Towards reproducible metabarcoding data: Lessons from an international cross-laboratory experiment.

Advances in high-throughput sequencing (HTS) are revolutionizing monitoring in marine environments by enabling rapid, accurate and holistic detection of species within complex biological samples. Research institutions worldwide increasingly employ HTS methods for biodiversity assessments. However, variance in laboratory procedures, analytical workflows and bioinformatic pipelines impede the transferability and comparability of results across research groups. An international experiment was conducted to assess the consistency of metabarcoding results derived from identical samples and primer sets using varying laboratory procedures. Homogenized biofouling samples collected from four coastal locations (Australia, Canada, New Zealand and the USA) were distributed to 12 independent laboratories. Participants were asked to follow one of two HTS library preparation workflows. While DNA extraction, primers and bioinformatic analyses were purposefully standardized to allow comparison, many other technical variables were allowed to vary among laboratories (amplification protocols, type of instrument used, etc.). Despite substantial variation observed in raw results, the primary signal in the data was consistent, with the samples grouping strongly by geographical origin for all data sets. Simple post hoc data clean-up by removing low-quality samples gave the best improvement in sample classification for nuclear 18S rRNA gene data, with an overall 92.81% correct group attribution. For mitochondrial COI gene data, the best classification result (95.58%) was achieved after correction for contamination errors. The identified critical methodological factors that introduced the greatest variability (preservation buffer, sample defrosting, template concentration, DNA polymerase, PCR enhancer) should be of great assistance in standardizing future biodiversity studies using metabarcoding.

An evaluation of strategies for restoring a degraded New Zealand scallop fishery using stochastic dynamic simulation modelling.

A decision-theoretical approach was used to evaluate strategies to rebuild a severely depleted scallop (Pecten novaezelandiae) populations in the Tasman Bay and Golden Bay of New Zealand. These strategies were: no intervention, cessation of seabed bottom contact fishing, and reduction of sediment and nutrient runoff from surrounding land through on-farm practices. Our approach combined outputs of estimated effects of on-farm practices on erosion and nutrient reduction with a stochastic dynamic model of the scallop populations. The most effective individual intervention is eliminating bottom contact fishing through dredging and trawling which increased scallop biomass on average by 73% compared to the no intervention scenario. Although on-farm practices have reduced sedimentation and nutrient runoff significantly (28-36% and 2%, respectively), they have no effect on scallop biomass if implemented individually and led to only marginal improvements in scallop biomass if implemented alongside cessation of bottom contact fishing (2-4%). Although our results showed, on average, substantial recovery in the scallop population when reducing seabed bottom contact and water pollution, the large uncertainty boundaries makes it unclear whether these improvements would be realized. The long-term success of such strategies will depend on the available habitat being able to sustain high densities of healthy scallop adults and recruits, a situation that has been posited in our analysis. Where scallop juvenile survival is compromised by sedimentation, nutrient pollution, or other exogenous influences, proposed interventions may be insufficient to aid recovery.

The risks of using molecular biodiversity data for incidental detection of species of concern.

Incidental detection of species of concern (e.g., invasive species, pathogens, threatened and endangered species) during biodiversity assessments based on high-throughput DNA sequencing holds significant risks in the absence of rigorous, fit-for-purpose data quality and reporting standards. Molecular biodiversity data are predominantly collected for ecological studies and thus are generated to common quality assurance standards. However, the detection of certain species of concern in these data would likely elicit interest from end users working in biosecurity or other surveillance contexts (e.g., pathogen detection in health-related fields), for which more stringent quality control standards are essential to ensure that data are suitable for informing decision-making and can withstand legal or political challenges. We suggest here that data quality and reporting criteria are urgently needed to enable clear identification of those studies that may be appropriately applied to surveillance contexts. In the interim, more pointed disclaimers on uncertainties associated with the detection and identification of species of concern may be warranted in published studies. This is not only to ensure the utility of molecular biodiversity data for consumers, but also to protect data generators from uncritical and potentially ill-advised application of their science in decision-making.

Trends in the detection of aquatic non-indigenous species across global marine, estuarine and freshwater ecosystems: A 50-year perspective.

