International Council for the Exploration of the Sea

Count data can be analyzed using generalized linear mixed models when observations are correlated in ways that require random effects. However, count data are often zero-inflated, containing more zeros than would be expected from the typical error distributions. We present a new package, glmmTMB, and compare it to other R packages that fit zero-inflated mixed models. The glmmTMB package fits many types of GLMMs and extensions, including models with continuously distributed responses, but here we focus on count responses. glmmTMB is faster than glmmADMB, MCMCglmm, and brms, and more flexible than INLA and mgcv for zero-inflated modeling. One unique feature of glmmTMB (among packages that fit zero-inflated mixed models) is its ability to estimate the Conway-Maxwell-Poisson distribution parameterized by the mean. Overall, its most appealing features for new users may be the combination of speed, flexibility, and its interface's similarity to lme4.

Summary Metacommunity ecology addresses the situation where sets of local communities are connected by the dispersal of a number of potentially interacting species. Aquatic systems (e.g. lentic versus lotic versus marine) differ from each other in connectivity and environmental heterogeneity, suggesting that metacommunity organisation also differs between major aquatic systems. Here, we review findings from observational field studies on metacommunity organisation in aquatic systems. Species sorting (i.e. species are ‘filtered’ by environmental factors and occur only at environmentally suitable sites) prevails in aquatic systems, particularly in streams and lakes, but the degree to which dispersal limitation interacts with such environmental control varies among different systems and spatial scales. For example, mainstem rivers and marine coastal systems may be strongly affected by ‘mass effects’ (i.e. where high dispersal rates homogenise communities to some degree at neighbouring localities, irrespective of their abiotic and biotic environmental conditions), whereas isolated lakes and ponds may be structured by dispersal limitation (i.e. some species do not occur at otherwise‐suitable localities simply because sites with potential colonists are too far away). Flow directionality in running waters also differs from water movements in other systems, and this difference may also have effects on the role of dispersal in different aquatic systems. Dispersal limitation typically increases with increasing spatial distance between sites, mass effects potentially increase in importance with decreasing distance between sites, and the dispersal ability of organisms may determine the spatial extents at which species sorting and dispersal processes are most important. A better understanding of the relative roles of species sorting, mass effects and dispersal limitation in affecting aquatic metacommunities requires the following: (i) characterising dispersal rates more directly or adopting better proxies than have been used previously; (ii) considering the nature of aquatic networks; (iii) combining correlative and experimental approaches; (iv) exploring temporal aspects of metacommunity organisation and (v) applying past approaches and statistical methods innovatively for increasing our understanding of metacommunity organisation.

Current global fisheries production of approximately 160 million tons is rising as a result of increases in aquaculture production. A number of climate-related threats to both capture fisheries and aquaculture are identified, but we have low confidence in predictions of future fisheries production because of uncertainty over future global aquatic net primary production and the transfer of this production through the food chain to human consumption. Recent changes in the distribution and productivity of a number of fish species can be ascribed with high confidence to regional climate variability, such as the El Niño-Southern Oscillation. Future production may increase in some high-latitude regions because of warming and decreased ice cover, but the dynamics in low-latitude regions are governed by different processes, and production may decline as a result of reduced vertical mixing of the water column and, hence, reduced recycling of nutrients. There are strong interactions between the effects of fishing and the effects of climate because fishing reduces the age, size, and geographic diversity of populations and the biodiversity of marine ecosystems, making both more sensitive to additional stresses such as climate change. Inland fisheries are additionally threatened by changes in precipitation and water management. The frequency and intensity of extreme climate events is likely to have a major impact on future fisheries production in both inland and marine systems. Reducing fishing mortality in the majority of fisheries, which are currently fully exploited or overexploited, is the principal feasible means of reducing the impacts of climate change.

