Search

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.

Species distribution models are widely used to identify potential and high‐quality habitat of endangered species to inform conservation decisions. However, their usefulness is constrained by the amount and quality of biodiversity data and the approaches for dealing with data deficiencies. Presence‐only data, used in presence/background modelling methods, are widely available but are often affected by sampling bias. Presence/absence modelling methods are less affected by biases, but data are less common. We modelled the distribution of a widely distributed, endangered species from Australia – the greater glider – and tested how predictions were influenced by data treatment and modelling framework. We collated available species data and fitted generalized linear models and boosted regression trees using presence/absence data, as well as using an augmented dataset that included additional presences alongside absences inferred from survey data. We also fitted presence/background models, adopting three common strategies for bias correction. We compared model performance quantitatively through evaluation metrics calculated internally and on held out data, and qualitatively by identifying areas of agreement of spatial predictions. We found that presence/background models with bias correction performed better than not corrected, though evaluation metrics did not favour a single strategy. Presence/absence models outperformed presence/background models in comparable metrics and delivered different spatial predictions. Importantly, differences in spatial predictions between models had the potential to substantially alter decisions about where to protect high‐quality habitat. The approach to inferring absences proved useful, as models fitted with these outperformed all other models. Dealing with sampling bias requires additional time and data management strategies, but we found that the time invested allowed improvement of models and more reliable predictions. Our results suggest that ancillary occurrence data and careful data handling can improve both presence/background and presence/absence models. Immunology research holds significant potential for enhanced inclusivity, but persistent challenges stem from limited awareness that improvement is needed in this field. At the 2023 Monash Sensory Science Exhibition, we presented tactile posters, using simple materials, for visually impaired participants to showcase our research on the pathogenesis and immunotherapy of rheumatoid arthritis. [ABSTRACT FROM AUTHOR]

Predictions of species' current and future ranges are needed to effectively manage species under environmental change. Species ranges are typically estimated using correlative species distribution models (SDMs), which have been criticized for their static nature. In contrast, dynamic occupancy models (DOMs) explicitily describe temporal changes in species' occupancy via colonization and local extinction probabilities, estimated from time series of occurrence data. Yet, tests of whether these models improve predictive accuracy under current or future conditions are rare. Using a long‐term data set on 69 Swiss birds, we tested whether DOMs improve the predictions of distribution changes over time compared to SDMs. We evaluated the accuracy of spatial predictions and their ability to detect population trends. We also explored how predictions differed when we accounted for imperfect detection and parameterized models using calibration data sets of different time series lengths. All model types had high spatial predictive performance when assessed across all sites (mean AUC > 0.8), with flexible machine learning SDM algorithms outperforming parametric static and DOMs. However, none of the models performed well at identifying sites where range changes are likely to occur. In terms of estimating population trends, DOMs performed best, particularly for species with strong population changes and when fit with sufficient data, while static SDMs performed very poorly. Overall, our study highlights the importance of considering what aspects of performance matter most when selecting a modelling method for a particular application and the need for further research to improve model utility. While DOMs show promise for capturing range dynamics and inferring population trends when fitted with sufficient data, computational constraints on variable selection and model fitting can lead to reduced spatial accuracy of predictions, an area warranting more attention. [ABSTRACT FROM AUTHOR]

Species distribution models (SDMs) are increasingly used in conservation and land‐use planning as inputs to describe biodiversity patterns. These models can be built in different ways, and decisions about data preparation, selection of predictor variables, model fitting, and evaluation all alter the resulting predictions. Commonly, the true distribution of species is unknown and independent data to verify which SDM variant to choose are lacking. Such model uncertainty is of concern to planners. We analyzed how 11 routine decisions about model complexity, predictors, bias treatment, and setting thresholds for predicted values altered conservation priority patterns across 25 species. Models were created with MaxEnt and run through Zonation to determine the priority rank of sites. Although all SDM variants performed well (area under the curve >0.7), they produced spatially different predictions for species and different conservation priority solutions. Priorities were most strongly altered by decisions to not address bias or to apply binary thresholds to predicted values; on average 40% and 35%, respectively, of all grid cells received an opposite priority ranking. Forcing high model complexity altered conservation solutions less than forcing simplicity (14% and 24% of cells with opposite rank values, respectively). Use of fewer species records to build models or choosing alternative bias treatments had intermediate effects (25% and 23%, respectively). Depending on modeling choices, priority areas overlapped as little as 10–20% with the baseline solution, affecting top and bottom priorities differently. Our results demonstrate the extent of model‐based uncertainty and quantify the relative impacts of SDM building decisions. When it is uncertain what the best SDM approach and conservation plan is, solving uncertainty or considering alterative options is most important for those decisions that change plans the most. Article impact statement: Methodological decisions on how species distribution models (SDMs) are built can alter locations of priority areas up to 90%. [ABSTRACT FROM AUTHOR]