Aim: The introduction of aquatic non-indigenous species (ANS) has become a major driver for global changes in species biogeography. We examined spatial patterns and temporal trends of ANS detections since 1965 to inform conservation policy and management. Location: Global. Methods: We assembled an extensive dataset of first records of detection of ANS (1965-2015) across 49 aquatic ecosystems, including the (a) year of first collection, (b) population status and (c) potential pathway(s) of introduction. Data were analysed at global and regional levels to assess patterns of detection rate, richness and transport pathways. Results: An annual mean of 43 (±16 SD) primary detections of ANS occurred-one new detection every 8.4 days for 50 years. The global rate of detections was relatively stable during 1965-1995, but increased rapidly after this time, peaking at roughly 66 primary detections per year during 2005-2010 and then declining marginally. Detection rates were variable within and across regions through time. Arthropods, molluscs and fishes were the most frequently reported ANS. Most ANS were likely introduced as stowaways in ships' ballast water or biofouling, although direct evidence is typically absent. Main conclusions: This synthesis highlights the magnitude of recent ANS detections, yet almost certainly represents an underestimate as many ANS go unreported due to limited search effort and diminishing taxonomic expertise. Temporal rates of detection are also confounded by reporting lags, likely contributing to the lower detection rate observed in recent years. There is a critical need to implement standardized, repeated methods across regions and taxa to improve the quality of global-scale comparisons and sustain core measures over longer time-scales. It will be fundamental to fill in knowledge gaps given that invasion data representing broad regions of the world's oceans are not yet readily available and to maintain knowledge pipelines for adaptive management.

Combining morpho-taxonomy and metabarcoding enhances the detection of non-indigenous marine pests in biofouling communities.

Marine infrastructure can favor the spread of non-indigenous marine biofouling species by providing a suitable habitat for them to proliferate. Cryptic organisms or those in early life stages can be difficult to distinguish by conventional morphological taxonomy. Molecular tools, such as metabarcoding, may improve their detection. In this study, the ability of morpho-taxonomy and metabarcoding (18S rRNA and COI) using three reference databases (PR2, BOLD and NCBI) to characterize biodiversity and detect non-indigenous species (NIS) in biofouling was compared on 60 passive samplers deployed over summer and winter in a New Zealand marina. Highest resolution of metazoan taxa was identified using 18S rRNA assigned to PR2. There were higher assignment rates to NCBI reference sequences, but poorer taxonomic identification. Using all methods, 48 potential NIS were identified. Metabarcoding detected the largest proportion of those NIS: 77% via 18S rRNA/PR2 and NCBI and 35% via COI/BOLD and NCBI. Morpho-taxonomy detected an additional 14% of all identified NIS comprising mainly of bryozoan taxa. The data highlight several on-going challenges, including: differential marker resolution, primer biases, incomplete sequence reference databases, and variations in bioinformatic pipelines. Combining morpho-taxonomy and molecular analysis methods will likely enhance the detection of NIS from complex biofouling.

Development of a real-time polymerase chain reaction assay for the detection of the invasive Mediterranean fanworm, Sabella spallanzanii, in environmental samples.

The Mediterranean fanworm, Sabella spallanzanii Gmelin 1791, was first detected in the Southern Hemisphere in the 1990s and is now abundant in many parts of southern Australia and in several locations around northern New Zealand. Once established, it can proliferate rapidly, reaching high densities with potential ecological and economic impacts. Early detection of new S. spallanzanii incursions is important to prevent its spread, guide eradication or control efforts and to increase knowledge on the species' dispersal pathways. In this study, we developed a TaqMan probe real-time polymerase chain reaction assay targeting a region of the mitochondrial cytochrome oxidase I gene. The assay was validated in silico and in vitro using DNA from New Zealand and Australian Sabellidae with no cross-reactivity detected. The assay has a linear range of detection over seven orders of magnitude with a limit of detection reached at 12.4 × 10-4 ng/µL of DNA. We analysed 145 environmental (water, sediment and biofouling) samples and obtained positive detections only from spiked samples and those collected at a port where S. spallanzanii is known to be established. This assay has the potential to enhance current morphological and molecular-based methods, through its ability to rapidly and accurately identify S. spallanzanii in environmental samples.

Invasion Expansion: Time since introduction best predicts global ranges of marine invaders.

