Accelerating joint species distribution modelling with Hmsc-HPC by GPU porting.

Joint species distribution modelling (JSDM) is a widely used statistical method that analyzes combined patterns of all species in a community, linking empirical data to ecological theory and enhancing community-wide prediction tasks. However, fitting JSDMs to large datasets is often computationally demanding and time-consuming. Recent studies have introduced new statistical and machine learning techniques to provide more scalable fitting algorithms, but extending these to complex JSDM structures that account for spatial dependencies or multi-level sampling designs remains challenging. In this study, we aim to enhance JSDM scalability by leveraging high-performance computing (HPC) resources for an existing fitting method. Our work focuses on the Hmsc R-package, a widely used JSDM framework that supports the integration of various dataset types into a single comprehensive model. We developed a GPU-compatible implementation of its model-fitting algorithm using Python and the TensorFlow library. Despite these changes, our enhanced framework retains the original user interface of the Hmsc R-package. We evaluated the performance of the proposed implementation across various model configurations and dataset sizes. Our results show a significant increase in model fitting speed for most models compared to the baseline Hmsc R-package. For the largest datasets, we achieved speed-ups of over 1000 times, demonstrating the substantial potential of GPU porting for previously CPU-bound JSDM software. This advancement opens promising opportunities for better utilizing the rapidly accumulating new biodiversity data resources for inference and prediction.

Resolving biology's dark matter: species richness, spatiotemporal distribution, and community composition of a dark taxon.

BACKGROUND: Zoology's dark matter comprises hyperdiverse, poorly known taxa that are numerically dominant but largely unstudied, even in temperate regions where charismatic taxa are well understood. Dark taxa are everywhere, but high diversity, abundance, and small size have historically stymied their study. We demonstrate how entomological dark matter can be elucidated using high-throughput DNA barcoding ("megabarcoding"). We reveal the high abundance and diversity of scuttle flies (Diptera: Phoridae) in Sweden using 31,800 specimens from 37 sites across four seasonal periods. We investigate the number of scuttle fly species in Sweden and the environmental factors driving community changes across time and space. RESULTS: Swedish scuttle fly diversity is much higher than previously known, with 549 putative specie) detected, compared to 374 previously recorded species. Hierarchical Modelling of Species Communities reveals that scuttle fly communities are highly structured by latitude and strongly driven by climatic factors. Large dissimilarities between sites and seasons are driven by turnover rather than nestedness. Climate change is predicted to significantly affect the 47% of species that show significant responses to mean annual temperature. Results were robust regardless of whether haplotype diversity or species-proxies were used as response variables. Additionally, species-level models of common taxa adequately predict overall species richness. CONCLUSIONS: Understanding the bulk of the diversity around us is imperative during an era of biodiversity change. We show that dark insect taxa can be efficiently characterised and surveyed with megabarcoding. Undersampling of rare taxa and choice of operational taxonomic units do not alter the main ecological inferences, making it an opportune time to tackle zoology's dark matter.

Landscape fragmentation overturns classical metapopulation thinking.

Habitat loss and isolation caused by landscape fragmentation represent a growing threat to global biodiversity. Existing theory suggests that the process will lead to a decline in metapopulation viability. However, since most metapopulation models are restricted to simple networks of discrete habitat patches, the effects of real landscape fragmentation, particularly in stochastic environments, are not well understood. To close this major gap in ecological theory, we developed a spatially explicit, individual-based model applicable to realistic landscape structures, bridging metapopulation ecology and landscape ecology. This model reproduced classical metapopulation dynamics under conventional model assumptions, but on fragmented landscapes, it uncovered general dynamics that are in stark contradiction to the prevailing views in the ecological and conservation literature. Notably, fragmentation can give rise to a series of dualities: a) positive and negative responses to environmental noise, b) relative slowdown and acceleration in density decline, and c) synchronization and desynchronization of local population dynamics. Furthermore, counter to common intuition, species that interact locally ("residents") were often more resilient to fragmentation than long-ranging "migrants." This set of findings signals a need to fundamentally reconsider our approach to ecosystem management in a noisy and fragmented world.

