Abstract
Systematic study of the palm (Arecaceae) flora from the Leeward Antilles marked both Sabal antillensis and Sabal lougheediana as endemic to Curaçao and Bonaire respectively. Although both species are emergent, obvious and charismatic elements of the Antillean flora, they were only described within the last decade. Sabal lougheediana is in demographic crisis as a result of constant ungulate pressure prohibiting seedling establishment, while S. antillensis is thriving as a result of careful management. Given the surprisingly recent discovery of these two species, and the severity of the conservation concern, genomic data can help inform whether these taxa are in fact well-circumscribed species, and help to guide further conservation actions. To investigate genomic diversity, we employed RADSeq data from samples throughout the range of both species. Results show significant genetic distance and fixation between the two taxa as currently circumscribed, as well as reduced genetic diversity and increased inbreeding in both species. Multivariate analysis of genetic distance and Bayesian clustering analysis both show clear and significant separation of the two taxa. Parallel assay of ex situ collections informs how future development of germplasm reserves can help support conservation of both species. We provide recommendations to help conserve these two unique and distinct species.
Apstrakto
Estudio sistemátiko di flora di palma (Arecaceae) di Islanan Abou a marka tantu Sabal antillensis komo Sabal lougheediana komo endémiko na Kòrsou i Boneiru respektivamente. Aunke tur dos espesie ta elementonan emergente, opvio i karismátiko di flora antiano, ta te den e último dékada a deskribí nan. Sabal lougheediana ta den krísis demográfiko komo resultado di preshon konstante di ungulado strobando stablesimentu di mata chikitu, miéntras S. antillensis ta prosperá komo resultado di maneho kuidadoso. Mirando e deskubrimentu resien sorprendente di e dos espesienan akí, i e severidat di e problema di konservashon, datonan genómiko por yuda informá si e táksònnan ei realmente ta espesienan bon definí, i ta yuda guia akshon di konservashon den futuro. Pa investigá diversidat genómiko, nos ta hasi uso di datonan di RADSeq for di muestranan di henter e área di distribushon di ámbos espesie. Resultadonan ta mustra distansia genétiko i fihamentu signifikante entre e dos táksònnan manera ta deskribí nan aktualmente, komo tambe diversidat genétiko redusí i oumento di endogamia den tur dos espesie. Tantu e análisis multivariante di distansia genétiko komo análisis di agrupashon bayesiano ta mustra separashon kla i signifikante di e dos táksònnan. Ensayonan paralelo di kolekshonnan ex situ ta informá kon futuro desaroyo di reserva di germoplasma por yuda sostené konservashon di tur dos espesie. Nos ta ofresé rekomendashon p.a. yuda konservá e dos espesienan úniko i distinto akí.
Abstract
Systematisch onderzoek van de palmflora (Arecaceae) van de Benedenwindse Antillen markeerde zowel Sabal antillensis als Sabal lougheediana als endemisch voor respectievelijk Curaçao en Bonaire. Hoewel beide soorten aspect-bepalende, typische en charismatische elementen van de Antilliaanse flora zijn, zijn ze pas in de afgelopen tien jaar beschreven. Sabal lougheediana verkeert in een demografische crisis als gevolg van de voortdurende druk van grazers die de vestiging van zaailingen verhindert, terwijl S. antillensis gedijt als resultaat van zorgvuldig beheer. Gezien de verrassend recente ontdekking van deze twee soorten, en de ernst van de bezorgdheid over het behoud, kunnen genomische gegevens helpen bepalen of deze taxa in feite goed omschreven soorten zijn, en helpen om verdere natuurbehoudsmaatregelen op te zetten. Om de genomische diversiteit te onderzoeken, hebben we RADSeq-gegevens gebruikt van monsters uit het hele verspreidingsgebied van beide soorten. De resultaten laten een aanzienlijke genetische afstand en fixatie zien tussen de twee taxa zoals momenteel omschreven, evenals een verminderde genetische diversiteit en een toegenomen inteelt bij beide soorten. Multivariate analyse van genetische afstand en Bayesiaanse clusteranalyse laten beide een duidelijke en significante scheiding tussen de twee taxa zien. Parallelle tests van ex situ collecties geven aan hoe de toekomstige ontwikkeling van kiemplasmareserves het behoud van beide soorten kan helpen ondersteunen. We geven aanbevelingen om deze twee unieke en verschillende soorten te helpen behouden.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
The Leeward Antillean flora, in the islands running east to west along the southern edge of the Caribbean, includes one confirmed native palm genus, Sabal Adans., with two species present, Sabal antillensis M.P.Griff. and Sabal lougheediana M.P.Griff. & Coolen (Fig. 1). Copernicia tectorum (Kunth) Mart. also occurs on Curaçao but is variously considered either native or introduced (Van Buurt 2009, Van Proosdij 2012). Sabal species on these islands were long known (Arnoldo 1954, 1964, Winkelman 1979, Van Proosdij 2001, 2012) but not given direct taxonomic attention until recently, only being described within the last decade (Griffith et al. 2017, 2019a). The two species show geographic and morphological separation and cohesiveness; S. antillensis, the Curaçao Palm, is only found on the slopes of Christoffelberg and hills to the west of that summit, and shows pronounced pachycauly and pendescent leaflets, while S. lougheediana, the Bonaire Palm, is restricted to a very small area of limestone terrace in southern Bonaire, does not show pronounced pachycauly, and holds leaflets erect; further differences, including habitat, habit, ecology, and anatomy, are detailed in Griffith et al. (2019a). Each species is limited to a small geographic range (Winkelman 1979; De Freitas et al. 2019). Thorough survey of each species in 1978 and 2018 found a robust, healthy demography for Sabal antillensis, with abundant seedlings and subadults (De Freitas et al. 2019), but signs of sustained herbivore pressure in Sabal lougheediana, leading to a mature stand with very few subadults and seedlings to replace senescing palms (cf. Maunder et al. 2002, Klimova et al. 2021). The numbers of mature S. antillensis increased over that 40-year period to well over 1,200 adult palms, whereas the S. lougheediana census fell precipitously from 31 to 25 adult palms (De Freitas et al. 2019). In addition, palms on Bonaire have been overexploited for thatch in earlier times (Winkelman 1979), although this cultural use is no longer an immediate concern (Griffith et al. 2019a).
