Department Aquatic Ecology

Species distribution models (SDMs) are central tools for predicting and forecasting how species respond to environmental changes, yet their reliability depends on accurately capturing ecological processes across spatial scales while accounting for uneven data availability and resolution. Models calibrated over spatially restricted extents are often prone to truncating species' niches, whereas broad-scale models typically rely on coarse datasets that overlook fine-scale habitat constraints. The spatially nested SDM (N-SDM) approach addresses this challenge by combining global- and regional-level models to couple large-scale drivers with detailed, high-resolution environmental data. Here, we present N-SDM 2.0, a reengineered and extended successor to the original N-SDM software – the first end-to-end high-performance computing platform for spatially nested species distribution modelling. The new version offers improved usability and modularity, along with an extended range of modelling features, while maintaining continuity for users of the previous release. New modelling features include additional nesting methods, background-point generation strategies, and cross-validation options. These developments are complemented by an improved workflow architecture, a refined settings system, streamlined data-preparation steps, and greater control over data usage across spatial levels as well as error handling. N-SDM 2.0 is designed for flexible integration into heterogeneous high-performance computing environments, supporting large-extent and high-resolution biodiversity forecasting across an extensive number of species and projection scenarios. Beyond pure technical advances, N-SDM 2.0 strengthens the conceptual foundation of spatially nested species distribution modelling, introducing new features that form an open and scalable framework for mapping species distributions and robustly projecting their responses to global environmental changes.

Translational participatory research for environmental management occupies a demanding space at the interface of societal needs and academic research. Projects in this space usually face the challenge of reconciling stakeholder involvement and real-world applicability with academic career requirements such as excellence in fundamental research and the attraction of substantial research funds. Here, we share our experiences from a 12-year translational participatory research initiative on the management of invasive gobies in central Europe. We describe the project, synthesize inputs, outputs, and outcomes, and summarize conditions and practices across political, administrative, societal, institutional, team, and individual levels that we found important and beneficial in our attempts to produce both academic and applied outcomes. Our impression is that an interaction of factors across these levels created windows of opportunity, and that the team constellation and setting allowed us to seize these opportunities for the sake of both scientific progress and societal impact. By reflecting from the perspective of a multidisciplinary group of natural scientists directly engaged in such a project, we aim to contribute a personal account to social science frameworks on translational, participatory, inter- and transdisciplinary research. Our retrospective potentially can inform policy makers, funders, and institutions who seek to create enabling environments for interface research, and is helpful to fellow natural scientists who want to prepare the ground for their own translational participatory research initiatives.

Seafloors are crucial to marine ecosystems for the functions and services they provide. Benthic organisms, vital to these ecosystems, are particularly vulnerable to climate change. Rising temperatures, ocean acidification, and shifting currents disrupt benthic species and communities, yet future related impact assessments remain limited. Here, we trained species distribution models with predictors from state of the art physical and biogeochemical marine models and a large database of species records (> 100,000 occurrences) to project the current and future distributions of ~350 benthic species (excluding cephalopods, invasive species, and commercially exploited species) and their related changes per site in diversity (α-) and community composition (β-diversity) over the Mediterranean Sea. We predicted most species to shift their distribution northwards for all future scenarios due to changes in water temperature and dissolved oxygen close to the seafloor, with up to 60% of species experiencing range contraction, 77% moving northwards, 20% experiencing range fragmentation (measured as range disjunctions in models' output), and 30% moving toward deeper waters over time. Cold-adapted species were more likely to face range contraction and shifts towards deeper waters, while warm-adapted species were more likely to face range expansions and shifts towards shallower waters. α-diversity increased in the Northern and decreased in the Southern Mediterranean, respectively. Changes in β-diversity within sites highlighted compositional changes (species turnover) in communities located in the Aegean and Tyrrhenian Seas, in deep parts of the Ionian Sea, and in coastal regions of the Adriatic Sea. Climate-smart, ecosystem-based Marine Spatial Planning can capitalize on the identified hotspots of species losses, gains, stability, and turnover. Prioritizing connectivity in regions of strong turnover and extending protected areas in regions with stable α-diversity and limited turnover is recommended for improved conservation actions.