Global Spore Sampling Project: A global, standardized dataset of airborne fungal DNA.

Novel methods for sampling and characterizing biodiversity hold great promise for re-evaluating patterns of life across the planet. The sampling of airborne spores with a cyclone sampler, and the sequencing of their DNA, have been suggested as an efficient and well-calibrated tool for surveying fungal diversity across various environments. Here we present data originating from the Global Spore Sampling Project, comprising 2,768 samples collected during two years at 47 outdoor locations across the world. Each sample represents fungal DNA extracted from 24 m3 of air. We applied a conservative bioinformatics pipeline that filtered out sequences that did not show strong evidence of representing a fungal species. The pipeline yielded 27,954 species-level operational taxonomic units (OTUs). Each OTU is accompanied by a probabilistic taxonomic classification, validated through comparison with expert evaluations. To examine the potential of the data for ecological analyses, we partitioned the variation in species distributions into spatial and seasonal components, showing a strong effect of the annual mean temperature on community composition.

Experimental evidence that root-associated fungi improve plant growth at high altitude.

Unravelling how species communities change along environmental gradients requires a dual understanding: the direct responses of the species to their abiotic surroundings and the indirect variation of these responses through biotic interactions. Here, we focus on the interactive relationships between plants and their symbiotic root-associated fungi (RAF) along stressful abiotic gradients. We investigate whether variations in RAF community composition along altitudinal gradients influence plant growth at high altitudes, where both plants and fungi face harsher abiotic conditions. We established a translocation experiment between pairs of Bistorta vivipara populations across altitudinal gradients. To separate the impact of shifting fungal communities from the overall influence of changing abiotic conditions, we used a root barrier to prevent new colonization by RAF following translocation. To characterize the RAF communities, we applied DNA barcoding to the root samples. Through the utilization of joint species distribution modelling, we assessed the relationship between changes in plant functional traits resulting from experimental treatments and the corresponding changes in the RAF communities. Our findings indicate that RAF communities influence plant responses to stressful abiotic conditions. Plants translocated from low to high altitudes grew more when they were able to associate with the resident high-altitude RAF compared to those plants that were not allowed to associate with the resident RAF. We conclude that interactions with RAF impact how plants respond to stressful abiotic conditions. Our results provide experimental support that interactions with RAF improve plant stress tolerance to altitudinal stressors such as colder temperatures and less nutrient availability.

Strategies for integrating scientific evidence in water policy and law in the face of uncertainty.

Understanding how human actions and environmental change affect water resources is crucial for addressing complex water management issues. The scientific tools that can produce the necessary information are ecological indicators, referring to measurable properties of the ecosystem state; environmental monitoring, the data collection process that is required to evaluate the progress towards reaching water management goals; mathematical models, linking human disturbances with the ecosystem state to predict environmental impacts; and scenarios, assisting in long-term management and policy implementation. Paradoxically, despite the rapid generation of data, evolving scientific understanding, and recent advancements in systems modeling, there is a striking imbalance between knowledge production and knowledge utilization in decision-making. In this paper, we examine the role and potential capacity of scientific tools in guiding governmental decision-making processes and identify the most critical disparities between water management, policy, law, and science. We demonstrate how the complex, uncertain, and gradually evolving nature of scientific knowledge might not always fit aptly to the legislative and policy processes and structures. We contend that the solution towards increased understanding of socio-ecological systems and reduced uncertainty lies in strengthening the connections between water management theory and practice, among the scientific tools themselves, among different stakeholders, and among the social, economic, and ecological facets of water quality management, law, and policy. We conclude by tying in three knowledge-exchange strategies, namely - adaptive management, Driver-Pressure-Status-Impact-Response (DPSIR) framework, and participatory modeling - that offer complementary perspectives to bridge the gap between science and policy.

Recommendations for quantitative uncertainty consideration in ecology and evolution.