(A)Sabal lougheediana (Kabana di Boneiru; Bonaire Palm) growing on a limestone terrace in Southeastern Bonaire. (B)Sabal antillensis (Kabana di Korsou; Curaçao Palm), growing on a chert hill in Northwestern Curaçao. Both species are dominant and emergent members of their respective florae and vegetation. Figure adapted with permission from Griffith et al. (2019a)
Given the geographic proximity of these two species, which excludes any other Sabal in these islands, a close relationship has been assumed (Griffith et al. 2019a). The genus includes 17 species from the southern United States through Northern South America. The most geographically proximal Sabal species to the Leeward Antilles is Sabal mauritiiformis (H.Karst.) Griseb. & H.Wendl., found in Venezuela and ranging north and west through Central America and Southern Mexico (Henderson et al. 1995). Beyond S. mauritiiformis, the nearest congener is S. causiarum (O.F.Cook) Becc., found in the Greater Antilles (Henderson et al. 1995). Sabal palmetto (Walter) Lodd. ex Schult. & Schult.f. is also used in civic landscaping on Curaçao, and could pose an invasive risk, or a vector for pests.
Conservation concern for the Leeward Antillean palm species has been put forward (De Freitas et al. 2019), even prior to these plants being determined to species level; Winkelman et al. (1979) and De Freitas et al. (2005) noted that the “Sabal spec.” on each island needed protection. The major herbivores are feral goats and donkeys, which consume entire seedlings and young palms. De Freitas et al. (2019) demonstrated the very positive effects of removing and excluding these herbivores from the palms’ habitat in Curaçao, and called for similar action in Bonaire. As a response, an emergency exclusion fence on Bonaire was completed in 2022 (Fig. 2), and this management effort is currently being expanded to encompass the native range of the Bonaire Palm. These efforts are now part of a broader official framework for resource conservation, cultural conservation and sustainable development for southern Bonaire (Engel et al. 2022). Additionally, ex situ living collections were developed to complement these in situ protections, to act as a germplasm reserve in case of pathogens, saltwater intrusion (ibid.), or other factors causing catastrophic losses.
Exclusion fence on Bonaire; outside of fence on left, inside on right. Overgrazing by introduced ungulates has limited the reproduction and recruitment of Sabal lougheediana. With the fence completed in 2022, this photo shows vegetation recovery underway. Survey in June 2023 located abundant one- and two-year seedlings within the exclosure
Related to these conservation actions are two questions. First, are these two species genomically distinct? Conservation investment in Sabal lougheediana in particular was greatly facilitated after its description as a unique and separate species, distinct from S. antillensis. Although we take the position that all native palm populations are worth conserving, our experience shows that confirmation or refutation of species status has bearing on the degree of effort put forward for protecting plant resources (cf. Morrison, 2009). While taxonomic descriptions are almost always morphologically-based (see examples in Article 38, International Code of Botanical Nomenclature; Turland et al. 2018), molecular data can confirm these descriptive classifications (Flores-Olvera et al. 2016). The fact that these two charismatic megaflora (cf. Crane 2015) species evaded formal description until so recently prompts questions about synonomy that can be addressed through molecular means. Discussion with local and international stakeholders highlighted a desire to evaluate these species boundaries through DNA data. So, here we test the hypothesis that Sabal antillensis and S. lougheediana are genomically distinct, as this has bearing on the degree of resources mobilized for their protection.
Second, are ex situ conservation efforts adequately representing the genomic diversity of each species? A revolution in rethinking and assaying botanic garden living collections for conservation genetic value is underway (e.g. Cibrian et al. 2013, Hoban et al. 2020), and genomic techniques can inform how well ex situ palm collections can carry forward the limited genetic diversity found in small populations (Diaz-Martin et al. 2023). Our study therefore aims to also test the hypothesis that ex situ collections of each species adequately represent in situ diversity. To answer both of these questions, we have gathered and genotyped in situ and ex situ population samples of the native palms from Bonaire and Curaçao.
Materials and methods
Sampling
In situ samples were collected in September 2021 (Bonaire) and January 2022 (Curaçao). Plants were collected from throughout the geographic range of both species (De Freitas et al. 2019), and involved removing up to 5 leaflets per adult plant, cutting these into short sections, and drying and storing at room temperature on silica gel. We selected plants for sampling based on geography, seeking to sample from all edges of the range as well as interior points (see maps on Fig. 3). Based on sampling guidelines from Nazareno et al. (2017), 44 in situ plants from 5 sites throughout the range of S. antillensis on Curaçao were used in the study, along with 18 in situ plants of S. lougheediana from two sites on Bonaire (Fig. 3); the main population of S. lougheediana occurs at Lima, and a single relict palm remains at Belnem. Seven ex situ samples came from plants growing at Montgomery Botanical Center derived from seeds collected on each island in 2017 and 2018. Each of the seven ex situ samples used in the study came from a separate maternal line (Griffith et al. 2020). Table 1 provides an overview of the sampling structure, and Appendix 1 provides details on the plants used in the study.