Aim: Accurate predictions of cetacean distributions are essential to their conservation but are limited by statistical challenges and a paucity of data. This study aimed at comparing the capacity of various statistical algorithms to deal with biases commonly found in nonsystematic cetacean surveys and to evaluate the potential for citizen science data to improve habitat modelling and predictions. An endangered population of humpback whales (Megaptera novaeangliae) in their breeding ground was used as a case study. Location: New Caledonia, Oceania. Methods: Five statistical algorithms were used to model the habitat preferences of humpback whales from 1,360 sightings collected over 14 years of nonsystematic research surveys. Three different background sampling approaches were tested when developing models from 625 crowdsourced sightings to assess methods accounting for citizen science spatial sampling bias. Model evaluation was conducted through cross‐validation and prediction to an independent satellite tracking dataset. Results: Algorithms differed in complexity of the environmental relationships modelled, ecological interpretability and transferability. While parameter tuning had a great effect on model performances, GLMs generally had low predictive performance, SVMs were particularly hard to interpret, and BRTs had high descriptive power but showed signs of overfitting. MAXENT and especially GAMs provided a valuable complexity trade‐off, accurate predictions and were ecologically intelligible. Models showed that humpback whales favoured cool (22–23°C) and shallow waters (0–100 m deep) in coastal as well as offshore areas. Citizen science models converged with research survey models, specifically when accounting for spatial sampling bias. Main conclusions: Marine megafauna distribution models present specific challenges that may be addressed through integrative evaluation, independent testing and appropriately tuned statistical algorithms. Specifically, controlling overfitting is a priority when predicting cetacean distributions for large‐scale conservation perspectives. Citizen science data appear to be a powerful tool to describe cetacean habitat. [ABSTRACT FROM AUTHOR]

Aim Managing fire is critical for the conservation of biodiversity in many ecosystems globally. To manage fire effectively, it is necessary to identify the temporal and spatial scales at which it affects a diverse range of species. This information is challenging to obtain for rare and threatened species for which data often are sparse, and in systems with long fire-return intervals (e.g. >100 years). We tested the effects of a century of fires on the distribution of 12 threatened bird species across a 100,000 km2 region in which 'long-unburnt' vegetation has been identified as important for the diversity of common species. Location Semi-arid mallee woodlands of south-eastern Australia. Methods We developed spatially explicit models to identify the effects of fire history and climatic factors on the distribution of 12 threatened bird species, including two globally endangered species, the Mallee Emu-wren ( Stipiturus mallee) and Black-eared Miner ( Manorina melanotis). Results Fire was a driver of distribution for all species. Four species were common in younger vegetation (<20 years post-fire) and 11 were most common in mid (20-60 years post-fire) to older (>60 years post-fire) vegetation. Species' distributions were further restricted to areas associated with particular vegetation types and climatic conditions. Main conclusions Comprehensive investigation of the response to fire by a range of threatened species highlights the importance of what is now recognized as mid-successional mallee vegetation (20-60 years post-fire), and that species' preferences for previously identified 'long-unburnt' vegetation extend to ≥60 years post-fire. Fire management conducted with incomplete knowledge, or which is focussed on introducing prescribed burns or suppressing fires for early/late-successional species alone, is unlikely to maximize biodiversity. Effective fire management for biodiversity requires the promotion of ecological processes that result in key successional stages at particular locations in the landscape. [ABSTRACT FROM AUTHOR]

