Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

A comparison of biomonitoring methodologies for surf zone fish communities

  • Zachary Gold ,

    Contributed equally to this work with: Zachary Gold, McKenzie Q. Koch

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Department of Ecology and Evolutionary Biology, University of California, Los Angeles, Los Angeles, CA, United States of America

  • McKenzie Q. Koch ,

    Contributed equally to this work with: Zachary Gold, McKenzie Q. Koch

    Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Department of Ecology and Evolutionary Biology, University of California, Los Angeles, Los Angeles, CA, United States of America

  • Nicholas K. Schooler,

    Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Marine Science Institute, University of California, Santa Barbara, Santa Barbara, CA, United States of America

  • Kyle A. Emery,

    Roles Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Marine Science Institute, University of California, Santa Barbara, Santa Barbara, CA, United States of America

  • Jenifer E. Dugan,

    Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Marine Science Institute, University of California, Santa Barbara, Santa Barbara, CA, United States of America

  • Robert J. Miller,

    Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Marine Science Institute, University of California, Santa Barbara, Santa Barbara, CA, United States of America

  • Henry M. Page,

    Roles Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Marine Science Institute, University of California, Santa Barbara, Santa Barbara, CA, United States of America

  • Donna M. Schroeder,

    Roles Conceptualization, Investigation, Methodology, Project administration, Resources, Supervision, Writing – original draft, Writing – review & editing

    Affiliation Bureau of Ocean Energy Management, Camarillo, CA, United States of America

  • David M. Hubbard,

    Roles Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Validation, Writing – original draft, Writing – review & editing

    Affiliation Marine Science Institute, University of California, Santa Barbara, Santa Barbara, CA, United States of America

  • Jessica R. Madden,

    Roles Conceptualization, Investigation, Methodology, Project administration, Writing – original draft, Writing – review & editing

    Affiliation Marine Science Institute, University of California, Santa Barbara, Santa Barbara, CA, United States of America

  • Stephen G. Whitaker,

    Roles Conceptualization, Data curation, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing – original draft, Writing – review & editing

    Affiliations Marine Science Institute, University of California, Santa Barbara, Santa Barbara, CA, United States of America, Channel Islands National Park, Ventura, CA, United States of America

  • Paul H. Barber

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Department of Ecology and Evolutionary Biology, University of California, Los Angeles, Los Angeles, CA, United States of America


Surf zones are highly dynamic marine ecosystems that are subject to increasing anthropogenic and climatic pressures, posing multiple challenges for biomonitoring. Traditional methods such as seines and hook and line surveys are often labor intensive, taxonomically biased, and can be physically hazardous. Emerging techniques, such as baited remote underwater video (BRUV) and environmental DNA (eDNA) are promising nondestructive tools for assessing marine biodiversity in surf zones of sandy beaches. Here we compare the relative performance of beach seines, BRUV, and eDNA in characterizing community composition of bony (teleost) and cartilaginous (elasmobranch) fishes of surf zones at 18 open coast sandy beaches in southern California. Seine and BRUV surveys captured overlapping, but distinct fish communities with 50% (18/36) of detected species shared. BRUV surveys more frequently detected larger species (e.g. sharks and rays) while seines more frequently detected one of the most abundant species, barred surfperch (Amphistichus argenteus). In contrast, eDNA metabarcoding captured 88.9% (32/36) of all fishes observed in seine and BRUV surveys plus 57 additional species, including 15 that frequent surf zone habitats. On average, eDNA detected over 5 times more species than BRUVs and 8 times more species than seine surveys at a given site. eDNA approaches also showed significantly higher sensitivity than seine and BRUV methods and more consistently detected 31 of the 32 (96.9%) jointly observed species across beaches. The four species detected by BRUV/seines, but not eDNA were only resolved at higher taxonomic ranks (e.g. Embiotocidae surfperches and Sygnathidae pipefishes). In frequent co-detection of species between methods limited comparisons of richness and abundance estimates, highlighting the challenge of comparing biomonitoring approaches. Despite potential for improvement, results overall demonstrate that eDNA can provide a cost-effective tool for long-term surf zone monitoring that complements data from seine and BRUV surveys, allowing more comprehensive surveys of vertebrate diversity in surf zone habitats.


Sandy beaches and their adjacent surf zones comprise ~30% of the world’s ice-free shoreline [1, 2]. Surf zones provide critical ecosystem services, supporting local marine biodiversity through the provisioning of forage habitat, refuge from predators, spawning sites, and nurseries for commercially and recreationally important fish species [1, 35]. Furthermore, sandy beaches and surf zones are important areas for recreation and tourism [4, 6, 7]. In California alone, the value of sandy beach ecosystem services in 2017 was estimated at $25.9 billion annually [79].

Despite their tremendous societal and ecological value, our understanding of the status, and spatial and temporal dynamics of surf zone fish communities in southern California and around the world is lacking [2], and sandy beaches and surf zones are rarely included in conservation management plans [10]. Sandy beaches and associated surf zone biological communities face both local and global anthropogenic stressors that threaten their biodiversity and ecosystem function [11]. Sea-level rise coupled with coastal armoring is contributing to coastal squeeze, compressing or eliminating sandy beaches and altering surf zone habitats [1215]. Coastal urban development and engineering are increasing erosion along shorelines, increasing turbidity and altering surf zone characteristics [1618]. Compounding these stressors, pollutants from stormwater, sewage, oil spills, and agricultural runoff often spill directly into surf zone habitats [11]. As urban development and climate change continues to impact these important coastal ecosystems, our ability to effectively manage sandy beaches hinges on accurate assessments and monitoring of the species and communities that depend on them [4, 11].

Traditional methods for monitoring surf zone ecosystems are based on surveys using nets, such as seines or bottom trawls, or hook and line fishing to capture surf zone fish [2, 4, 19]. Net, and hook and line surveys are advantageous as they can provide detailed information on size, sex, and age structure of fish populations, and are not influenced by poor underwater visibility. However, these capture surveys have known biases that limit their reliability and repeatability. Hook and line fishing surveys are often species-specific due to the choice of tackle and bait, and observer skill affects capture rates [20]. Wave and weather conditions can affect seine surveys by reducing the capture efficiency of nets and creating hazards to researchers in heavy surf (Table 1). Seines are also sensitive to slight variation in mesh size, width of opening, and speed of implementation, impacting repeatability and comparability of results [10, 21]. Seines are also less effective for sampling large, fast-moving species [22, 23] as well as small benthic fishes, such as flatfish (Families Pleuronectidae and Paralichthyidae), that pass through or under the nets. In addition, both these techniques are highly labor-intensive, and can be destructive, often injuring or killing captured specimens [24] (Table 1).

Alternative surf zone biomonitoring approaches rely on visual surveys, either via SCUBA or snorkel transects or baited remote underwater video (BRUV) units [2, 25, 26]. BRUVs are increasingly used to overcome diver avoidance behavior [19, 2729], instead employing baited video cameras that record fish passing through the field of view, allowing for non-invasive measurements of fish diversity, abundance, and behavior. However, BRUV surveys also have biases that limit their reliability and repeatability (Table 1). Large waves, inclement weather, light conditions, and drifting macrophytes, can all reduce visibility and impair species identification and detection [30, 31]. BRUV methods are also sensitive to bait choice, length and location of deployment [10, 21], may not attract planktotrophic and herbivorous fish that are not attracted to the bait, and are poor at detecting cryptic species [21]. Moreover, they are challenging to deploy by kayak or swimming in the surf zone, and can require processing of hundreds of hours of underwater video [26]. Together, these limitations affect the reliably and effectiveness of visual monitoring approaches of surf zone fish communities, highlighting the need for new approaches.

A promising new approach for surveying the diversity of coastal marine ecosystems is environmental DNA (eDNA) metabarcoding [32, 33]. eDNA refers to the collection, capture, sequencing, and identification of DNA shed from organisms inhabiting a particular ecosystem [34, 35]. Studies indicate that eDNA metabarcoding is highly sensitive and provides an accurate, practical, and cost-effective method of monitoring marine biodiversity [3640].

Studies of eDNA highlight some key advantages relative to seining and BRUVs (Table 1). In particular, eDNA identifies a broad diversity of marine life, frequently detecting more species than other methods [4144], including cryptic, rare, invasive, and endangered species [4549], and is effective across a variety of marine ecosystems, including coral reefs [50, 51], kelp forests [52, 53], estuaries [31, 54, 55], and coastal oceans [39, 56, 57]. eDNA is largely independent of developmental stage, allowing for the detection of larval and juvenile life stages, identifying potential nursery grounds [32]. In addition, eDNA samples are simple to collect, encouraging student led and community science, and are also cost effective, permitting increased sampling efforts across both time and space [32, 5860].

Yet eDNA also has limitations. For example, the need for molecular expertise and laboratory space to process samples may limit some research groups and monitoring agencies where such resources are not already available [36]. Additionally, eDNA does not provide key information needed for fishery and stock assessments (e.g., size, age, sex), and it is unclear whether eDNA results accurately reflect the relative abundance of marine species [31, 40, 61, 62].

There are also unresolved questions about the fate and transport of eDNA, particularly in highly dynamic coastal marine ecosystems. For example, previous studies report spatial resolution of eDNA in nearshore marine environments is on the scale of 50–1000 m [52, 53, 6369] and temporal resolution is on the scale of hours to days [67, 70, 71], complicating the ecological interpretation of detected community assemblages [72]. However, these studies were not conducted in surf zone ecosystems which are strongly affected by wave driven longshore transport and nearshore currents with higher velocities (e.g., rip currents) and tides compared to the subtidal ecosystems previously studied, potentially integrating ecological signatures over greater space and time, and mixing species detections across ecosystems [2].

Although eDNA and BRUV surveys hold promise for monitoring surf zone habitats, evaluating how well these methods perform compared to traditional seine surveys and each other is a crucial information gap [26, 73, 74]; to date, only two studies [52, 75] employed eDNA to assess fish biodiversity in surf zones habitats. To address this gap, we compared the ability of seine, BRUV, and eDNA methods to describe surf zone fish communities using a series of surveys where we simultaneously employed all three methodologies at 18 open coast surf zones associated with beaches in southern California. We compared these results to assess how the different survey methods performed in surf zone habitats, information critical to resource managers charged with monitoring these important coastal ecosystems.