Evolutionary impact assessment is a framework for quantifying the effects of harvest-induced evolution on the utility generated by fish stocks. CREDIT: N. KEVITIYAGALA/SCIENCE Darwinian evolution is the driving process of innovation and adaptation across the world’s biota. Acting on top of natural selection, human-induced selection pressures can also cause rapid evolution. Sometimes such evolution has undesirable consequences, one example being the spreading resistance to antibiotics and pesticides, which causes suffering and billion-dollar losses annually (1). A comparable anthropogenic selection pressure originates from fishing, which has become the main source of mortality in many fish stocks, and may exceed 1

Researchers frequently use automated model selection methods such as backwards elimination to identify variables that are independent predictors of an outcome under consideration. We propose using bootstrap resampling in conjunction with automated variable selection methods to develop parsimonious prediction models. Using data on patients admitted to hospital with a heart attack, we demonstrate that selecting those variables that were identified as independent predictors of mortality in at least 60%% of the bootstrap samples resulted in a parsimonious model with excellent predictive ability.

While the physical dimensions of climate change are now routinely assessed through multimodel intercomparisons, projected impacts on the global ocean ecosystem generally rely on individual models with a specific set of assumptions. To address these single-model limitations, we present standardized ensemble projections from six global marine ecosystem models forced with two Earth system models and four emission scenarios with and without fishing. We derive average biomass trends and associated uncertainties across the marine food web. Without fishing, mean global animal biomass decreased by 5% (±4% SD) under low emissions and 17% (±11% SD) under high emissions by 2100, with an average 5% decline for every 1 °C of warming. Projected biomass declines were primarily driven by increasing temperature and decreasing primary production, and were more pronounced at higher trophic levels, a process known as trophic amplification. Fishing did not substantially alter the effects of climate change. Considerable regional variation featured strong biomass increases at high latitudes and decreases at middle to low latitudes, with good model agreement on the direction of change but variable magnitude. Uncertainties due to variations in marine ecosystem and Earth system models were similar. Ensemble projections performed well compared with empirical data, emphasizing the benefits of multimodel inference to project future outcomes. Our results indicate that global ocean animal biomass consistently declines with climate change, and that these impacts are amplified at higher trophic levels. Next steps for model development include dynamic scenarios of fishing, cumulative human impacts, and the effects of management measures on future ocean biomass trends.

On the basis of synchronization of three carbon-14 (14C)-dated lacustrine sequences from Sweden with tree ring and ice core records, the absolute age of the Younger Dryas-Preboreal climatic shift was determined to be 11,450 to 11,390 +/- 80 years before the present. A 150-year-long cooling in the early Preboreal, associated with rising Delta14C values, is evident in all records and indicates an ocean ventilation change. This cooling is similar to earlier deglacial coolings, and box-model calculations suggest that they all may have been the result of increased freshwater forcing that inhibited the strength of the North Atlantic heat conveyor, although the Younger Dryas may have begun as an anomalous meltwater event.

Abstract Ecological phenomena are often measured in the form of count data. These data can be analyzed using generalized linear mixed models (GLMMs) when observations are correlated in ways that require random effects. However, count data are often zero-inflated , containing more zeros than would be expected from the standard error distributions used in GLMMs, e.g., parasite counts may be exactly zero for hosts with effective immune defenses but vary according to a negative binomial distribution for non-resistant hosts. We present a new R package, glmmTMB , that increases the range of models that can easily be fitted to count data using maximum likelihood estimation. The interface was developed to be familiar to users of the lme4 R package, a common tool for fitting GLMMs. To maximize speed and flexibility, estimation is done using Template Model Builder ( TMB ), utilizing automatic differentiation to estimate model gradients and the Laplace approximation for handling random effects. We demonstrate glmm TMB and compare it to other available methods using two ecological case studies. In general, glmm TMB is more flexible than other packages available for estimating zero-inflated models via maximum likelihood estimation and is faster than packages that use Markov chain Monte Carlo sampling for estimation; it is also more flexible for zero-inflated modelling than INLA , but speed comparisons vary with model and data structure. Our package can be used to fit GLMs and GLMMs with or without zero-inflation as well as hurdle models. By allowing ecologists to quickly estimate a wide variety of models using a single package, glmm TMB makes it easier to find appropriate models and test hypotheses to describe ecological processes.

