Carsten Friedrich Dormann

Pollination is a key ecological process sustaining biodiversity and food security, yet global patterns of plant–pollinator specialization remain unresolved. Using a global dataset of quantitative plant–pollinator networks (>3,400 networks, >110,000 interactions), we show that specialization is latitudinally structured, but not as a single directional pattern. We find no consistent monotonic increase or decrease in specialization towards the Equator. Instead, network-level and pollinator specialization peak at low northern latitudes near the tropical–subtropical boundary, whereas plant specialization is hemispherically asymmetric. Climate is most often the best-supported predictor of specialization compared with latitude, biodiversity and environmental productivity. Network-level specialization declines with increasing mean annual temperature, with strong group-specific differences. Precipitation exerts nonlinear effects: plant specialization peaks at intermediate precipitation but is lower at the wettest sites, whereas pollinator specialization is lower at intermediate precipitation, although often in interaction with temperature. Pollinator groups diverge strongly, especially in contrasting precipitation-related patterns between birds and insects. These findings show that global plant–pollinator specialization is structured by latitude and climate in complex and taxon-specific ways, with implications for uneven responses of pollination networks to future climatic change.

Environmental sciences aim to provide unbiased data for decision making, yet conservation management, and personal decisions often rely on perception shaped by experiences, information exchange, media exposure, and values. When monitoring data differs from people’s perceptions, unintended consequences (e.g. overlooked degradation, unacknowledged conservation successes) may occur. We compare long-term monitoring data of tree and bird abundance with self-reported perceptions, revealing mismatches in 74% of the cases as well as their drivers. Phenomena like shifted baselines and change blindness were prevalent. Importantly, when environmental data reported population increases, these were often overlooked or minimized, but when species declined, the two datasets were more consistently aligned. This highlights the need to base decision-making on long-term datasets that account for the mismatches in perception and monitoring data. Our study has implications for conservation science, restoration, and land management, recommending integration of long-term monitoring data to mitigate change blindness and shifted baselines.

The UN Decade on Ecosystem Restoration aims to stop biodiversity losses ¹ . Approximately 60% of tropical forests have already been lost or severely degraded ² , making restoration essential to achieve conservation goals. Recovery trajectories of trees have been studied intensively ^3,4 , but a comprehensive understanding of biodiversity recovery is lacking. Here we analyse recovery trajectories across trophic levels including 16 taxonomic groups from three kingdoms in a lowland tropical forest by investigating resistance to perturbation, recovery times and return rates to old-growth forest conditions. Abundance and diversity regained more than 90% and composition approximately 75% similarity to old-growth forests within 30 years, but full recovery takes several decades. Mobile animal communities acting as seed dispersers or pollinators had high resistance levels and recovered faster than trees or tree seedlings. Return rates contributed 1–2.5 times more than resistance to the recovery times of species composition. Taxon-specific recovery times could not be explained by simple mechanisms (life-history strategies, trophic level or mobility). We show the enormous potential of protecting naturally recovering secondary forests to stop and reverse biodiversity losses.

Making robust and generalizable predictions within ecological systems such as forests remains challenging due to limited data availability and the slow pace of environmental change. To address this, we integrate a semi-empirical environmental process model (PRELES) to support deep learning approaches, specifically artificial neural networks (ANNs). We replicate and extend previous work on process-guided neural networks (PGNN) by introducing new model types and conducting a comprehensive hyperparameter optimisation within systematic nested cross-validation analyses in both data-thinning and extrapolative scenarios. Results show that both data-driven ANNs and PGNNs consistently outperform the stand-alone process model, while PGNNs provide additional advantages over ANNs in data-sparse settings and under transfer scenarios to unseen, changing climatic conditions. We further estimate the generalisation error for data-driven models as a function of the amount of training data, allowing for guidance on model suitability under different data availability. A variable importance analysis using accumulated local effects reveals that both PGNNs and ANNs learn simple, physically plausible relationships, whereas PRELES exhibits a strong bias toward boreal conditions and limited ability to predict unseen, climatically divergent sites.

1. Meta-analyses are a reliable method for a quantitative research synthesis. They are, however, prone to specific biases that can be introduced in the process. Such a bias could exist if primary literature produces similar results if coming from the same authors. Authorship network bias is the non-independence of effect sizes introduced by the overlap of authors of primary studies. If not accounted for, it can severely impact the quality of meta-analysis and the conclusions drawn from it. 2. To account for such non-independence, multilevel models with author clusters as an additional hierarchy level were recently suggested. We propose a new method for the detection of non-independent effect sizes based on authorship networks and for their correction. 3. An analysis of simulated data demonstrates the effectiveness of the here-suggested new method. We further applied our new method to nine exemplary meta-analyses. 4. Our new method for detection and effective correction can be easily integrated in existing meta-analysis workflows, using the functionality already offered by R’s metafor package. 5. Our goal is to enhance the reliability of meta-analyses by highlighting potential authorship network bias and offering a method to address this often-overlooked bias.

