published data from Serranilla, one of the most isolated cays of the SFBR (Triffleman et al., 1991). The reported foraminiferal assemblages in sediments from Serranilla were similar to those from Albuquerque (e.g., Archaias, Amphistegina, Quinqueloculina, Rotorbinella, among others) but were more dominated by Rotorbinella. As a result, FI values ranged from 2 to 7, reflecting deeper sampling and the influence of current-induced upwelling that suppresses coral-reef development on Serranilla Bank (Fig. 8).

The lagoon material of Albuquerque was provided through SCUBA diving, and some of the samples (samples CAC-1 to CAC-8) were collected during the ecological survey of moderately to well-preserved coral patch reefs. Approximately a quarter of the lagoon bottom is covered with patch reefs, dominated by the star corals Orbicella spp. (formerly considered Montastraea spp.) and the staghorn coral Acropora cervicornis (Díaz et al., 1996; Geister & Díaz, 2007). In contrast, the shallow lagoonal reefs around the cays are characterized by dense stands of the moose-horn coral Acropora palmata and the brain coral Diploria strigosa (Díaz et al., 1996, Geister & Díaz, 2007). Thus, the prevalence of coral and common occurrence of Am. gibbosa across Albuquerque Cay reflect the suitability of water quality to support coral reefs and their potential for recovery after bleaching or storm events.

Although the sediments are similarly medium-coarse grained biogenic sands, some environmental aspects such as the type of substrate and wave exposure vary considerably in the cays of Albuquerque. For example, the substrate in North Cay is related to a more exposed shoreline, with dominance of sands and rubble, moderately developed seagrasses in the east flank, and scattered hardgrounds. The substrate in the South Cay is dominated by sands and rubble in the northwestern flank, next to the 250 m wide channel that separates it from the North Cay (Fig. 1). Moreover, seagrass cover and hardgrounds are more extensive in the South Cay, with restricted ponds in some parts of the southeastern flank. This higher variability in the intertidal ecological niches, and a more effective communication with the reef lagoon settings, might also explain the differences in the evenness and diversity estimations between the two cays and the greater similarity of South Cay assemblages to reef-lagoon assemblages (Figs. 6, 7).

Finally, the foraminiferal assemblages in the reef-lagoon samples might also reflect some methodological biases because sampling took place during surveys with other research goals. For instance, samples CTB1–3 were collected in sandy substrates of the northern reef lagoon on areas chosen for shark observation. The EST samples complemented a series of bathymetrical studies in the cays (results in Martínez-Clavijo et al., 2021). Only the CAC samples were collected as a part of an ecological survey on reef patches with good-moderate conservation.

We recommend expanding the foraminiferal surveys along the SFBR, including unexplored areas such as Courtown (Cayo Bolívar) and Quitasueño cays and heavily populated islands such as San Andrés. The isolation of most of these cays makes them vulnerable not only to natural phenomena (e.g., hurricanes, strong oceanic currents) but also to human-induced problems (e.g., marine pollution, unregulated fisheries, ocean acidification). This holistic approach eventually will lead to suitable comparisons with the previous FI estimations in the Colombian Caribbean coast (Velasquez et al., 2011; Rodríguez et al., 2023). In addition, genetic studies in the living foraminiferal assemblages could be another promissory research line, considering the unique location and ecological particularities of the SFBR. Future studies also should include specific foraminiferal samplings (including rose Bengal staining of the material), preferably along water-depth transects that comprise the range of ecological niches of the cay. In addition, foraminiferal counting of the samples should be also standardized.

Comprising one of most isolated insular bodies the Western Caribbean, the Albuquerque cay (San Andrés Archipelago, Colombia) represented a good opportunity to not only expand the knowledge baseline of the Caribbean foraminiferal assemblages but also to continue testing its reliability as bioindicators in the reefal settings. Results of a foraminiferal survey in this area illustrated normal reef-lagoon and shore-face assemblages dominated by miliolids (Quinqueloculina), symbiont-bearing forms (Archaias, Amphistegina, Asterigerina, Laevipeneroplis, Sorites), and notable heterotrophic taxa (Cymbaloporetta, Rotorbinella). Stress-tolerant taxa (Elphidium, Cribroelphidium) and agglutinated forms (Clavulina, Textularia) were also present in low abundances. Moreover, diversity analysis indicated differences in the composition of the assemblages in both sandy cays and the reef lagoon settings, probably related to depth and hydrodynamic conditions associated with degree of exposure to open Caribbean waters. Finally, calculations of the FoRAM Index (FI) based on assemblage data indicated generally suitable water quality for carbonate production and reef accretion in Albuquerque Cay waters. These FI values were compared with FI calculations based on data from other insular bodies in the area, demonstrating the potential that foraminifers provide as bioindicators in this region of the Caribbean.