Study sites

To compare the effectiveness of seine, BRUV, and eDNA survey techniques for monitoring surf zone bony (teleost) and cartilaginous (elasmobranch) fish communities, we deployed the three survey techniques contemporaneously at 18 sandy beach sites across southern California, USA (Fig 1; S1 Table in S2 File); 14 on the California Channel Islands and 4 on the mainland. These represent novel fish community surveys for all but three of the mainland sites, providing important baseline data for fish assemblages. To maximize comparability, we surveyed surf zones using all three methods at each location on the same day using the methods described below. All surveys were conducted between August 15, 2018 and November 2, 2018. At one site, Soledad beach, on Santa Rosa Island, we were unable to conduct the BRUV surveys due to hazardous surf conditions.

Fig 1. Site map.

Map of the study region showing mainland sites, Northern Channel Islands sites, and Catalina Island sites on the coast of southern California, USA. Black dots and numbers correspond to site names. 1– Dangermond, 2 –R Beach, 3 –Santa Claus, 4 –Santa Monica, 5 –Cuyler Harbor, 6 –Sandy Point, 7 –Soledad, 8 –Bechers Bay, 9 –Water Canyon, 10 –Southeast Anchorage, 11 –Ford Point, 12 –China Camp, 13 –Forney Cove, 14 –Christy Beach, 15 –Coches Prietos, 16 –Emerald Bay, 17 –Little Harbor, 18 –Ben Weston. The base map was created with Natural Earth Dataset (

Beach seine surveys

Beach seine surveys were employed using methods modified from the California Department of Fish and Wildlife (Monterey, CA, USA) (Carlisle, Schott & Abrahamson, 1960) using a 15.3 m (50 ft) x 1.8 m (6 ft) seine net (10 mm knotless nylon mesh, 2 m poles, 20 m leader ropes) with a bag, floats, and weighted lead line. At each site, we conducted seine hauls in the surf zone at four locations spaced haphazardly along the beach. For each seine haul, two researchers opened the beach seine parallel to shore in approximately 1.5 m of water. Keeping the weighted line flush with the bottom, we dragged the seine perpendicular to the shoreline until reaching the beach. Fish were then immediately removed from the seine, placed in aerated 1 m x 0.5 m x 0.5 m live wells, identified, enumerated, measured (total and standard length) on glazed (smooth) fish boards, and released alive at the site of capture in accordance University of California Santa Barbara’s Institutional Animal Care and Use (IACUC) protocol #943. Any fish that appear to be severely injured, moribund, or that did not recover from the stress of trapping were euthanized using Tricaine methanosulfonate (MS-222), a non-inhaled agents approved in the “AVMA Guidelines for the Euthanasia of Animals: 2013 Edition” for finfish [76].

Baited remote underwater video (BRUV) surveys

We conducted BRUV surveys following methods modified from Vargas-Fonseca et al. [77] and Borland et al. [25]. Each BRUV consisted of a GoPro Hero2 camera (GoPro Inc., San Mateo, California, USA, 2020) mounted on a five kg weight with a line and float attached for ease of retrieval. We then attached a bait bag containing ~152 g of frozen squid (Loligo sp.) to the weight with a PVC pipe, positioning it one meter in front of the camera. Snorkelers deployed three haphazardly spaced BRUV units along the outer edge of the surf zone at a depth of greater than two meters within two hours of low tide after conducting sein hauls, except for at sites where sufficient personnel allowed for concurrent sampling. We deployed each BRUV for one hour, producing three hours of video per beach. We reviewed videos to determine fish abundance, species richness, and community composition, using the MaxN statistic, the maximum number of individuals of one species in one frame during the hour-long footage [78].

Environmental DNA (eDNA) surveys

We collected three replicate 0.5 L samples of seawater (herein referred to as sample replicates) using sterile collapsible enteral feeding bags at each site. We then gravity filtered samples through 0.2 μm Sterivex filters following the methods of Gold et al. [79] (See Supplement for detailed methods description), storing filters at -20˚C prior to extraction via a modified Qiagen DNAeasy Blood and Tissue kit (Qiagen Inc., Gernmantown, MD, USA) [80]. Four field blanks consisting of 1 L of tap water were filtered in the field to serve as field and extraction controls. We amplified eDNA samples in triplicate using both 12S MiFish Universal teleost (MiFish-U) and elasmobranch (MiFish-E) primer sets [81], and then prepared sequencing libraries preparation followed Gold et al. [79] using Nextera Unique Dual Indices (Illumina, San Diego, CA, USA). Each unique PCR reaction included both positive and negative PCR controls; negative controls substituted molecular grade water in place of the DNA extraction, and we used either American alligator (Alligator mississippiensis) or dromedary (Camelus dromedarius), species non-native to California, for positive controls. In total, 2 positive controls, 5 PCR negative controls, and 4 field blanks were sequenced. We pooled all samples in equimolar concentrations by primer set except negative controls (5μL were added given no quantifiable DNA), resulting in a MiFish-U and a MiFish-E library which were separately sequenced on NextSeq PE 2 x 150 bp mid-output at the Technology Center for Genomics & Bioinformatics at the University of California–Los Angeles (UCLA) with 20% PhiX added to both sequencing runs.

Reference barcode generation

To supplement the California Current Large Marine Ecosystem reference database [79], we generated a MiFish 12S Universal Teleost specific reference sequence for white seabass (Atractoscion nobilis). Tissue was acquired from the California Current Cooperative Fisheries Investigation collections and extracted using a Qiagen DNAeasy Blood and Tissue kit (Qiagen Inc., Gernmantown, MD, USA). Tissue was amplified following the same PCR protocol for MiFish 12S Universal Teleost primer set described above. PCR products were purified using ExoSAP-IT (Affymetrix, Cleveland, OH, USA) and Sanger sequenced in both directions using BigDye chemistry (Applied Biosystems Inc, Foster City, CA, USA) at Laragen Inc., (Culver City, CA, USA) following Gold et al. [79].

eDNA bioinformatics

We processed the resulting eDNA metabarcoding sequences using the Anacapa Toolkit (version 1) (Curd et al., 2018) [119], conducting quality control, amplicon sequence variant (ASV) parsing, and taxonomic assignment. Taxonomy was assigned using the Bayesian Lowest Common Ancestor classifier [82] and a curated reference database composed of fishes from the California Current Large Marine Ecosystem supplemented with the generated Atractoscion nobilis sequence following [79]; See detailed description in supplement. We processed each sequencing library twice using two different barcoding reference libraries. First, to assign taxonomy to marine mammalian and avian species, we used the CRUX-generated-12S database, comprised of reference barcodes for all publicly available 12S barcodes. Second, we used a curated metabarcoding database specific to California coastal marine fish to generate taxonomic assignments for fishes. We employed a Q score cutoff of 30 and Bayesian taxonomic cutoff score of 60 following the methods of Gold et al. [79]. The resulting taxonomic tables were transferred into R for further processing (R Core Team, 2020a). Reads from multiple ASVs with identical assigned taxonomic paths were summed together following the methods of Gold et al. [79]. For example, both ASV 1 and ASV 3 were assigned to Atherinops affinis with 1.5 million reads and 1.4 million reads respectively and were thus all reads assigned to both ASVs were summed.

We employed a multifaceted decontamination approach described in Gold et al. [68] developed by Kelly et al. [65] to remove field contamination, lab contamination, and index hopping [65, 8385]. Through the decontamination process we implemented a hierarchical site occupancy modeling framework to distinguish occupancy rates across multiple sample bottle and technical replicate detections [65]. Only ASVs detected in at least two technical replicates in a site were kept. From these processes, we obtained decontaminated eDNA species-by-sample community tables with counts of total sequence reads. We then summed the total reads of ASVs by assigned taxonomy including multiple ASVs from the two MiFish markers employed (e.g. summed 41 ASVs assigned to Black surfperch Embiotica jacksoni). From these processes, we obtained decontaminated eDNA species-by-sample community tables with counts of total sequence reads. We note that this approach does not use sequenced negative controls or field blanks to correct reads as previous work has demonstrated that this frequently removes the most abundant ASVs which arise as a result of index hopping [65].

We transformed the eDNA read counts into eDNA index scores according to Kelly et al. [65], which normalizes the read count per technical PCR replicate per species. This index was computed by first calculating the proportional relative abundance of each species in each technical PCR replicate. The relative abundance was then divided by the maximum relative abundance for a given species across all samples (e.g. highest observed abundance for Northern Anchovy is 15% of all total reads which serves as the denominator), yielding the eDNA index score, which ranges from 0 to 1 and allows for comparisons of relative abundance for specific taxa across samples. The goal of applying this transformation is to account for the effects of amplification bias across taxa [8688].

We acknowledge that such an approach loses information about relative abundance between taxa in a sample. In particular, this approach likely over-inflates the abundance of rare taxa [88]. However, recent metabarcoding frameworks have highlighted that the compositional nature of metabarcoding [89] alongside species-specific amplification biases [87, 90], impair our ability to make adequate inferences from raw sequence counts or proportions between taxa without ancillary information on either the underlying target taxa abundances or amplification efficiencies [87, 9093]. Thus, in the absence of such information, we follow the conservative methods advised by Kelly, Shelton, and Gallego [88] and employ the eDNA index, erring on the side of amplification efficiency bias being the largest contributor to observed differences in read counts across fish species from eDNA samples [72, 87, 94].

Data analysis

To explore the relative efficacy of seines, BRUV, and eDNA surveys for characterizing surf zone fish communities, we compared the total number of teleost and elasmobranch species identified by each method using the phyloseq (version 1.28.0) and vegan packages (version 2.5–7) [95, 96] in R (version 3.6.1 [97]). Comparisons were made in two ways: 1) all detected fish taxa and 2) only surf zone fish taxa. Surf zone taxa were determined using habitat descriptions from and the literature [4, 98100] (S2 Table in S2 File). We determined and visualized the overlapping and unique fish species detected by each survey method across all 18 sites using the VennDiagram package (version 1.6.20) [101], comparing species richness of each method using Analysis of Variance (ANOVA) and post-hoc Tukey tests using the vegan package [96].