DNA extraction, library preparation and sequencing
DNA extraction, library preparation and sequencing were carried out by FLORAGENEX, INC (Oregon, USA). In summary DNA was extracted from fresh silica dried leaf material, quantified via a Qubit Quant iT dsDNA HS Assay system (Invitrogen), and then normalized to 500ng of purified DNA for restriction site associated DNA sequencing (RADseq sensu Flanagan and Jones 2018) using the sbfI restriction enzyme and then sequenced using the Illumina HiSeq 4000 genomic sequencer for 150 single end run/low density scan.
Assembly of RADseq data
First, Illumina reads were assessed for quality using FastQC 0.12.1 (Andrews et al. 2014). Quality checking, filtering and de novo assembly of the single-end reads was performed using ipyrad 0.9.87 (Eaton and Overcast 2020), in ipyrad all parameters were set to default unless stated otherwise. The initial filtering of data and quality control of data in ipyrad was completed as follows; filtering for adaptor sequences were done strictly (filter_adapters = 2), then only reads were retained with a strict minimum phred quality score (phred_Qscore_offset = 43). The first and last 5 bases of all reads were removed (max_low_qual_bases = 5), then reads with less than 50 bp discarded (filter_min_trim_len = 50). For de novo assembly, both minimum depth for statistical and minimum depth for majority-rule base calling was set to default (mindepth_statistical and mindepth_majrule = 6). The maximum cluster depth within samples was set to 10,000 (maxdepth = 10,000), clustering threshold for de novo assembly remained at default (clust_threshold = 0.85), the maximum number of alleles per site in consensus sequences was remained at default (max_alleles_consens = 2), maximum number of uncalled bases (max_alleles_consens) in consensus sequences and maximum heterozygotes in consensus sequences (max_Hs_consens) were both left to default at 0.05. The minimum number of samples per locus was set to 35 (min_samples_locus = 35), so each SNP would be present across a minimum of 35 samples, which corresponded to 50% minimum samples per locus (one sample failed to meet quality threshold for assembly and was discarded) to ensure effective population genotyping with 50% missing data or less (Shafer et al. 2016). The maximum number of SNPs per locus (max_SNPs_locus = 0.2), maximum number of indels per locus (max_Indels_locus = 5) was to set 5 and finally the maximum number heterozygous sites per locus (max_shared_Hs_locus = 0.5) remained at default. Lastly, the first five bases of all loci were trimmed (trim_loci = 5).
Location of sampling sites on Curaçao (left) and Bonaire (right). The Belnem site was known to host Sabal lougheediana in past decades (Winkelman 1979) but the habitat is converted to residential development, and only a single wild palm is extant there. Lagun = two adjacent hilltops, Seru Pasku and Seru Para Mira
Genetic diversity, distance, and multivariate analysis
Descriptive statistics for genomic data were calculated with GenAlEx version 6.51b2 (Peakall and Smouse 2012). Comparative estimates of genetic distance (Nei 1978), analysis of molecular variance (AMOVA), and calculation of pairwise fixation index (Fst) were performed in GenAlEx with 10,000 permutations. Multivariate analysis of genetic distance (Orloci 1978; Huff et al. 1993) were performed in GenAlEx version 6.51b2 (Peakall and Smouse 2012). This PCA-based visualization method follows recent studies in population diversity and species delimitation using genomic data (Sunde et al. 2020, Piñeros et al. 2022), and resolves results similarly to other visualization methods (Azizi et al. 2017). An advantage of this multivariate analysis is that it does not require assignment of samples to a priori groups (such as required for DAPC), and thereby can avoid potential bias (Miller et al. 2020; Thia 2023), especially for species delimitation.
Population genetic structure
STRUCTURE v.2.3.4 (Pritchard et al., 2000) was used to determine the genetic structure and identify the most likely number of distinct genetic groups in the two Sabal species. STRUCTURE used a Bayesian algorithm to cluster samples into K distinct genetic groups by minimizing deviations from Hardy–Weinberg and linkage equilibrium within each cluster for K = 1–5 using 1,000,000 Markov chain Monte Carlo (MCMC) iterations after a burnin of 100,000 steps. Each analysis was repeated 60 times for each consecutive value of K. Results from STRUCTURE were then visualized using StructureSelector (Li and Liu 2018).
Assay of ex situ conservation effectiveness
We evaluated genetic distance and Fst values between ex situ collections and their source populations using GenAlEx. To estimate in situ alleles represented in ex situ collections, we calculated in situ private alleles as compared to ex situ collection for each species in GenAlEx, following methods established in Namoff et al. (2010) and Griffith et al. (2015, 2020). This provides an assay of how effective ex situ collections capture and steward diversity present in wild populations (Hoban et al. 2020).