Ecological and evolutionary studies are currently failing to achieve complete and consistent reporting of model-related uncertainty. We identify three key barriers - a focus on parameter-related uncertainty, obscure uncertainty metrics, and limited recognition of uncertainty propagation - which have led to gaps in uncertainty consideration. However, these gaps can be closed. We propose that uncertainty reporting in ecology and evolution can be improved through wider application of existing statistical solutions and by adopting good practice from other scientific fields. Our recommendations include greater consideration of input data and model structure uncertainties, field-specific uncertainty standards for methods and reporting, and increased uncertainty propagation through the use of hierarchical models.

LIES of omission: complex observation processes in ecology.

Advances in statistics mean that it is now possible to tackle increasingly sophisticated observation processes. The intricacies and ambitious scale of modern data collection techniques mean that this is now essential. Methodological research to make inference about the biological process while accounting for the observation process has expanded dramatically, but solutions are often presented in field-specific terms, limiting our ability to identify commonalities between methods. We suggest a typology of observation processes that could improve translation between fields and aid methodological synthesis. We propose the LIES framework (defining observation processes in terms of issues of Latency, Identifiability, Effort and Scale) and illustrate its use with both simple examples and more complex case studies.

Novel community data in ecology-properties and prospects.

New technologies for monitoring biodiversity such as environmental (e)DNA, passive acoustic monitoring, and optical sensors promise to generate automated spatiotemporal community observations at unprecedented scales and resolutions. Here, we introduce 'novel community data' as an umbrella term for these data. We review the emerging field around novel community data, focusing on new ecological questions that could be addressed; the analytical tools available or needed to make best use of these data; and the potential implications of these developments for policy and conservation. We conclude that novel community data offer many opportunities to advance our understanding of fundamental ecological processes, including community assembly, biotic interactions, micro- and macroevolution, and overall ecosystem functioning.

Evaluating the predictive performance of presence-absence models: Why can the same model appear excellent or poor?

When comparing multiple models of species distribution, models yielding higher predictive performance are clearly to be favored. A more difficult question is how to decide whether even the best model is "good enough". Here, we clarify key choices and metrics related to evaluating the predictive performance of presence-absence models. We use a hierarchical case study to evaluate how four metrics of predictive performance (AUC, Tjur's R 2, max-Kappa, and max-TSS) relate to each other, the random and fixed effects parts of the model, the spatial scale at which predictive performance is measured, and the cross-validation strategy chosen. We demonstrate that the very same metric can achieve different values for the very same model, even when similar cross-validation strategies are followed, depending on the spatial scale at which predictive performance is measured. Among metrics, Tjur's R 2 and max-Kappa generally increase with species' prevalence, whereas AUC and max-TSS are largely independent of prevalence. Thus, Tjur's R 2 and max-Kappa often reach lower values when measured at the smallest scales considered in the study, while AUC and max-TSS reaching similar values across the different spatial levels included in the study. However, they provide complementary insights on predictive performance. The very same model may appear excellent or poor not only due to the applied metric, but also how predictive performance is exactly calculated, calling for great caution on the interpretation of predictive performance. The most comprehensive evaluation of predictive performance can be obtained by evaluating predictive performance through the combination of measures providing complementary insights. Instead of following simple rules of thumb or focusing on absolute values, we recommend comparing the achieved predictive performance to the researcher's own a priori expectations on how easy it is to make predictions related to the same question that the model is used for.

Alternative measures of trait-niche relationships: A test on dispersal traits in saproxylic beetles.