We propose a model for growing networks based on a finite memory of the nodes. The model shows stylized features of real-world networks: power-law distribution of degree, linear preferential attachment of new links, and a negative correlation between the age of a node and its link attachment rate. Notably, the degree distribution is conserved even though only the most recently grown part of the network is considered. As the network grows, the clustering reaches an asymptotic value larger than that for regular lattices of the same average connectivity and similar to the one observed in the networks of movie actors, coauthorship in science, and word synonyms. These highly clustered scale-free networks indicate that memory effects are crucial for a correct description of the dynamics of growing networks.

Bottom trawling is the most widespread human activity affecting seabed habitats. Here, we collate all available data for experimental and comparative studies of trawling impacts on whole communities of seabed macroinvertebrates on sedimentary habitats and develop widely applicable methods to estimate depletion and recovery rates of biota after trawling. Depletion of biota and trawl penetration into the seabed are highly correlated. Otter trawls caused the least depletion, removing 6% of biota per pass and penetrating the seabed on average down to 2.4 cm, whereas hydraulic dredges caused the most depletion, removing 41% of biota and penetrating the seabed on average 16.1 cm. Median recovery times posttrawling (from 50 to 95% of unimpacted biomass) ranged between 1.9 and 6.4 y. By accounting for the effects of penetration depth, environmental variation, and uncertainty, the models explained much of the variability of depletion and recovery estimates from single studies. Coupled with large-scale, high-resolution maps of trawling frequency and habitat, our estimates of depletion and recovery rates enable the assessment of trawling impacts on unprecedented spatial scales.

In the context of growing networks, we introduce a simple dynamical model that unifies the generic features of real networks: scale-free distribution of degree and the small-world effect. While the average shortest path length increases logarithmically as in random networks, the clustering coefficient assumes a large value independent of system size. We derive analytical expressions for the clustering coefficient in two limiting cases: random [C approximately (ln N)(2)/N] and highly clustered (C=5/6) scale-free networks.

European eels (Anguilla anguilla) undertake a approximately 5000-kilometer (km) spawning migration from Europe to the Sargasso Sea. The larvae are transported back to European waters by the Gulf Stream and North Atlantic Drift. However, details of the spawning migration remain unknown because tracking eels in the Atlantic Ocean has, so far, eluded study. Recent advances in satellite tracking enable investigation of migratory behavior of large ocean-dwelling animals. However, sizes of available tags have precluded tracking smaller animals like European eels. Here, we present information about the swimming direction, depth, and migratory behavior of European eels during spawning migration, based on a miniaturized pop-up satellite archival transmitter. Although the tagging experiment fell short of revealing the full migration to the Sargasso Sea, the data covered the first 1300 km and provided unique insights.

study area was trawled, and 86% was not trawled. Trawling activity was aggregated; the most intensively trawled areas accounting for 90% of activity comprised 77% of footprint on average. Regional swept area ratio (SAR; ratio of total swept area trawled annually to total area of region, a metric of trawling intensity) and footprint area were related, providing an approach to estimate regional trawling footprints when high-resolution spatial data are unavailable. If SAR was ≤0.1, as in 8 of 24 regions, there was >95% probability that >90% of seabed was not trawled. If SAR was 7.9, equal to the highest SAR recorded, there was >95% probability that >70% of seabed was trawled. Footprints were smaller and SAR was ≤0.25 in regions where fishing rates consistently met international sustainability benchmarks for fish stocks, implying collateral environmental benefits from sustainable fishing.

Reconstructions of Mediterranean ocean temperature fields back to 1950 show a proxy relationship between heat content changes in the North Atlantic and the Western Mediterranean Deep Water (WMDW) formed in the Gulf of Lions in winter, because of consistent air‐sea heat fluxes over these areas, strongly correlated to the North Atlantic Oscillation (NAO).