Results
Sequencing and de-novo assembly
Illumina sequencing resulted in a total of 141,632,568 raw single end reads that ranged from 651,045 to 3,056,161 (mean = 65,1045) base pairs per sample. Initial quality control and filtering in ipyrad resulted in a total of 140,968,975 reads passing the filters that ranged from 647,621 to 3,041,828 (mean = 1,985,478.5) filtered reads. De-novo assembly of the reads using ipyrad generated 951,254 total clusters which ranged from 9,054 to 32,752 (mean cluster depth = 13,397.94) sequence clusters per sample, with 7,092 to 22,443 (average cluster depth = 9,707.3) being high depth clusters (defined as containing six or more reads for a minimum clustering depth). Considering only SNPs that were present in at least 50% of all individuals (minimum samples per locus), the final output from ipyrad resulted in 6,634 SNPs (uploaded to NCBI GenBank BioProject ID: PRJNA1073954 see Appendix 1).
Genetic diversity, distance, and multivariate analysis
The mean number of alleles per SNP ranged from 0.945 (Belnem, Bonaire) to 1.413 (Seru Gracia, Curaçao) (Table 2). The Lima population, i.e., the main group on Bonaire, had the greatest number of private alleles (0.153). Observed heterozygosity (Ho) ranged from Ho=0.014 for S. lougheediana in Belnem to Ho=0.083 for S. antillensis at Montgomery (ex situ), with a mean of Ho=0.042. Inbreeding coefficient (Fis) varied from Fis = -1.0 and indicated complete outbreeding in a single plant located at Belnem, Bonaire to Fis = 0.62 in the Viewpoint, Curaçao population, showing high levels of inbreeding. Comparing only the in situ plants between the two species (i.e. between the two islands) AMOVA found that the greatest genomic variation was within populations (76%), with 24% variation between the two species. Considering only in situ plants, the Fst between islands (i.e., between species) was 0.243. Parsed by population, Fst values ranged from 0.050 (between Seru Gracia and Viewpoint, two proximal sites on Curaçao; Fig. 3) to 0.370 (between Seru Boosman on Curaçao and the entire population on Bonaire) (Table 3). Nei’s Genetic Distance values (Table 3) were closest between Seru Gracia and Viewpoint (0.016) and furthest between Bonaire and the population above Lagun (0.122). Multivariate analysis of genetic distance data is shown in Fig. 4. The first two axes account for 18% of the variation in the dataset. The first axis separates the two species into two distinct clusters that correspond to island groups, regardless of in situ or ex situ source. Populations from different hilltop locations in Curaçao are resolvable along the second axis.
Population structure
The results from STRUCTURE show that both the ΔK statistic and parsimony index suggested that the most likely number of genetic groups was K = 2 (ΔK = 8714.3764 (Fig. 5). In situ populations from Curaçao and Bonaire formed separate genetic clusters with some admixture in Curaçao, indicating gene flow from Bonaire to Curaçao (Fig. 5). Ex situ samples also assorted based on island source, with admixture in Curaçao and no evidence of admixture in Bonaire, indicating evidence of gene flow from Bonaire to Curaçao. Overall, these results show a similar pattern found in the multivariate analysis (Fig. 4).
Assay of ex situ conservation effectiveness
Fst and Nei’s Genetic Distance statistics show closest affinity of the ex situ collections to their source populations. The Bonaire ex situ collection’s least distance (0.054) and lowest Fst (0.178) is to the Bonaire population (Table 3). The Curaçao ex situ collection is closest to Seru Bientu (0.028 NGD / 0.077 Fst; Table 3). The 4 ex situ plants (= 4 maternal lines) from Bonaire capture 81% of all alleles from the island. The 3 ex situ plants (= 3 maternal lines) from Curaçao capture only 69% of alleles from the in situ plants on Curaçao, but capture 82% of the alleles from Seru Bientu, the specific source population for the ex situ seeds.
Discussion
The results presented above show both a very significant genomic separation between the two native palm species of the Leeward Antilles, and a close affinity of ex situ collections with their source populations. With the above data in place, we can now discuss genetic diversity of these species and address the two questions asked in the introduction – are these distinct species, and are they well represented in ex situ collections?
Genetic diversity and inbreeding
The genetic diversity measure for both species indicated that both S. lougheediana (Ho =0.014 to 0.028) and S. antillensis (Ho =0.033 to 0.083) show low levels of genetic diversity. It is possible that the low diversity is a consequence of isolation. Placing these results in a context highlights the imperilment of both taxa. For example, expansive Butia eriospatha (Martius ex Drude) Beccari populations in Southern Brazil showed much higher Ho measures (between 0.21 and 0.55), yet were still considered at conservation risk due to limited recruitment (Nazareno and dos Reis 2014). Another Brazilian palm, Acrocomia emensis (Toledo) Lorenzi showed Ho values between 0.13 and 0.92, and was being targeted for conservation action due to limited genetic diversity (Neiva et al. 2016). Finally, Copernicia prunifera (Miller) HE Moore showed Ho values ranging from 0.37 to 0.39 across 14 populations in Northeastern Brazil (Costa et al. 2022). Given that species’ high economic value, it was recommended to conserve these existing populations, and set aside habitat reserves to steward greater genetic diversity. By comparison, these examples serve to illustrate the relatively very reduced genetic diversity of the native Leeward Antillean palms, and highlight the need for conserving these native species.