To examine survey method performance on a site-by-site basis, we calculated and compared the overlap of presence/absence site-species detections [97, 102]. Here, we define a site-species detection as the detection of a species at a given site (e.g., Top smelt detected at Bechers Bay). Comparisons of site-species detections were conducted for both the surf zone fishes and all fishes, observed by seine, BRUV, and eDNA, respectively. We estimated sample coverage, the fraction of the total incidence probabilities of the discovered species for a set of sampling units, from rarefaction and extrapolation models for species richness (Hill number q = 0) for each method using the iNext package (version 2.0.20) [103].

To determine whether the presence or absence of a species is a true reflection of biological reality or due to issues in the sampling process. [104, 105], we also conducted a site-occupancy analysis of species detections at each site following the methods of Chambert et al. [85] as implemented by Kelly et al. [65, 68]. We note this site occupancy analysis is separate from the method implemented in the decontamination process and was conducted on the final quality controlled data set. The binomial model yields the likelihood that a taxon detected is truly present in the sample. The model, implemented in Stan for R [104], depends upon three parameters: 1) the commonness of a taxon in the dataset (denoted Psi), 2) the probability of a detection given that the taxon is truly present (true positive; denoted P11), and 3) the probability of a detection given that the taxon was not truly present (false positive; denoted P10). The probability of occurrence function used was the following:

Where K is the number of samples taken within a site and N is the number of species detections within a site (See Supplemental methods for detailed description). For each species we calculated the number of detections out of the number of replicate surveys taken at each site.

We emphasize that the true occupancy of any given species at any site is unknown. Here, we are estimating the true positive (P11) and false positive (P10) of species being present at a given site using detections across repeated sampling events. Importantly, we make the explicit assumption that any detection of a species by a method is a real detection of that species.

In addition to probability of occurrence we also calculated the mean sensitivity, the proportion of true positive detections correctly identified as positive using the following equation for each species:

We also calculated the mean specificity, the proportion of true negative detections correctly identified as negative, using the following equation for each species:

We then compared the probability of occupancy, mean sensitivity, and mean specificity of each method across all species detected [106]. We further compared differences in the eDNA-derived probability of occurrence of surf zone and non-surf zone associated species to test if occupancy rates are a potential function of transport dynamics.

To analyze differences in the composition of surf zone fish detected among methods and across sites, we conducted a PERMANOVA and companion multivariate homogeneity of group dispersions on Jaccard-Binary dissimilarity indices based on presence/absence data using the adonis and betadisper functions in the vegan package [96]. The PERMANOVA was conducted using the following model:

We excluded the Soledad site on Santa Rosa Island given the lack of a BRUV survey. We further visualized community beta diversity among sampling methods using a constrained canonical analysis of principal components (CAP) through the vegan package [95, 97].

Lastly, to assess the ability of eDNA to capture relative abundance, we compared mean eDNA index scores to both the average catch counts per seine as well as average MaxN counts per BRUV station using species-specific linear regressions. Similarly, we compared BRUV-derived average MaxN counts against average seine counts. We focused our analyses on species detected jointly by each method at three or more sites. We note that comparing uncorrected compositional results (eDNA metabarcoding data) to estimates of absolute abundance (BRUV MaxN and seine counts) is inherently flawed [31, 87]. We present such results here to highlight the caveats of such approaches and discuss their merits (See Discussion).


Our beach seine surveys captured a total of 1,359 individuals in 72 hauls across all sites (4 hauls per site). Seven of the 72 hauls produced 0 individuals. In total, seining detected 24 species of fish from 24 genera, 13 families, and two classes (S3 Table in S2 File). On average, we captured 4.0 species (± 2.5 standard deviation, range 0–9), and 75.5 ± 82.8 individuals per site (range 0–325 individual fishes).

Our BRUV surveys detected a total of 1,114 individual fishes in 51 BRUV deployments (3 replicate deployments per site). In total, BRUV surveys detected 30 species, 30 genera, 21 families, and two classes (S4 Table in S2 File). An average of 6.3 ± 3.2 species (range 2–16 species) and 65.5 ± 65.5 total individuals (range 13–236 individuals) were recorded per site.

We successfully generated a reference barcode sequence for Atractoscion nobilis (S5 Table in S2 File).

Sequencing of the 54 eDNA samples yielded a total of 4,839,3365 MiFish elasmobranch reads and 16,835,329 MiFish teleost reads that passed the initial Illumina NextSeq quality controls across all samples. After decontamination and site occupancy modeling, we retained 3,652,862 reads and 915 ASVs from MiFish Elasmobranch primer set and 14,159,828 reads and 1,967 ASVs from MiFish Teleost primer set, representing 89 species of fish from 79 genera, 46 families, and two classes across sites. On average we observed 34.7 ± 11.1 SD species per site (range 17–59 species) (S1 Fig in S1 File). We note that in addition to fish we detected 8 species of birds and 5 species of marine mammals not discussed here.

Furthermore, six species of fish were detected in our field and negative controls when sequenced, despite the fact that no bands were visualized in PCR products during gel electrophoresis. Five of these six species were in the top 7 most abundant ASVs and in total represented 0.05% of all reads across all samples (n = 89,521). Given the prevalence of these species across all samples and technical replicates we did not use blanks for decontamination purposes [65].

Species assemblages characterized by each method

We found variable agreement in the assemblages of species captured by each survey method across all 18 sites (Fig 2 & S2 Fig in S1 File). Seine and BRUV captured distinct, but overlapping surf zone fish assemblages, sharing only 50% (18/36) of fishes species. Seine surveys detected 6 species of fishes not observed in BRUV surveys, including 2 species of croakers (Family Sciaenidae), 2 species of surfperches (Family Embiotocidae), and two planktotrophic coastal-pelagic species (families Clupeidae and Atherinidae). In contrast, BRUV surveys detected 12 fish species not observed in seines, including 3 species of elasmobranchs, 6 species of rocky reef associated species, and 2 coastal-pelagic predator species.

Fig 2. Venn diagram of eDNA, seine, and BRUV species detections.

Environmental DNA methods captured the majority (30/36) of fish species detected by both BRUV and seine surveys, only failing to identify six fish species found in the other two survey methods. In addition eDNA identified 58 additional fish species missed by seine and BRUV methods. In contrast, BRUV and seine surveys only captured 50% of species detected by both methods, showing strong difference in the species detected by each method. This was largely driven by the unique detection of elasmobranchs as well as nearshore pelagic and rocky reef carnivorous fishes in BRUV surveys compared to the unique detection of surfperches (Family Embiotocidae), grunts (Family Sciaenidae), and planktivorous nearshore pelagic species in seine surveys.

In contrast, eDNA detected the majority (88.9%, 32 out of 36) of species found in seine and BRUV surveys (Fig 2). Similarly, when only focusing on surf zone fish (S2 Table in S2 File), eDNA detected 93.1% (27 out of 29) of species detected in seine and BRUV surveys (S2 Fig in S1 File). eDNA methods failed to detect four species found in the seine and BRUV surveys including three species of surfperch, the most abundant and widespread family (Embiotocidae) detected in the seine surveys. Undetected species include the walleye surfperch (Hyperprosopon argenteum), silver surfperch (Hyperprosopon ellipticum), barred surfperch (Amphistichus argenteus) and kelp pipefish (Syngnathus californiensis). However, eDNA surveys detected 57 fish species not detected in seine or BRUV surveys (S6 & S7 Tables in S2 File), including 15 surf-zone associated species and 42 species more typically associated with reef and pelagic habitats (S2 Table in S2 File). Thus, eDNA had high overlap with both BRUV and seine surveys in addition to capturing additional surf zone and nearshore marine fishes.

Composition of detected taxa varied significantly among survey methods (Fig 3 and S3 Fig in S1 File; CAP ANOVA p < 0.001) driven by biases in detection of specific taxa. Seines and BRUVs commonly detected barred surfperch (Amphistichus argenteus), whereas eDNA only could not resolve surfperches below family level. Similarly, eDNA and BRUV surveys more frequently detected leopard shark (T. semifasciata) and California bat ray (M. californica) compared to seine surveys. In contrast, eDNA detected many more species than BRUVs or seines, including opaleye (Girella nigricans), northern anchovy (Engraulis mordax), giant kelpfish (Heterostichus rostratus), and dwarf perch (Micrometrus minimus) (Fig 3).

Fig 3. Constrained analysis of principal components.

Constrained Analysis of Principal Components (CAP) analysis was conducted on Jaccard binary dissimilarities of fish assemblages of all species detected across surveys. Survey method explained 42.6% of the total variation observed in the composition of detected taxa while site explained an additional 28.3% (PERMANOVA p <0.0001). We found no significant difference in homogeneity of dispersions across sites (betadisper > 0.05). BRUV and eDNA approaches more frequently detected leopard sharks (Triakis semifasciata) and California bat ray (Myliobatis californica) compared to seine surveys. Both seine and BRUV surveys detected Barred surfperch (Amphistichs argenteus) while eDNA methods could only achieve family level resolution for this taxon. eDNA approaches more consistently detected opaleye (Girella nigricans), northern anchovy (Engraulis mordax), giant kelpfish (Heterostichus rostratus), and dwarf perch (Micrometrus minimus).

In total, survey method explained 42.6% of the total variation observed in the composition of detected taxa, while site explained an additional 28.3% (PERMANOVA p <0.0001). We found no significant difference in homogeneity of dispersions across methods or sites (betadisper > 0.05) (S8 Table in S2 File). We also found similar differences in fish communities between survey methods when we limited our comparisons to only taxa observed by both visual and eDNA methods. Survey method explained 33.2% of the total variation observed in the composition of detected taxa, while site explained an additional 36.8% (PERMANOVA, p< 0.001). However, eDNA had significantly lower dispersion than seines across all sites (homogeneity of dispersions p< 0.001) (S9 Table in S2 File).

