There are no whole truths in meta-analyses: all their truths are half truths

In a recent letter, Thomsen & Wernberg (2015) rean-alyzed data compiled for our recent paper (Lyonset al., 2014). In that paper, we examined the effectsof macroalgal blooms and macroalgal mats on sevenimportant measures of community structure and eco-system functioning and explored several ecologicaland methodological factors that might explain someof the variation in the observed effects. Thomsen &Wernberg (2015) re-analyzed two small subsets of the data, focusing on experimental studies examining effects of blooms/mats on invertebrate abundance.Their analyses revealed two interesting patterns.First, they showed that macroalgal blooms reducedthe abundance of communities that Thomsen andWernberg categorized as ‘mainly infauna’, whileincreasing the abundance of communities categorized as ‘mainly epifauna’. Second, they showed that theimpacts of macroalgal blooms on ‘mainly infauna’communities increased with algal density in experiments that included multiple levels of algal density.These findings, as well as the conclusions that Thomsen & Wernberg (2015) draw from them, are largely consistent with our own expectations and interpretations. However, we also feel that some caution is required when interpreting the results of their analyses.

Horizon scanning for invasive alien species with the potential to threaten biodiversity in Great Britain

Invasive alien species (IAS) are considered one of the greatest threats to biodiversity, particularly through their interactions with other drivers of change. Horizon scanning, the systematic examination of future potential threats and opportunities, leading to prioritization of IAS threats is seen as an essential component of IAS management. Our aim was to consider IAS that were likely to impact on native biodiversity but were not yet established in the wild in Great Britain. To achieve this, we developed an approach which coupled consensus methods (which have previously been used for collaboratively identifying priorities in other contexts) with rapid risk assessment. The process involved two distinct phases: 1. Preliminary consultation with experts within five groups (plants, terrestrial invertebrates, freshwater invertebrates, vertebrates and marine species) to derive ranked lists of potential IAS. 2. Consensus-building across expert groups to compile and rank the entire list of potential IAS. Five hundred and ninety-one species not native to Great Britain were considered. Ninety-three of these species were agreed to constitute at least a medium risk (based on score and consensus) with respect to them arriving, establishing and posing a threat to native biodiversity. The quagga mussel, Dreissena rostriformis bugensis, received maximum scores for risk of arrival, establishment and impact; following discussions the unanimous consensus was to rank it in the top position. A further 29 species were considered to constitute a high risk and were grouped according to their ranked risk. The remaining 63 species were considered as medium risk, and included in an unranked long list. The information collated through this novel extension of the consensus method for horizon scanning provides evidence for underpinning and prioritizing management both for the species and, perhaps more importantly, their pathways of arrival. Although our study focused on Great Britain, we suggest that the methods adopted are applicable globally.

Ecological equivalence of species within phytoplankton functional groups

1.There are tens of thousands of species of phytoplankton found throughout the tree of life. Despite this diversity, phytoplankton are often aggregated into a few functional groups according to metabolic traits or biogeochemical role. We investigate the extent to which phytoplankton species dynamics are neutral within functional groups. 2.Seasonal dynamics in many regions of the ocean are known to affect phytoplankton at the functional group level leading to largely predictable patterns of seasonal succession. It is much more difficult to make general statements about the dynamics of individual species. 3.We use a 7 year time-series at station L4 in the Western English Channel with 57 diatom and 17 dinoflagellate species enumerated weekly to test if the abundance of diatom and dinoflagellate species vary randomly within their functional group envelope or if each species is driven uniquely by external factors. 4.We show that the total biomass of the diatom and dinoflagellate functional groups is well predicted by irradiance and temperature and quantify trait values governing the growth rate of both functional groups. The biomass dynamics of the functional groups are not neutral and each has their own distinct responses to environmental forcing. Compared to dinoflagellates, diatoms have faster growth rates, and grow faster under lower irradiance, cooler temperatures, and higher nutrient conditions. 5.The biomass of most species vary randomly within their functional group biomass envelope, most of the time. As a consequence, modelers will find it difficult to predict the biomass of most individual species. Our analysis supports the approach of using a single set of traits for a functional group and suggests that it should be possible to determine these traits from natural communities.