Structure (Pritchard et al. 2000) histogram of genetic cluster membership generated from analysis of SNPs for K = 2, at which ΔK = 8,714.3764 (plot at lower left)
In S. antillensis there is a difference between the measures of heterozygosity expected and observed, which indicates inbreeding may be playing a part in the population dynamics. For S. lougheediana although the single individual at Belnem shows complete outbreeding, the inbreeding coefficient (FIS) indicates the populations fall closely to being within Hardy-Weinberg equilibrium, whereas populations of S. antillensis show a higher degree of inbreeding (FIS = -0.015 to 0.466). This inbreeding, combined with such low level of allelic diversity indicates a potential bottleneck. This indicated bottleneck is consistent with the known history of the S. antillensis census, which shows a large recovery from a greatly reduced population over the last 40 years (Winkelman 1979; De Freitas et al. 2019). While populations of S. antillensis from Curaçao show minimal admixture with S. lougheediana, the low allelic diversity (≤ 0.028) indicates that any potential gene flow is likely ancestral. While Sabal flowers are well established as entemophilous (Koptur et al. 2020), consistent and strong east-northeast prevailing winds (Chadee and Clarke 2014) could have potentially vectored pollen or pollen-laden insects from Bonaire to Curaçao. With the current genomic evidence suggesting historically larger past palm populations, both the pollen source and destination may have been more amenable to this gene flow than the current situation. It is also possible that the structure results reflect incomplete lineage sorting, as the two species may be closely related. While that hypothesis is outside the scope of the current study, it remains an interesting hypothesis for future testing.
Distinct species endemic to each island
Since these two prominent, highly-visible species evaded formal description until so recently (Griffith et al. 2017, 2019a) discussion with stakeholders prompts questions about synonomy: are these two species genetically distinct? Genomic data presented demonstrate a significant separation between these island taxa, with an Fst of 0.243. This level of fixation is greater than Fst values observed between many conspecific palm groups. For example, comparison among different date palm (Phoenix dactylifera L.) cultivars show Fst values between 0.01 and 0.16 (Gros-Balthazard et al. 2017; Saboori et al. 2021), The Fst value (0.243) reported here for the Leeward Antillean palms is more in line with published comparisons among palm species. For example, Mauritiella Burret species show an Fst values of 0.23 between species (Torres Jiménez et al. 2021), and Phoenix L. species show Fst values between 0.29 and 0.33 (Gros-Balthazard et al. 2017). Within the genus Sabal, observed Fst between S. antillensis and S. lougheediana exceeds values seen between S. palmetto (Walter) Lodd. ex Schult. & Schult.f., and S. etonia Swingle (Fst = 0.195) or between S. palmetto and S. miamiensis Zona (Fst = 0.161) (Grinage et al. in prep.). Fixation values therefore support the two taxa as distinct species. This distinct separation between S. lougheediana and S. antillensis on each respective island is also resolvable in the multivariate analysis in Fig. 4 and the Bayesian clustering in Fig. 5. In Fig. 4, the first axis clearly and significantly separates the two species, and the second axis separates intraspecific diversity. We infer that genetic diversity is higher for Curaçao’s S. antillensis than Bonaire’s S. lougheediana due to the much higher populations sizes on Curaçao and the observed recent bottleneck for S. lougheediana.
We concur with Lesica and Lavin (2023) that molecular data used to inform taxonomy must utilize robust sampling, or otherwise risk destabilizing nomenclature. While Sabal species have seen a number of molecular taxonomic studies (Ramp 1989; Goldman et al. 2011; Heyduk et al. 2016), the current work is the first to explore the Sabal of the Leeward Antilles, employs a thorough sampling (Nazareno et al. 2017), and employs more data than previous work. Results presented here therefore robustly confirm the hypothesis that Sabal antillensis and Sabal lougheediana are genomically distinct, contributing to nomenclatural stability. Such stability is a worthy goal on its own, but also ultimately fosters more effective conservation (Ely et al. 2017).
Ex situ conservation effectiveness
Parallel assays of ex situ collections alongside in situ population diversity studies can help guide and inform actions that can increase conservation effectiveness (Griffith et al. 2019b). The data presented here can help answer the question: are ex situ conservation efforts adequately representing the genetic diversity of each species? Genetic distance and fixation measures (Table 3) show how the ex situ collections share the highest affinity with their respective source populations, giving a first level of confidence in the results. This is also resolvable in the multivariate analysis (Fig. 4) and clustering analysis (Fig. 5). Assay of private alleles shows that these collections capture 81% (Bonaire) or 69% (Curaçao) of the observed in situ alleles. Compared to a widely-used global ex situ plant conservation target of 70% genetic diversity conserved per species (Convention on Biological Diversity 2020; Griffith et al. 2021a), the Bonaire collection meets the goal and the Curaçao collection just barely fails to achieve the goal. However, when compared to just its source population (Seru Bientu), the Curaçao ex situ collection captures 82% of in situ alleles. The tight relationship between the Curaçao collection and the hilltop population it was collected from are also reflected in genetic distance and fixation statistics (Table 3), as well as multivariate analysis (Fig. 4).