Detection rates of species across methods

Detection rates of species also differed significantly among survey methods (Fig 4 & S4-S8 Figs in S1 File) with eDNA having a significantly higher sensitivity (98.5%) than both seine (96.7%) and BRUV (96.2%) surveys across all taxa (ANOVA, p< 0.0001). Likewise, eDNA had significantly higher probability of occupancy (47.2%) at the site level than both seine (24.9%) and BRUV (28.6%) surveys (ANOVA, p< 0.0001) as well as having significantly higher specificity (71.4%) than seines (66.4%) (ANOVA, p = 0.01). However, we observed no difference in specificity between BRUV (69.1%) and eDNA or seine surveys at the site level (ANOVA, p >0.5) (Fig 5). Furthermore, we found no significant difference in probability of occupancy for species known to inhabit surf zone habitats (52.4%) than non-surf zone associated species (42.6%) detected with eDNA methods (ANOVA, p = 0.06) (S8 Fig in S1 File).

Fig 4. Heatmap of surf zone fishes jointly detected between surveys.

Teleost species in black font and elasmobranch species in blue font. Environmental DNA approaches more frequently detected 26 of 27 known surf zone species detected by either BRUV or seine surveys. Only Leopard shark Triakis semifasciata was more frequently detected by BRUV surveys. Aggregating surfperches with identical 12S barcodes to family level assignment (surfperch), eDNA detected surfperches at the same rate at seines. Aggregating to genus level assignment, eDNA detected Syngathus sp. More frequently than either BRUV of seine surveys.

Fig 5. Probability of occupancy, specificity, and sensitivity of eDNA, seine, and BRUV surveys.

Environmental DNA surveys had higher probability of occupancy and sensitivity than BRUV and seine surveys. eDNA had significantly higher specificity than seine surveys. We found no difference in specificities between BRUV and eDNA and seine surveys. Probability of occurrence is a measure of how likely a species is present at a site as a function of the commonness of the species as well as the true positive and false positive detection rates of the method surveyed. Sensitivity is the proportion of true positive species detections correctly identified as true positive detections. Specificity is the proportion of true negative species detections identified as negative detections.

The three methods yielded different levels of detection both overall and of individual species of surf zone fish. Our eDNA samples more consistently detected 96.9% (31/32) of all species jointly observed by either BRUV or seines. Leopard shark (Triakis semifasciata) was the only species more frequently detected with BRUV (15/18) than eDNA methods (14/18). When grouping surfperch species with identical 12S reference barcodes at the family level, eDNA detected surfperches as frequently as seine surveys (17/18) (Fig 4). When grouping all Syngathus sp. pipefish species with identical 12S reference barcodes at the genus level, eDNA detected Syngathus sp. More frequently than seine and BRUV surveys.

Comparing only BRUV and seine surveys, our BRUV surveys detected elasmobranchs and flatfishes (Families Pleuronectidae and Paralichthyidae) more frequently than seine surveys. In contrast, seine surveys more frequently detected barred surfperch (Amphistichus argenteus), walleye surfperch (Hyperprosopon argenteum), California corbina (Menticirrhus undulatus), northern anchovy (Engraulis mordax), giant kelpfish (Heterostichus rostratus), and kelp pipefish (Syngnathus californiensis) than BRUV surveys.

Across all sites, eDNA had higher sample coverage estimates (98.6%) than both BRUV (89.6%) and seine (85.2%) surveys (Fig 6). From species rarefaction curves of all species surveyed at the site level, we estimate that both BRUV and seine surveys would have to be conducted at more than 100 sites to achieve similar sample coverage estimates as eDNA at the 18 sites surveyed here.

Fig 6. Sample coverage estimates of eDNA, seine, and BRUV surveys.

Across all sites, environmental DNA surveys had an estimate sample coverage of 98.9%, higher than the sample coverage of BRUV (89.6%) and seine (85.2%) surveys. Shaded area represents 95% confidence intervals. Sample rarefaction curves across sites suggest BRUV and seine surveys would have to be conducted at more than 100 sites to achieve similar sample coverage estimates to eDNA surveys conducted at the 18 sites surveyed here.

Comparisons of relative abundance among survey methods

Estimates of relative abundance varied significantly among the three survey methods and were generally not correlated. We found a significant positive relationship between BRUV MaxN values and seine counts (R2 = 0.31, p = 0.031, S11 Table in S2 File, S9 Fig in S1 File) for only one species, topsmelt (Atherniops affinis). Likewise, there was a significant positive relationship between seine counts and eDNA index scores for only two species, topsmelt, (R2 = 0.27, p = 0.03, S12 Table in S2 File, S10 Fig in S1 File), and California corbina, Menticirrhus undulatus (R2 = 0.91, p < 0.001, S12 Table in S2 File). Similarly, there was a significant positive relationship between BRUV MaxN and eDNA index for three species (kelp bass, Paralbrax clathratus, shovelnose guitarfish, Psuedobatos productus, and round stingray Urobatis halleri) (respective R2: 0.33, 0.41, and 0.94, p < 0.01, S13 Table in S2 File, S11 Fig in S1 File).


Despite extreme methodological differences, seine, BRUV, and eDNA surveys captured largely overlapping, but distinct fish assemblages in surf zone habitats with notable taxonomic biases. Seines more consistently detected surfperches, including the most abundant fished species, barred surfperch (Amphistichus argenteus) while BRUV surveys efficiently revealed larger predatory species, particularly elasmobranchs as previously documented [107]. eDNA captured the highest species richness of all three methods, including the majority of species detected by seine and BRUV surveys (32/36). The mismatch in fish assemblages sampled by each method made comparisons of relative abundance difficult, highlighting a key challenge of comparing survey methods [43].

Importantly, similar to other studies (see [32, 33, 35, 41, 43, 68, 108] among others), we found that eDNA analysis had higher sensitivity than the two traditional methods, and more frequently detected nearly all jointly observed species at a given site. Our results suggest that seine, BRUV, and eDNA approaches are complementary and their use in tandem provides the most accurate characterization of surf zone fish communities. Recent studies using two of these three methods reached similar conclusions [107109].

Species assemblages characterized by each method

Only half of fish species detected by seine and/or BRUV surveys overlapped (18/36) indicating that these methods target different species assemblages. Compared to BRUV surveys, seine surveys captured additional surfperches and croakers associated with surf zone habitats as well as planktivorous coastal pelagic species. In contrast, BRUV surveys detected a greater number of elasmobranch and rocky reef species, particularly carnivores and scavengers, suggesting that fish are attracted from adjacent habitats to the bait, or our current understanding of species’ surf zone habitat utilization is limited. Combined, our results align well with previous findings from tropical shorelines indicating that BRUV and seines capture distinct, but overlapping fish assemblages in surf zone habitats [107].

Our finding that eDNA approaches detected nearly 90% (32 out of 36) of fish species observed using seine and BRUV methods, with higher overlap in detected fish assemblages. Importantly, eDNA approaches also captured an additional 15 surf zone species not observed by our seine or BRUV methods, including the federally listed northern tidewater goby (Eucyclogobius newberryi) and commercially-fished species of management concern, such as the flathead grey mullet (Mugil cephalus), black croaker (Cheilotrema saturnum), white croaker (Genyonemus lineatus), and Pacific sanddab (Citharichthys sordidus) [4]. Furthermore, eDNA detected a wide array of elasmobranchs that are typically underrepresented in most traditional sampling approaches [108, 110112] including angel shark (Squatina californica), horn shark (Heterodontus francisci), California butterfly ray (Gymnura marmorata), and broadnose sevengill shark (Notorynchus cepedianus). As such, eDNA should be viewed as a valuable complement both seine and BRUV surveys.

The failure of eDNA to detect four common species captured by seine and BRUV surveys was due to the limitations of the 12S MiFish-U primers, particularly for the surfperches and pipefishes [79]. Both Embiotocidae and Syngnathidae are diverse, recent radiations [113, 114] and the MiFish-U primers perform poorly in such cases, such as rockfish in the genus Sebastes [115]. Failure to detect three of six surfperch species and is likely a result of insufficient genetic variation within the 12S gene region bounded by the MiFish 12S primer set, leading to many surfperches only being resolved at higher taxonomic ranks (e.g. Embiotocidae) [79]. Importantly, we note all three species had shared corresponding 12S reference barcodes [79]. Likewise, kelp pipefish (Syngnathus californiensis) and bay pipefish (Syngnathus leptorhynchus) have identical 12S reference barcodes and thus could not be resolved to species level [114]. However, given that we were able to resolve surfperches to family level and pipefishes to genus level resolution with eDNA suggests we were able to capture shed DNA from these taxa and that the application of alternative primer sets could have led to improved results [79, 114, 115].

Although the MiFish-U primer set is imperfect, our results demonstrate the value of generating reference barcode sequences for all taxa in the study system. At the request of an anonymous reviewer, we generated a MiFish-U reference sequence for white seabass (Atractoscion nobilis) which allowed us to successfully detect the species at multiple sites unlike in previous applications of eDNA in coastal Southern California waters [52, 67, 79, 116, 117]. Without the barcode, the species could only be resolved to Family level and was not included in analyses. These results add further support to the importance of reference barcode generation for successful eDNA efforts [68, 118, 119].

eDNA captured a strong signature of surf zone fish assemblages including an additional 15 species of surf zone fishes not observed by seine and BRUV approaches, highlighting the utility of eDNA biomonitoring to improve estimates of total fish diversity in coastal monitoring surveys. eDNA also detected an additional 42 native coastal marine fishes not detected by our seine and BRUV surveys (S6, S7 Tables in S2 File). Although many of these species are unlikely to inhabit surf zone habitats directly [4, 100, 120], our study beaches were adjacent to rocky reef kelp forests, rocky intertidal habitats, and estuaries. Our detections of additional native fish species highlight the capacity for movement of both fish and eDNA across pelagic and inshore habitats [63]. Given the potential for transport on the scale of tens to thousands of meters, the detection of fishes from adjacent habitats in eDNA samples is to be expected [66], thus highlighting a potential shortcoming of eDNA approaches, and the need for better understanding of spatial and temporal variability in the dispersal of eDNA within and across ecosystems. Despite the need to better characterize the fate and transport of eDNA, our results still demonstrate that such eDNA approaches can be highly informative of surf zone communities [52], particularly on longer open coast beaches that are not located adjacent to rocky subtidal or intertidal habitats.