Evaluating machine-learning techniques for recruitment forecasting of seven North East Atlantic fish species

The effect of different factors (spawning biomass, environmental conditions) on recruitment is a subject of great importance in the management of fisheries, recovery plans and scenario exploration. In this study, recently proposed supervised classification techniques, tested by the machine-learning community, are applied to forecast the recruitment of seven fish species of North East Atlantic (anchovy, sardine, mackerel, horse mackerel, hake, blue whiting and albacore), using spawning, environmental and climatic data. In addition, the use of the probabilistic flexible naive Bayes classifier (FNBC) is proposed as modelling approach in order to reduce uncertainty for fisheries management purposes. Those improvements aim is to improve probability estimations of each possible outcome (low, medium and high recruitment) based in kernel density estimation, which is crucial for informed management decision making with high uncertainty. Finally, a comparison between goodness-of-fit and generalization power is provided, in order to assess the reliability of the final forecasting models. It is found that in most cases the proposed methodology provides useful information for management whereas the case of horse mackerel is an example of the limitations of the approach. The proposed improvements allow for a better probabilistic estimation of the different scenarios, i.e. to reduce the uncertainty in the provided forecasts.

An Objective Framework to Test the Quality of Candidate Indicators of Good Environmental Status

Large efforts are on-going within the EU to prepare the Marine Strategy Framework Directive’s (MSFD) assessment of the environmental status of the European seas. This assessment will only be as good as the indicators chosen to monitor the eleven descriptors of good environmental status (GEnS). An objective and transparent framework to determine whether chosen indicators actually support the aims of this policy is, however, not yet in place. Such frameworks are needed to ensure that the limited resources available to this assessment optimize the likelihood of achieving GEnS within collaborating states. Here, we developed a hypothesis-based protocol to evaluate whether candidate indicators meet quality criteria explicit to the MSFD, which the assessment community aspires to. Eight quality criteria are distilled from existing initiatives, and a testing and scoring protocol for each of them is presented. We exemplify its application in three worked examples, covering indicators for three GEnS descriptors (1, 5 and 6), various habitat components (seaweeds, seagrasses, benthic macrofauna and plankton), and assessment regions (Danish, Lithuanian and UK waters). We argue that this framework provides a necessary, transparent and standardized structure to support the comparison of candidate indicators, and the decision-making process leading to indicator selection. Its application could help identify potential limitations in currently available candidate metrics and, in such cases, help focus the development of more adequate indicators. Use of such standardized approaches will facilitate the sharing of knowledge gained across the MSFD parties despite context-specificity across assessment regions, and support the evidence-based management of European seas.

An Objective Framework to Test the Quality of Candidate Indicators of Good Environmental Status

Large efforts are on-going within the EU to prepare the Marine Strategy Framework Directive’s (MSFD) assessment of the environmental status of the European seas. This assessment will only be as good as the indicators chosen to monitor the eleven descriptors of good environmental status (GEnS). An objective and transparent framework to determine whether chosen indicators actually support the aims of this policy is, however, not yet in place. Such frameworks are needed to ensure that the limited resources available to this assessment optimize the likelihood of achieving GEnS within collaborating states. Here, we developed a hypothesis-based protocol to evaluate whether candidate indicators meet quality criteria explicit to the MSFD, which the assessment community aspires to. Eight quality criteria are distilled from existing initiatives, and a testing and scoring protocol for each of them is presented. We exemplify its application in three worked examples, covering indicators for three GEnS descriptors (1, 5 and 6), various habitat components (seaweeds, seagrasses, benthic macrofauna and plankton), and assessment regions (Danish, Lithuanian and UK waters). We argue that this framework provides a necessary, transparent and standardized structure to support the comparison of candidate indicators, and the decision-making process leading to indicator selection. Its application could help identify potential limitations in currently available candidate metrics and, in such cases, help focus the development of more adequate indicators. Use of such standardized approaches will facilitate the sharing of knowledge gained across the MSFD parties despite context-specificity across assessment regions, and support the evidence-based management of European seas.

Clustering in non-parametric multivariate analyses.

