Metabarcoding outperforms traditional electrofishing in decapod and fish inventories, paving the way for enhanced biodiversity monitoring in the Caribbean

537 Metabarcoding outperforms traditional electrofishing in decapod and fish inventories, paving the way for enhanced biodiversity monitoring in the Caribbean Thomas Baudry1, Valentin Vasselon2, Carine Delaunay1, Alexandre Arqué3, Fabian Rateau4, Géraldine Lala3, Claire Maurice-Madelon5, Frédéric Grandjean1 1 Université de Poitiers, Laboratoire Écologie et Biologie des Interactions, UMR CNRS 7267 Equipe Ecologie Evolution Symbiose, 3 rue Jacques Fort, Poitiers Cedex, France 2 SCIMABIO-Interface, 5 B Rue des 4 Vents, Thonon-les-Bains, France 3 Officedel'EaudeMartinique(ODE),140BoulevarddelaPointedesNègres,Fort-de-France,Martinique(Fr) 4 OfficeFrançaisdelaBiodiversité(OFB),19rueduBelAir,PointeDesgrottes,LesTroisIlets,Martinique(Fr) 5 Directiondel'Environnement,del'AménagementetduLogementdeMartinique(DEAL),PointedeJaham,Schoelcher,Martinique(Fr) Correspondingauthor:ThomasBaudry([email protected]) Copyright: © Thomas Baudry et al. This is an open access article distributed under terms of the Creative Commons Attribution License (Attribution 4.0 International – CC BY 4.0). Research Article Abstract Environmental DNA (eDNA) metabarcoding revolutionized the biodiversity monitoring in aquatic ecosystems, giving access to taxonomic lists in a non-disruptive way. Although the method has limits, such as reduced taxonomic resolution for certain groups and difficulties in estimating species abundance, it has proven its effectiveness in many contexts. In Martinique, a Caribbean island, traditional methods like electrofishing (TEF) are known to be stressful for organisms, non-selective and disruptive for the ecosystem, and have been progressively abandoned for routine monitoring. The aim of this project was to explore the possibility of using the eDNA-based metabarcoding method for the detection of fish and decapods in Martinique streams, by first validating it with TEF. We selected 14 stations, a representative panel of the river diversity, and performed TEF and eDNA-based monitoring to compare both, based on the species richness. Then, from eDNA taxonomic inventories, we assessed the ecological state of the studied stations, using Simpson index and investigated how stations abiotic characteristics shape assemblages. Here, we confirmed the eDNA metabarcoding method is a reliable tool for monitoring fish and decapods, confirming most of the taxa caught by TEF and revealing the presence of additional (native and/or invasive) species. We faced some issues in discriminating some genetically close species (e.g. Sicydium sp.) potentially leading to under-representation in community assemblages, but not in functional diversity. Additional efforts are needed to raise standardized protocols, but we encourage stakeholders to join such an initiative to shed light on the rich biodiversity in sometimes poorly studied regions and to face invasions. Key words: biodiversity hotspot, ecological assessment, environmental DNA, Martinique island, method validation Academic editor: Bernd Hänfling Received: 28 February 2025 Accepted: 25 September 2025 Published: 3 November 2025 Citation: Baudry T, Vasselon V, Delaunay C, Arqué A, Rateau F, Lala G, Maurice-Madelon C, Grandjean F (2025) Metabarcoding outperforms traditional electrofishing in decapod and fish inventories, paving the way for enhanced biodiversity monitoring in the Caribbean. Metabarcoding and Metagenomics 9: e151675. https:// doi.org/10.3897/mbmg.9.151675 Metabarcoding and Metagenomics 9: 537–558 (2025) DOI: 10.3897/mbmg.9.151675 538 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean Introduction Freshwater ecosystems are of high importance, providing habitat for at least 6% of known species (and probably many more to discover) on < 0.8% of the total Earth surface (Michelet 2017). Nevertheless, unsustainable human activity (i.e. use of pesticides, urbanization, dredging and draining, for example) has had a considerable impact on these freshwater ecosystems in recent years, with an estimated loss of 84% of the biodiversity since 1970 and almost one species out of three threatened with extinction, all taxa combined (Magurran 2009; WWF 2020). These losses weaken the environment by affecting the distribution and composition of native communities, sometimes disrupting migratory patterns and associated life cycles (Magurran 2009; Engman and Ramirez 2012). Tropical islands are particularly vulnerable, characterized by low species diversity combined to high endemism (Nivet et al. 2010) on small areas (Myers et al. 2000). For instance, Martinique is a rugged island, located in Lesser Antilles archipelago, presenting a wide variety of landscapes and terrestrial ecosystems with more than 70 permanent rivers (as well as many non-permanent ones and tropical forests wetlands), justifying its place in one the 25 hotspots of biodiversity (Anadón-Irizarry et al. 2012). These disruptions also make the ecological niche much more permeable to invaders, introduced mainly via aquaristic activities and aquaculture, one of the major factors in current biodiversity loss (Gherardi et al. 2008; Nunes et al. 2015; Rodríguez-Barreras et al. 2020). For all these reasons, there is an urgent need for conservation of these freshwater ecosystems, based on environmental protection programs (Flitcroft et al. 2019). However, these programs require a good knowledge of the environment and the distribution of native and native species, which calls for inventories with accurate species identification and assessment (endemic, rare, endangered or invasive). Biological inventories of macro-organisms in aquatic environments were initially based on traditional methods like direct capture, electrofishing, or baited traps, known to be non-selective, time-consuming and particularly disruptive for the environment (Hänfling et al. 2016; Wang et al. 2021), making them increasingly controversial. Traditional electrofishing (TEF) for example, which is largely used in freshwater environment, uses electric currents to temporarily stun fish, making them easier