Detection rates of species across methods

In addition to the differences in fish assemblages captured by each method, we found substantial differences in the detection frequency of jointly observed species across sites between these methods. Overall, we found that eDNA had higher frequency of detection of nearly all species (31/32) jointly detected by either of the seine and BRUV methods (S6, S7 Tables in S2 File). This higher rate of detection also resulted in eDNA having significantly higher sensitivity than both seines and BRUV surveys. Furthermore, results from species rarefaction curves suggest that eDNA surveys capture a larger proportion of the total fish diversity across sites than seine and BRUV surveys. Importantly, our results suggest that additional BRUV and seine surveys should be deployed across more sandy beach sites rather than additional deployments at the same site to maximize fish diversity across the region. In contrast, our results suggest that the current eDNA deployment of three sample replicates with three technical replicates was sufficient to adequately capture diversity across the region, providing a baseline sampling regime for future eDNA deployments for monitoring fish diversity in surf zone ecosystems.

One possible explanation for the differences in site-species detection frequency across methods is poor taxonomic resolution or erroneous assignment across methods. The Anacapa Toolkit provides confidence scores around each taxonomic rank of assignment, providing information on the accuracy of eDNA identifications [119]. However, such confidence scores are not readily available for data from seine and BRUV surveys, where taxonomic identification depends on the presence of easily observed morphological characteristics and the resolution of video still captures. For example, topsmelt (Atherinops affinis) and California grunion (Leurethes tenuis) are morphologically very similar, with the potential for misidentification, particularly under low visibility conditions for BRUV surveys.

The variation in temporal and spatial scales sampled by each of the three survey methods may also drive differences in site-species detections [26, 63, 67, 70, 71, 107, 108]. Beach seines survey a small spatial area (here 15.3 m x 1.8 m x 2m) at 0 to 1.5 m depth at a single instantaneous snapshot of sampling [31, 120]. In contrast, BRUV units were deployed for an hour at 2–3 m depth and likely attracted species across tens to hundreds of square meters [26, 73, 107, 108, 121]. Although the spatial and temporal scales of eDNA methods in marine systems are still an active area of research, previous studies have found that eDNA integrates across spatial scales from 50–1,000 meters and degrades in situ between 2 and 12 hours, although laboratory experiments suggest degradation rates on the order of days [53, 63, 67, 71, 122]. Thus, the ecological integration time of each of these surveys is substantially different and likely contributes to the differences we observed in species detections [43].

Differences in species detection among methods are also likely driven by the dynamics of eDNA. eDNA shedding rates can vary among [123] and within species [124], driven by differences in physiology and behavior. Increased shedding rates result in higher eDNA detection probabilities, thus biasing which species are successfully detected within surf zone ecosystems. For example, eDNA methods have the potential to be biased during spawning events when high DNA concentrations are released [125]. Likewise, the interaction between high water transport within and potentially variable degradation rates across species or environmental conditions (temperature, UV, etc.) could influence detection probabilities [64, 70, 126].

We found that eDNA captured a wide variety of species not typically associated with surf zone habitats, suggesting transport of eDNA from offshore and other intertidal habitats and some level of spatial integration of eDNA measurements. Interestingly, we found that species known to inhabit surf zone habitats had similar probability of occupancy than species known to associate with further offshore habitats. Previous work has found that eDNA signatures were able to distinguish surf zone and adjacent subtidal kelp forest ecosystems from differences in fish assemblage composition as well as relative abundance estimates [52]. However, our results suggests that eDNA detection is not biased towards species known to generally inhabit surf zone habitats. Our observational study design precluded directly testing whether species recently inhabiting the surf zone upon sample collection had higher rates of detection as many of the species known to inhabit surf zones use such habitats at different frequencies and durations. Future additional research using field based exclusions of specific taxa alongside on eDNA dispersal tracking could allow for modeled adjustment of eDNA data to account for transportation dynamics.

Relative abundance

Given the observed low overlap in species detections across survey methods, assessing the capability of eDNA approaches to estimate relative abundance was challenging, particularly since eDNA surveys frequently detected a species at multiple sites where seine and BRUV surveys did not detect that species at all. This presents a core challenge of comparing eDNA to capture and visual surveys when the true abundance of species is unknown (Table 1) [43]. However, given that the ability to estimate relative abundance is a function of the ability to detect a given species, our results suggest that eDNA approaches are more sensitive and better suited than capture and visual survey methods to estimate abundance [85, 104, 105]. This result, however, is highly dependent on the ability of eDNA approaches to resolve a given taxa. Here eDNA approaches using the MiFish-U primer set failed to resolve the most abundant surf zone species from both seine and BRUV surveys, the surfperches (Family Embiotocidae).

Recent work from studies with greater survey overlap show promise for estimating relative abundance using eDNA approaches [31, 40, 62, 87, 94, 127], particularly when accounting for the underlying mechanisms that relate observed sequence read counts to the underlying biology and biomass of detected species [72, 90]. Here we note that we did not account for such underlying mechanisms in our comparisons of relative abundance which may explain why we observed such poor correlations. Such poor correlations between our simple eDNA index transformation and seine and BRUV counts are expected given the compositional nature of metabarcoding [89], as such a transformation does not address the many underlying non-linear factors that affect the relationship between true environmental abundance and observed reads obtained from eDNA metabarcoding sequencing [87, 90, 92]. Future work should account for the suite of mechanisms that affect observed sequence reads including transport, residence time, and variation in species specific shedding and degradation rates of eDNA [72, 94, 126, 128]. Importantly such efforts must account for the role of amplification efficiency for biasing metabarcoding results using joint models of metabarcoding data, amplification efficiency estimates, and species specific absolute abundance estimates [88, 90, 92, 127].

Choosing a survey method

All survey methods have biases, and the more a particular survey method is used allows the determination of such biases. For example, diver avoidance behavior is a well-established bias of visual SCUBA surveys [19, 2729]. Likewise, results of this study showed that each method had distinct advantages and disadvantages. BRUVs are more likely to capture large mobile species than seines, and eDNA captured more total diversity than BRUVs or seines. As such, method selection will largely be a function of the goals of a study, and whether detection of specific taxa or total diversity is a priority.

However, an important consideration when employing eDNA or BRUV data compared to seine surveys (without photographic documentation of hauls) is that the DNA sequences and ASV tables generated by eDNA and the video footage produced by BRUVs are permanent records of what was present at a particular time [32] (Table 1). For eDNA in particular, as reference databases are improved, eDNA sequence data can be reanalyzed to test for the presence of previously missed or poorly resolved taxa, e.g. white seabass. In addition, bio-archived eDNA samples or extractions can be revisited for future resequencing and management and biomonitoring applications (e.g., species invasions) [129]. The ability for future analyses of a given ecosystem at a specific time highlights the advantages of applying multiple approaches, where eDNA can provide robust and accurate taxonomic information that can be updated over time while carefully deployed stereo-video approaches (not deployed here given challenging surf conditions) and seine hauls can provide size structure and biomass estimates with demonstrated utility [26, 107].


There is a growing need to survey threatened surf zone and beach ecosystems in the face of global change [11]. Our results suggest that seine, BRUV, and eDNA approaches are complementary techniques for surveying fish diversity in open coast surf zone habitats. eDNA is a relatively quick, effective, and nondestructive approach to surveying marine wildlife, compared to capture and visual surveys of dynamic surf zone habitat (Table 1). Given the cost effectiveness and ability to automate collection and processing, eDNA methods could provide an approach to increase the scope and scale of surf zone ecosystem monitoring across time and space [33, 36]. The ease of sample collection in this challenging habitat could allow researchers, marine resource managers, and community scientists to conduct surveys more frequently and in more places, better characterizing surf zone biodiversity and dynamics [24, 59, 60, 130]. Furthermore, the ability to archive eDNA samples for future use provides an important resource for comparative analyses of ecosystem change [32, 129] and for making use of advances in reference libraries [68, 118].

Although we demonstrated that eDNA provides more robust species detections in surf zone habitats, eDNA cannot provide information on sex ratios or population size structure that can be obtained from seine and BRUV surveys, information critical to resource management [2, 120]. Thus, eDNA cannot be viewed as a wholesale replacement for other survey methods, but instead as a complementary tool for biomonitoring surf zone ecosystems [109]. Nevertheless, adding eDNA surveys to traditional monitoring programs or conducting them on their own when and where other methods are untenable has the potential to greatly enhance our knowledge of surf zone fish communities, providing a new source of comprehensive and detailed information needed for management and preservation of these vital coastal ecosystems in the face of global change.


We thank the Channel Islands National Park, The Nature Conservancy, UC Natural Reserve System, CSUCI Santa Rosa Island Field Station, USC Wrigley Institute and NOAA Channel Islands National Marine Sanctuary for access and use of facilities. We thank Laura Beresford, Francesca Puerzer, Justin Hoesterey, and Russel Johnson for their assistance in the field. We thank Beverly Shih, Nikita Sridhar and Lauren Man for their help in library preparation of eDNA samples. We thank Garfield Kwan for generously providing white seabass tissue.


