Tancredi Caruso

Seabird populations fluctuate dramatically in response to perturbations and human activity, such as the current highly pathogenic avian influenza (HPAI) panzootic and fishery practices. Seabirds vector marine nutrients into terrestrial and coastal environments causing cascading effects through ecosystems, but quantifying the inter-related mechanisms underpinning marine-terrestrial cascades in response to changes in seabird populations is challenging. Here we simulate decreases and increases in seabird density, as have occurred due to HPAI and fishing practices respectively, through turf transplant experiments and stratified surveys to quantify marine nutrient cascades and show how they regulate terrestrial ecosystem processes and biodiversity. We use structural equation modelling to quantify the relative strengths of each hypothesised mechanistic pathway in the ecological network. This analysis identified that seabirds shape terrestrial plant and invertebrate biodiversity primarily indirectly, through modifying soil nutrient availability. Our findings highlight the ecosystem changes that are likely occurring, and the pathways through which they operate, due to acute declines and chronic increases in seabirds caused by HPAI, fishing practices or other drivers. Significance statement: We simulate increases and decreases in seabird populations to test their impact on terrestrial ecosystem functioning and biodiversity in Atlantic islands. We show using stratified surveys, modelling and manipulation experiments that invertebrate community diversity and ecosystem processes are driven by cascades of ocean-derived nutrients into soils and through plants via deposition of pellets and faeces from inland-nesting seabirds. Our findings demonstrate how catastrophic decreases in seabird populations, as has occurred due to the avian influenza panzootic, and increases due to marine fisheries, drive terrestrial ecosystem function and biodiversity.

In population ecology, the Jacobian matrix links species interactions to local dynamics and stability, but is often treated as a single realization despite being state-dependent and uncertain. We propose an ensemble perspective, replacing one Jacobian with a maximum-entropy distribution of matrices consistent with accessible structure. We use the ensemble entropy as a measure of the number of Jacobians compatible with the same large-scale organization and hypothesize that higher entropy promotes robustness by preserving dynamically similar interaction structures under parameter changes. We test this hypothesis in a Lotka-Volterra model by perturbing parameters and comparing trajectory divergence with entropy and Jacobian trace, a measure of local contraction. Higher entropy identifies states with lower short-term divergence and ranks responses better than trace. This advantage persists under moderate magnitude errors but disappears when interactions are misspecified. Consistently, across nutrient switches in two predator-prey chemostat experiments, higher pre-switch entropy is associated with lower cycle-to-cycle divergence.

Soil food webs contribute significantly to ecosystem carbon fluxes, but how carbon flux through them is affected by both intensification of land management and climate extremes, such as drought, is poorly understood and quantified. Here, we quantified in situ the energetics of soil food webs under experimental drought across grasslands with contrasting management intensity, which allowed us to test how management shapes the response to drought. Immediately after drought, total ecosystem respiration and food web energy fluxes were reduced relative to controls, whereas the fungal to bacterial energy flux ratio increased. Intensive management affected all these variables but affected only the responses of some of these variables to drought. Food web stability was also affected, with increased return time under intensive management and drought. Although we found that soil moisture, ecosystem respiration, stability, and food web and fungal to bacterial energy flux recovered after two months, nematodes and bacteria remained negatively impacted, while all other microarthropod groups increased. These findings show that while total fluxes and dynamical stability may have recovered from the drought, both community structure and pathways of energy flux remained sensitive to past drought. These long-lasting effects may be critical in regulating pathways of below-ground carbon flux and ecosystem carbon emissions from soil.

Endophytic bacteria are increasingly recognised for their roles in plant health through symbiosis. However, methodological challenges, such as inconsistent root sterilisation, inefficient microbial DNA extraction, and co-amplification of plant organellar DNA, limit accurate characterisation of these communities, especially in wild grassland plants and non model plant in general. To address this, we developed and tested a streamlined protocol for bacterial endophyte detection from wild grassland plant roots, encompassing surface sterilisation of roots, DNA extraction, clamping of plant internal mitochondrial and chloroplast DNA, and 16S rRNA amplicon sequencing. Our approach minimises plant DNA contamination and yields high-quality microbial profiles. The protocol is adaptable and specific to grassland plant species, offering a standardised foundation for endophyte studies in wild and non-model plants. Graphical (Haskins and Ajaz, 2026) https://BioRender.com/47gd2xr

Summary: There is uncertainty in whether there is a common pattern of nestedness and modularity in plant–arbuscular mycorrhizal (AM) fungi associations, partly because of limitations arising from the use of null models that randomly rewire the observed connections to test for non‐random patterns in the network. Here, we overcome these limitations by generating null association matrices using maximum entropy network modelling, and specifically the bipartite binary configuration model (BiCM) with degree distributions as soft constraints. This was used to test the hypothesis that nestedness and modularity are prevalent in plant–AM fungi associations. In contrast to past findings, we found most plant–AM fungi associations were anti‐nested and modular. This pattern was almost universal, being consistent across habitat types, multiple spatial scales, and multiple levels of plant node aggregation, from communities and species to populations. Anti‐nestedness can easily emerge from modularity when network patterns are determined by the identity of the plant and AM fungal nodes. Our findings emphasize the need for experiments that test the factors that cause the observed network structure and how that structure determines the function and stability of plant–AM fungi association networks.

