Marine Flora of French Polynesia: An Updated List Using DNA Barcoding and Traditional Approaches

Simple Summary The French Polynesian islands represent a unique insular system in the Pacific Ocean. Previous surveys of the marine flora of French Polynesia were mostly established on traditional morphology-based taxonomy. DNA barcoding allowed us to provide a major revision of French Polynesian marine flora, with an updated total of 702 species from French Polynesia, including 119 species of Chlorophyta, 169 Cyanobacteria, 92 Ochrophyta, 320 Rhodophyta, and 2 species of seagrasses (Alismatales)—nearly a two-fold increase from previous estimates. In addition to improving and refining our knowledge of French Polynesian marine flora, this study also provides a valuable DNA barcode reference library for identification purposes and future taxonomic and conservation studies. A significant part of the diversity uncovered corresponds to unidentified lineages, which will need careful taxonomic examination. Abstract Located in the heart of the South Pacific Ocean, the French Polynesian islands represent a remarkable setting for biological colonization and diversification, because of their isolation. Our knowledge of this region’s biodiversity is nevertheless still incomplete for many groups of organisms. In the late 1990s and 2000s, a series of publications provided the first checklists of French Polynesian marine algae, including the Chlorophyta, Rhodophyta, Ochrophyta, and Cyanobacteria, established mostly on traditional morphology-based taxonomy. We initiated a project to systematically DNA barcode the marine flora of French Polynesia. Based on a large collection of ~2452 specimens, made between 2014 and 2023, across the five French Polynesian archipelagos, we re-assessed the marine floral species diversity (Alismatales, Cyanobacteria, Rhodophyta, Ochrophyta, Chlorophyta) using DNA barcoding in concert with morphology-based classification. We provide here a major revision of French Polynesian marine flora, with an updated listing of 702 species including 119 Chlorophyta, 169 Cyanobacteria, 92 Ochrophyta, 320 Rhodophyta, and 2 seagrass species—nearly a two-fold increase from previous estimates. This study significantly improves our knowledge of French Polynesian marine diversity and provides a valuable DNA barcode reference library for identification purposes and future taxonomic and conservation studies. A significant part of the diversity uncovered from French Polynesia corresponds to unidentified lineages, which will require careful future taxonomic investigation.


French Polynesia
French Polynesia consists of strings of islands positioned in the heart of the South Pacific Ocean, arranged into five archipelagos: the Society, Marquesas, Austral, Gambier, and Tuamotu Islands [1] (Figure 1). French Polynesian island chains resulted from linear volcanic hotspot activities on the Pacific plate that began during the Eocene (47.4 Ma, Tuamotu archipelago [2]; but see [3]). The five archipelagos feature some 120 emerged islands and more than 500 seamounts [4], dispersed over an area of 2,500,000 km 2 [5], forming a total land area of 3521 km 2 with a combined coastline of 5830 km [6]. French Polynesian emerged islands present a variety of stages of the tropical volcanic island life cycle, ranging from high islands (such as Tahiti and Moorea) to atolls (such as Rangiroa and Tikehau) and guyot (or seamount), with variable climates linked to their geographical position. French Polynesia is the world's largest contiguous exclusive economic zone (EEZ) with a total of 5,030,000 km 2 [7].

Relatively Poor Biodiversity
Chiefly imputable on the insularity, remoteness, and smallness of its islands, French Polynesia supports a relatively low terrestrial and marine biodiversity [8][9][10][11]. Some authors noted that due to geographic isolation, some algal families were poorly represented in French Polynesia [12]. Contrasting to the absence of particular families, levels of endemism in some seaweed groups can be remarkably high [9,13]. In comparison to their terrestrial counterpart (893 indigenous plant species and 1700 introduced species [9]), the marine flora biodiversity is significantly poorer. The latest published works on French Polynesian marine flora reported a total of 332 species of marine macroalgae (198 Rhodophyta, 32 Ochrophyta, 82 Chlorophyta [14][15][16]), 117 Cyanobacteria [17], and 2 seagrasses [18]. Since then, only a limited number of taxonomic studies have been conducted in French Polynesia, e.g., [13,[19][20][21].
The same year, Payri and Meinesz [48,49] listed 346 taxa of French Polynesian algae, followed by Abbott [50], who reported on a new species of Valoniopsis from Huahine Island. The majority of earlier works on French Polynesian marine plants were regrouped in a comprehensive checklist by Payri and N'Yeurt [17], which also included species from Moorea Island reported by Payri [51] in her doctoral thesis dissertation. A richly illustrated field guide to French Polynesian marine flora directed at the general public was published by Payri, N'Yeurt, and Orempüller [52]. At this time, the University of French Polynesia's phycological herbarium (officially recognized as UPF) was created by Antoine De Ramon N'Yeurt and Claude Payri to house the growing number of algal specimens stemming from ongoing work on French Polynesian flora. These efforts culminated in the publication of a three-part scientific marine flora for French Polynesia, detailing the Ochrophyta, Chlorophyta, and Rhodophyta [14][15][16]. Three new species of Rhodophyta from the Marquesas and Tuamotu archipelagos resulting from this work were published in [53]. The number of endemic species identified from the region, prior to molecular studies, appears remarkably low.

DNA Barcoding Approach
There are several challenges to a DNA barcoding approach that have been discussed in [80], including the completeness and reliability of the reference library (e.g., non-and mis-labeled sequences) and the limited resolution of some markers. Gene libraries are often biased towards temperate regions, potentially neglecting tropical biodiversity in poorly sampled areas. There is an extensive mislabeling on GenBank that results in numerous polytomic species names. It is, therefore, a priority to augment the available GenBank records with high-quality sequences from all biogeographic areas, providing as much metadata as possible along with the sequences, while avoiding assigning names to submissions that do not align with any known sequences.

Objective of the Study
The present study aimed to re-assess the species diversity of French Polynesian marine flora using DNA barcoding and to deliver a revised checklist for the marine flora of French Polynesia including three macroalgal classes (Rhodophyta, Ochrophyta, Chlorophyta), the Cyanobacteria (Cyanobacteria), and the seagrasses (Alismatales).

Study Site and Sampling
This study was conducted between 2014 and 2023. Marine floral specimens were collected in the subtidal zone via SCUBA or snorkeling and in the shallow intertidal zone on foot and through wading or snorkeling. Specimens identified by Antoine De Ramon N'Yeurt using morphological methods while working with the Moorea BIOCODE Project in 2008 were also included in the present study.
Specimens were initially identified at the genus level and, when possible, at the species level based on the morphology. In situ and ex situ photographs were taken of most specimens collected. After field collection, for each taxa, specimens were (1) mounted as herbarium vouchers, (2) fixed in formaldehyde (4% formaldehyde in seawater) for anatomical examination, (3) and small fragments (~1 cm 2 ) were preserved either in fine silica gel (~0.2-1 mm; ref. 1.01905.1000; Merck KGaA, Damstadt, Germany) or in ethanol (absolute EtOH) for molecular analyses. For the macroalgae, morphospecies determination was done using monographs and identification manuals from the region [14][15][16] as well as based on the most recent publications for the identified taxa. Detailed morphological and anatomical examinations and final identifications were carried out at the University of French Polynesia using a binocular and a light microscope D2000 (Leica Microsystems, Wetzlar, Germany), equipped with a Canon EOS 600D. For the Cyanobacteria, morphological observations consisted of microscopic, ultrastructural, and morphometric analyses. Fresh samples of Cyanobacteria were examined immediately following collection using a light microscope DMZ50 (Leica Microsystems, Wetzlar, Germany) to locate representative subsamples and taxonomically uniform colonies. Fresh and semi-permanent slides were prepared. Observations and measurements were made using a light microscope D2000 (Leica Microsystems, Wetzlar, Germany), equipped with a Canon EOS 600D. Measurements were carried out with Sigma-Scan Image analytical software (Sausalito, CA, USA) and Motic Images Plus (Motic Group, Hong Kong, China), using a calibrated ocular micrometer and in-scale projections and photomicrographs. Phenotype determination was performed using the available monographs and identification manuals (see [94]) as well as more recent publications, e.g., [95][96][97][98][99][100]. Some taxonomic revisions with designation changes were introduced following our phylogenetic reconstructions. A voucher number was assigned to each sample together with the date of collection and deposited into the Phycological Herbarium of the University of French Polynesia (UPF). A total of 2435 (1823 macroalgae + 612 Cyanobacteria + 10 seagrasses) specimens were collected (Supplementary Table S1).

DNA Extraction, PCR Amplification and Sequencing
Total genomic DNA was extracted from tissue samples dried in silica gel using either (1) a cetyl-trimethyl ammonium bromide extraction method [101] or (2)  For the CTAB protocol, the silica-dried portion was ground in liquid nitrogen using a mortar and pestle. Extraction of total genomic DNA was carried out using the protocol from OmniPrep for plant tissue (G-Biosciences, St. Louis, MO, USA). For the MagPurix protocol, algal material was directly processed, without grinding, within PLA buffer, and incubated for 4 h at 60 • C. The MagPurix DNA Extraction Kit was run on the MagPurix 12A automated nucleic acid extraction system (Zinexts Life Science Corporation, New Taipei City, Taiwan). The DNA extract (final volume of 100 µL) was stored at −24 • C.

Sequence Alignment, Phylogenetic Reconstruction, Molecular-Assisted Taxonomic Identification
Nucleotide sequences newly generated were firstly BLASTed against the genomic BLAST database on NCBI BLAST (Basic Local Alignment Search Tool). BLASTing results allowed for the confirmation that newly generated sequences originated from algal DNA material (and not from microbial contaminants or epiphytes) and were more or less in line with preliminary field identification. All nucleotide sequences available on NCBI GenBank were downloaded for the corresponding marker and taxon, either at the generic level or at a higher taxonomic level (e.g., family level), if doubt on the generic identification existed. Several names assigned to sequences deposited on GenBank are inaccurate or were deposited in GB with a preliminary/different species label prior to final publication) and have not been updated since their publications. Sequences names from GenBank were therefore curated whenever possible based on the most up-to-date taxonomic information available on AlgaeBase (e.g., through updating sequences names with currently accepted species names) and the latest molecular taxonomic studies (e.g., those that delivered amendments on species names assigned to lineages). For large datasets (e.g., greater than 200 sequences) we kept a single representative sequence per haplotype, defined as >99.9% similar, using the CD-HIT program [118] run on a local computer.
Nucleotide sequences newly generated in this study were added to the GenBank downloaded sequence datasets and aligned using MUSCLE v.3.5 [119], with default parameters implemented in the eBioX software package v.1.5.1 [120].
Maximum-likelihood phylogenetic trees were reconstructed from each marker and each class using a best fit substitution model and an SPR branch swapping algorithm in PhyML v.3.0 [121].
Molecular-assisted taxonomic identification was based on placement of newly generated sequences within the phylogenetic trees and sequence similarity. Whenever newly generated sequences did not position clearly within a clade (i.e., a cluster of sequences corresponding to a given species) and diverged from the closest sequences by >1%, these lineages were considered as unidentified species and given the specific epithet "sp.#FP". Finally, we returned to the morphological observations to ensure that molecular-assisted taxonomic identification corresponded to morphological data, at least at the genus level.

Species Identification
Based on molecular analyses, we have identified a total of 352 lineages/species: 1 Alismatales, 53 Cyanobacteria, 150 Rhodophyta, 77 Ochrophyta, and 71 Chlorophyta. We could confirm a total of 80 names from previous checklists (7 Cyanobacteria, 26 Rhodophyta, 14 Ochrophyta, 33 Chlorophyta). Our molecular phylogenetic study disclosed a total of 227 new lineages/species for French Polynesia: 115 Rhodophyta, 30 Chlorophyta, 61 Ochrophyta, 20 Cyanobacteria, 1 Alismatales. Among these 227 new lineages, 75 matched a name through the barcoding approach, and 152 did not (i.e., unidentified species) (113 Rhodophyta, 16 Chlorophyta, 23 Ochrophyta, 37 Cyanobacteria). Lineages labelled as "sp.#FP" diverged from the closest sequences from GenBank by >1%. It should be noted that these lineages do not necessarily correspond to new species, but from a barcoding approach, it strictly implies these sequences do not match sequenced species for these particular markers. Further analyses are needed to determine if these lineages should be defined as new species. Also, for those that match a current name, one needs to verify if the identifications of the names in GenBank are accurately based on type specimens or on secondary collections that could have been misidentified. Hence, these numbers need to be taken with caution at this time. We did not get molecular confirmation for 356 names from the previous checklist (170 Rhodophyta, 48 Chlorophyta, 15 Ochrophyta, 115 Cyanobacteria).

Discussion
This study aimed to revisit the marine floral biodiversity of French Polynesia through DNA barcoding. We discuss the present marine floral biodiversity findings and the value, utility, and challenges of barcoding an entire regional marine flora. Some of the species collected have been illustrated in Figure 2.

French Polynesian Marine Floral Species Diversity
Until the last update on regional flora from French Polynesia by N'Yeurt and Payri [16], the marine floral diversity consisted of 430 species: 198 Rhodophyceae, 32 Ochrophyta, 83 Chlorophyceae, and 117 Cyanobacteria. As mentioned earlier, a limited number of molecular-based studies have been conducted in French Polynesia in the past. Taking into account the diversity documented in past works on the French Polynesian flora and the diversity uncovered in the current study, we provide the numbers of 702 species, 320 Rhodophyta, 119 Chlorophyta, 92 Ochrophyta, and 169 Cyanobacteria. Moreover, it is very likely that an important fraction of the names previously documented and not verified with the barcoding approach are misapplied names due to the limitations of traditional methods in resolving morphologically identical species. Previous molecularbased studies have already confirmed this with some pan-tropical genera: Lobophora [13], Gibsmithia [55], and Sargassum [74]. The case of the genus Lobophora perhaps illustrates this best; previous works provided the widely-distributed name Lobophora variegata, which in fact corresponds to 37 different pseudo-cryptic species that would have been very difficult to discern using traditional techniques [123]. At the family level, e.g., Scytosiphonaceae, among the six previously identified names, four were confirmed via barcoding, six were new, and another two have not been confirmed via barcoding, leaving their previous identification as disputable.
From our study (Table 1), the level of marine floral species endemism is 11% (only including currently accepted names) and extends to 28% when accounting for all new lineages (i.e., unidentified species). These results, however, need to be taken with caution, as floristic species richness and single-island endemic species richness can be unknown, with dynamic figures dependent on sampling effort [124]. This is especially true of French Polynesia, with its vast geographical area and diversity of habitats.

French Polynesian Marine Floral Species Diversity
Until the last update on regional flora from French Polynesia by N'Yeurt and Payri [16], the marine floral diversity consisted of 430 species: 198 Rhodophyceae, 32 Ochrophyta, 83 Chlorophyceae, and 117 Cyanobacteria. As mentioned earlier, a limited number of molecular-based studies have been conducted in French Polynesia in the past. Taking into account the diversity documented in past works on the French Polynesian flora and the diversity uncovered in the current study, we provide the numbers of 702 species, 320 Rhodophyta, 119 Chlorophyta, 92 Ochrophyta, and 169 Cyanobacteria. Moreover, it is very likely that an important fraction of the names previously documented and not verified with the barcoding approach are misapplied names due to the limitations of traditional methods in resolving morphologically identical species. Previous molecular-based studies have already confirmed this with some pan-tropical genera: Lobophora [13], Gibsmithia [55], and Sargassum [74]. The case of the genus Lobophora perhaps illustrates this

Matching Barcode Data to Morphological-Based Identifications
Matching sequences to previously identified names turned out to be more challenging than expected. We were able to match sequences onto only 80 names (~18.6%) of the 430 names that were previously identified based on morphological data only. In the process, we unveiled a large potentially unknown species diversity (227 new lineages/species), including 75 previously described species and 152 unknown lineages. Accordingly, important taxonomic work remains to be conducted on French Polynesian marine flora to confirm if these new lineages and species reported through barcoding represent actual new taxa. The Rhodophyta in particular will require extensive studies, with no less than 113 unmatched lineages, i.e., not closely matching sequences from GenBank for our reference markers. This work would be all the more important considering that biogeographically speaking, species diversity generally decreases as one moves east from the Indo-Pacific centers of distribution [13,128,129].

Challenges of Barcoding an Entire Marine Flora
Barcoding an entire marine flora can be a complex and challenging task, with several obstacles to take into account. Difficulties include the availability of a comprehensive barcode reference library, a lack of new specimens for rare or seasonal species, a lack of taxonomic expertise, and difficulties in the amplification and matching of barcode data [130]. Based on our own study, we have highlighted some of the major challenges we have faced when barcoding an entire flora that one should be aware of when conducting such studies. These difficulties are inherent to (1) data collection, (2) molecular analyses, and (3) DNA-based identification.

Data Collection Challenges: Spatial Coverage
Marine flora are found in a wide range of habitats, from the intertidal zone to the deep ocean. Conducting a survey that covers all of these habitats can be logistically difficult and time consuming. Considering that French Polynesia consists of some 120 islands with a coastline of 5830 km dispersed over an area of 2,500,000 km 2 , comprehensive sampling covering the whole region is a daunting and unmanageable task. Moreover, some species are locally very limited in geographic distribution (e.g., Dasya palmatifida from Afaahiti in Tahiti; Stypopodium australasicum from Rapa Island) and, consequently, inconvenient to sample. In our study, we were able to sample a total of 12 islands (c. 10% of all islands). While our study covered intertidal to subtidal areas (down to 60 m depth), sampling was chiefly conducted in shallow depths. We are therefore missing the biodiversity below these depths (e.g., mesophotic depths), which, as illustrated in recent studies, may vary from that of shallower depths (e.g., [131]); also, microhabitats were not sampled. It is evident that we are still missing an important fraction of the total algal diversity from French Polynesia.

Data Collection Challenges: Seasonal Coverage
The abundance and distribution of seaweeds can vary seasonally, depending on factors such as temperature, light, and nutrient availability. Conducting a survey over multiple seasons can help to account for this variability but can also be logistically challenging. Considering that there are two main seasons in French Polynesia (hot and humid from November to March, and drier and cooler from April to October), sampling efforts would need to be doubled, without counting the flora that thrive during inter-seasons. This is especially true for genera of brown algae, for instance, Colpomenia, Rosenvingea, Hydroclathrus, and Sargassum [14,132]. Many genera (Hydroclathrus, Rosenvingea, Pseudochnoospora) belonging to the family Scytosiphonaceae are blooming seasonally after the cool season between September and December [51]. The abundance of some Dictyotaceae species varies strongly between seasons, such as the genus Padina, which has the highest cover during the austral summer [19].

Data Collection Challenges: Taxonomic Expertise
Marine algae are a diverse group of organisms, with many different species that can be difficult to distinguish from one another. Taxonomic expertise is required to accurately identify the different species, which can be time consuming and resource intensive. Marine plant taxonomists usually specialize in specific groups and will develop a sharp eye in the field for their group of expertise. So, while a molecular approach will allow us to uncover hidden diversity, sampling will first and foremost demand from collectors a high discriminatory capacity between taxonomic assemblages in the field, and primary field identification and classification of the collected specimens will require a high level of taxonomic expertise from several specialists for each group to meaningfully synthesize morphological and molecular data [133]. Another critical challenge is that the availability of trained morphological taxonomists is in a worldwide decline due to a lack of recognition and funding for this discipline of research [133].

Data Collection Challenges: Sampling Methods
Marine algae are attached to rocks, the substratum, or other organisms, making it difficult to collect samples without damaging the habitat or the flora itself. Additionally, some epiphytic and turf species are microscopic (e.g., Ceramiales, Rhodophyta), rare, or occur in mixed assemblages or low densities, making them difficult to find, sort, and sample.

Molecular Analyses: DNA Quality and Quantity
Obtaining good-quality DNA from all the species in a marine flora can be challenging, as some species may be difficult to collect, rare, or preserved in a way that is not suitable for DNA extraction. We were able to generate 1007 sequences from 2452 specimens (1823 macroalgae, 619 Cyanobacteria, and 10 seagrasses) from which we extracted DNA. Additionally, some species may have a low DNA yield, which can make barcode amplification difficult. 4.3.6. DNA-Based Identification: Markers and Reference Library DNA barcoding approaches rely on accurate species identification, and if the reference database contains errors or incomplete information, it can lead to misidentification. As discussed at length in [80], the current challenge with DNA barcoding is the necessity to have a good reference dataset. The reference database is largely lacking in many groups, since not all known species have been sequenced yet, and not all species have been described, with many regions still hosting novelties that remain to be identified. Another major issue is the choice of markers. Currently, no universal marker is used across and within marine floral phyla. For the Rhodophyta and Phaeophyta, rbcL is by far the best reference marker, while tuf A is preferred for the Chlorophyta [80,111]. Nevertheless, the reference library is far from complete for many taxa within each phylum. For instance, within the Rhodophyta, an rbcL library for the Corallinophycidae is largely lacking, and within this group, cox1 and psbA are preferred markers; similarly, in the Chlorophyta, the 44 tuf A reference library does not include the Cladophorales, for which nuclear markers are used (e.g., 18S, 28S, ITS). In this sense, [80] highlighted the need to complement the algal GenBank reference libraries, for instance, with mitochondrial markers such as cox1 and cox3. As a result, there may be difficulties in identifying and classifying all the species in a given marine flora using a single marker per phylum. The marker selection and reference library completeness and quality are likely the main limitations of a standardized approach to DNA barcoding.

DNA-Based Identification: Variation within Species
While DNA barcoding is highly accurate, there can be significant genetic variability within a species, especially across latitudinal gradients, which can make it difficult to distinguish between closely related species in under-sampled regions [134,135]. This can lead to misidentification or the creation of artificial species groups, if the threshold to separate 'species' is incorrectly applied. This has led some taxonomists to suggest considering species as 'discrete evolutionary units' rather than finite taxa [133].

Cost and Time
Barcoding an entire flora can be a costly and time-consuming process, as each sample must be collected, processed, sequenced, and analyzed. The cost can also vary depending on the sequencing technology used and the number of samples and markers analyzed. We have calculated an average price ranging from USD 50 to 100 per specimen from collection to sequencing.

Utility of DNA Barcoding to Identify Cryptic Species and Ill-Defined Species
DNA barcoding can be especially useful in identifying cryptic and pseudo-cryptic species (i.e., species difficult to distinguish based on morphological characteristics alone) new to science and ill-defined species (e.g., lacking reproductive features to make firm taxonomic decisions). Cryptic species are difficult to find and study in the wild. Through using DNA barcoding, we can identify and distinguish between different species with a high degree of accuracy, even if they have very similar morphological characteristics. Furthermore, DNA barcoding can be used to confirm the identity of species whose names had been previously misapplied.

Conserving the Biodiversity of French Polynesian Marine Flora through DNA Barcoding
As shown in this study, French Polynesia is home to a diverse range of marine plants, including various species of Cyanobacteria, Chlorophyta, Rhodophyta, and Ochrophyta (but only two species of seagrasses). Many of these species are important for ecological reasons, such as providing habitat for other marine organisms, as well as for economic purposes, such as being used for food [136][137][138] and other commercial purposes [139,140]. However, some of these species are threatened by factors such as pollution, coastal development, tourism, and the proliferation of some brown macroalgae (e.g., Dictyotales, Sargassaceae [141]) and Cyanobacteria [20,142]. Marine floral biodiversity is threatened globally, and rapid surveying and monitoring are needed before species are irremediably lost. DNA barcoding can be used to help conserve biodiversity through accurately identifying different species and understanding their genetic relationships [44]. Using DNA barcoding in conjunction with traditional taxonomic expertise and systematic habitat sampling results in a quicker, more precise understanding of species diversity. This knowledge can then be used to inform conservation efforts, such as identifying which species are most in need of protection and which areas are most important for conservation.
In addition to its conservation applications, DNA barcoding can also be used for other purposes, such as detecting the presence of invasive species. The number of alien marine species recorded in the various coastal countries and islands of the world has continued to increase, particularly since the second half of the 20th century, in connection with the increase in maritime traffic and aquaculture exchanges [143]. The morpho-taxonomic methods traditionally employed by many marine monitoring programs are laborious, expensive, and require taxonomic expertise that is often lacking or in decline. In addition, the morpho-taxonomic approach is often unable to detect microscopic invasive species or those at the larval stage, which limits the capacity for eradication, especially at the first stage of introduction. The recent development of molecular tools (e.g., metabarcoding), in particular high-throughput DNA sequencing such as environmental DNA monitoring [143], offers enormous advantages in marine monitoring due to its dual capacity to detect invasive species at any stage of development, while producing a comprehensive and holistic view of all biological communities present in any type of environmental sample (water, sediment, biofilm/biofouling [144,145]). However, it implies that reference molecular data are available, and its effectiveness depends on the choice of genetic markers used. These techniques are particularly effective at detecting targeted species, including when they are present in low density, and make it possible to massively increase the number of sites studied.
They are thus complementary to traditional techniques [146]. A survey based on eDNA metabarcoding would have revealed 22 species of introduced seaweeds-of which only one species was confirmed with barcoding, Solieria filiformis)-including 21 Rhodophyta and 1 Ochrophyta (Colpomenia sinuosa) [147]. The latter study defined "introduced" as "nonnative" based on the information available on the database WORMS [148] (i.e., to determine a species geographic origin). Nevertheless, comprehensive large-scale phylogeographic analyses are necessary to conclusively determine the natural geographic range of a given species [149]. For instance, [148] listed the cosmopolitan species C. sinuosa as an introduced species in French Polynesia. However, a global phylogeographic study conducted on this taxon showed that it consisted of a species complex with high genetic diversity mainly associated with its geographic distribution [150]. In our study, C. sinuosa belonged to two groups within the C. sinuosa complex, both broadly distributed in the Pacific, thus not supporting the hypothesis of an introduction. On the other hand, another species, Gracilaria caudata, with a clear native range in the Atlantic [151], was identified from Tahiti in our study-but not in [147].

Conclusions
This study demonstrated how DNA barcoding is a powerful tool that can help to document and conserve the biodiversity of marine flora worldwide. DNA barcoding can help to create a more accurate picture of biodiversity and inform conservation efforts to protect local species. Our revised listing of French Polynesian marine flora now contains a total of 670 species, including 2 species of Alismatales, 146 Cyanobacteria, 315 Rhodophyta, 92 Ochrophyta, and 115 Chlorophyta, a nearly two-fold increase from previous estimates. Further systematic studies (both molecular and using traditional taxonomic expertise) will be needed to validate the taxonomic identity of the numerous new molecular lineages reported in this study. The DNA database presented in this study has the potential to serve as a valuable reference library for identification purposes, making a significant contribution to the advancement of molecular taxonomy, ecological research, and biodiversity studies in French Polynesia.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/biology12081124/s1, Table S1: Detailed information of specimens sequenced in this study; Table S2: Detailed information of markers and primers sequence used in this study; Table S3: Detailed information on polymerase chain reaction steps used for each gene studied.

Data Availability Statement:
The authors declare that all data supporting the findings of this study are available within the paper and its Supplementary Materials. Phylogenetic trees generated for taxonomic identification are available upon request.