  1. 1. Luijendijk A, Hagenaars G, Ranasinghe R, Baart F, Donchyts G, Aarninkhof S. The state of the world’s beaches. Scientific reports. 2018;8: 1–11.
  2. 2. Olds AD, Vargas-Fonseca E, Connolly RM, Gilby BL, Huijbers CM, Hyndes GA, et al. The ecology of fish in the surf zones of ocean beaches: A global review. Fish and Fisheries. 2018;19: 78–89.
  3. 3. Klein YL, Osleeb JP, Viola MR. Tourism-generated earnings in the coastal zone: A regional analysis. Journal of Coastal Research. 2004;20: 1080–1088.
  4. 4. Allen LG, Horn MH. The ecology of marine fishes: California and adjacent waters. Univ of California Press; 2006.
  5. 5. McLachlan A, Defeo O, Jaramillo E, Short AD. Sandy beach conservation and recreation: Guidelines for optimising management strategies for multi-purpose use. Ocean and Coastal Management. 2013;71: 256–268.
  6. 6. Schlacher TA, Dugan J, Schoeman DS, Lastra M, Jones A, Scapini F, et al. Sandy beaches at the brink. Diversity and Distributions. 2007;13: 556–560.
  7. 7. Dugan JE, Hubbard DM, Nielsen KJ, Altstatt J, Bursek J. Baseline Characterization of Sandy Beach Ecosystems Along the South Coast of California. 2015 pp. 113–113. Available:
  8. 8. Dugan J, Hubbard D. Macrofauna communities of exposed sandy beaches on the Southern California mainland and Channel Islands. 2000. pp. 339–346. Available:
  9. 9. National Marine Fisheries Service. Fisheries economics of the United States, 2014. NOAA Technical Memorandum. 2014; 175 pp.-175 pp.
  10. 10. Schlacher TA, Schoeman DS, Dugan J, Lastra M, Jones A, Scapini F, et al. Sandy beach ecosystems: Key features, sampling issues, management challenges and climate change impacts. Marine Ecology. 2008;29: 70–90.
  11. 11. Defeo O, McLachlan A, Schoeman DS, Schlacher TA, Dugan J, Jones A, et al. Threats to sandy beach ecosystems: A review. Estuarine, Coastal and Shelf Science. 2009;81: 1–12.
  12. 12. Dugan JE, Emery KA, Alber M, Alexander CR, Byers JE, Gehman AM, et al. Generalizing Ecological Effects of Shoreline Armoring Across Soft Sediment Environments. Estuaries and Coasts. 2018;41: 180–196.
  13. 13. Schooler NK, Dugan JE, Hubbard DM. No lines in the sand: Impacts of intense mechanized maintenance regimes on sandy beach ecosystems span the intertidal zone on urban coasts. Ecological Indicators. 2019;106: 105457–105457.
  14. 14. Barnard PL, Dugan JE, Page HM, Wood NJ, Hart JAF, Cayan DR, et al. Multiple climate change-driven tipping points for coastal systems. Scientific Reports. 2021;11: 1–13. pmid:34330962
  15. 15. Jaramillo E, Dugan J, Hubbard D, Manzano M, Duarte C. Ranking the ecological effects of coastal armoring on mobile macroinvertebrates across intertidal zones on sandy beaches. Science of the Total Environment. 2021;755: 142573–142573. pmid:33039935
  16. 16. Manning LM, Peterson CH, Fegley SR. Degradation of surf-fish foraging habitat driven by persistent sedimentological modifications caused by beach nourishment. Bulletin of Marine Science. 2013;89: 83–106.
  17. 17. Peterson CH, Bishop MJ, D’Anna LM, Johnson GA. Multi-year persistence of beach habitat degradation from nourishment using coarse shelly sediments. Science of the Total Environment. 2014;487: 481–492. pmid:24802271
  18. 18. Parkinson RW, Ogurcak DE. Beach nourishment is not a sustainable strategy to mitigate climate change. Estuarine, Coastal and Shelf Science. 2018;212: 203–209.
  19. 19. Carlisle JG, Schott JW, Abrahamson NJ. The barred surf perch in Southern California, Calif. Dept Fish and Fish Bull. 1960;109.
  20. 20. Kuriyama PT, Branch TA, Hicks AC, Harms JH, Hamel OS. Investigating three sources of bias in hook-and-line surveys: Survey design, gear saturation, and multispecies interactions. Canadian Journal of Fisheries and Aquatic Sciences. 2019;76: 192–207.
  21. 21. Andradi-Brown DA, Macaya-Solis C, Exton DA, Gress E, Wright G, Rogers AD. Assessing caribbean shallow and mesophotic reef fish communities using Baited-Remote Underwater Video (BRUV) and diver-operated video (DOV) survey techniques. PLoS ONE. 2016;11: e0168235–e0168235. pmid:27959907
  22. 22. Baker R, Sheaves M. Visual surveys reveal high densities of large piscivores in shallow estuarine nurseries. Marine Ecology Progress Series. 2006;323: 75–82.
  23. 23. Ceni G, Vieira JP. Looking through a dirty glass: How different can the characterization of a fish fauna be when distinct nets are used for sampling? Zoologia. 2013;30: 499–505.
  24. 24. Kelly RP, Port J a., Yamahara KM, Martone RG, Lowell N, Thomsen PF, et al. Harnessing DNA to improve environmental management. Science. 2014;344. pmid:24970068
  25. 25. Borland HP, Schlacher TA, Gilby BL, Connolly RM, Yabsley NA, Olds AD. Habitat type and beach exposure shape fish assemblages in the surf zones of ocean beaches. Marine Ecology Progress Series. 2017;570: 203–211.
  26. 26. Schramm KD, Harvey ES, Goetze JS, Travers MJ, Warnock B, Saunders BJ. A comparison of stereo-BRUV, diver operated and remote stereo-video transects for assessing reef fish assemblages. Journal of Experimental Marine Biology and Ecology. 2020;524: 151273–151273.
  27. 27. Hodgson G, Maun L, Shuman C. Reef Check Survey Manual. Reef Check, Institute of the Environment, University of California, Los Angeles, CA. 2004.
  28. 28. Kushner DJ, Rassweiler A, McLaughlin JP, Lafferty KD. A multi-decade time series of kelp forest community structure at the California Channel Islands. Ecology. 2013;94: 2655–2655.
  29. 29. Lindfield SJ, Harvey ES, McIlwain JL, Halford AR. Silent fish surveys: bubble-free diving highlights inaccuracies associated with SCUBA-based surveys in heavily fished areas. Börger L, editor. Methods in Ecology and Evolution. 2014;5: 1061–1069.
  30. 30. Lowry M, Folpp H, Gregson M, Mckenzie R. A comparison of methods for estimating fish assemblages associated with estuarine artificial reefs. Brazilian Journal of Oceanography. 2011;59: 119–131.
  31. 31. Shelton AO, Kelly RP, O’Donnell JL, Park L, Schwenke P, Greene C, et al. Environmental DNA provides quantitative estimates of a threatened salmon species. Biological Conservation. 2019;237: 383–391.
  32. 32. Deiner K, Bik HM, Mächler E, Seymour M, Lacoursière‐Roussel A, Altermatt F, et al. Environmental DNA metabarcoding: Transforming how we survey animal and plant communities. Molecular ecology. 2017;26: 5872–5895. pmid:28921802
  33. 33. Beng KC, Corlett RT. Applications of environmental DNA (eDNA) in ecology and conservation: opportunities, challenges and prospects. Biodiversity and Conservation. 2020;29: 2089–2121.
  34. 34. Taberlet P, Coissac E, Hajibabaei M, Rieseberg LH. Environmental DNA. Molecular Ecology. 2012;21: 1789–1793. pmid:22486819
  35. 35. Taberlet P, Bonin A, Zinger L, Coissac E. Environmental DNA: For biodiversity research and monitoring. Oxford University Press; 2018. p. 253.
  36. 36. Bohmann K, Evans A, Gilbert MTP, Carvalho GR, Creer S, Knapp M, et al. Environmental DNA for wildlife biology and biodiversity monitoring. Trends in Ecology & Evolution. 2014;29: 358–367. pmid:24821515
  37. 37. Doi H, Inui R, Akamatsu Y, Kanno K, Yamanaka H, Takahara T, et al. Environmental DNA analysis for estimating the abundance and biomass of stream fish. Freshwater Biology. 2017;62: 30–39.
  38. 38. Closek CJ, Santora JA, Starks HA, Schroeder ID, Andruszkiewicz EA, Sakuma KM, et al. Marine Vertebrate Biodiversity and Distribution Within the Central California Current Using Environmental DNA (eDNA) Metabarcoding and Ecosystem Surveys. Frontiers in Marine Science. 2019;6: 732–732.
  39. 39. Djurhuus A, Closek CJ, Kelly RP, Pitz KJ, Michisaki RP, Starks HA, et al. Environmental DNA reveals seasonal shifts and potential interactions in a marine community. Nature Communications. 2020;11: 1–9. pmid:31937756
  40. 40. Stoeckle MY, Adolf J, Charlop-Powers Z, Dunton KJ, Hinks G, Vanmorter SM. Trawl and eDNA assessment of marine fish diversity, seasonality, and relative abundance in coastal New Jersey, USA. ICES Journal of Marine Science. 2021;78: 293–304.
  41. 41. Thomsen PF, Kielgast J, Iversen LL, Møller PR, Rasmussen M, Willerslev E. Detection of a diverse marine fish fauna using environmental DNA from seawater samples. PloS one. 2012;7: e41732. pmid:22952584
  42. 42. Valentini A, Taberlet P, Miaud C, Civade R, Herder J, Thomsen PF, et al. Next‐generation monitoring of aquatic biodiversity using environmental DNA metabarcoding. Molecular ecology. 2016. pmid:26479867
  43. 43. Kelly RP, Closek CJ, O’Donnell JL, Kralj JE, Shelton AO, Samhouri JF. Genetic and Manual Survey Methods Yield Different and Complementary Views of an Ecosystem. Front Mar Sci. 2017;3.
  44. 44. Stat M, Huggett MJ, Bernasconi R, DiBattista JD, Berry TE, Newman SJ, et al. Ecosystem biomonitoring with eDNA: metabarcoding across the tree of life in a tropical marine environment. Scientific Reports. 2017;7: 12240. pmid:28947818
