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...

Marine organisms are simultaneously exposed to anthropogenic stressors with likely interactive effects, including synergisms in which the combined effects of multiple stressors are greater than the sum of individual effects. Early life stages of marine organisms are potentially vulnerable to the stressors associated with global change, but identifying general patterns across studies, species and response variables is challenging. This review represents the first meta-analysis of multistressor studies to target early marine life stages (embryo to larvae), particularly between temperature, salinity and pH as these are the best studied. Knowledge gaps in research on multiple abiotic stressors and early life stages are also identified. The meta-analysis yielded several key results: (1) Synergistic interactions (65% of individual tests) are more common than additive (17%) or antagonistic (17%) interactions. (2) Larvae are generally more vulnerable than embryos to thermal and pH stress. (3) Survival is more likely than sublethal responses to be affected by thermal, salinity and pH stress. (4) Interaction types vary among stressors, ontogenetic stages and biological responses, but they are more consistent among phyla. (5) Ocean acidification is a greater stressor for calcifying than noncalcifying larvae. Despite being more ecologically realistic than single-factor studies, multifactorial studies may still oversimplify complex systems, and so meta-analyses of the data from them must be cautiously interpreted with regard to extrapolation to field conditions. Nonetheless, our results identify taxa with early life stages that may be particularly vulnerable (e.g. molluscs, echinoderms) or robust (e.g. arthropods, cnidarians) to abiotic stress. We provide a list of recommendations for future multiple stressor studies, particularly those focussed on early marine life stages.

With marine biodiversity declining globally at accelerating rates, maximising the effectiveness of conservation has become a key goal for local, national and international regulators. Marine protected areas (MPAs) have been widely advocated for conserving and managing marine biodiversity yet, despite extensive research, their benefits for conserving non-target species and wider ecosystem functions remain unclear. Here, we demonstrate that MPAs can increase the resilience of coral reef communities to natural disturbances, including coral bleaching, coral diseases, Acanthaster planci outbreaks and storms. Using a 20-year time series from Australia's Great Barrier Reef, we show that within MPAs, (1) reef community composition was 21-38% more stable; (2) the magnitude of disturbance impacts was 30% lower and (3) subsequent recovery was 20% faster that in adjacent unprotected habitats. Our results demonstrate that MPAs can increase the resilience of marine communities to natural disturbance possibly through herbivory, trophic cascades and portfolio effects.