Plastics and other artificial materials pose new risks to the health of the ocean. Anthropogenic debris travels across large distances and is ubiquitous in the water and on shorelines, yet, observations of its sources, composition, pathways, and distributions in the ocean are very sparse and inaccurate. Total amounts of plastics and other man-made debris in the ocean and on the shore, temporal trends in these amounts under exponentially increasing production, as well as degradation processes, vertical fluxes, and time scales are largely unknown. Present ocean circulation models are not able to accurately simulate drift of debris because of its complex hydrodynamics. In this paper we discuss the structure of the future integrated marine debris observing system (IMDOS) that is required to provide long-term monitoring of the state of this anthropogenic pollution and support operational activities to mitigate impacts on the ecosystem and on the safety of maritime activity. The proposed observing system integrates remote sensing and in situ observations. Also, models are used to optimize the design of the system and, in turn, they will be gradually improved using the products of the system. Remote sensing technologies will provide spatially coherent coverage and consistent surveying time series at local to global scale. Optical sensors, including high-resolution imaging, multi-and hyperspectral, fluorescence, and Raman technologies, as well as SAR will be used to measure different types of debris. They will be implemented in a variety of platforms, from hand-held tools to ship-, buoy-, aircraft-, and satellite-based sensors. A network of in situ observations, including reports from volunteers, citizen scientists and ships of opportunity, will be developed to provide data for calibration/validation of remote sensors and to monitor the spread of plastic pollution and other marine debris. IMDOS will interact with other observing systems monitoring physical, chemical, and biological processes in the ocean and on shorelines as well as the state of the ecosystem, maritime activities and safety, drift of sea ice, etc. The synthesized data will support innovative multi-disciplinary research and serve a diverse community of users.

Protoplanetary disks are believed to accrete onto their central T Tauri star because of magnetic stresses. Recently published shearing box simulations indicate that Ohmic resistivity, ambipolar diffusion and the Hall effect all play important roles in disk evolution. In the presence of a vertical magnetic field, the disk remains laminar between 1-5au, and a magnetocentrifugal disk wind forms that provides an important mechanism for removing angular momentum. Questions remain, however, about the establishment of a true physical wind solution in the shearing box simulations because of the symmetries inherent in the local approximation. We present global MHD simulations of protoplanetary disks that include Ohmic resistivity and ambipolar diffusion, where the time-dependent gas-phase electron and ion fractions are computed under FUV and X-ray ionization with a simplified recombination chemistry. Our results show that the disk remains laminar, and that a physical wind solution arises naturally in global disk models. The wind is sufficiently efficient to explain the observed accretion rates. Furthermore, the ionization fraction at intermediate disk heights is large enough for magneto-rotational channel modes to grow and subsequently develop into belts of horizontal field. Depending on the ionization fraction, these can remain quasi-global, or break-up into discrete islands of coherent field polarity. The disk models we present here show a dramatic departure from our earlier models including Ohmic resistivity only. It will be important to examine how the Hall effect modifies the evolution, and to explore the influence this has on the observational appearance of such systems, and on planet formation and migration.

Abstract Application of environmental DNA (e DNA ) analysis has attracted the attention of researchers, advisors and managers of living marine resources and biodiversity. The apparent simplicity and cost‐effectiveness of e DNA analysis make it highly attractive as species distributions can be revealed from water samples. Further, species‐specific analyses indicate that e DNA concentrations correlate with biomass and abundance, suggesting the possibility for quantitative applications estimating abundance and biomass of specific organisms in marine ecosystems, such as for stock assessment. However, the path from detecting occurrence of an organism to quantitative estimates is long and indirect, not least as e DNA concentration depends on several physical, chemical and biological factors which influence its production, persistence and transport in marine ecosystems. Here, we provide an overview of basic principles in relation to e DNA analysis with potential for marine fisheries application. We describe fundamental processes governing e DNA generation, breakdown and transport and summarize current uncertainties about these processes. We describe five major challenges in relation to application in fisheries assessment, where there is immediate need for knowledge building in marine systems, and point to apparent weaknesses of e DNA compared to established marine fisheries monitoring methods. We provide an overview of emerging applications of interest to fisheries management and point to recent technological advances, which could improve analysis efficiency. We advise precaution against exaggerating the present scope for application of e DNA analysis in fisheries monitoring, but also argue that with informed insights into strengths and limitations, e DNA analysis can become an integrated tool in fisheries assessment and management.