Conservation implications and recommendations
Significant conservation actions were set into motion following the description and especially the redlisting of Sabal lougheediana as Critically Endangered (Griffith and Coolen 2021). The results presented here, by confirming the uniqueness of both species, further support and justify those actions, as well as maintenance and expansion of herbivore management and exclusion on both islands (De Freitas et al. 2019; Engel et al. 2022). Plans to develop a Sabalpalm Park to encompass the entire current range of S. lougheediana on Bonaire will complement the protection offered by Christoffelpark which protects S. antillensis on Curaçao. The results also confirm that existing documented ex situ collections at Montgomery Botanical Center (22 maternal lines comprising 127 plants of Sabal antillensis, and 13 maternal lines comprising 149 plants of Sabal lougheediana) form the core of a genetic reserve in the case of unanticipated losses on either site, but also indicate that further development of such collections and metacollections is necessary. Such ex situ reserves should be grown at a number of sites to distribute risk of loss (Griffith et al. 2019b), as these taxa are at very few ex situ sites currently (Griffith et al. 2021b). Targeted efforts to augment the Curaçao ex situ collection with seeds collected from the other hilltop sites in Christoffelpark are recommended, in order to fully represent the breadth of genetic diversity throughout the species (Griffith et al. 2015). Efforts by nurseries on both islands to greatly increase availability of these native species for local horticulture is a critical action to augment and protect these species, both by serving as a directly accessible backup population as well as providing a marketable alternative to the import of nonnative Sabal. Making both species available in the international horticulture trade will also help reduce the threat of poaching (Kay et al. 2011), as evidence for poaching has been observed in the Sabal lougheediana population. Related to the observed gene flow from Bonaire to Curaçao, we also strongly recommend against planting nonnative Sabal species in landscapes on these islands. Besides the risk of inadvertent admixture (Klonner et al. 2017), the high risk of invasion of nonnative Sabal and potential introduction of pests and pathogens are major concerns (Van Buurt and Debrot 2012). The authors have observed nonnative Sabal palmetto seedlings germinating proximally to imported landscape palms at the Curaçao airport, demonstrating a clear potential for invasion. Following that observation, finally, we envision and encourage each island community to broadly and robustly adopt each of these two native palms, respectively, as a mainstay for civic, public, and private landscapes – a vivid, living symbol for native culture and conservation.
Data availability
SNP data was uploaded to NCBI GenBank BioProject ID: PRJNA1073954.
References
Andrews S et al (2014) FastQC. A quality control tool for high throughput sequence data. Babraham Bioinformatics
Arnoldo FM (1954) Zakflora, Wat in Het wild groeit en bloeit op Curaçao, Aruba en Bonaire. Natuurwetenschappelijke Werkgroep Nederlandse Antillen, Curaçao, pp 82–83
Arnoldo FM (1964) Zakflora, Wat in Het wild groeit en bloeit op Curaçao, Aruba en Bonaire. Natuurwetenschappelijke Werkgroep Nederlandse Antillen, Curaçao, p 17
Azizi Z, Moradi Shahrbabak H, Moradi Shahrbabak M (2017) Comparison of PCA and DAPC methods for analysis of Iranian Buffalo population structure using SNPchip90k data. Iran J Anim Sci 48(2):153–161
Chadee XT, Clarke RM (2014) Large-scale wind energy potential of the Caribbean region using near-surface reanalysis data. Renew Sustain Energy Rev 30:45–58
Cibrian-Jaramillo A, Hird A, Oleas N, Ma H, Meerow AW, Francisco-Ortega J, Griffith MP (2013) What is the conservation value of a plant in a botanic garden? Using indicators to improve management of ex situ collections. Bot Rev 79:559–577
Convention on Biological Diversity 2020 Targets Convention on Biological Diversity, Montreal. http://www.cbd.int/gspc/targets.shtml
Costa MF, Morales-Marroquín JA, de Araújo Batista CE, Alves-Pereira A, de Almeida Vieira F, Zucchi MI (2022) Population genomics of the neotropical palm Copernicia prunifera (Miller) HE Moore: implications for conservation. PLoS ONE 17(11):e0276408
Crane P (2015) Can we save the charismatic megaflora? Oryx 49(3):377–378
de Freitas JA, Nijhof BSJ, Rojer AC, Debrot AO (2005) Landscape ecological vegetation map of the island of Bonaire (Southern Caribbean). Royal Netherlands Academy of Arts and Sciences, Amsterdam, 64 pp. & 2 maps
De Freitas J, Camilleri J, van Eijk S, Posno V, Valdes I, Coolen Q, van Blerk J, Griffith MP (2019) Sabalpalm (Sabal Antillensis) recovery over 40 years: lessons for successful palm conservation. Palms 63(2):57–68
Diaz-Martin Z, Fant J, Havens K, Cinea W, Lima JMT, Griffith MP (2023) Current management practices do not adequately safeguard endangered plant species in conservation collections. Biol Conserv 280:109955
Eaton DAR, Overcast I (2020) Ipyrad: interactive assembly and analysis of RADseq datasets. Bioinformatics 36(8):2592–2594
Ely CV, de Loreto Bordignon SA, Trevisan R, Boldrini II (2017) Implications of poor taxonomy in conservation. J Nat Conserv 36:10–13
Engel S, Johnson J, Bertuol P, van der Ploeg J (2022) Bonaire’s Southern Wetlands Management Plan. STINAPA, OLB & Cargill, Kralendijk