to capture and study (Pusey et al. 1998). While effective, this method can be stressful or even harmful to the organisms and potentially disruptive to the fragile ecosystems, harboring small populations, which researchers seek to understand and protect (Snyder 2003; Dolan and Miranda 2004). Finally, the labor-intensive and time-consuming nature of these inventories limits the scale of their use, especially in larger or more remote areas (Hense et al. 2010; Evans et al. 2017). Despite these drawbacks, traditional methods remain invaluable for certain research aims, particularly when precise biometric data or population densities are needed (Thomsen and Willerslev 2015; Evans et al. 2017). In recent years, freshwater inventories have undergone a revolution with the emergence of monitoring techniques based on the detection of DNA shed by organisms in the water (i.e. skin cells, mucus or feces) dubbed ‘environmental DNA’ (Ficetola et al. 2008). This approach offers the possibility to detect targeted species even with low population density (i.e. invasive and/or rare, endangered endemic species) without the need to observe it, at any stage of 539 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean life, regardless of their size or ecology (i.e. very small or even microscopic, and sometimes cryptic, living underground or in disconnected ditches) (Ficetola et al. 2008; Thomsen and Willerslev 2015). The eDNA-based method rapidly gained traction in the field of biodiversity assessments, representing a promising alternative or complement to traditional methods, thanks to its low disruptiveness, the ease of implementing it on field at large scale and its ability to detect a broad spectrum of species when using metabarcoding (Valentini et al. 2016; Pont et al. 2018; Taberlet et al. 2018). However, while eDNA metabarcoding presents significant opportunities, it is not without challenges and drawbacks. First, there is a need for a resolutive barcode to discriminate taxa. For instance, MiFish, Teleo and 12S-V5 primers (designed respectively by Riaz et al. 2011; Miya et al. 2015; Valentini et al. 2016) all target a quite short fragment (65 - 175 bp) in a highly conserved genetic region (12S rRNA) and mlCOIintF/ jgHCO2198 primers (Leray et al. 2013) target a 313 bp fragment of the cytochrome oxidase sub-unit I (COI). Then, robust reference databases are essential, containing at least genetic sequences for all species within the study area, and additionally, species susceptible to be introduced (Schenekar et al. 2020; Marques et al. 2021). Such reference databases, coupled with a good knowledge on the studied environment and its native communities, are essential for comprehensive and accurate biodiversity assessments (Schenekar et al. 2020; Marques et al. 2021). Moreover, eDNA-based methods now reached high sensitivity yields and controls samples (i.e. for cross-station or lab contaminations) are therefore crucial, in addition to rigorous protocols, to ensure sample integrity, optimization of species discrimination and avoidance of false positives. As eDNA-based methods grow in popularity regarding their operationality for biodiversity assessment, they are more and more integrated into legal monitoring frameworks and decision making (Morisette et al. 2021; Adams et al. 2024; Kelly et al. 2024). Validation through comparisons with traditional methods is crucial to ensure the reliability of eDNA in biodiversity assessments. While eDNA metabarcoding has many advantages, it should complement, not replace, traditional approaches, which provide key context on species abundance, age structure, and health (Evans et al. 2017). By integrating eDNA with traditional techniques, researchers can achieve a more holistic understanding of ecosystems. For instance, eDNA can be used for rapid initial surveys to identify species presence across large areas, while traditional methods can be employed for more detailed studies in key locations (Evans et al. 2017; Baudry et al. 2023). This combined approach allows for both broad-scale biodiversity assessments and in-depth investigations of particular species or habitats. In this study, we investigated the potential of eDNA metabarcoding for freshwater fish and decapods’ long-term monitoring in Martinique, using respectively MiFish (Miya et al. 2015) and MiDeca (Komai et al. 2019) primers. We compared species richness and taxonomic diversity revealed by both TEF and eDNA-based method, across 14 stations selected to ensure representative coverage of all the island’s rivers. We hypothesized that eDNA-based metabarcoding method would reveal species richness similar to - or higher –than TEF, especially by detecting rare, cryptic or elusive taxa. We also expected the Simpson index to be a meaningful indicator of ecological quality, reflecting patterns of species dominance and community homogenization, for instance, in southern sites where we expected lower diversity and a stronger dominance of invasive species. 540 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean Materials and methods Study area and sampling sites Martinique is a Caribbean island of 1128 km2 belonging to Lesser Antilles (14°39'00"N, 61°00'54"W) and dominated by a rainy tropical climate, leading to a vast hydrographic network encompassing 70 main permanent rivers, fed by at least as many tributaries (Baudry et al. 2021). The presence of the Montagne Pelée volcano (1397 m height) in the northern part of the island induces a difference in hydromorphologies, with northern rivers characterized by steep slopes and strong waterflows and inversely, southern rivers being larger and slow-moving. In this study, we sampled 14 locations, spread over the territory, from the north to the south (Fig. 1), for the most accurate overview of the biodiversity in presence. Eight locations were selected in the northern part (CERon, COULeuvre, CARbet, Maison ROUsse, LORrain, Fonds St-Jacques, BASsignac and GUEs) and six in the southern part of the island (FRançois, SainT-Esprit, Petit BOUrg, LOWinsky, MADeleine and DORmante). Each station was sampled in April 2023 by both eDNA filtration method (first, to minimize contamination risks) and