Non-parametric multivariate analyses of complex ecological datasets are widely used. Following appropriate pre-treatment of the data inter-sample resemblances are calculated using appropriate measures. Ordination and clustering derived from these resemblances are used to visualise relationships among samples (or variables). Hierarchical agglomerative clustering with group-average (UPGMA) linkage is often the clustering method chosen. Using an example dataset of zooplankton densities from the Bristol Channel and Severn Estuary, UK, a range of existing and new clustering methods are applied and the results compared. Although the examples focus on analysis of samples, the methods may also be applied to species analysis. Dendrograms derived by hierarchical clustering are compared using cophenetic correlations, which are also used to determine optimum  in flexible beta clustering. A plot of cophenetic correlation against original dissimilarities reveals that a tree may be a poor representation of the full multivariate information. UNCTREE is an unconstrained binary divisive clustering algorithm in which values of the ANOSIM R statistic are used to determine (binary) splits in the data, to form a dendrogram. A form of flat clustering, k-R clustering, uses a combination of ANOSIM R and Similarity Profiles (SIMPROF) analyses to determine the optimum value of k, the number of groups into which samples should be clustered, and the sample membership of the groups. Robust outcomes from the application of such a range of differing techniques to the same resemblance matrix, as here, result in greater confidence in the validity of a clustering approach.

Clustering in non-parametric multivariate analyses.

Non-parametric multivariate analyses of complex ecological datasets are widely used. Following appropriate pre-treatment of the data inter-sample resemblances are calculated using appropriate measures. Ordination and clustering derived from these resemblances are used to visualise relationships among samples (or variables). Hierarchical agglomerative clustering with group-average (UPGMA) linkage is often the clustering method chosen. Using an example dataset of zooplankton densities from the Bristol Channel and Severn Estuary, UK, a range of existing and new clustering methods are applied and the results compared. Although the examples focus on analysis of samples, the methods may also be applied to species analysis. Dendrograms derived by hierarchical clustering are compared using cophenetic correlations, which are also used to determine optimum  in flexible beta clustering. A plot of cophenetic correlation against original dissimilarities reveals that a tree may be a poor representation of the full multivariate information. UNCTREE is an unconstrained binary divisive clustering algorithm in which values of the ANOSIM R statistic are used to determine (binary) splits in the data, to form a dendrogram. A form of flat clustering, k-R clustering, uses a combination of ANOSIM R and Similarity Profiles (SIMPROF) analyses to determine the optimum value of k, the number of groups into which samples should be clustered, and the sample membership of the groups. Robust outcomes from the application of such a range of differing techniques to the same resemblance matrix, as here, result in greater confidence in the validity of a clustering approach.

Historical comparisons reveal multiple drivers of decadal change of an ecosystem engineer at the range edge

Biogenic reefs are important for habitat provision and coastal protection. Long-term datasets on the distribution and abundance of Sabellaria alveolata (L.) are available from Britain. The aim of this study was to combine historical records and contemporary data to (1) describe spatiotemporal variation in winter temperatures, (2) document short-term and long-term changes in the distribution and abundance of S. alveolata and discuss these changes in relation to extreme weather events and recent warming, and (3) assess the potential for artificial coastal defense structures to function as habitat for S. alveolata. A semi-quantitative abundance scale (ACFOR) was used to compare broadscale, long-term and interannual abundance of S. alveolata near its range edge in NW Britain. S. alveolata disappeared from the North Wales and Wirral coastlines where it had been abundant prior to the cold winter of 1962/1963. Population declines were also observed following the recent cold winters of 2009/2010 and 2010/2011. Extensive surveys in 2004 and 2012 revealed that S. alveolata had recolonized locations from which it had previously disappeared. Furthermore, it had increased in abundance at many locations, possibly in response to recent warming. S. alveolata was recorded on the majority of artificial coastal defense structures surveyed, suggesting that the proliferation of artificial coastal defense structures along this stretch of coastline may have enabled S. alveolata to spread across stretches of unsuitable natural habitat. Long-term and broadscale contextual monitoring is essential for monitoring responses of organisms to climate change. Historical data and gray literature can be invaluable sources of information. Our results support the theory that Lusitanian species are responding positively to climate warming but also that short-term extreme weather events can have potentially devastating widespread and lasting effects on organisms. Furthermore, the proliferation of coastal defense structures has implications for phylogeography, population genetics, and connectivity of coastal populations.

Historical comparisons reveal multiple drivers of decadal change of an ecosystem engineer at the range edge

Biogenic reefs are important for habitat provision and coastal protection. Long-term datasets on the distribution and abundance of Sabellaria alveolata (L.) are available from Britain. The aim of this study was to combine historical records and contemporary data to (1) describe spatiotemporal variation in winter temperatures, (2) document short-term and long-term changes in the distribution and abundance of S. alveolata and discuss these changes in relation to extreme weather events and recent warming, and (3) assess the potential for artificial coastal defense structures to function as habitat for S. alveolata. A semi-quantitative abundance scale (ACFOR) was used to compare broadscale, long-term and interannual abundance of S. alveolata near its range edge in NW Britain. S. alveolata disappeared from the North Wales and Wirral coastlines where it had been abundant prior to the cold winter of 1962/1963. Population declines were also observed following the recent cold winters of 2009/2010 and 2010/2011. Extensive surveys in 2004 and 2012 revealed that S. alveolata had recolonized locations from which it had previously disappeared. Furthermore, it had increased in abundance at many locations, possibly in response to recent warming. S. alveolata was recorded on the majority of artificial coastal defense structures surveyed, suggesting that the proliferation of artificial coastal defense structures along this stretch of coastline may have enabled S. alveolata to spread across stretches of unsuitable natural habitat. Long-term and broadscale contextual monitoring is essential for monitoring responses of organisms to climate change. Historical data and gray literature can be invaluable sources of information. Our results support the theory that Lusitanian species are responding positively to climate warming but also that short-term extreme weather events can have potentially devastating widespread and lasting effects on organisms. Furthermore, the proliferation of coastal defense structures has implications for phylogeography, population genetics, and connectivity of coastal populations.

Depth-related gradients in community structure and relatedness of bivalves and isopods in the Southern Ocean.

Despite increased research over the last decade, diversity patterns in Antarctic deep-sea benthic taxa and their driving forces are only marginally known. Depth-related patterns of diversity and distribution of isopods and bivalves collected in the Atlantic sector of the Southern Ocean are analysed. The data, sampled by epibenthic sledge at 40 deep-sea stations from the upper continental slope to the hadal zone (774 – 6348 m) over a wide area of the Southern Ocean, comprises 619 species of isopods and 81 species of bivalves,. There were more species of isopods than bivalves in all samples, and species per station varied from 2 to 85 for isopods and from 0 to 18 for bivalves. Most species were rare, with 72% of isopod species restricted to one or two stations, and 45% of bivalves. Among less-rare species bivalves tended to have wider distributions than isopods. The species richness of isopods varied with depth, showing a weak unimodal curve with a peak at 2000 – 4000 m, while the richness of bivalves did not. Multivariate analyses indicate that there are two main assemblages in the Southern Ocean, one shallow and one deep. These overlap over a large depth-range (2000 – 4000 m). Comparing analyses based on the Sørensen resemblance measure (presence/absence) and Γ+ (presence/absence incorporating relatedness among species) indicates that rare species tend to have other closely related species within the same depth band. Analysis of relatedness among species indicates that the taxonomic variety of bivalves tends to decline at depth, whereas that of isopods is maintained. This, it is speculated, may indicate that the available energy at depth is insufficient to maintain a range of bivalve life-history strategies

A new procedure to optimize the selection of groups in a classification tree

Agglomerative cluster analyses encompass many techniques, which have been widely used in various fields of science. In biology, and specifically ecology, datasets are generally highly variable and may contain outliers, which increase the difficulty to identify the number of clusters. Here we present a new criterion to determine statistically the optimal level of partition in a classification tree. The criterion robustness is tested against perturbated data (outliers) using an observation or variable with values randomly generated. The technique, called Random Simulation Test (RST), is tested on (1) the well-known Iris dataset [Fisher, R.A., 1936. The use of multiple measurements in taxonomic problems. Ann. Eugenic. 7, 179–188], (2) simulated data with predetermined numbers of clusters following Milligan and Cooper [Milligan, G.W., Cooper, M.C., 1985. An examination of procedures for determining the number of clusters in a data set. Psychometrika 50, 159–179] and finally (3) is applied on real copepod communities data previously analyzed in Beaugrand et al. [Beaugrand, G., Ibanez, F., Lindley, J.A., Reid, P.C., 2002. Diversity of calanoid copepods in the North Atlantic and adjacent seas: species associations and biogeography. Mar. Ecol. Prog. Ser. 232, 179–195]. The technique is compared to several standard techniques. RST performed generally better than existing algorithms on simulated data and proved to be especially efficient with highly variable datasets.