Aim Species distribution models (SDMs) are currently the most widely used tools in ecology for evaluating the suitability of environments for biodiversity in the face of future environmental change. In this study we seek to provide an assessment of the predictive performance of SDMs over time. How well do SDMs predict for future time periods and what factors influence predictive performance? Innovation We used a historical spatially explicit database of 1.8 million occurrence records for 318 tetrapod species from across continental Australia over the period 1950-2013. We fitted distribution models for each species to data from four multi-decadal time slices and used these to predict the species distributions up to 60 years after the data collection period for the fitted models. We evaluated predictions against observed data from the relevant time period. Predictions were made assuming either complete knowledge of changes in climatic and environmental conditions or assuming the environment and climate remained unchanged between the fitting and evaluation time periods. We used generalized linear mixed models to model variation in the predictive performance of SDMs over time in relation to a variety of factors, including the length of time between fitting and evaluation, species traits, taxonomic group and attributes of the dataset used to fit models. Main conclusions We found that most models provided useful predictions even when the period between model fitting and evaluation was 60 years (area under the receiver operator characteristic curve > 0.7 in 80% of the species evaluated). Variation in predictive performance over time was strongly related to the species range breadth (models for species with broad geographical ranges tended to perform worse than models for locally restricted species) and to the environmental coverage of occupancy data. Conversely, taxonomic group, habitat preferences and body size were not highly influential in describing the variation in predictive performance over time. [ABSTRACT FROM AUTHOR]

Presence-only records may provide data on the distributions of rare species, but commonly suffer from large, unknown biases due to their typically haphazard collection schemes. Presence-absence or count data collected in systematic, planned surveys are more reliable but typically less abundant., We proposed a probabilistic model to allow for joint analysis of presence-only and survey data to exploit their complementary strengths. Our method pools presence-only and presence-absence data for many species and maximizes a joint likelihood, simultaneously estimating and adjusting for the sampling bias affecting the presence-only data. By assuming that the sampling bias is the same for all species, we can borrow strength across species to efficiently estimate the bias and improve our inference from presence-only data., We evaluate our model's performance on data for 36 eucalypt species in south-eastern Australia. We find that presence-only records exhibit a strong sampling bias towards the coast and towards Sydney, the largest city. Our data-pooling technique substantially improves the out-of-sample predictive performance of our model when the amount of available presence-absence data for a given species is scarce, If we have only presence-only data and no presence-absence data for a given species, but both types of data for several other species that suffer from the same spatial sampling bias, then our method can obtain an unbiased estimate of the first species' geographic range. [ABSTRACT FROM AUTHOR]

Data limitations can lead to unrealistic fits of predictive species distribution models (SDMs) and spurious extrapolation to novel environments. Here, we want to draw attention to novel combinations of environmental predictors that are within the sampled range of individual predictors but are nevertheless outside the sample space. These tend to be overlooked when visualizing model behaviour. They may be a cause of differing model transferability and environmental change predictions between methods, a problem described in some studies but generally not well understood. We here use a simple simulated data example to illustrate the problem and provide new and complementary visualization techniques to explore model behaviour and predictions to novel environments. We then apply these in a more complex real-world example. Our results underscore the necessity of scrutinizing model fits, ecological theory and environmental novelty. [ABSTRACT FROM AUTHOR]

The main role of conservation planning is to design reserve networks to protect biodiversity in situ. Research within the field of conservation planning has focused on the development of theories and tools to design reserve networks that protect biodiversity in an efficient and representative manner. Whilst much progress has been made in this regard, there has been limited assessment of the sensitivity of conservation planning outcomes to uncertainty associated with the datasets used for conservation planning. Predicted species distribution data are commonly used for conservation planning because the alternatives (e.g. survey data) are incomplete or biased spatially. However, there may be considerable uncertainty associated with the use of predicted species distribution data, particularly given the variety of approaches available to generate a dataset from such predictions for use in conservation planning. These approaches range from using the probabilistic data directly to using a threshold identified a priori or a posteriori to convert the probabilistic data to presence/absence data. We assess the sensitivity of conservation planning outcomes to different uses of predicted species distribution data. The resulting reserve networks differed, and had different expected species representation. The choice of approach will depend on how much risk a conservation planner is willing to tolerate and how much efficiency can be sacrificed. [Copyright &y& Elsevier]