Flanagan SP, Jones AG (2018) Substantial differences in bias between single-digest and double-digest RAD-seq libraries: a case study. Mol Ecol Resour 18(2):264280
Flores-Olvera H, Zumaya S, Borsch T (2016) Two new species of Iresine (Amaranthaceae: Gomphrenoideae) from Mexico supported by morphological and molecular characters. Willdenowia 46(1):165–174
Goldman DH, Klooster MR, Griffith MP, Fay MF, Chase MW (2011) A preliminary evaluation of the ancestry of a putative Sabal hybrid (Arecaceae: Coryphoideae), and the description of a new nothospecies, Sabal× brazoriensis, vol 27. Phytotaxa, pp 8–25
Griffith MP, Coolen QT (2021) Sabal Lougheediana. The IUCN Red List of Threatened Species (2021): e.T207984529A207984533
Griffith MP, Calonje M, Meerow AW, Tut F, Kramer AT, Hird A, Magellan TM, Husby CE (2015) Can a botanic garden cycad collection capture the genetic diversity in a wild population? Int J Plant Sci 176(1):1–10
Griffith MP, De Freitas J, Barros M, Noblick LR (2017) Sabal Antillensis (Arecaceae): a new palmetto species from the Leeward Antilles. Phytotaxa 303(1):56–64
Griffith MP, Coolen Q, Barros M, Noblick LR (2019a) Sabal Lougheediana (Arecaceae), a critically endangered, endemic palm species from Bonaire. Phytotaxa 420(2):95–101
Griffith MP, Beckman E, Calicrate T, Clark JR, Clase T, Deans S, Dosmann M, Fant J, Gratacos X, Havens K, Hoban S (2019b) Toward the metacollection: safeguarding plant diversity and coordinating conservation collections. Botanic Gardens Conservation International-US
Griffith MP, Clase T, Toribio P, Piñeyro YE, Jimenez F, Gratacos X, Sanchez V, Meerow A, Meyer A, Kramer A, Fant J (2020) Can a botanic garden metacollection better conserve wild plant diversity? A case study comparing pooled collections with an ideal sampling model. Int J Plant Sci 181(5):485–496
Griffith MP, Cartwright F, Dosmann M, Fant J, Freid E, Havens K, Jestrow B, Kramer AT, Magellan TM, Meerow AW, Meyer A (2021a) Ex situ conservation of large and small plant populations illustrates limitations of common conservation metrics. Int J Plant Sci 182(4):263–276
Griffith MP, Meyer A, Grinage A (2021b) Global ex situ conservation of palms: living treasures for research and education. Front Forests Global Change 4:711414
Gros-Balthazard M, Galimberti M, Kousathanas A, Newton C, Ivorra S, Paradis L, Vigouroux Y, Carter R, Tengberg M, Battesti V, Santoni S (2017) The discovery of wild date palms in Oman reveals a complex domestication history involving centers in the Middle East and Africa. Curr Biol 27(14):2211–2218
Henderson A, Galeano G, Bernal R (1995) Field guide to the palms of the Americas, vol 5390. Princeton University Press
Heyduk K, Trapnell DW, Barrett CF, Leebens-Mack J (2016) Phylogenomic analyses of species relationships in the genus Sabal (Arecaceae) using targeted sequence capture. Biol J Linn Soc 117(1):106–120
Hoban S, Callicrate T, Clark J, Deans S, Dosmann M, Fant J, Gailing O, Havens K, Hipp AL, Kadav P, Kramer AT (2020) Taxonomic similarity does not predict necessary sample size for ex situ conservation: a comparison among five genera. Proceedings of the Royal Society B, 287(1926), p.20200102
Huff DR, Peakall R PE Smouse 1993 RAPD variation within and among natural populations of outcrossing buffalograss Buchloe dactyloides (Nutt) Engelm. Theor Appl Genet 86:927–934
Jiménez MF, Prata E, Zizka A, Cohn-Haft M, de Oliveira AV, Emilio T, Chazot N, Couvreur TL, Kamga SM, Sonké B, Cano Á (2021) Phylogenomics of the palm tribe Lepidocaryeae (Calamoideae: Arecaceae) and description of a new species of Mauritiella. Syst Bot 46(3):863–874.
Kay J, Strader AA, Murphy V, Nghiem-Phu L, Calonje M, Griffith MP (2011) Palma Corcho: a case study in botanic garden conservation horticulture and economics. HortTechnology 21(4):474–481
Klimova A, Ortega-Rubio A, Gutiérrez-Rivera N, Garcillán PP (2021) Genomic insights into the historical population dynamics and spatial differentiation of an endangered island endemic palm, Brahea Edulis. Int J Plant Sci 182(8):695–711
Klonner G, Dullinger I, Wessely J, Bossdorf O, Carboni M, Dawson W, Essl F, Gattringer A, Haeuser E, van Kleunen M, Kreft H (2017) Will climate change increase hybridization risk between potential plant invaders and their congeners in Europe? Divers Distrib 23(8):934–943
Koptur S, Barrios B (2020) Are native palms Pollinator hogs? A field experiment in Pine Rocklands of Southern Florida. Nat Areas J 40(2):142–149
Lesica P, Lavin M (2023) Will molecular phylogenetics help decrease nomenclatural instability? Am J Bot 110(8):e16219
Li YL, Liu JX (2018) StructureSelector: a web based software to select and visualize the optimal number of clusters using multiple methods. Mol Ecol Resour 18:176–177
Maunder M, Page W, Mauremootoo J, Payendee R, Mungroo Y, Maljkovic A, Vericel C, Lyte B (2002) The decline and conservation management of the threatened endemic palms of the Mascarene Islands. Oryx 36(1):56–65
Miller JM, Cullingham CI, Peery RM (2020) The influence of a priori grouping on inference of genetic clusters: simulation study and literature review of the DAPC method. Heredity 125(5):269–280
Morrison III, Lohr WR, Duchen JL, Wilches P, Trujillo R, Mair D, M. and, Renner SS (2009) The impact of taxonomic change on conservation: does it kill, can it save, or is it just irrelevant? Biol Conserv 142(12):3201–3206
Namoff S, Husby CE, Francisco-Ortega J, Noblick LR, Lewis CE, Griffith MP (2010) How well does a botanical garden collection of a rare palm capture the genetic variation in a wild population? Biol Conserv 143(5):1110–1117