Deer abundances and distributions have expanded at increasing rates in several regions across the globe due to human-driven land use change, uncontrolled introductions and reintroductions, and insufficient top-down control by human or natural predators. Yet, the impact of unsustainable density of deer on the belowground component of ecosystems is understudied. Ireland is a good model for anthropic environmental context in which the population of large herbivores is not regulated by large predators nor planned hunting management. Through a combination of camera trapping paired with soil and vegetation surveys conducted with high-throughput technology, we monitored 50 sites in Irish woodlands to study the effect of varying sika deer ( Cervus nippon ) relative densities on soil biota. Site-level deer intensity of use correlated with vegetation through an increase in litter and moss quantity and reduction in understory vegetation. Furthermore, the effect covaried with soil pH and cascaded on soil microarthropod, bacterial and fungal communities, with more opportunistic taxa linked to high relative deer densities. We also found that variation in deer intensity of use directly affects vertebrate communities, which shifted towards a dominance of foxes, squirrels, cats and dogs – species more commonly associated to human-dominated landscapes – and a reduction in birds and mustelids, which also cascaded on soil biota. Our results thus support the hypothesis that deer overabundance has a strong potential to drive the structure of soil biota mostly through indirect effects via vertebrate fauna and vegetation, acting as a powerful ecosystem engineer. Future studies will have to experimentally tease apart the mechanisms underpinning the patterns documented in this study.

Drought can have long‐lasting legacy effects on terrestrial ecosystems via persistent shifts in soil microbial community structure and function. Yet, the role drought intensity plays in the formation of soil‐mediated drought legacies and in determining plant and microbial responses to subsequent droughts is unknown. Here, we evaluate how soil‐mediated drought legacies shaped by the intensity of an initial drought event influence plant and microbial communities in the following year and their response to a subsequent experimental drought. We determined these responses in two model grassland communities with contrasting resource acquisition strategies. We found that the intensity of the initial (i.e., past) drought shaped the composition, network structure and functioning of soil microbial communities, with stronger effects on prokaryotes than fungi. Moreover, drought intensity determined soil‐mediated legacy effects on plant responses to a subsequent drought: increasing past drought intensity decreased the drought resistance of the slow‐strategy plant community and reduced productivity overshoot in the fast‐strategy community after re‐wetting. Our findings demonstrate that increasing drought intensity can lead to distinct legacies in soil microbial community composition and function with impacts on plant responses to future droughts.

1.) Increasing atmospheric CO2 concentrations alongside more frequent and severe droughts are key global change factors impacting litter decomposition and global carbon cycles. However, we have a poor understanding of how both drought and elevated CO2 impact interactions between initial litter chemical properties and abiotic and biotic components of the decomposition environment, especially under field conditions. 2.) We close this gap by testing how drought and elevated atmospheric CO2 concentrations modify rates of litter decomposition by acting through two pathways, the decomposition environment and initial litter properties. For this analysis, we used two long-term manipulative field experiments in mature oak woodlands: the Birmingham Institute of Forest Research Free-Air CO2 Enrichment (BIFoR FACE) experiment and the Oak Observatory at the Observatoire de Haute-Provence (O3HP) drought experiment that reduces seasonal precipitation. In each experiment, litter bags were deployed in a reciprocal transplant design whereby control and experimental litter (i.e. litter derived from plots under either drought or elevated CO2 conditions) were incubated in control and experimental plots at the respective experiment. We measured litter mass loss, functional groups, C:N ratios, moisture content, and microbial and mesofauna communities. 3.) We found that drought and elevated CO2 concentrations modified mass loss rates through different pathways. In the drought experiment, litter placed in droughted conditions decomposed slower as the decomposition environment shaped mass loss rates more than initial litter properties. By contrast, both decomposition environment and initial litter properties impacted litter C:N, functional groups, and microbial communities during the experiment. In the elevated CO2 experiment, experimental litter derived from elevated CO2 plots corresponded with slower mass loss over the first three harvests as the decomposition environment did not impact litter properties or decomposers. 4.) Synthesis: We found that drought and elevated atmospheric CO2 can modify mass loss rates through distinct pathways. Further research is needed to disentangle the relative roles of climate and decomposers with decomposition rates under drought and elevated CO2, but our experiment highlights that litter decomposition will be reshaped by global change factors through multiple mechanisms.

Motivation: Here, we make available a second version of the BioTIME database, which compiles records of abundance estimates for species in sample events of ecological assemblages through time. The updated version expands version 1.0 of the database by doubling the number of studies and includes substantial additional curation to the taxonomic accuracy of the records, as well as the metadata. Moreover, we now provide an R package (BioTIMEr) to facilitate use of the database.Main Types of Variables Included: The database is composed of one main data table containing the abundance records and 11 metadata tables. The data are organised in a hierarchy of scales where 11,989,233 records are nested in 1,603,067 sample events, from 553,253 sampling locations, which are nested in 708 studies. A study is defined as a sampling methodology applied to an assemblage for a minimum of 2 years.Spatial Location and Grain: Sampling locations in BioTIME are distributed across the planet, including marine, terrestrial and freshwater realms. Spatial grain size and extent vary across studies depending on sampling methodology. We recommend gridding of sampling locations into areas of consistent size.Time Period and Grain: The earliest time series in BioTIME start in 1874, and the most recent records are from 2023. Temporal grain and duration vary across studies. We recommend doing sample‐level rarefaction to ensure consistent sampling effort through time before calculating any diversity metric.Major Taxa and Level of Measurement: The database includes any eukaryotic taxa, with a combined total of 56,400 taxa.Software Format: csv and. SQL.