  45. 45. Simpfendorfer C, Kyne P, Noble T, Goldsbury J, Basiita R, Lindsay R, et al. Environmental DNA detects Critically Endangered largetooth sawfish in the wild. Endangered Species Research. 2016;30: 109–116.
  46. 46. Weltz K, Lyle JM, Ovenden J, Morgan JAT, Moreno DA, Semmens JM. Application of environmental DNA to detect an endangered marine skate species in the wild. PLoS ONE. 2017;12: e0178124–e0178124. pmid:28591215
  47. 47. Lafferty KD, Benesh KC, Mahon AR, Jerde CL, Lowe CG. Detecting Southern California’s white sharks with environmental DNA. Frontiers in Marine Science. 2018;5: 1–6.
  48. 48. Leblanc F, Belliveau V, Watson E, Coomber C, Simard N, Di Bacco C, et al. Environmental DNA (eDNA) detection of marine aquatic invasive species (AIS) in Eastern Canada using a targeted species-specific qPCR approach. Management of Biological Invasions. 2020;11: 201–217.
  49. 49. Thomas AC, Tank S, Nguyen PL, Ponce J, Sinnesael M, Goldberg CS. A system for rapid eDNA detection of aquatic invasive species. Environmental DNA. 2020;2: 261–270.
  50. 50. West K, Travers MJ, Stat M, Harvey ES, Richards ZT, DiBattista JD, et al. Large‐scale eDNA metabarcoding survey reveals marine biogeographic break and transitions over tropical north‐western Australia. Diversity and Distributions. 2021.
  51. 51. Marwayana ON, Gold Z, Meyer CP, Barber PH. Environmental DNA in a global biodiversity hotspot: Lessons from coral reef fish diversity across the Indonesian archipelago. Environmental DNA. 2022;4: 222–238.
  52. 52. Monuki K, Barber PH, Gold Z. eDNA captures depth partitioning in a kelp forest ecosystem. PLoS ONE. 2021;16: e0253104–e0253104. pmid:34735443
  53. 53. Port JA O’Donnell JL, Romero‐Maraccini OC, Leary PR, Litvin SY, Nickols KJ, et al. Assessing vertebrate biodiversity in a kelp forest ecosystem using environmental DNA. Molecular ecology. 2016;25: 527–541. pmid:26586544
  54. 54. Stoeckle MY, Soboleva L, Charlop-Powers Z. Aquatic environmental DNA detects seasonal fish abundance and habitat preference in an urban estuary. PLoS ONE. 2017;12. pmid:28403183
  55. 55. Schwentner M, Zahiri R, Yamamoto S, Husemann M, Kullmann B, Thiel R. eDNA as a tool for non-invasive monitoring of the fauna of a turbid, well-mixed system, the Elbe estuary in Germany. PLoS ONE. 2021;16: e0250452–e0250452. pmid:33861810
  56. 56. Thomsen PF, Møller PR, Sigsgaard EE, Knudsen SW, Jørgensen OA, Willerslev E. Environmental DNA from seawater samples correlate with trawl catches of subarctic, deepwater fishes. PLoS ONE. 2016;11. pmid:27851757
  57. 57. Andruszkiewicz EA, Starks HA, Chavez FP, Sassoubre LM, Block BA, Boehm AB. Biomonitoring of marine vertebrates in Monterey Bay using eDNA metabarcoding. Doi H, editor. PLoS ONE. 2017;12: e0176343–e0176343. pmid:28441466
  58. 58. Biggs J, Ewald N, Valentini A, Gaboriaud C, Dejean T, Griffiths RA, et al. Using eDNA to develop a national citizen science-based monitoring programme for the great crested newt (Triturus cristatus). Biological Conservation. 2015;183: 19–28.
  59. 59. Freiwald J, Meyer R, Caselle JE, Blanchette CA, Hovel K, Neilson D, et al. Citizen science monitoring of marine protected areas: Case studies and recommendations for integration into monitoring programs. Marine Ecology. 2018;39: e12470–e12470.
  60. 60. Meyer RS, Ramos MM, Lin M, Schweizer TM, Gold Z, Ramos DR, et al. The CALe DNA program: Citizen scientists and researchers inventory California’s biodiversity. California Agriculture. 2021;75: 20–32.
  61. 61. Ushio M, Murata K, Sado T, Nishiumi I, Takeshita M, Iwasaki W, et al. Demonstration of the potential of environmental DNA as a tool for the detection of avian species. Scientific reports. 2018;8: 4493. pmid:29540790
  62. 62. Di Muri C, Handley LL, Bean CW, Li J, Peirson G, Sellers GS, et al. Read counts from environmental DNA (eDNA) metabarcoding reflect fish abundance and biomass in drained ponds. Metabarcoding and Metagenomics. 2020;4: 97–112.
  63. 63. O’Donnell JL, Kelly RP, Shelton AO, Samhouri JF, Lowell NC, Williams GD. Spatial distribution of environmental DNA in a nearshore marine habitat. PeerJ. 2017;5: e3044. pmid:28265513
  64. 64. Shogren AJ, Tank JL, Andruszkiewicz E, Olds B, Mahon AR, Jerde CL, et al. Controls on eDNA movement in streams: Transport, Retention, and Resuspension /704/158/2464 /704/242 /45/77 article. Scientific Reports. 2017;7. pmid:28698557
  65. 65. Kelly RP, Gallego R, Jacobs-Palmer E. The effect of tides on nearshore environmental DNA. PeerJ. 2018;6: e4521. pmid:29576982
  66. 66. Andruszkiewicz EA, Koseff JR, Fringer OB, Ouellette NT, Lowe AB, Edwards CA, et al. Modeling environmental DNA transport in the coastal ocean using Lagrangian particle tracking. Frontiers in Marine Science. 2019;6: 477–477.
  67. 67. Ely T, Barber PH, Man L, Gold Z. Short-lived detection of an introduced vertebrate eDNA signal in a nearshore rocky reef environment. PLoS ONE. 2021;16: e0245314–e0245314. pmid:34086697
  68. 68. Gold Z, Wall AR, Schweizer TM, Pentcheff ND, Curd EE, Barber PH, et al. A manager’s guide to using eDNA metabarcoding in marine ecosystems. PeerJ. 2022;10: e14071–e14071. pmid:36405018
  69. 69. Yamamoto S, Masuda R, Sato Y, Sado T, Araki H, Kondoh M, et al. Environmental DNA metabarcoding reveals local fish communities in a species-rich coastal sea. Scientific reports. 2017;7: 40368. pmid:28079122
  70. 70. Collins RA, Wangensteen OS, O’Gorman EJ, Mariani S, Sims DW, Genner MJ. Persistence of environmental DNA in marine systems. Communications Biology. 2018;1: 1–11. pmid:30417122
  71. 71. Murakami H, Yoon S, Kasai A, Minamoto T, Yamamoto S, Sakata MK, et al. Correction to: Dispersion and degradation of environmental DNA from caged fish in a marine environment (Fisheries Science, (2019), 85, 2, (327–337), 10.1007/s12562-018-1282-6). Fisheries Science. 2019;85: 1109–1109.
  72. 72. Harrison JB, Sunday JM, Rogers SM. Predicting the fate of eDNA in the environment and implications for studying biodiversity. Proceedings of the Royal Society B: Biological Sciences. 2019;286: 20191409–20191409. pmid:31744434
  73. 73. French B, Wilson S, Holmes T, Kendrick A, Rule M, Ryan N. Comparing five methods for quantifying abundance and diversity of fish assemblages in seagrass habitat. Ecological Indicators. 2021;124: 107415–107415.
  74. 74. Gutierrez-Rodriguez A, Stukel MR, Lopes dos Santos A, Biard T, Scharek R, Vaulot D, et al. High contribution of Rhizaria (Radiolaria) to vertical export in the California Current Ecosystem revealed by DNA metabarcoding. ISME Journal. 2019;13: 964–976. pmid:30538274
  75. 75. Stoeckle MY, Das Mishu M, Charlop-Powers Z. Improved Environmental DNA Reference Library Detects Overlooked Marine Fishes in New Jersey, United States. Frontiers in Marine Science. 2020;7: 226–226.
  76. 76. Leary SL, Underwood W, Anthony R, Cartner S, Corey D, Grandin T, et al. AVMA guidelines for the euthanasia of animals: 2013 edition. American Veterinary Medical Association Schaumburg, IL; 2013.
  77. 77. Vargas-Fonseca E, Olds AD, Gilby BL, Connolly RM, Schoeman DS, Huijbers CM, et al. Combined effects of urbanization and connectivity on iconic coastal fishes. Diversity and Distributions. 2016;22: 1328–1341.
  78. 78. Ellis DM, DeMartini EE. Evaluation of a video camera technique for indexing abundances of juvenile pink snapper, Pristipomoides filamentosus, and other Hawaiian insular shelf fishes. Fishery Bulletin. 1995;93: 67–77.
  79. 79. Gold Z, Curd EE, Goodwin KD, Choi ES, Frable BW, Thompson AR, et al. Improving metabarcoding taxonomic assignment: A case study of fishes in a large marine ecosystem. Molecular Ecology Resources. 2021;21: 2546–2564. pmid:34235858
  80. 80. Spens J, Evans AR, Halfmaerten D, Knudsen SW, Sengupta ME, Mak SS, et al. Comparison of capture and storage methods for aqueous macrobial eDNA using an optimized extraction protocol: advantage of enclosed filter. Methods in Ecology and Evolution. 2017;8: 635–645.
  81. 81. Miya M, Sato Y, Fukunaga T, Sado T, Poulsen JY, Sato K, et al. 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. 2015;2: 150088. pmid:26587265
  82. 82. Gao X, Lin H, Revanna K, Dong Q. A Bayesian taxonomic classification method for 16S rRNA gene sequences with improved species-level accuracy. BMC bioinformatics. 2017;18: 247. pmid:28486927
  83. 83. Goldberg CS, Turner CR, Deiner K, Klymus KE, Thomsen PF, Murphy MA, et al. Critical considerations for the application of environmental DNA methods to detect aquatic species. 2016.
  84. 84. Costello M, Fleharty M, Abreu J, Farjoun Y, Ferriera S, Holmes L, et al. Characterization and remediation of sample index swaps by non-redundant dual indexing on massively parallel sequencing platforms. BMC genomics. 2018;19: 332. pmid:29739332