Abstract Recent advances in the application of stock identification methods have revealed inconsistencies between the spatial structure of biological populations and the definition of stock units used in assessment and management. From a fisheries management perspective, stocks are typically assumed to be discrete units with homogeneous vital rates that can be exploited independently of each other. However, the unit stock assumption is often violated leading to spatial mismatches that can bias stock assessment and impede sustainable fisheries management. The primary ecological concern is the potential for overexploitation of unique spawning components, which can lead to loss of productivity and reduced biodiversity along with destabilization of local and regional stock dynamics. Furthermore, ignoring complex population structure and stock connectivity can lead to misperception of the magnitude of fish productivity, which can translate to suboptimal utilization of the resource. We describe approaches that are currently being applied to improve the assessment and management process for marine fish in situations where complex spatial structure has led to an observed mismatch between the scale of biological populations and spatially-defined stock units. The approaches include: (i) status quo management, (ii) “weakest link” management, (iii) spatial and temporal closures, (iv) stock composition analysis, and (v) alteration of stock boundaries. We highlight case studies in the North Atlantic that illustrate each approach and synthesize the lessons learned from these real-world applications. Alignment of biological and management units requires continual monitoring through the application of stock identification methods in conjunction with responsive management to preserve biocomplexity and the natural stability and resilience of fish species.

Censored quantile regression offers a valuable supplement to Cox propor-tional hazards model for survival analysis. Existing work in the literature of-ten requires stringent assumptions, such as unconditional independence of the survival time and the censoring variable or global linearity at all quantile lev-els. Moreover, some of the work use recursive algorithms making it challeng-ing to derive asymptotic normality. To overcome these drawbacks, we propose a new locally weighted censored quantile regression approach that adopts the redistribution-of-mass idea and employs a local reweighting scheme. Its validity only requires conditional independence of the survival time and the censoring variable given the covariates, and linearity at the particular quantile level of interest. Our method leads to a simple algorithm that can be conveniently im-plemented with R software. Applying recent theory of M-estimation with infinite dimensional parameters, we establish the consistency and asymptotic normality of the proposed estimator. The proposed method is studied via simulations and is illustrated with the analysis of an acute myocardial infarction dataset.

European eels (Anguilla anguilla) spawn in the remote Sargasso Sea in partial sympatry with American eels (Anguilla rostrata), and juveniles are transported more than 5000 km back to the European and North African coasts. The two species have been regarded as classic textbook examples of panmixia, each comprising a single, randomly mating population. However, several recent studies based on continental samples have found subtle, but significant, genetic differentiation, interpreted as geographical or temporal heterogeneity between samples. Moreover, European and American eels can hybridize, but hybrids have been observed almost exclusively in Iceland, suggesting hybridization in a specific region of the Sargasso Sea and subsequent nonrandom dispersal of larvae. Here, we report the first molecular population genetics study based on analysis of 21 microsatellite loci in larvae of both Atlantic eel species sampled directly in the spawning area, supplemented by analysis of European glass eel samples. Despite a clear East-West gradient in the overlapping distribution of the two species in the Sargasso Sea, we only observed a single putative hybrid, providing evidence against the hypothesis of a wide marine hybrid zone. Analyses of genetic differentiation, isolation by distance, isolation by time and assignment tests provided strong evidence for panmixia in both the Sargasso Sea and across all continental samples of European eel after accounting for the presence of sibs among newly hatched larvae. European eel has declined catastrophically, and our findings call for management of the species as a single unit, necessitating coordinated international conservation efforts.