Plant functional traits are fundamental to ecosystem dynamics and Earth system processes, but their global characterization is limited by available field surveys and trait measurements. Recent expansions in biodiversity data aggregation—including vegetation surveys, citizen science observations, and trait measurements—offer new opportunities to overcome these constraints. Here we demonstrate that combining these diverse data sources with high-resolution Earth observation data enables accurate modeling of key plant traits at up to 1 km ² resolution. Our approach achieves correlations up to 0.63 (15 of 31 traits exceeding 0.50) and improved spatial transferability, effectively bridging gaps in under-sampled regions. By capturing a broad range of traits with high spatial coverage, these maps can enhance understanding of plant community properties and ecosystem functioning, while serving as tools for modeling global biogeochemical processes and informing conservation efforts. Our framework highlights the power of crowdsourced biodiversity data in addressing longstanding extrapolation challenges in global plant trait modeling, with continued advancements in data collection and remote sensing poised to further refine trait-based understanding of the biosphere.

In the web of life, every interaction between species tells a story of cooperation, conflict, or chance. For centuries, ecologists have charted these stories to better understand phenomena such as pollination and disease. We have been using the lens of network science to distill them into topological patterns such as nestedness or modularity. Yet, like in the old buddhist parable of the blind monks and the elephant, these separate views have often missed the bigger picture. In this Perspective, we review the history of topological studies of interaction networks, challenge the “black-and-white” paradigm, and point out a way forward: the Integrative Theory of Interaction Networks (ITIN). This theory unveils the logic of how interaction networks assemble through sequential, rule-bound processes shaped mainly by resource dissimilarity and interaction type. We show that compound topologies are likely the norm in function-oriented, taxonomically-inclusive, well-sampled networks. ITIN is deductive, predictive, and testable, mapping from first principles to real-world systems, thus helping us grasp “the whole elephant”. As ecologists confront the polycrisis, from irreparable biodiversity loss to global health threats, ITIN offers a coherent framework to understand, predict, and restore the threads of the web of life.

Pollination is a key ecological process sustaining biodiversity and food security, yet global patterns of plant–pollinator specialisation have remained unresolved. Using the largest global dataset of quantitative networks (>3,400 networks, >110,000 interactions), we show that the latitudinal specialisation gradient (LSG) exists, but it is non-linear, hemispherically asymmetric, and strongly taxon-dependent. Network-level and pollinator specialisation were lowest in the tropics and peaked at northern mid-latitudes, whereas plants tended to become more specialised toward higher latitudes. Climate consistently outperformed latitude, species richness, and environmental productivity as a predictor of these patterns. Specialisation declined with increasing temperature, rose with moderate rainfall before declining at the wettest sites, and increased with temperature seasonality, but plants and pollinators responded differently to these drivers. Functional groups diverged strongly: ectothermic insects were most specialised in cooler, seasonal climates, while birds showed weaker links to latitude but reduced specialisation in wetter regions. These findings demonstrate that climate, rather than latitude or species richness, structures global variation in specialisation. Because warmer and less seasonal climates promote generalisation, climate change is likely to disrupt the most specialised pollination systems, unevenly across taxa and regions, with important consequences for biodiversity and ecosystem stability.

Recreation (i.e., hiking and biking) and hunting can occur simultaneously in time and space, and both sources of disturbance affect wildlife behavior, leading to reactions resembling anti‐predator behavior. However, the additive effects of lethal and non‐lethal human disturbances on wildlife are only beginning to be understood, and research on the impact of hunting on non‐target species is limited. Recreation and hunting commonly co‐occur in areas where wildlife is present, and understanding their combined effects on wildlife behavior is crucial for protected area management. Using records from 122 camera traps placed along trails and in surrounding forests, we assessed the effect of varying intensities of hunting and recreation over space and time on the temporal activity of red deer ( Cervus elaphus ), roe deer ( Capreolus capreolus ), wild boar ( Sus scrofa ), red fox ( Vulpes vulpes ), and Eurasian lynx ( Lynx lynx ) in the Bavarian Forest National Park, Germany. We documented the relative abundance of these species on trails versus in forests and applied Bayesian models to assess how hunting and recreation influenced wildlife nocturnality. Our results suggest that hunting is a strong driver behind wildlife temporal behavior. Hunting amplified avoidance of non‐lethal recreation and potentially impacts species interactions. Red deer exhibited the most pronounced temporal avoidance of both hunting and recreational activity, increasing nocturnality and trail avoidance as these disturbances increased. Red deer were more diurnal in the non‐hunting zone and decreased nocturnal activity with increasing distance from the hunting zone. Wild boar and non‐hunted species exhibited moderate or negligible responses. However, high hunting effort led to species not targeted by hunting (roe deer and red fox) increasing their temporal avoidance of recreational activities, with wild boar and roe deer avoiding trails more strongly. In the context of protected area management, our results suggest that strictly reducing hunting in space and time while concentrating recreation in certain areas to create disturbance‐free habitat year‐round has great potential to reduce the temporal avoidance of humans by wildlife, thereby fostering nature conservation goals by protecting natural processes.

In the past decades, studies on interaction networks may have zoomed in too much on arbitrary subsets of ecological communities. Consequently, the current “nestedness paradigm” is biased towards restricted taxonomic groups. This bias leads to a myopic understanding of ecological functions and ecosystem services, with serious consequences for basic and applied ecology.