  85. 85. Chambert T, Pilliod DS, Goldberg CS, Doi H, Takahara T. An analytical framework for estimating aquatic species density from environmental DNA. Ecology and Evolution. 2018;8: 3468–3477. pmid:29607039
  86. 86. Shelton AO O’Donnell JL, Samhouri JF, Lowell N, Williams GD, Kelly RP. A framework for inferring biological communities from environmental DNA. Ecological Applications. 2016;26: 1645–1659. pmid:27755698
  87. 87. Shelton AO, Gold ZJ, Jensen AJ, D’Agnese E, Andruszkiewicz Allan E, Van Cise A, et al. Toward quantitative metabarcoding. Ecology. 2022; e3906–e3906. pmid:36320096
  88. 88. Kelly RP, Shelton AO, Gallego R. Understanding PCR Processes to Draw Meaningful Conclusions from Environmental DNA Studies. Scientific Reports. 2019;9: 1–14. pmid:31431641
  89. 89. Gloor GB, Macklaim JM, Pawlowsky-Glahn V, Egozcue JJ. Microbiome datasets are compositional: And this is not optional. Frontiers in Microbiology. 2017;8: 2224–2224. pmid:29187837
  90. 90. McLaren MR, Willis AD, Callahan BJ. Consistent and correctable bias in metagenomic sequencing experiments. eLife. 2019;8: e46923–e46923. pmid:31502536
  91. 91. McLaren MR, Nearing JT, Willis AD, Lloyd KG, Callahan BJ. Implications of taxonomic bias for microbial differential-abundance analysis. 2022. Available:
  92. 92. Silverman JD, Bloom RJ, Jiang S, Durand HK, Dallow E, Mukherjee S, et al. Measuring and mitigating PCR bias in microbiota datasets. PLoS Computational Biology. 2021;17: e1009113–e1009113. pmid:34228723
  93. 93. Egozcue JJ, Graffelman J, Ortego MI, Pawlowsky-Glahn V. Some thoughts on counts in sequencing studies. NAR Genomics and Bioinformatics. 2020;2: 1–10. pmid:33575638
  94. 94. Yates MC, Wilcox TM, Stoeckle M, Heath DD. Interspecific allometric scaling in eDNA production among northwestern Atlantic bony fishes reflects physiological allometric scaling. Environmental DNA. 2022.
  95. 95. McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PloS one. 2013;8: e61217. pmid:23630581
  96. 96. Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, et al. vegan: Community Ecology Package. R package version 2.5–6. 2019. 2020.
  97. 97. R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing V, Austria. 2016. URL 2018.
  98. 98. Froese R, Pauly D. FishBase. 2010.
  99. 99. Kells VA, Rocha LA, Allen LG. A field guide to coastal fishes: from Alaska to California. JHU Press; 2016.
  100. 100. Love MS, Passarelli JK. Miller and Lea’s Guide to the Coastal Marine Fishes of California. UCANR Publications; 2020.
  101. 101. Chen H, Boutros PC. VennDiagram: A package for the generation of highly-customizable Venn and Euler diagrams in R. BMC Bioinformatics. 2011;12: 1–7. pmid:21269502
  102. 102. Wickham H. ggplot2: elegant graphics for data analysis. Springer; 2016.
  103. 103. Hsieh TC, Ma KH, Chao A. iNEXT: an R package for rarefaction and extrapolation of species diversity (Hill numbers). Methods in Ecology and Evolution. 2016;7: 1451–1456.
  104. 104. Royle JA, Link WA. Generalized site occupancy models allowing for false positive and false negative errors. Ecology. 2006;87: 835–841. pmid:16676527
  105. 105. Schmidt BR, Kery M, Ursenbacher S, Hyman OJ, Collins JP. Site occupancy models in the analysis of environmental DNA presence/absence surveys: a case study of an emerging amphibian pathogen. Methods in Ecology and Evolution. 2013;4: 646–653.
  106. 106. Parikh R, Mathai A, Parikh S, Sekhar GC, Thomas R. Understanding and using sensitivity, specificity and predictive values. Indian journal of ophthalmology. 2008;56: 45. pmid:18158403
  107. 107. Shah Esmaeili Y, Corte G, Checon H, Gomes T, Lefcheck J, Amaral A, et al. Comprehensive assessment of shallow surf zone fish biodiversity requires a combination of sampling methods. Marine Ecology Progress Series. 2021;667: 131–144.
  108. 108. Jeunen G-J, Urban L, Lewis R, Knapp M, Lamare M, Rayment W, et al. Marine environmental DNA (eDNA) for biodiversity assessments: a one-to-one comparison between eDNA and baited remote underwater video (BRUV) surveys. Authorea Preprints. 2020; 486941–486941.
  109. 109. Stat M, John J, DiBattista JD, Newman SJ, Bunce M, Harvey ES. Combined use of eDNA metabarcoding and video surveillance for the assessment of fish biodiversity. Conservation Biology. 2019;33: 196–205. pmid:30004598
  110. 110. Bassett DK, Montgomery JC. Investigating nocturnal fish populations in situ using baited underwater video: With special reference to their olfactory capabilities. Journal of Experimental Marine Biology and Ecology. 2011;409: 194–199.
  111. 111. Bakker J, Wangensteen OS, Chapman DD, Boussarie G, Buddo D, Guttridge TL, et al. Environmental DNA reveals tropical shark diversity in contrasting levels of anthropogenic impact. Scientific Reports. 2017;7: 16886. pmid:29203793
  112. 112. Boussarie G, Bakker J, Wangensteen OS, Mariani S, Bonnin L, Juhel JB, et al. Environmental DNA illuminates the dark diversity of sharks. Science Advances. 2018;4: eaap9661–eaap9661. pmid:29732403
  113. 113. Longo G, Bernardi G. The evolutionary history of the embiotocid surfperch radiation based on genome-wide RAD sequence data. Molecular Phylogenetics and Evolution. 2015;88: 55–63. pmid:25858559
  114. 114. Nester GM, De Brauwer M, Koziol A, West KM, DiBattista JD, White NE, et al. Development and evaluation of fish eDNA metabarcoding assays facilitate the detection of cryptic seahorse taxa (family: Syngnathidae). Environmental DNA. 2020;2: 614–626.
  115. 115. Min MA, Barber PH, Gold Z. MiSebastes: An eDNA metabarcoding primer set for rockfishes (genus Sebastes). Conservation Genetics Resources. 2021;13: 447–456.
  116. 116. Gold Z, Sprague J, Kushner DJ, Marin EZ, Barber PH. eDNA metabarcoding as a biomonitoring tool for marine protected areas. PLoS ONE. 2021;16: 2020.08.20.258889–2020.08.20.258889. pmid:33626067
  117. 117. Lamy T, Pitz KJ, Chavez FP, Yorke CE, Miller RJ. Environmental DNA reveals the fine-grained and hierarchical spatial structure of kelp forest fish communities. Scientific Reports. 2021;11: 1–13. pmid:34262101
  118. 118. Collins RA, Trauzzi G, Maltby KM, Gibson TI, Ratcliffe FC, Hallam J, et al. Meta-Fish-Lib: A generalised, dynamic DNA reference library pipeline for metabarcoding of fishes. Journal of Fish Biology. 2021;99: 1446–1454. pmid:34269417
  119. 119. Curd E, Gold Z, Kandlikar G, Gomer J, Ogden M, Pipes L, et al. Anacapa: an environmental DNA toolkit for processing multilocus metabarcode datasets. Methods in Ecology and Evolution. 2018; in revision.
  120. 120. Allen LG, Pondella DJ. Surf zone, coastal pelagic zone, and harbors. The Ecology of Marine Fishes: California and Adjacent Waters. 2006; 149–166.
  121. 121. Cole VJ, Harasti D, Lines R, Stat M. Estuarine fishes associated with intertidal oyster reefs characterized using environmental DNA and baited remote underwater video. Environmental DNA. 2022;4: 50–62.
  122. 122. Yamamoto S, Minami K, Fukaya K, Takahashi K, Sawada H, Murakami H, et al. Environmental DNA as a “snapshot” of fish distribution: A case study of Japanese jack mackerel in Maizuru Bay, Sea of Japan. PLoS ONE. 2016;11. pmid:26933889
  123. 123. Andruszkiewicz Allan E, Zhang WGC, Lavery A, F. Govindarajan A. Environmental DNA shedding and decay rates from diverse animal forms and thermal regimes. Environmental DNA. 2021;3: 492–514.
  124. 124. Thalinger B, Rieder A, Teuffenbach A, Pütz Y, Schwerte T, Wanzenböck J, et al. The Effect of Activity, Energy Use, and Species Identity on Environmental DNA Shedding of Freshwater Fish. Frontiers in Ecology and Evolution. 2021;9: 73–73.
  125. 125. Tillotson MD, Kelly RP, Duda JJ, Hoy M, Kralj J, Quinn TP. Concentrations of environmental DNA (eDNA) reflect spawning salmon abundance at fine spatial and temporal scales. Biological Conservation. 2018;220: 1–11.
  126. 126. Barnes MA, Turner CR. The ecology of environmental DNA and implications for conservation genetics. Conservation Genetics. 2016;17: 1–17.
  127. 127. Shelton AO, Ramón-Laca A, Wells A, Clemons J, Chu D, Feist BE, et al. Environmental DNA provides quantitative estimates of Pacific hake abundance and distribution in the open ocean. Proceedings of the Royal Society B. 2022;289: 20212613–20212613. pmid:35317670
  128. 128. Shogren AJ, Tank JL, Egan SP, Bolster D, Riis T. Riverine distribution of mussel environmental DNA reflects a balance among density, transport, and removal processes. Freshwater Biology. 2019;64: 1467–1479.
  129. 129. Jarman SN, Berry O, Bunce M. The value of environmental DNA biobanking for long-term biomonitoring. Nature Ecology and Evolution. 2018;2: 1192–1193. pmid:29967486
  130. 130. Goodwin K, Certner R, Strom M, Arzayus F, Bohan M, Busch S, et al. NOAA ‘Omics White Paper: Informing the NOAA ‘Omics Strategy and Implementation Plan. 2020. Available: