Predictive models are central to many scientific disciplines and vital for informing management in a rapidly changing world. However, limited understanding of the accuracy and precision of models transferred to novel conditions (their 'transferability') undermines confidence in their predictions. Here, 50 experts identified priority knowledge gaps which, if filled, will most improve model transfers. These are summarized into six technical and six fundamental challenges, which underlie the combined need to intensify research on the determinants of ecological predictability, including species traits and data quality, and develop best practices for transferring models. Of high importance is the identification of a widely applicable set of transferability metrics, with appropriate tools to quantify the sources and impacts of prediction uncertainty under novel conditions. Predicting the UnknownPredictions facilitate the formulation of quantitative, testable hypotheses that can be refined and validated empirically [1]. Predictive models have thus become ubiquitous in numerous scientific disciplines, including ecology [2], where they provide means for mapping species distributions, explaining population trends, or quantifying the risks of biological invasions and disease outbreaks (e.g., [3,4]). The practical value of predictive models in supporting policy and decision making has therefore grown rapidly (Box 1) [5]. With that has come an increasing desire to predict (see Glossary) the state of ecological features (e.g., species, habitats) and our likely impacts upon them [5], prompting a shift from explanatory models to anticipatory predictions [2]. However, in many situations, severe data deficiencies preclude the development of specific models, and the collection of new data can be prohibitively costly or simply impossible [6]. It is in this context that interest in transferable models (i.e., those that can be legitimately projected beyond the spatial and temporal bounds of their underlying data [7]) has grown.Transferred models must balance the tradeoff between estimation and prediction bias and variance (homogenization versus nontransferability, sensu [8]). Ultimately, models that can Highlights Models transferred to novel conditions could provide predictions in data-poor scenarios, contributing to more informed management decisions.The determinants of ecological predictability are, however, still insufficiently understood.Predictions from transferred ecological models are affected by species' traits, sampling biases, biotic interactions, nonstationarity, and the degree of environmental dissimilarity between reference and target systems.We synthesize six technical and six fundamental challenges that, if resolved, will catalyze practical and conceptual advances in model transfers.We propose that the most immediate obstacle to improving understanding lies in the absence of a widely applicable set of metrics for assessing transferability, and that encouraging the development of models grounded in well-established mech...

Population connectivity refers to the exchange of individuals among populations: it affects gene flow, regulates population size and function, and mitigates recovery from natural or anthropogenic disturbances. Many populations in the deep sea are spatially fragmented, and will become more so with increasing resource exploitation. Understanding population connectivity is critical for spatial management. For most benthic species, connectivity is achieved by the planktonic larval stage, and larval dispersal is, in turn, regulated by complex interactions between biological and oceanographic processes. Coupled biophysical models, incorporating ocean circulation and biological traits, such as planktonic larval duration (PLD), have been used to estimate population connectivity and generate spatial management plans in coastal and shallow waters. In the deep sea, knowledge gaps in both the physical and biological components are delaying the effective use of this approach. Here, we review the current efforts in conservation in the deep sea and evaluate (1) the relevance of using larval dispersal in the design of marine protected areas and (2) the application of biophysical models in the study of population connectivity. Within biophysical models, PLD can be used to estimate dispersal distance. We propose that a PLD that guarantees a minimum dispersal distance for a wide range of species should be used in the planning of marine protected areas in the deep sea. Based on a review of data on species found at depths >200 m, a PLD of 35 and 69 days ensures a minimum distance for 50 and 75%, respectively, of eurybathic and deep-sea species. We note that more data are required to enhance accuracy and address the high variability in PLD between and within taxonomic groups, limiting generalizations that are often appealing to decision-makers. Given the imminent expansion of resource exploitation in the deep sea, data relevant to spatial management are needed urgently.

Aim To demonstrate the application of predictive species distribution modelling methods to habitat mapping and assessment of percentage area-based conservation targets.Location The NE Atlantic deep sea (UK and Irish extended continental shelf limits).Methods MaxEnt modelling of three listed habitats (Lophelia pertusa (Linnaeus, 1758) reef (LpReef), Pheronema carpenteri (WyvilleThomson, 1869) aggregations (PcAggs) and Syringammina fragilissima (Brady, 1883) aggregations (SfAggs)), with some pre-selection of variables by generalized additive modelling. Models are validated using repeated 70/30 build/test data splits using AUC and threshold-dependent assessment methods. Predicted distribution maps are used to assess the adequacy of existing area closures for the protection of listed habitats and to assess percentage representation of each community within existing MPA networks.Results Model performances are rated as fair (LpReef), excellent (PcAggs) and good (SfAggs). Current closures are focused on the protection of cold-water coral reef and incidentally capture some SfAggs suitable environments, but largely fail to protect PcAggs. Considering the wider network of MPAs in the study region, approximately 23% (LpReef), 2% (PcAggs) and 6% (SfAggs) of the area predicted as suitable for each habitat respectively is contained within an MPA.Main conclusions To date, decisions on area closures for the protection of 'listed' deep-sea habitats have been based on maps of recorded presence of species that are taken as being indicative of that habitat. Predictive habitat modelling may provide a useful method of better estimating the extent of listed habitats, providing direction for future MPA establishment and a means of assessing MPA network effectiveness against politically set percentage targets. Given the coarse resolution of the model, percentages should be taken as maximal figures, with habitat occurrence likely to be less prevalent in reality.

Video and image data are regularly used in the field of benthic ecology to document biodiversity. However, their use is subject to a number of challenges, principally the identification of taxa within the images without associated physical specimens. The challenge of applying traditional taxonomic keys to the identification of fauna from images has led to the development of personal, group, or institution level reference image catalogues of operational taxonomic units (OTUs) or morphospecies. Lack of standardisation among these reference catalogues has led to problems with observer bias and the inability to combine datasets across studies. In addition, lack of a common reference standard is stifling efforts in the application of artificial intelligence to taxon identification. Using the North Atlantic deep sea as a case study, we propose a database structure to facilitate standardisation of morphospecies image catalogues between research groups and support future use in multiple front-end applications. We also propose a framework for coordination of international efforts to develop reference guides for the identification of marine species from images. The proposed structure maps to the Darwin Core standard to allow integration with existing databases. We suggest a management framework where high-level taxonomic groups are curated by a regional team, consisting of both end users and taxonomic experts. We identify a mechanism by which overall quality of data within a common reference guide could be raised over the next decade. Finally, we discuss the role of a common reference standard in advancing marine ecology and supporting sustainable use of this ecosystem.