Nazareno AG, dos Reis MS (2014) At risk of population decline? An ecological and genetic approach to the threatened palm species Butia Eriospatha (Arecaceae) of Southern Brazil. J Hered 105(1):120–129
Nazareno AG, Bemmels JB, Dick CW, Lohmann LG (2017) Minimum sample sizes for population genomics: an empirical study from an amazonian plant species. Mol Ecol Resour 17(6):1136–1147
Neiva DS, Melo Júnior AF, Oliveira DA, Royo VA, Brandão MM, Menezes EV (2016) Acrocomia emensis (Arecaceae) genetic structure and diversity using SSR molecular markers. Genet Mol Res 15(1):1–11
Orloci L 1978 Multivariate analysis in vegetation research. 2nd ed. Junk, The Hague
Peakall PE R Smouse 2012 GenAlEx 6.5: genetic analysis in Excel: population genetic software for teaching and research—an update. Bioinformatics 28:2537–2539
Piñeros VJ, del Pedraza-Marrón R, Betancourt-Resendes C, Calderón-Cortés I, Betancur-R N, R. and, Domínguez-Domínguez O (2022) Genome-wide species delimitation analyses of a silverside fish species complex in central Mexico indicate taxonomic over-splitting. BMC Ecol Evol 22(1):108
Pritchard JK, Stephens M P Donnelly 2000 inference of population structure using multilocus genotype data. Genetics 155:945–959
Ramp PF (1989) Natural history of Sabal minor: demography, population genetics and reproductive ecology. Tulane University, Graduate Program in Biomedical Sciences
Saboori S, Noormohammadi Z, Sheidai M, Marashi S (2021) Insight into date palm diversity: genetic and morphological investigations, vol 39. Plant Molecular Biology Reporter, pp 137–145
Shafer ABA et al (2016) Bioinformatic processing of RAD-seq data dramatically impacts downstream population genetic inference. Methods Ecol Evol 8:8
Sunde J, Yıldırım Y, Tibblin P, Forsman A (2020) Comparing the performance of microsatellites and RADseq in population genetic studies: analysis of data for pike (Esox lucius) and a synthesis of previous studies. Front Genet 11:218
Thia JA (2023) Guidelines for standardizing the application of discriminant analysis of principal components to genotype data. Mol Ecol Resour 23(3):523–538
Turland NJ, Wiersema JH, Barrie FR, Greuter W, Hawksworth DL, Herendeen PS, Knapp S, Kusber W-H, Li D-Z, Marhold K, May TW, McNeill J, Monro AM, Prado J, Price MJ, Smith GF (eds) 2018: International Code of nomenclature for algae, fungi, and plants (Shenzhen Code) adopted by the Nineteenth International Botanical Congress Shenzhen, China, July 2017. Regnum Vegetabile 159. Glashütten: Koeltz Botanical Books. https://doi.org/10.12705/Code.2018
Van Buurt G (2009) A short natural history of Curaçao. In Crossing Shifting Boundaries, Language and Changing Political status in Aruba, Bonaire and Curaçao. Proceedings of the ECICC Conference, Dominica (Vol. 1, pp. 229–256)
Van Buurt G, Debrot AO (2012) Introduced agricultural pests, plant and animals diseases and vectors in the Dutch Caribbean, with an alert species list (No. C193/11). IMARES
van Proosdij ASJ (2001) Arnoldo’s Zakflora, wat in het wild groeit en bloeit op Aruba, Bonaire en Curaçao, 3rd edition. Walburg Pers, Zutphen
van Proosdij ASJ (2012) Arnoldo’s Zakflora, wat in het wild groeit en bloeit op Aruba, Bonaire en Curaçao, 4th edition. Walburg Pers, Zutphen
Winkelman JE (1979) Inventarisatie Sabal spec. (Cabana) (fam. Palmae) op Curaçao en Bonaire N. A. Thesis. Landbouwhogeschool, Wageningen, 14 pp
Acknowledgements
We thank Carmabi, Section Environment and Nature - Openbaar Lichaam Bonaire, and CARGILL for permission to study, collect and export specimens; USDA for permission to import specimens; Cyrill Kooistra, Caren Eckrich, Peter Montanus, Frank van Slobbe, and Mark Vermeij, for information, advice, guidance, and discussion. We thank Seacology and CARGILL for funds to construct the exclusion fence on Bonaire, and Openbaar Lichaam Bonaire for permission to construct. We thank Johan van Blerk and Tera Barra Foundation for propagating seedlings for the local nursery trade, and Troy Doedens of Coldpalm for making seedlings available internationally to reduce the threat of poaching. Funding for fieldwork and genomic analysis was provided by Montgomery Botanical Center, through the Plant Exploration Fund. We are grateful for two anonymous reviewers who improved the manuscript.
Funding
Funding for fieldwork and genomic analysis was provided by Montgomery Botanical Center, through the Plant Exploration Fund. Funding for bioinformatic analysis was provided by Montgomery Botanical Center through a grant from Tim Gregory.
Author information
Authors and Affiliations
Contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by M.P.G., Q.C., E.K., A.G. A.v.P., and J.C. The first draft of the manuscript was written by M.P.G. and all authors drafted and commented on the final manuscript and the revision. All authors read and approved the final manuscript and the revision.
Corresponding author
Ethics declarations
Competing interests
The authors have no relevant financial or non-financial interests to disclose.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Clugston, J.A.R., Coolen, Q., Houtepen, E. et al. Genomic patterns of native palms from the Leeward Antilles confirm single-island endemism and guide conservation priorities. Conserv Genet (2024). https://doi.org/10.1007/s10592-024-01618-6
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s10592-024-01618-6