TEF, following the protocols described below. At each of the stations, the physico-chemical characteristics (pH, temperature, oxygen concentration and conductivity) (Suppl. material 1: S1) were measured using a Hanna® HI98129 instrument. eDNA sampling Filtration were performed on-site, before TEF as said above, through 0.45 μm nitrocellulose filters (Sartorius® 47 mm diameter), using a hand-operated vacuum pump (NalgeneTM) together with a 1L-filtration unit (NalgeneTM), as described in Baudry et al. (2021). Sampling was carried out along a transect starting from the riverbank outward, or in flowing sections, depending on the rivers considered, with two independent eDNA samples taken per station. Each sample was filtered until clogging occurred, typically between one and 2.5 liters per replicate. They were then removed and placed (folded in quarters) into 1.5 mL tubes filled with 1 mL of absolute (99%) molecular-grade ethanol, using sterile forceps. To avoid potential field cross-contamination, sampling material was decontaminated using 20% bleach and thoroughly rinsed using tap water after each sampling and a blank control sample (1L of distilled water) was done. All eDNA samples were stored in a cooling bag until their return to the laboratory, where they were stored at 4 °C, until eDNA extraction, showing satisfactory yields if processed quickly and ease of use on-field (Renshaw et al. 2015; Majaneva et al. 2018). Traditional Electro-Fishing (TEF) sampling The chosen protocol was adapted from Lefrancois et al. (2024), using a SmithRoot LR-24 backpack electrofisher, set on 250V (direct current) or 500V (pulsed current) (10 to 16 Hz and 15 to 25 A, respectively), depending on conductivity and hydrological characteristics of the stations. Briefly, with reasonable human effort (3 or 4 operators), the objective was to detect the maximal fish and crustacean species richness, following past observations (Lim et al. 2002; Baudry et al. 2024), with a minimal habitat disturbance. Each station was surveyed 541 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean using spot-fishing, a method involving short, targeted electrofishing burst (~15 seconds per spot) over discrete areas (~10 m2), and progressing upstream for approximately fifty meters. If all the species known to be present at the station (based on historical data) were not captured, additional spot-fishing was carried out, targeting certain micro-habitats that were favorable to certain species (for example, stumps and banks for eels). Once caught, the specimens (fish and crustaceans) were directly taken back to the riverbank to be identified, sorted and sampled. These operations must be carried out very efficiently, as the water in the tanks heats up quickly in such a tropical climate, which can lead to significant losses. As the aim here was to verify the species’ presence (not to quantify), we have therefore chosen to handle the individuals as little as possible, taking Figure 1. The hydrological network of Martinique, in Lesser Antilles, with the location of the 14 stations sampled during this study, highlighted with their code preceded by their location (Nfor northern part and Sfor southern part). 542 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean no biometric measurements (except for taxa of interest such as the American eel - Anguilla rostrata). Individuals were therefore identified visually, down to species level (or genus level for juvenile individuals). A maximum of 5 individuals per species and per station were non-destructively sampled, by taking mucus or a fin fragment from fish, or by taking a P4 leg fragment from crustaceans. This was intended to, first, confirm the species identification and then, contribute to complete the genetic database (see sections below). The samples were then referenced (station, species and date) and preserved in absolute (99%) molecular-grade ethanol, in 1.5 mL tubes, until the DNA was extracted in the laboratory. Lab analysis (e)DNA extraction DNA (and eDNA) extractions were performed in dedicated rooms, different from that used for PCRs preparations, with benches, tools and surfaces bleach-disinfected before processing samples. From tissue, DNA was extracted using Qiagen DNeasy Blood & Tissue Kit, following manufacturers’ guidelines. Concerning the extraction from filters, some minor modifications were applied, following Baudry et al. (2021): ¼ of each filter was cut into small pieces, using sterilized forceps and scissors and dried for thirty minutes (to evaporate the ethanol), into a 2 mL Eppendorf tube. Lysis reagents (450 µL of ATL buffer and 50 µL Proteinase K) were added, submerging the filter fragment, and then vortexed before incubation at 56 °C for 3 hours. The following steps (washing) were done as described by the manufacturer, until the elution, in 60 µL of AE buffer (instead of 200 µL), to concentrate the eDNA. For both DNA and eDNA, the extraction yields were measured (concentration and absorbance ratios) using the Implen® N60/N50 nanophotometer (Implen GmbH, Munchen, Germany). Sanger sequencing First, to ensure the species identity of fish and crustaceans caught, the COI gene was sequenced, using the universal primers FishF1-TCAACCAACCACAAAGACATTGGCAC, FishF2-TCGACTAATCATAAAGATATCGGCAC and FishR1-TAGACTTCTGGGTGGCCAAAGAATCA, FishR2-5′ACTTCAGGGTGACCGAAGAATCAGAA for fish (Ward et al. 2005) and LCO1490-GGTCAACAAATCATAAAGATATTGG and HCO2198TGATTTTTTGGTCACCCTGAAGTTTA for decapods (Folmer et al. 1994). For each individual, Polymerase Chain Reaction (PCR) was performed following Chucholl et al. (2015) with minor modifications: 2.5 min at 95 °C for initial denaturing, followed by 35 cycles of 45 sec at 95 °C, 1 min at 50 °C and 1 min at 72 °C, and finally 10 min at 72 °C for final elongation. PCR products were purified and 1/10 diluted before sequencing, in both forward and reverse direction, on an Applied Biosystems SeqStudio Genetic Analyzer (Waltham, U.S.A.). Once the species’ identity was verified, DNA extracts were used to complete the genetic database for both fish (12S rRNA) and crustaceans (16S rRNA). The protocol used was the same as described before, with metabarcoding primers MiFish (Miya et al. 2015) and MiDeca (Komai et al. 2019) (see just below). 