Strategies for managing biological invasions are often based on the premise that characteristics of invading species and the invaded environment are key predictors of the invader's distribution. Yet, for either biological traits or environmental characteristics to explain distribution, adequate time must have elapsed for species to spread to all potential habitats. We compiled and analyzed a database of natural history and ecological traits of 138 coastal marine invertebrate species, the environmental conditions at sites to which they have been introduced, and their date of first introduction. We found that time since introduction explained the largest fraction (20%) of the variability in non-native range size, while traits of the species and environmental variables had significant, but minimal, influence on non-native range size. The positive relationship between time since introduction and range size indicates that non-native marine invertebrate species are not at equilibrium and are still spreading, posing a major challenge for management of coastal ecosystems.

Using temporal sampling to improve attribution of source populations for invasive species.

Numerous studies have applied genetic tools to the identification of source populations and transport pathways for invasive species. However, there are many gaps in the knowledge obtained from such studies because comprehensive and meaningful spatial sampling to meet these goals is difficult to achieve. Sampling populations as they arrive at the border should fill the gaps in source population identification, but such an advance has not yet been achieved with genetic data. Here we use previously acquired genetic data to assign new incursions as they invade populations within New Zealand ports and marinas. We also investigated allelelic frequency change in these recently established populations over a two-year period, and assessed the effect of temporal genetic sampling on our ability to assign new incursions to their population of source. We observed shifts in the allele frequencies among populations, as well as the complete loss of some alleles and the addition of alleles novel to New Zealand, within these recently established populations. There was no significant level of genetic differentiation observed in our samples between years, and the use of these temporal data did alter the assignment probability of new incursions. Our study further suggests that new incursions can add genetic variation to the population in a single introduction event as the founders themselves are often more genetically diverse than theory initially predicted.

Using habitat suitability index and particle dispersion models for early detection of marine invaders.

Eradication and control of invasive species are often possible only if populations are detected when they are small and localized. To be efficient, detection surveys should be targeted at locations where there is the greatest risk of incursions. We examine the utility of habitat suitability index (HSI) and particle dispersion models for targeting sampling for marine pests. Habitat suitability index models are a simple way to identify suitable habitat when species distribution data are lacking. We compared the performance of HSI models with statistical models derived from independent data from New Zealand on the distribution of two nonindigenous bivalves: Theora lubrica and Musculista senhousia. Logistic regression models developed using the HSI scores as predictors of the presence/absence of Theora and Musculista explained 26.7% and 6.2% of the deviance in the data, respectively. Odds ratios for the HSI scores were greater than unity, indicating that they were genuine predictors of the presence/ absence of each species. The fit and predictive accuracy of each logistic model were improved when simulated patterns of dispersion from the nearest port were added as a predictor variable. Nevertheless, the combined model explained, at best, 46.5% of the deviance in the distribution of Theora and correctly predicted 56% of true presences and 50% of all cases. Omission errors were between 6% and 16%. Although statistical distribution models built directly from environmental predictors always outperformed the equivalent HSI models, the gain in model fit and accuracy was modest. High residual deviance in both types of model suggests that the distributions realized by Theora and Musculista in the field data were influenced by factors not explicitly modeled as explanatory variables and by error in the environmental data used to project suitable habitat for the species. Our results highlight the difficulty of accurately predicting the distribution of invasive marine species that exhibit low habitat occupancy and patchy distributions in time and space. Although the HSI and statistical models had utility as predictors of the likely distribution of nonindigenous marine species, the level of spatial accuracy achieved with them may be well below expectations for sensitive surveillance programs.

A risk-based predictive tool to prevent accidental introductions of nonindigenous marine species.

Preventing the introduction of nonindigenous species (NIS) is the most efficient way to avoid the costs and impacts of biological invasions. The transport of fouling species on ship hulls is an important vector for the introduction of marine NIS. We use quantitative risk screening techniques to develop a predictive tool of the abundance and variety of organisms being transported by ocean-going yachts. We developed and calibrated an ordinal rank scale of the abundance of fouling assemblages on the hulls of international yacht hulls arriving in New Zealand. Fouling ranks were allocated to 783 international yachts that arrived in New Zealand between 2002 and 2004. Classification tree analysis was used to identify relationships between the fouling ranks and predictor variables that described the maintenance and travel history of the yachts. The fouling ranks provided reliable indications of the actual abundance and variety of fouling assemblages on the yachts and identified most (60%) yachts that had fouling on their hulls. However, classification tree models explained comparatively little of the variation in the distribution of fouling ranks (22.1%), had high misclassification rates (approximately 43%), and low predictive power. In agreement with other studies, the best model selected the age of the toxic antifouling paint on yacht hulls as the principal risk factor for hull fouling. Our study shows that the transport probability of fouling organisms is the result of a complex suite of interacting factors and that large sample sizes will be needed for calibration of robust risk models.