This research was possible thanks to the Comisión Colombiana del Oceano (CCO) and the other Colombian governmental institutions, local authorities, universities, and non-governmental organizations that participated in the Expedición Científica SeaFlower 2018. To them (e.g., Juliana Sintura, Daniel Barrios, Lina Barrios, Maria Fernanda Maya, Angie Montoya, Rafael Bortolin†), and the Creole collaborators, we express our gratitude for their enthusiasm and commitment with the future of the archipelago. Logistic support was also provided by the Universidad Industrial de Santander (UIS) and Stratos Consultoría Geológica. G.D.P. thanks to the laboratory staff of itt-Oceaneon, Universidade do Vale do Rio dos Sinos (UNISINOS University) for the SEM images. Pamela Hallock (University of South Florida) and an anonymous reviewer provided valuable comments to improve an earlier version of this study.  Appendices A and  B can be found linked to the online version of this article.

Appendix A. Foraminiferal data of the Albuquerque Cay. Thanatocoenoses and biocoenoses. Type of tests, percentage abundances and pie charts. Diversity indexes, PERMANOVA test, and FI estimation. FI estimations in Albuquerque, Old Providence, and other insular bodies of the San Andres Archipelago (published data of Triffleman et al., 1991; Sánchez et al., 2019).

Appendix B. Foraminiferal data and FI estimations of the Old Providence Island.

Appendix C. Faunal reference list.

Amphistegina gibbosa d’Orbigny, 1839

Archaias angulatus (Fichtel & Moll, 1798)

Articulina spp. d’Orbigny, 1826

Asterigerina carinata d’Orbigny, 1839

Borelis pulchra (d’Orbigny, 1839)

Clavulina angularis d’Orbigny, 1826

Cribroelphidium gunteri (Cole, 1931)

Cribroelphidium poeyanum (d’Orbigny, 1839)

Cymbaloporetta squammosa (d’Orbigny, 1839)

Discorbis spp. Lamarck, 1804

Eggerella bradyi (Cushman, 1911)

Elphidium spp. Montfort, 1808

Eponides repandus (Fichtel & Moll, 1798)

Heterostegina depressa d’Orbigny, 1826

Laevipeneroplis bradyi (Cushman, 1930)

Laevipeneroplis proteus (d’Orbigny, 1839)

Massilina spp. Schlumberger, 1893

Miliolinella subrotunda (Montagu, 1803)

Nodosaria spp. Lamarck, 1816

Parasorites orbitolitoides (Hofker, 1930)

Peneroplis pertusus (Forsskål in Niebuhr, 1775)

Peneroplis planatus (Fichtel & Moll, 1798)

Planogypsina acervalis (Brady, 1884)

Pseudotriloculina kerimbatica (Heron-Allen & Earland, 1915)

Pseudotriloculina lecalvezae (Kaasschieter, 1961)

Pyrgo johnsoni Cushman, 1935

Quinqueloculina agglutinans d’Orbigny, 1839

Quinqueloculina angulostriata Cushman & Valentine, 1930

Quinqueloculina lamarckiana d’Orbigny, 1839

Quinqueloculina poeyana d’Orbigny, 1839

Quinqueloculina philippinensis Cushman, 1921

Reophax spp. Montfort, 1808

Rotorbinella rosea (d’Orbigny in Guérin-Méneville, 1832)

Sorites orbiculus (Forsskål in Niebuhr, 1775)

Sorites variabilis Lacroix, 1940

Spiroloculina antillarum d’Orbigny, 1839

Spiroloculina atlantica Cushman, 1947

Textularia agglutinans d’Orbigny, 1839

Textularia oviedoiana (d’Orbigny, 1839)

Triloculina spp. d’Orbigny, 1826