543 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean The sequences obtained (12S rRNA and 16S rRNA) were cleaned and trimmed using Geneious Pro R10 software (https://www.geneious.com; Kearse et al. 2012). They were added to MIDORI2, a newly curated database for eukaryotic taxonomic assignments (Leray et al. 2022), giving a complete database for fish and crustaceans assignments in Martinique (Lim et al. 2002; Baudry et al. 2024). Within this database, an additional curation was done to remove sequences containing ambiguities (N with maxambig = 0) or of poor quality (length – minlength = 150 - or homopolymer – maxhomop = 10). Metabarcoding amplifications To enable the sequencing of all samples in a single Illumina run, 2-steps PCRs were performed, using MiFish-U-FGTCGGTAAAACTCGTGCCAGC and MiFish-U-RCATAGTGGGGTATCTAATCCCAGTTTG primers targeting a 175 bp fragment of the mitochondrial 12S rRNA gene (for fish, Miya et al. 2015) and MiDeca-FGGACGATAAGACCCTATAAA and MiDeca-RACGCTGTTATCCCTAAAGT primers targeting a 164 bp fragment of the mitochondrial 16S rRNA gene (for crustaceans, Komai et al. 2019), including adapters (forward: TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG and reverse: GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAG). PCR reactions were set up in a sterile room, decontaminated every night by UV-light treatment. Each eDNA sample was amplified four times, representing eight PCR reactions per station. PCR reactions were carried out in a 25 µL final volume containing: 12.5 µL of KAPA HiFi HotStart ReadyMix (Roche), 5 µL of each primer with index (final concentration 0.2 µM) and 2.5 µL of template. Each PCR plate contained one negative control (i.e. no-template DNA), to assess for potential contamination during the amplification and three positive mock controls (for each taxa, fish and decapods). The first mock sample corresponds to an equimolar mix of DNA from individuals representing 19 species of decapods and 27 species of fish (Suppl. material 1: S2). In the second and third mocks, DNA were mixed using different quantities to simulate variation of taxa relative abundance. Mock samples were used as positive control during the PCR amplifications’ step and were used to calibrate the taxonomic inference from the genetic reference database (see below). Amplifications programs were: activation at 95 °C for 3 min followed by 35 cycles of 98 °C for 30 sec, 65 °C (60 °C for MiDeca) for 30 sec and 72 °C for 30 sec, and finally 72 °C for 5 min, for final extension. PCR products were visualized on 1.5% agarose gels and then pooled together per station - resulting in two sequencing results per station. They were then sent to PGTB sequencing platform in Bordeaux (France) for quality check (using TapeStation, Agilent, USA), library preparation (2nd PCR) and sequencing on Illumina NextSeq 2000 (U.S.A.), using 2 x 150 pb kit. Bioinformatics and data analyses Illumina sequencing handling Reads generated by Illumina NextSeq 2000 sequencing were handled using DADA2 package (v1.30.0; Callahan et al. 2016) implemented in R (v4.3.2; R Development Core Team, 2023). Primers were removed, reads with Ns were 544 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean pre-filtered and quality profiles were inspected. Considering these profiles, especially the quality score in end sequenced reads, and expected lengths, reads were filtered and trimmed. They were then dereplicated and pair-ended merged, using the error model implemented in DADA2. Finally, chimera were removed from the final sequence table and taxonomic assignments (using the curated database) were done using mothur (Schloss et al. 2009) with RDP classifier using the classify.seqs() command with method=wang, iters=1000 and cutoff=75 parameters. To limit the interpretation of low abundant erroneous DNA reads related to potential contaminants, PCR amplification or sequencing errors, we added additional filtering steps. Amplicons Sequences Variants (ASV) produced by DADA2 pipeline represented by < 10 reads in a sample and then, the ones representing < 0.1% of the total number of reads were removed from the analyses. After these filtering steps, we summed the reads for each ASV at each station and converted these counts into relative proportions by dividing each ASV’s read count by the total read count of that station. No filtering criteria based on representativity of taxa in a minimum of replicates per station was used, and all taxa were conserved in the final taxonomic inventory. Data analyses All statistical and graphical analyses were performed in the R environment (v4.3.2; R Development Core Team 2023). Before each statistical treatment, when appropriate, data normality and variance homogeneity were verified using ShapiroWilk and Bartlett tests, respectively. Maps were generated using QGIS 2.18 (Las Palmas) software (QGIS Team Development 2016): Martinique map was imported from the database ©IGN and the streams from BD Carthage® and BD Topo®. We first analyzed stations’ characteristics and searched for correlations between physico-chemical parameters (altitude, conductivity, pH, temperature, oxygen concentration) depending on the geographical situation of the considered station, based on a principal component analysis (PCA) (FactoMineR and factoextra packages; Lê et al. 2008; Kassambara and Mundt 2020). Contribution of each variable was visualized using fviz_contrib() and fviz_pca_var() functions and they were then projected on the factorial axes using ggplot2 (Wickham 2016). For eDNA, after taxonomic assignment of ASVs, fish and decapods taxonomic lists were produced and taxa sorted according to their known occurrence in freshwater or marine environments. For example, Caranx sp.