Functional trait approaches are common in ecology, but a lack of clear hypotheses on how traits relate to environmental gradients (i.e., trait-niche relationships) often makes uncovering mechanisms difficult. Furthermore, measures of community functional structure differ in their implications, yet inferences are seldom compared among metrics. Community-weighted mean trait values (CWMs), a common measure, are largely driven by the most common species and thus do not reflect community-wide trait-niche relationships per se. Alternatively, trait-niche relationships can be estimated across a larger group of species using hierarchical joint species distribution models (JSDMs), quantified by a parameter Γ. We investigated how inferences about trait-niche relationships are affected by the choice of metric. Using deadwood-dependent (saproxylic) beetles in fragmented Finnish forests, we followed a protocol for investigating trait-niche relationships by (1) identifying environmental filters (climate, forest age, and deadwood volume), (2) relating these to an ecological function (dispersal ability), and (3) identifying traits related to this function (wing morphology). We tested 18 hypothesized dispersal relationships using both CWM and Γ estimates across these environmental gradients. CWMs were more likely than Γ to show support for trait-niche relationships. Up to 13% of species' realized niches were explained by dispersal traits, but the directions of effects were consistent with fewer than 11%-39% of our 18 trait-niche hypotheses (depending on the metric used). This highlights the difficulty in connecting morphological traits and ecological functions in insects, despite the clear conceptual link between landscape connectivity and flight-related traits. Caution is thus warranted in hypothesis development, particularly where apparent trait-function links are less clear. Inferences differ when CWMs versus Γ estimates are used, necessitating the choice of a metric that reflects study questions. CWMs help explain the effects of environmental gradients on community trait composition, whereas the effects of traits on species' niches are better estimated using hierarchical JSDMs.

The role of seasonality in shaping the interactions of honeybees with other taxa.

The Eltonian niche of a species is defined as its set of interactions with other taxa. How this set varies with biotic, abiotic and human influences is a core question of modern ecology. In seasonal environments, the realized Eltonian niche is likely to vary due to periodic changes in the occurrence and abundance of interaction partners and changes in species behavior and preferences. Also, human management decisions may leave strong imprints on species interactions. To compare the impact of seasonality to that of management effects, honeybees provide an excellent model system. Based on DNA traces of interaction partners archived in honey, we can infer honeybee interactions with floral resources and microbes in the surrounding habitats, their hives, and themselves. Here, we resolved seasonal and management-based impacts on honeybee interactions by sampling beehives repeatedly during the honey-storing period of honeybees in Finland. We then use a genome-skimming approach to identify the taxonomic contents of the DNA in the samples. To compare the effects of the season to the effects of location, management, and the colony itself in shaping honeybee interactions, we used joint species distribution modeling. We found that honeybee interactions with other taxa varied greatly among taxonomic and functional groups. Against a backdrop of wide variation in the interactions documented in the DNA content of honey from bees from different hives, regions, and beekeepers, the imprint of the season remained relatively small. Overall, a honey-based approach offers unique insights into seasonal variation in the identity and abundance of interaction partners among honeybees. During the summer, the availability and use of different interaction partners changed substantially, but hive- and taxon-specific patterns were largely idiosyncratic as modified by hive management. Thus, the beekeeper and colony identity are as important determinants of the honeybee's realized Eltonian niche as is seasonality.

Mixed effects of a national protected area network on terrestrial and freshwater biodiversity.

Protected areas are considered fundamental to counter biodiversity loss. However, evidence for their effectiveness in averting local extinctions remains scarce and taxonomically biased. We employ a robust counterfactual multi-taxon approach to compare occupancy patterns of 638 species, including birds (150), mammals (23), plants (39) and phytoplankton (426) between protected and unprotected sites across four decades in Finland. We find mixed impacts of protected areas, with only a small proportion of species explicitly benefiting from protection-mainly through slower rates of decline inside protected areas. The benefits of protection are enhanced for larger protected areas and are traceable to when the sites were protected, but are mostly unrelated to species conservation status or traits (size, climatic niche and threat status). Our results suggest that the current protected area network can partly contribute to slow down declines in occupancy rates, but alone will not suffice to halt the biodiversity crisis. Efforts aimed at improving coverage, connectivity and management will be key to enhance the effectiveness of protected areas towards bending the curve of biodiversity loss.

Local eukaryotic and bacterial stream community assembly is shaped by regional land use effects.

