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...

Aim Correlative species distribution models (SDMs) often involve some degree of projection into novel covariate space (i.e. extrapolation), because calibration data may not encompass the entire space of interest. Most methods for identifying extrapolation focus on the range of each model covariate individually. However, extrapolation can occur that is well within the range of univariate variation, but which exhibits novel combinations between covariates. Our objective was to develop a tool that can detect, distinguish and quantify these two types of novelties: novel univariate range and novel combinations of covariates.Location Global, Australia, South Africa. Methods We developed a new multivariate statistical tool, based on the Mahalanobis distance, which measures the similarity between the reference and projection domains by accounting for both the deviation from the mean and the correlation between variables. The method also provides an assessment tool for the detection of the most influential covariates leading to dissimilarity. As an example application, we modelled an Australian shrub (Acacia cyclops) widely introduced to other countries and compared reference data, global distribution data and both types of model extrapolation against the projection globally and in South Africa. ResultsThe new tool successfully detected and quantified the degree of dissimilarity for points that were either outside the univariate range or formed novel covariate combinations (correlations) but were still within the univariate range of covariates. For A. cyclops, more than half of the points (6617 of 10,785) from the global projection space that were found to lie within the univariate range of reference data exhibited distorted correlations. Not all the climate covariates used for modelling contributed to novelty equally over the geographical space of the model projection.Main conclusions Identifying non-analogous environments is a critical component of model interrogation. Our extrapolation detection (ExDet) tool can be used as a quantitative method for exploring novelty and interpreting the projections from correlative SDMs and is available for free download as standalone software from http://www.climond.org/exdet.

Increasing population has posed insurmountable challenges to agriculture in the provision of future food security, particularly in the Middle East and North Africa (MENA) region where biophysical conditions are not well-suited for agriculture. Iran, as a major agricultural country in the MENA region, has long been in the quest for food self-sufficiency, however, the capability of its land and water resources to realize this goal is largely unknown. Using very high-resolution spatial data sets, we evaluated the capacity of Iran’s land for sustainable crop production based on the soil properties, topography, and climate conditions. We classified Iran’s land suitability for cropping as (million ha): very good 0.4% (0.6), good 2.2% (3.6), medium 7.9% (12.8), poor 11.4% (18.5), very poor 6.3% (10.2), unsuitable 60.0% (97.4), and excluded areas 11.9% (19.3). In addition to overarching limitations caused by low precipitation, low soil organic carbon, steep slope, and high soil sodium content were the predominant soil and terrain factors limiting the agricultural land suitability in Iran. About 50% of the Iran’s existing croplands are located in low-quality lands, representing an unsustainable practice. There is little room for cropland expansion to increase production but redistribution of cropland to more suitable areas may improve sustainability and reduce pressure on water resources, land, and ecosystem in Iran.

The founding population in most new species introductions, or at the leading edge of an ongoing invasion, is likely to be small. Severe Allee effects-reductions in individual fitness at low population density-may then result in a failure of the species to colonize, even if the habitat could support a much larger population. Using a simulation model for plant populations that incorporates demography, mating systems, quantitative genetics, and pollinators, we show that Allee effects can potentially be overcome by transient hybridization with a resident species or an earlier colonizer. This mechanism does not require the invocation of adaptive changes usually attributed to invasions following hybridization. We verify our result in a case study of sequential invasions by two plant species where the outcrosser Cakile maritima has replaced an earlier, inbreeding, colonizer Cakile edentula (Brassicaceae). Observed historical rates of replacement are consistent with model predictions from hybrid-alleviated Allee effects in outcrossers, although other causes cannot be ruled out.species colonization | mating system | model | Cakile | sea-rockets

The germination of a population of seeds was modelled using the concept of hydrotime or hydrothermal time. Typically, a Normal distribution for base water potential (Ψ b(g) ) was used within these models to relate variation in Ψ b(g) to the variation in time to germination of a given fraction of seeds. We sought to examine empirically the validity of this assumption, to compare the fit of alternative distributions and make recommendations for improved germination modelling procedures. Eight statistical distributions (Gumbel, Weibull, Normal, Log-Normal, Logistic, Loglogistic, Inverse Normal and Gamma) were fitted to data for four weed species Hordeum spontaneum, Phalaris minor, Heliotropium europaeum and Raphanus raphanistrum. Methods for incorporating each of these distributions into hydrotime are presented. For three species (H. spontaneum, P. minor and H. europaeum), the Normal distribution gave the worst fit (with AIC values: À124.2, À296.9 and À264.5, respectively) and would lead to biased predictions, whereas the Loglogistic distribution consistently provided the best explanation of Ψ b(g) variation in these species (with AIC values: À188.6, À326.2 and À272.1 respectively). All distributions failed to provide an unbiased description of the observed distribution of Ψ b(g) in R. raphanistrum. The Normal distribution is not necessarily the best function for base water potential in hydrothermal models and, indeed, may give much more biased predictions than alternative functions. The 'best' distribution may vary with species. The distribution of Ψ b(g) within a seed sample should therefore be examined and an appropriate equation selected, before using a model to make predictions.