– a marine genus resulting from the human consumption – was removed from the dataset here for species richness calculations. Then, we decided to pool all ASVs related to Loricariidae sp. together, as they are genetically and morphologically very close, making their identification difficult. Moreover, many of them are sold for aquarium trade. The influence of the method (TEF vs. eDNA) on those results (species richness) was analyzed based on an analysis of variance (ANOVA), considering a station effect. Species richness was then plotted for each station to visualize those assemblage differences individually. From eDNA data, Simpson index was calculated using phyloseq package (McMurdie and Holmes 2013) and then community structure comparison between station was tested, for fish, decapods and both mixed, using Bray-Curtis dissimilarity index implemented in phyloseq and ape packages (Paradis and Schliep 2019) and visualized using Non-metric MultiDimensional Scaling (NMDS). 545 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean Influence of the exposition (north/south) of the station was assessed using a permutational multivariate analysis of variance (PERMANOVA) using adonis2() from the vegan package (Oksanen et al. 2022). Then, canonical ordination methods were used to investigate the influence of metadata (north/south exposition, temperature, pH, conductivity and oxygen) on species assemblages, using vegan package (Oksanen et al. 2022), which include Redundancy Analysis (RDA) and Canonical Correspondence Analysis (CCA) (Legendre and Legendre 2012). After a gradient length analysis, using detrended correspondence analysis (DCA) (length < 3, suggesting a linear response of taxa to environmental gradients; Lepš and Šmilauer 2003), RDA models were run on Hellinger-transformed ASV abundance data (Legendre and Gallagher 2001). Conductivity, temperature, pH, oxygen concentration, and north/south exposure were used as predictors. Global model significance was assessed via permutation tests (999 permutations), and the contribution of individual variables was evaluated using both sequential (Type I) and marginal (Type III) tests. Finally, the correlation between assemblages’ dissimilarity and geographical distances (Suppl. material 1: S3) between each station was tested with a Mantel’s test together with a Spearman’s rank correlation test, using vegan (Oksanen et al. 2022) and geosphere (Hijmans 2022) packages. Results Environmental characterization of sampling sites Conductivity and temperature were the most influential variables, contributing respectively 30.79% and 28.61% to the variation explained by the axis 1 (Fig. 2A). Inversely, pH played a major role (65.48%) in shaping the axis 2 (Fig. 2A). As expected, temperature and conductivity were negatively correlated with altitude, and oxygen concentration is positively correlated with altitude (Fig. 2B). All stations seemed to exhibit variable pH values, but the north cluster was mainly characterized by lower temperatures and conductivity than the cluster of stations from the south (Fig. 2C). Bioinformatics and dataset clean-up In total, for the 14 stations studied (without the mocks), 11,656,029 reads were generated for fish (mean 832,573.5 ± 142,312.1 per station) and 14,813,701 reads for decapods (mean 1,058,121.5 ± 408,818.2 per station). After data filtering, taxonomic assignment and curation, average sample read counts per station was 570,082.9 ± 124,442.6 for fish and 963,464.9 ± 388,726.8 for decapods (Suppl. material 1: S4). Comparison of TEF vs. eDNA The number of species captured (TEF) ranged from one to seven for fish and three to seven for decapods, while eDNA-based method detected between four and eleven species of fish and four and nine of decapods (Fig. 3). That said, the eDNA-based method detected significantly more fish species than TEF (7.93 ± 1.94 vs. 4.28 ± 1.68 using TEF; F = 53.41, p < 0.001) (Fig. 3A) and more decapod species (6.93 ± 1.68 vs. 5.36 ± 1.28 using TEF; F = 6.47, p < 0.05) (Fig. 3B). 552 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean confirming most of the species caught by TEF and revealing the presence of additional (native or sometimes more worryingly invasive) species. This metabarcoding method showed limitations in discriminating some genetically close species (e.g. Sicydium sp.), potentially leading to under-representation of communities’ assemblages when it comes to calculate biodiversity indices, but not affecting the functional diversity in presence. That said, such metabarcoding data allowed to appreciate the ecological state of the stations studied, in a non-disruptive way, using different biodiversity indices (species richness, Simpson and Bray-Curtis), and to investigate how stations’ characteristics (using NMDS, PERMANOVA and RDA analysis) shape assemblages across Martinique. Finally, even if the eDNA-based method offers the ability to early detect invasive species or the discovery of new suitable areas for endemic and/or rare native species, traditional methods, such as TEF, remain indispensable for certain aims, for example genetic studies, and both methodologies could be used in a complementary way. Harmonization of eDNA protocols is crucial to maximize the effectiveness of biodiversity studies and we encourage stakeholders to join such an initiative to shed light on the rich biodiversity occurring in poorly studied regions and to facilitate the fight against invasive species, one of the leading causes of biodiversity loss. Acknowledgments We thank Marion Labeille working at Sentinelle Lab (Guadeloupe, Lesser Antilles) for TEF expertise and support. Part of the experiments (Illumina sequencing) were performed at the PGTB (doi:10.15454/1.5572396583599417E 12) with the help of Préscillia Alves-Gomes and Erwan Guichoux. Additional information Conflict of interest The authors have declared that no competing interests exist. Ethical statement No ethical statement was reported. Use of AI No use of AI was reported. Funding We warmly thank the Office Français de la Biodiversité (OFB), the Office de l’Eau (ODE) de Martinique and the Direction de l’Environnement, de l’Aménagement et du Logement (DEAL) de Martinique for financing TB’s post-doctoral contract, as well as the functioning of the InCrust project. This work was also supported by the Centre National de la Recherche Scientifique (CNRS) and the University of Poitiers, for lab facilities and intramural funds. Author contributions TB: Conceptualization, Methodology, Data curation, Formal