With anticipated expansion of agricultural areas for food production and increasing intensity of pressures stemming from land-use, it is critical to better understand how species respond to land-use change. This is particularly true for microbial communities which provide key ecosystem functions and display fastest responses to environmental change. However, regional land-use effects on local environmental conditions are often neglected, and, hence, underestimated when investigating community responses. Here we show that the effects stemming from agricultural and forested land use are strongest reflected in water conductivity, pH and phosphorus concentration, shaping microbial communities and their assembly processes. Using a joint species distribution modelling framework with community data based on metabarcoding, we quantify the contribution of land-use types in determining local environmental variables and uncover the impact of both, land-use, and local environment, on microbial stream communities. We found that community assembly is closely linked to land-use type but that the local environment strongly mediates the effects of land-use, resulting in systematic variation of taxon responses to environmental conditions, depending on their domain (bacteria vs. eukaryote) and trophic mode (autotrophy vs. heterotrophy). Given that regional land-use type strongly shapes local environments, it is paramount to consider its key role in shaping local stream communities.

Digital twins: dynamic model-data fusion for ecology.

Digital twins (DTs) are an emerging phenomenon in the public and private sectors as a new tool to monitor and understand systems and processes. DTs have the potential to change the status quo in ecology as part of its digital transformation. However, it is important to avoid misguided developments by managing expectations about DTs. We stress that DTs are not just big models of everything, containing big data and machine learning. Rather, the strength of DTs is in combining data, models, and domain knowledge, and their continuous alignment with the real world. We suggest that researchers and stakeholders exercise caution in DT development, keeping in mind that many of the strengths and challenges of computational modelling in ecology also apply to DTs.

High-resolution 3D forest structure explains ecomorphological trait variation in assemblages of saproxylic beetles.

Climate, topography and the 3D structure of forests are major drivers affecting local species communities. However, little is known about how the specific functional traits of saproxylic (wood-living) beetles, involved in the recycling of wood, might be affected by those environmental characteristics.Here, we combine ecological and morphological traits available for saproxylic beetles and airborne laser scanning (ALS) data in Bayesian trait-based joint species distribution models to study how traits drive the distributions of more than 230 species in temperate forests of Europe.We found that elevation (as a proxy for temperature and precipitation) and the proportion of conifers played important roles in species occurrences while variables related to habitat heterogeneity and forest complexity were less relevant. Furthermore, we showed that local communities were shaped by environmental variation primarily through their ecological traits whereas morphological traits were involved only marginally. As predicted, ecological traits influenced species' responses to forest structure, and to other environmental variation, with canopy niche, wood decay niche and host preference as the most important ecological traits. Conversely, no links between morphological traits and environmental characteristics were observed. Both models, however, revealed strong phylogenetic signal in species' response to environmental characteristics.These findings imply that alterations of climate and tree species composition have the potential to alter saproxylic beetle communities in temperate forests. Additionally, ecological traits help explain species' responses to environmental characteristics and thus should prove useful in predicting their responses to future change. It remains challenging, however, to link simple morphological traits to species' complex ecological niches. Read the free Plain Language Summary for this article on the Journal blog.

Macrosystem community change in lake phytoplankton and its implications for diversity and function.

Aim: We use lake phytoplankton community data to quantify the spatio-temporal and scale-dependent impacts of eutrophication, land-use and climate change on species niches and community assembly processes while accounting for species traits and phylogenetic constraints. Location: Finland. Time period: 1977-2017. Major taxa: Phytoplankton. Methods: We use hierarchical modelling of species communities (HMSC) to model metacommunity trajectories at 853 lakes over four decades of environmental change, including a hierarchical spatial structure to account for scale-dependent processes. Using a "region of common profile" approach, we evaluate compositional changes of species communities and trait profiles and investigate their temporal development. Results: We demonstrate the emergence of novel and widespread community composition clusters in previously more compositionally homogeneous communities, with cluster-specific community trait profiles, indicating functional differences. A strong phylogenetic signal of species responses to the environment implies similar responses among closely related taxa. Community cluster-specific species prevalence indicates lower taxonomic dispersion within the current dominant clusters compared with the historically dominant cluster and an overall higher prevalence of smaller species sizes within communities. Our findings denote profound spatio-temporal structuring of species co-occurrence patterns and highlight functional differences of lake phytoplankton communities. Main conclusions: Diverging community trajectories have led to a nationwide reshuffling of lake phytoplankton communities. At regional and national scales, lakes are not single entities but metacommunity hubs in an interconnected waterscape. The assembly mechanisms of phytoplankton communities are strongly structured by spatio-temporal dynamics, which have led to novel community types, but only a minor part of this reshuffling could be linked to temporal environmental change.

Spatial cumulant models enable spatially informed treatment strategies and analysis of local interactions in cancer systems.

Theoretical and applied cancer studies that use individual-based models (IBMs) have been limited by the lack of a mathematical formulation that enables rigorous analysis of these models. However, spatial cumulant models (SCMs), which have arisen from theoretical ecology, describe population dynamics generated by a specific family of IBMs, namely spatio-temporal point processes (STPPs). SCMs are spatially resolved population models formulated by a system of differential equations that approximate the dynamics of two STPP-generated summary statistics: first-order spatial cumulants (densities), and second-order spatial cumulants (spatial covariances). We exemplify how SCMs can be used in mathematical oncology by modelling theoretical cancer cell populations comprising interacting growth factor-producing and non-producing cells. To formulate model equations, we use computational tools that enable the generation of STPPs, SCMs and mean-field population models (MFPMs) from user-defined model descriptions (Cornell et al. Nat Commun 10:4716, 2019). To calculate and compare STPP, SCM and MFPM-generated summary statistics, we develop an application-agnostic computational pipeline. Our results demonstrate that SCMs can capture STPP-generated population density dynamics, even when MFPMs fail to do so. From both MFPM and SCM equations, we derive treatment-induced death rates required to achieve non-growing cell populations. When testing these treatment strategies in STPP-generated cell populations, our results demonstrate that SCM-informed strategies outperform MFPM-informed strategies in terms of inhibiting population growths. We thus demonstrate that SCMs provide a new framework in which to study cell-cell interactions, and can be used to describe and perturb STPP-generated cell population dynamics. We, therefore, argue that SCMs can be used to increase IBMs' applicability in cancer research.

Bayesian Modeling of Sequential Discoveries.

We aim at modeling the appearance of distinct tags in a sequence of labeled objects. Common examples of this type of data include words in a corpus or distinct species in a sample. These sequential discoveries are often summarized via accumulation curves, which count the number of distinct entities observed in an increasingly large set of objects. We propose a novel Bayesian method for species sampling modeling by directly specifying the probability of a new discovery, therefore, allowing for flexible specifications. The asymptotic behavior and finite sample properties of such an approach are extensively studied. Interestingly, our enlarged class of sequential processes includes highly tractable special cases. We present a subclass of models characterized by appealing theoretical and computational properties, including one that shares the same discovery probability with the Dirichlet process. Moreover, due to strong connections with logistic regression models, the latter subclass can naturally account for covariates. We finally test our proposal on both synthetic and real data, with special emphasis on a large fungal biodiversity study in Finland. Supplementary materials for this article are available online.

Understanding and applying biological resilience, from genes to ecosystems.

The natural world is under unprecedented and accelerating pressure. Much work on understanding resilience to local and global environmental change has, so far, focussed on ecosystems. However, understanding a system's behaviour requires knowledge of its component parts and their interactions. Here we call for increased efforts to understand 'biological resilience', or the processes that enable components across biological levels, from genes to communities, to resist or recover from perturbations. Although ecologists and evolutionary biologists have the tool-boxes to examine form and function, efforts to integrate this knowledge across biological levels and take advantage of big data (e.g. ecological and genomic) are only just beginning. We argue that combining eco-evolutionary knowledge with ecosystem-level concepts of resilience will provide the mechanistic basis necessary to improve management of human, natural and agricultural ecosystems, and outline some of the challenges in achieving an understanding of biological resilience.