Effects of Self-Guided Snorkeling Trails on Corals in a Tropical Marine Park.

Underwater trails are intended as interpretative tools in marine parks, but concentrating divers and snorkelers in defined areas may negatively affect the surrounding environment. We examined spatial and temporal patterns in the effects of use of underwater trails on coral reef flats in the Great Barrier Reef Marine Park, Australia. Changes in benthic assemblages were assessed on two new trails used by snorkelers, two unused (control) trails, and two undisturbed areas. Total percent coral cover, numbers of broken colonies, and living coral fragments were counted 6 months before and 6 months after the new trails began to be used. Spatial patterns of effects around concentrated nodes of use were determined by stratified sampling around and away from the interpretative signs within each trail. Despite comparatively low levels of use (approximately 15 snorkelers per trail per week), snorkelers caused significant damage to corals along the trails. Branching corals (non-Acropora branching corals and Millepora spp.) were most affected. More damage occurred near the interpretative signs than elsewhere on the trails. The numbers of broken branches and damaged coral colonies in the snorkeling trails increased rapidly but stabilized within 2 months of the commencement of use. There was no significant change in overall benthic assemblages within the trails after 6 months of use by snorkelers. Although concentrating snorkelers within confined trails caused increased damage to corals, the effects can be mitigated by appropriate design and placement of the trails and by managing the behavior of snorkelers. Interpretative information should warn users about the damage they may cause when swimming along the trails. Managing the behavior of snorkelers in the water is likely to be more effective in reducing damage than simply applying fixed limits to the amount of use the trails receive.

RESUMEN: Los senderos submarinos tienen la intención de servir como herramientas interpretativas en los parques marinos, pero la concentración de buzos y buceadores libres en áreas definidas puede tener un efecto negativo en el ambiente de los alrededores. Examinamos los patrones temporales y espaciales de los efectos del uso de senderos submarinos en relieves de arrecifes de coral en el arrecife del Parque Marino de la Gran Barrera de Arrecifes en Australia. Los cambios en los ensamblajes bénticos fueron evaluados en dos senderos nuevos usados por buceadores libres, dos senderos sin usar (controles) y dos áreas sin perturbar. Evaluamos el porcentaje de cobertura de coral, el número de colonias rotas, y fragmentos de coral vivo 6 meses antes y seis meses después de que se iniciara el uso de los senderos nuevos. Los patrones espaciales de los efectos alrededor de nódulos concentrados de uso fueron determinados por un muestreo estratificado alrededor y en sitios lejanos de las señales interpretativas de cada sendero. A pesar de los niveles significativamente bajos de uso (∼15 buceadores libres/sendero/semana), los buceadores libres causaron un daño significativo a los corales a lo largo de los senderos. Los corales ramificados (spp. de corales que no pertenecen a Acropora o Millepora) fueron los más afectados. Se observó más daño cerca de las señales interpretativas que en cualquier otra parte del sendero. Los números de ramas rotas y de colonias de coral dañadas en los senderos se incrementaron rápidamente, pero se estabilizaron a los dos meses de haber iniciado el uso de los senderos. No hubo un cambio significativo dentro de los senderos después de seis meses de uso. A pesar de que la concentración de buceadores libres dentro de senderos confinados incrementó el daño a los corales, los efectos pueden ser mitigados mediante el diseño y la ubicación apropiada de los senderos y mediante el manejo de la conducta de los buceadores libres. La información interpretativa debería prevenir a los usuarios sobre el daño que ellos mismos pueden ocasionar cuando nadan en estos senderos. El manejo de la conducta de los buceadores libres en el agua probablemente sea más efectivo en la reducción del daño que la simple aplicación de límites fijos a la cantidad de uso que un sendero puede recibir.