analysis, Investigation, Software, Validation, Visualization, Funding acquisition, Writing – original draft, Writing – review & editing. VV: Validation, Software, Supervision, Writing – review & editing. CD: 553 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean Investigation, Validation, Writing – review & editing. AA: Funding acquisition, Administration, Writing – review & editing. FR: Funding acquisition, Administration, Writing – review & editing. GL: Funding acquisition, Administration, Writing – review & editing. CMM: Funding acquisition, Administration, Writing – review & editing. FG: Conceptualization, Validation, Supervision, Funding acquisition, Administration, Writing – review & editing. Author ORCIDs Thomas Baudry https://orcid.org/0000-0001-5699-6837 Valentin Vasselon https://orcid.org/0000-0001-5038-7918 Fabian Rateau https://orcid.org/0000-0003-1857-3387 Frédéric Grandjean https://orcid.org/0000-0002-8494-0985 Data availability All data generated or analyzed during this study are included in this published article (and its supplementary information files), are accessible in Zenodo repository (doi: 10.5281/zenodo.17227561) and available from the corresponding author upon reasonable request. References Adams AJ, Kamoroff C, Daniele NR, Grasso RL, Halstead BJ, Kleeman PM, Mengelt C, Powelson K, Seaborn T, Goldberg CS (2024) From eDNA to decisions using a multi-method approach to restoration planning in streams. Scientific Reports 14: 14335. https://doi.org/10.1038/s41598-024-64612-5 Allan EA, Zhang WGC, Lavery AF, Govindarajan A (2021) Environmental DNA shedding and decay rates from diverse animal forms and thermal regimes. Environmental DNA 3: 492–514. https://doi.org/10.1002/edn3.141 Anadón-Irizarry V, Wege DC, Upgren A, Young R, Boom B, León YM, Arias Y, Koenig K, Morales AL, Burke W, Pérez-Leroux A, Levy C, Koenig S, Gape L, Moore P (2012) Sites for priority biodiversity conservation in the Caribbean Islands Biodiversity Hotspot. Journal of Threatened Taxa 04: 2806–2844. https://doi.org/10.11609/JoTT.o2996.2806-44 Baudry T (2022) Étude des impacts de l’écrevisse exotique envahissante Cherax quadricarinatus sur les hydrosystèmes de Martinique. University of Poitiers. Baudry T, Mauvisseau Q, Goût J, Arqué A, Delaunay C, Smith-ravin J, Sweet M (2021) Mapping a super-invader in a biodiversity hotspot, an eDNA-based success story. Ecological Indicators 126: 107637. https://doi.org/10.1016/j.ecolind.2021.107637 Baudry T, Mauvisseau Q, Arqué A, Goût JP, Delaunay C, de Boer HJ, Grandjean F (2023) Environmental DNA survey to detect an endemic cryptic fish, Anablepsoidescryptocallus, in tropical freshwater streams. Aquatic Conservation 33: 325–335. https:// doi.org/10.1002/aqc.3916 Baudry T, Smith J, Alexandre R, Jean A, Goût P, Cucherousset J, Marc J, Frédéric P (2024) Trophic niche of the invasive Cherax quadricarinatus and extent of competition with native shrimps in insular freshwater food webs. Biological Invasions 26: 3227–3241. https://doi.org/10.1007/s10530-024-03373-8 Belle CC, Stoeckle BC, Geist J (2019) Taxonomic and geographical representation of freshwater environmental DNA research in aquatic conservation. Aquatic Conservation 29: 1996–2009. https://doi.org/10.1002/aqc.3208 Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP (2016) DADA2: High-resolution sample inference from Illumina amplicon data. Nature Methods 13: 581–583. https://doi.org/10.1038/nmeth.3869 554 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean Chucholl C, Mrugała A, Petrusek A (2015) First record of an introduced population of the southern lineage of white-clawed crayfish (Austropotamobius ‘italicus’) north of the Alps. Knowledge and Management of Aquatic Ecosystems 416: 1–8. https://doi. org/10.1051/kmae/2015006 Cucherousset J, Olden JD (2011) Ecological Impacts of Non-native Freshwater Fishes. Fisheries 36: 215–230. https://doi.org/10.1080/03632415.2011.574578 Di Muri C, Lawson Handley L, Bean CW, Li J, Peirson G, Sellers GS, Walsh K, Watson HV, Winfield IJ, Hänfling B (2020) Read counts from environmental DNA (eDNA) metabarcoding reflect fish abundance and biomass in drained ponds. Metabarcoding and Metagenomics 4: e56959. https://doi.org/10.3897/mbmg.4.56959 Dolan CR, Miranda LE (2004) Injury and Mortality of Warmwater Fishes Immobilized by Electrofishing. North American Journal of Fisheries Management 24: 118–127. https://doi.org/10.1577/M02-115 Dubreuil T, Baudry T, Mauvisseau Q, Arqué A, Courty C, Delaunay C, Sweet M, Grandjean F (2021) The development of early monitoring tools to detect aquatic invasive species: eDNA assay development and the case of the armored catfish Hypostomusrobinii. Environmental DNA 4: 349–362. https://doi.org/10.1002/edn3.260 Engman AC, Ramirez A (2012) Fish assemblage structure in urban streams of Puerto Rico: The importance of reach-and catchment-scale abiotic factors. Hydrobiologia 693: 141–155. https://doi.org/10.1007/s10750-012-1100-6 Evans NT, Shirey PD, Wieringa JG, Mahon AR, Lamberti GA (2017) Comparative cost and effort of fish distribution detection via environmental DNA analysis and electrofishing. Fisheries 42: 90–99. https://doi.org/10.1080/03632415.2017.1276329 Ficetola GF, Miaud C, Pompanon F, Taberlet P (2008) Species detection using environmental DNA from water samples. Biology Letters 4: 423–425. https://doi. org/10.1098/rsbl.2008.0118 Flitcroft R, Cooperman MS, Harrison IJ, Juffe-Bignoli D, Boon PJ (2019) Theory and practice to conserve freshwater biodiversity in the Anthropocene. Aquatic Conservation 29: 1013–1021. https://doi.org/10.1002/aqc.3187 Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R (1994) DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology 3: 294–299. Gherardi F, Bertolino S, Bodon M, Casellato S, Cianfanelli S, Ferraguti M, Lori E, Mura G, Nocita A, Riccardi N, Rossetti G, Rota E, Scalera R, Zerunian S, Tricarico E (2008) Animal xenodiversity in Italian inland waters: Distribution, modes of arrival, and pathways. Biological Invasions 10: 435–454. https://doi.org/10.1007/s10530-007-9142-9 Hänfling B, Handley LL, Read DS, Hahn C, Li J, Nichols P, Blackman RC, Oliver A, Winfield IJ (2016) Environmental DNA metabarcoding of lake fish communities reflects long-term data from established survey methods. Molecular Ecology 25: 3101–3119. https://doi.org/10.1111/mec.13660 Hense Z, Martin RW, Petty JT (2010) Electrofishing capture efficiencies for common stream fish species to support watershed-scale studies in the Central Appalachians. North American Journal of Fisheries Management 30: 1041–1050. https://doi. org/10.1577/M09-029.1 Hijmans R (2022) geosphere: Spherical Trigonometry. Kassambara A, Mundt F (2020) Factoextra: Extract and Visualize the Results of Multivariate Data Analyses. https://cran.r-project.org/package=factoextra Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper A, Markowitz S, Duran C, Thierer T, Ashton B, Meintjes P, Drummond A (2012) Gene- 555 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean ious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28: 1647–1649. https://doi. org/10.1093/bioinformatics/bts199 Kelly RP, Lodge DM, Lee KN, Theroux S, Sepulveda AJ, Scholin CA, Craine JM, Andruszkiewicz Allan E, Nichols KM, Parsons KM, Goodwin KD, Gold Z, Chavez FP, Noble RT, Abbott CL, Baerwald MR, Naaum AM, Thielen PM, Simons AL, Jerde CL, Duda JJ, Hunter ME, Hagan JA, Meyer RS, Steele JA, Stoeckle MY, Bik HM, Meyer CP, Stein E, James KE, Thomas AC, Demir-Hilton E, Timmers MA, Griffith JF, Weise MJ, Weisberg SB (2024) Toward a national eDNA strategy for the United States. Environmental DNA 6: e432. https://doi.org/10.1002/edn3.432 Komai T, Gotoh RO, Sado T, Miya M (2019) Development of a new set of PCR primers for eDNA metabarcoding decapod crustaceans. Metabarcoding and Metagenomics 3: e33835. https://doi.org/10.3897/mbmg.3.33835 Lê S, Josse J, Husson F (2008) FactoMineR: A package for multivariate analysis. Journal of Statistical Software 25: 1–18. https://doi.org/10.18637/jss.v025.i01 Lefrancois E, Labeille M, Marquès J, Robert M, Valentini A (2024) Validation of an eDNA-based method for surveying fish and crustacean communities in the rivers of the French West Indies. Hydrobiologia 851: 3249–3269. https://doi.org/10.1007/ s10750-024-05476-8 Legendre P, Legendre L (2012) Numerical Ecology (3rd English edn.). Elsevier, Amsterdam, The Netherlands. Legendre P, Gallagher ED (2001) Ecologically meaningful transformations for ordination of species data. Oecologia 129: 271–280. https://doi.org/10.1007/s004420100716 Lepš J, Šmilauer P (2003) Multivariate analysis of ecological data using CANOCO. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/CBO9780511615146 Leray M, Yang JY, Meyer CP, Mills SC, Agudelo N, Ranwez V, Boehm JT, Machida RJ (2013) A new versatile primer set targeting a short fragment of the mitochondrial COI region for metabarcoding metazoan diversity: application for characterizingcoralreeffishgut contents. Frontiers in Zoology 10: 34. https://doi. org/10.1186/1742-9994-10-34 Leray M, Knowlton N, Machida RJ (2022) MIDORI2: A collection of quality controlled, preformatted, and regularly updated reference databases for taxonomic assignment of eukaryotic mitochondrial sequences. Environmental DNA 4: 894–907. https://doi. org/10.1002/edn3.303 Lim P, Meunier FJ, Keith PA, Noël PY (2002) Muséum national d’Histoire naturelle, Paris, 124 pp. (Patrimoines naturels; 51) Atlas des poissons et des crustacés d’eau douce de la Martinique, 120 pp. Macher T-H, Schütz R, Yildiz A, Beermann AJ, Leese F (2023) Evaluating five primer pairs for environmental DNA metabarcoding of Central European fish species based on mock communities. Metabarcoding and Metagenomics 7: e103856. https://doi. org/10.3897/mbmg.7.103856 Magurran AE (2009) Threats to freshwater fish. Science 325: 1215. https://doi. org/10.1126/science.1177215 Majaneva M, Diserud OH, Eagle SHC, Boström E, Hajibabaei M, Ekrem T (2018) Environmental DNA filtration techniques affect recovered biodiversity. Scientific Reports 8: 1–11. https://doi.org/10.1038/s41598-018-23052-8 Marques V, Milhau T, Albouy C, Dejean T, Manel S, Mouillot D, Juhel J-B (2021) GAPeDNA: Assessing and mapping global species gaps in genetic databases for eDNA metabarcoding. Diversity & Distributions 27: 1880–1892. https://doi.org/10.1111/ddi.13142 556 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean McMurdie PJ, Holmes S (2013) phyloseq: An R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE 8: e61217. https://doi. org/10.1371/journal.pone.0061217 Michelet P (2017) La biodiversité des milieux aquatiques continentaux en France métropolitaine: état des lieux et menaces. Annales des Mines-Responsabilité et environnement 2: 36–39. https://doi.org/10.3917/re1.086.0036 Miya M, Sato Y, Fukunaga T, Sado T, Poulsen JY, Sato K, Minamoto T, Yamamoto S, Yamanaka H, Araki H, Kondoh M, Iwasaki W (2015) MiFish, a set of universal PCR primers for metabarcoding environmental DNA from fishes: Detection of more than 230 subtropical marine species. Royal Society Open Science 2: 150088. https://doi. org/10.1098/rsos.150088 Morisette J, Burgiel S, Brantley K, Daniel WM, Darling J, Davis J, Franklin T, Gaddis K, Hunter M, Lance R, Leskey T, Passamaneck Y, Piaggio A, Rector B, Sepulveda A, Smith M, Stepien CA, Wilcox T (2021) Strategic considerations for invasive species managers in the utilization of environmental DNA (eDNA): Steps for incorporating this powerful surveillance tool. Management of Biological Invasions 12: 747–775. https://doi. org/10.3391/mbi.2021.12.3.15 Myers N, Mittermeier RA, Mittermeier CG, da Fonseca GAB, Kent J (2000) Biodiversity hotspots for conservation priorities. Nature 403: 853–858. https://doi. org/10.1038/35002501 Nivet C, Mc Key D, Legris C (2010) Connaissance et gestion des écosystèmes tropicaux. Résultats du programme de recherche «écosystèmes Tropicaux» 2005-2010. Paris. Nunes AL, Tricarico E, Panov VE, Cardoso AC, Katsanevakis S (2015) Pathways and gateways of freshwater invasions in Europe. Aquatic Invasions 10: 359–370. https:// doi.org/10.3391/ai.2015.10.4.01 Oksanen J, Simpson G, Blanchet F, Kindt R, Legendre P, Minchin P, O’Hara R, Solymos P, Stevens M, Szoecs E, Wagner H, Barbour M, Bedward M, Bolker B, Borcard D, Carvalho G, Chirico M, De Caceres M, Durand S, Evangelista H, FitzJohn R, Friendly M, Furneaux B, Hannigan G, Hill M, Lahti L, McGlinn D, Ouellette M, Ribeiro Cunha E, Smith T, Stier A, Ter Braak C, Weedon J (2022) vegan: Community Ecology Package.: R package version 2.6–4. https://cran.r-project.org/package=vegan Oliveira Carvalho C, Pazirgiannidi M, Ravelomanana T, Andriambelomanana F, Schrøder-Nielsen A, Ready JS, de Boer H, Fusari C-E, Mauvisseau Q (2024) Multi-method survey rediscovers critically endangered species and strengthens Madagascar’s freshwater fish conservation. Scientific Reports 14: 20427. https://doi.org/10.1038/s41598-024-71398-z Paradis E, Schliep K (2019) ape 5.0: An environment for modern phylogenetics and evolutionary analyses in R. Bioinformatics 35: 526–552. https://doi.org/10.1093/bioinformatics/bty633 Pont D, Rocle M, Valentini A, Civade R, Jean P, Maire A, Roset N, Schabuss M, Zornig H, Dejean T (2018) Environmental DNA reveals quantitative patterns of fish biodiversity in large rivers despite its downstream transportation. Scientific Reports 8: 1–13. https://doi.org/10.1038/s41598-018-28424-8 Pusey BJ, Kennard MJ, Arthur JM, Arthington AH (1998) Quantitative sampling of stream fish assemblages: Singlevs multiple-pass electrofishing. Australian Journal of Ecology 23: 365–374. https://doi.org/10.1111/j.1442-9993.1998.tb00741.x QGIS Team Development (2016) QGIS Geographic Information System. Open Source Geospatial Foundation Project. R Development Core Team (2023) R: a language and environment for statistical computing. https://www.r-project.org/ 557 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean Renshaw MA, Olds BP, Jerde CL, Mcveigh MM, Lodge DM (2015) The room temperature preservation of filtered environmental DNA samples and assimilation into a phenol-chloroform-isoamyl alcohol DNA extraction. Molecular Ecology Resources 15: 168–176. https://doi.org/10.1111/1755-0998.12281 Riaz T, Shehzad W, Viari A, Pompanon F, Taberlet P, Coissac E (2011) ecoPrimers:inference of new DNA barcode markers from whole genome sequence analysis. Nucleic Acids Research 39: e145. https://doi.org/10.1093/nar/gkr732 Rodríguez-Barreras R, Zapata-Arroyo C, Falcón LW, Olmeda MDL (2020) An island invaded by exotics: A review of freshwater fish in Puerto Rico. Neotropical Biodiversity 6: 42–59. https://doi.org/10.1080/23766808.2020.1729303 Schenekar T, Schletterer M, Lecaudey LA, Weiss SJ (2020) Reference databases, primer choice, and assay sensitivity for environmental metabarcoding: Lessons learnt from a re-evaluation of an eDNA fish assessment in the Volga headwaters. River Research and Applications 36: 1004–1013. https://doi.org/10.1002/rra.3610 Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB, Lesniewski RA, Oakley BB, Parks DH, Robinson CJ, Sahl JW, Stres B, Thallinger GG, Van Horn DJ, Weber CF (2009) Introducing mothur: Open-source, platform-independent, community-supported software for describing and comparing microbial communities. Applied and Environmental Microbiology 75: 7537–7541. https://doi. org/10.1128/AEM.01541-09 Snyder DE (2003) Electrofishing and its harmful effects on fish. US Department of the Interior USGSurvey (Eds), 149 pp. Taberlet P, Bonin A, Zinger L, Coissac E (2018) Environmental DNA: For biodiversity research and monitoring. Oxford, 272 pp. https://doi.org/10.1093/ oso/9780198767220.001.0001 Taylor DS (2012) Twenty-Four Years in the Mud: What Have We Learned About the Natural History and Ecology of the Mangrove Rivulus, Kryptolebias marmoratus? Integrative and Comparative Biology 52: 724–736. https://doi.org/10.1093/icb/ics062 Thomsen PF, Willerslev E (2015) Environmental DNA - An emerging tool in conservation for monitoring past and present biodiversity. Biological Conservation 183: 4–18. https://doi.org/10.1016/j.biocon.2014.11.019 Valentini A, Taberlet P, Miaud C, Civade R, Herder J, Thomsen PF, Bellemain E, Besnard A, Coissac E, Boyer F, Gaboriaud C, Jean P, Poulet N, Roset N, Copp GH, Geniez P, Pont D, Argillier C, Baudoin JM, Peroux T, Crivelli AJ, Olivier A, Acqueberge M, Le Brun M, Møller PR, Willerslev E, Dejean T (2016) Next-generation monitoring of aquatic biodiversity using environmental DNA metabarcoding. Molecular Ecology 25: 929–942. https://doi.org/10.1111/mec.13428 Wang S, Yan Z, Hänfling B, Zheng X, Wang P, Fan J, Li J (2021) Methodology of fish eDNA and its applications in ecology and environment. The Science of the Total Environment 755: 142622. https://doi.org/10.1016/j.scitotenv.2020.142622 Ward RD, Zemlak TS, Innes BH, Last PR, Hebert PDNH (2005) DNA barcoding Australia’s fish species. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 360: 1847–1857. https://doi.org/10.1098/ rstb.2005.1716 Wickham H (2016) ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York. https://ggplot2.tidyverse.org WWF (2020) Living planet report 2020 - Bending the curve of biodiversity loss. In: Almond REA, Grooten M, Petersen T (Eds) WWF, Gland, Switzerland, 164 pp. https:// www.wwf.org.uk/sites/default/files/2020-09/LPR20_Full_report.pdf 558 Metabarcoding and Metagenomics 9: 537–558 (2025), DOI: 10.3897/mbmg.9.151675 Thomas Baudry et al.: Metabarcoding outperforms electrofishing for freshwater inventories in Caribbean Supplementary material 1 Supplementary information Authors: Thomas Baudry, Valentin Vasselon, Carine Delaunay, Alexandre Arqué, Fabian Rateau, Géraldine Lala, Claire Maurice-Madelon, Frédéric Grandjean Data type: docx Explanation note: S1. Biotic and abiotic characteristics of the 14 stations studied. S2. Presentation of the use of mock communities, used as positive and calibration controls in the present study. S3. Output tables generated after using the packages phyloseq, ape, vegan and geosphere. S4. Number of reads generated with Illumina NextSeq 2000 and reads count after each curation step, for both decapods and fish and for each station. S5. Results of the redundancy analyses (RDA) investigating the effects of Exposition (North/South), Temperature, pH, Conductivity and Oxygen concentration on community assemblages (fish, decapods and both mixed). Copyright notice: This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0/). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Link: https://doi.org/10.3897/mbmg.9.151675.suppl1