Forecasting ecosystems under climate change requires multi-species approaches that account for ecological interactions through ecological networks. While previous projections have used network approaches mainly focusing on extinction-cascade directionality as a driver of ecological change, few multi-species projections have considered the resilience mechanisms of ecological networks to species extinctions. These mechanisms are link-loss sensitivity (the inverse of a network’s ability to withstand loss of links when interaction partners go extinct) and realisation of rewiring potential (the capacity to reallocate lost interactions to novel or existing partners). Using a new quantitative framework, we simulate compositional and network changes across globally distributed, in-situ sampled mutualistic frugivory networks. These simulations are based on species-specific climate-driven extinction cascades that incorporate both resilience mechanisms simultaneously. To assess the impact of network resilience mechanisms on biodiversity loss and network connectedness, we introduce and explore two-dimensional resilience landscapes defined by link-loss sensitivity and rewiring potential. We evaluate outcomes under optimistic and pessimistic resilience assumptions, and both low- and high-emission climate change scenarios to improve understanding of the potential consequences of extinction cascades across the Earth. Additionally, we compare climate driven projections with baseline scenarios that remove the least- and most-connected species in each network. In contrast to traditional models emphasizing bottom-up or top-down cascades, we account for bi-directional extinction processes to better reflect ecological reality. Our findings suggest that ignoring ecological network resilience mechanisms may significantly underestimate ecological change, particularly in terms of biodiversity loss and shifts in network connectedness under climate change.
This paper presents an evolutionary game-theoretic model to explore the transition of human economies between non-symbiotic and symbiotic regimes, driven by the level of renewable natural resource availability. Drawing analogies to biological ecosystems, we formalize a threshold-driven function that triggers regime shifts when resource stocks dip below a critical limit. Through replicator dynamics grounded in the Price equation, our results show that, in a non-symbiotic regime, populations move from competitive to predatory behaviors, overshooting the planet's regenerative capacity. Conversely, a symbiotic regime promotes cooperation and resource-sharing, thus reducing the risk of ecological collapse. Simulations indicate that, under the normalized ecological-footprint indicator used in the model, a predator-dominated world would demand renewable-resource services equivalent to approximately 2.30 times the system's regenerative capacity. In contrast, a mutualist-dominated scenario reduces this demand to approximately 0.99 times regenerative capacity, conditional on lower depletion intensity and greater responsiveness to resource scarcity. The findings highlight how strategic choices shape ecological footprints, offering pathways for policy interventions to encourage collective solutions and minimize exploitation pressures.
To improve the accuracy and integrity of carbon offset evaluations, this study proposes an integrated framework using System Dynamics modeling that accounts for both carbon stock and flow in natural and managed ecosystems. We introduce the Atmosphere-Biomass-Soil (ABS) model, which simulates carbon budgets by integrating fluxes among atmospheric, biomass, and soil pools, while considering growth, sequestration, inputs, and harvests. Three case studies are examined: (1) Japanese average monoculture vs. locally practiced polyculture (synecoculture), (2) global temperate monoculture vs. standardized synecoculture, and (3) tropical paddy fields, rubber plantations, and forest regeneration vs. synecoculture in Indonesia, with simulated beta diversity scenarios. Results show that conventional monoculture generally yields positive carbon budgets due to high inputrelated emissions and limited retention. In contrast, synecoculture consistently achieves net-negative budgets by reducing emissions and enhancing sequestration. In temperate zones, converting monoculture to synecoculture can offset carbon equivalent to natural forests within 6-18 years. In tropical contexts, synecoculture shows earlystage offset benefits over paddy and rubber systems; however, its long-term advantage narrows relative to highproductivity systems like rubber plantations, whose offset potential may be overstated due to short product lifespans and limited recyclability. Beyond simulation, we propose a theory-driven framework for optimizing ecosystem sampling, incorporating normalized proxies, offset capacity metrics, and a cost-aware strategy based on statistical confidence. This framework enables robust and comprehensive assessments of ecosystem multifunctionality while reducing the risk of overvaluation and greenwashing.
Understanding and identifying causal relationships within complex, dynamic ecosystems is essential for elucidating ecological mechanisms and guiding effective ecosystem management. Conventional statistical approaches predominantly quantify correlations among multiple variables, yet fall short of capturing causality. Existing causal-discovery tools in ecology either focus on pairwise interactions, thereby overlooking emergent effects arising from the simultaneous influence of multiple drivers, or fail to provide a direct metric for the strength of causal control. Consequently, there is an urgent need for a framework that can simultaneously reconstruct causal networks among numerous variables and furnish quantitative assessments of causal importance. To address this challenge, we developed a dual-strategy ecological causal-discovery (DS-ECD) model founded on Granger causality that integrated a forward local-search strategy with a backward global-search strategy to detect causal links in long-term time-series data. The local strategy employed a forward greedy search to construct an information set capturing significant individual-level causal relationships, while the global strategy utilized a backward one-step search to uncover group-level causal interactions. In addition, we introduced the Relative Causality-Driven Intensity (RCDI) metric to quantify causal strength by decomposing direct and indirect effects, which complemented existing causal-discovery tools. Simulation experiments demonstrated robust model performance under high noise and high dimensionality. Accordingly, DS-ECD was deployed in ecosystems across different scenarios, rapidly revealing intuitive causal networks together with their associated RCDI values. Being purely data-driven, the approach promptly delivers transparent causal graphs and quantitative intensity values, facilitating cross-validation with experimental or process-model results and offering a reference for ecosystem-management practices.
The Antarctic marine ecosystem forms a tightly coupled food web linking phytoplankton, krill, and whales, and is highly sensitive to human disturbance. Historical overexploitation of whales was expected to trigger a trophic cascade that increased krill abundance, yet observations revealed a paradoxical decline, a phenomenon known as the krill paradox. To explore the mechanisms underlying this unexpected pattern, we develop and analyze a harvested tri-trophic model that captures feedbacks among phytoplankton, krill, and whales. The model incorporates Holling type I functional responses and harvesting at multiple trophic levels to examine how exploitation and nonlinear interactions shape ecosystem outcomes. Analytical results show that the system can admit up to four positive equilibria. Using numerical continuation methods, we identify and characterize various bifurcations, including saddle-node, Hopf, degenerate Hopf, Bogdanov-Takens, zero-Hopf, and higher-order nilpotent types. Within ecologically realistic parameter ranges, these bifurcations generate multiple stable states, oscillations, and chaotic dynamics. Our findings suggest that the krill paradox may emerge from intrinsic ecological complexity and nonlinear feedbacks rather than simple trophic release, highlighting how multistability and long-term unpredictability can influence the resilience of marine ecosystems.
Montane stream ecosystems are increasingly exposed to pressures from human land use and climate change. Although periphytic algae are important primary producers in stream ecosystems, their responses to these pressures across elevational gradients remain poorly understood. This study examined the elevational pattern of periphytic algal biomass and identified the key environmental variables and functional traits associated with biomass variation. The results revealed a consistent positive association between periphytic algal biomass and elevation. Farmland percentage (FP) and water temperature (WT) emerged as the principal upstream factors shaping biomass variation along the elevational gradient, whereas salinity and total dissolved solids (TDS) acted as key intermediate variables linking these upstream factors to biomass responses. In low-elevation streams, higher FP and WT accounted for lower periphytic algal biomass, partly through elevating salinity and TDS. In addition, greater landscape fragmentation further inhibited periphytic algal biomass at lower elevations. Cell size and life form were the pivotal functional traits associated with biomass variation and were highly responsive to environmental change. Periphytic algal communities characterized by smaller cell size, stronger attachment capacity, and greater colonization ability exhibited higher biomass at higher elevations. These findings provide new insights into how human land use and climate change shape periphytic algal communities in montane streams, and highlight the importance of functional traits for understanding variation in stream primary production under ongoing global change.
Coiba National Park, a UNESCO World Heritage Site, supports a highly diverse coastal marine ecosystem. This study characterized its structure and functioning by developing an Ecopath mass-balance model for 2021-2022, integrating 42 functional groups. Model evaluation and energy-flow analysis allowed us to assess total consumption, respiration, exports, and detrital pathways, as well as indicators of ecosystem development (ascendancy, development capacity, and system overhead) and trophic structure. The results indicate that the ecosystem is in a developmental phase, with an energetic imbalance typical of early successional stages and moderate efficiency in converting energy into biomass. Lower trophic levels showed strong connectivity, whereas higher levels were sparsely represented, highlighting structural vulnerability to stressors. These findings provide essential insights for conservation planning and underscore the need for management strategies that strengthen functional groups with low connectivity to enhance ecosystem resilience.
This study investigates the active matter dynamics of Escherichia coli bacteria in a polluted fluid environment using computational simulation modeling. Focusing on the interplay between individual bacterial flagella propulsion and emergent collective behavior, we extend our analysis to include spatial clustering of bacteria and its correlation with local fluid flow properties such as pressure and velocity magnitude. The simulations reveal the significant influence of the fluid’s shear-thinning, non-Newtonian rheology on both microscale propulsion and macroscale bacterial aggregation patterns. Quantitative clustering analysis highlights how bacterial swarming adapts dynamically to heterogeneous flow conditions, demonstrating hallmark features of active matter systems. These insights advance understanding of bacterial motility in complex fluids and can inform the design of microfluidic devices and synthetic active materials.
Planktonic systems form complex interaction networks that are pivotal to marine ecosystem health and stability. However, the long-term responses of these interactions to multiple stressors, especially in dynamic coastal and estuarine ecosystems, remain poorly understood. We applied an information-theoretic modeling approach to monthly abundance data of 74 plankton taxa (2000-2020) from Helgoland Roads (German Bight, North Sea) to reconstruct dynamic interaction networks. Our aim was to explore community-and system-level network properties and their relationships with potential environmental drivers. On the community-level, we found that interaction types encompassing predator-prey relationships among species strengthened under high light availability and nitrogen-to-phosphorus ratios of 20-25, whereas interaction types involving both predator-prey and competitive relationships among species (e.g., mixotrophs that can both prey on and compete with autotrophs) intensified under low light and nutrient availability. The latter interactions intensified particularly after 2006. Since then, links from grazers to primary producers have also strengthened, closely associated with a decline in ambient nitrogen-to-phosphorus ratios slightly below the Redfield ratio of 16:1. The system-level properties of connectedness and resilience provided an abstract view of network behavior, revealing that the system's maturation process was non-linearly driven by salinity and nutrient availability. Consistent with complex-systems theory, the interplay between connectedness and resilience indicated that under low environmental variability, the system became highly sensitive to small nutrient changes, potentially triggering a reorganization of the interaction structure. These results highlight the system's dynamic response to environmental changes and advance understanding of plankton network functionality in coastal ecosystems.
The Sustainable Development Goals (SDGs) urge all parties to strengthen the resilience of the food systems in their respective countries. The purpose of this study is to establish feedback loops and offer new insights into how supply chain members manage complex disruptions across four key subsystems: the agricultural subsystem, the agricultural trade subsystem, the food subsystem, and the food waste disposal subsystem. The paper employs qualitative modelling based on in-depth interviews, participant observation, and the use of literature databases. The analysis of pilot entry-level studies demonstrates that this locked and fragmented system of distribution networks and food consumption is crucial for reconciliation and should highlight the benefits of applying the SES approach to food management projects to promote the metabolic sustainability of local food system projects. The findings suggest that the local flow of the food system should be supported by the development of a comprehensive database, extensive stakeholder participation both upstream and downstream, a reduction in reliance on imported food, and market innovation at the production-consumption stage on both urban and rural scales. Enhancements in synergies at the grassroots level, including farmer's groups at the farm level, can offset the negative effects of initiation prior to linking to the subsequent subsystem. The SES framework developed in this study may make it easier for other scientists to examine the complexity of food systems from a transdisciplinary perspective.
Species Distribution Models (SDMs) are widely used to analyze the relationship between species occurrence and environmental factors, offering critical ecological and evolutionary insights. However, the complexity of SDMs, coupled with high-dimensional environmental data, can hinder model interpretability, especially machine learning (ML) based SDMs. To address this, we propose and evaluate an interpretable modeling approach that integrates Feature Selection (FS) techniques to enhance both transparency and predictive performance of ML based SDMs.In the present study, we predict the distribution of seven bird species by evaluating the impacts of six univariate filter- based FS methods (each tested with two thresholds), six wrapper methods, two multivariate approaches, and ensemble wrappers on the classification performance and interpretability. We employed four black-box ML classifiers: Extreme Gradient Boosting (XGB), Decision Tree (DT), Random Forest (RF), and Light Gradient Boosting Machine (LGBM) as well three performance criteria: accuracy, Kappa, and F1-score. Moreover, we used four interpretability techniques (Lime, Shapley additive explanations, Accumulated Local Effects, and Global surrogate) to analyze feature importance and understand how selected variables influence predictions.The findings indicate that wrapper methods outperformed both filter methods and their corresponding ensembles. Further- more, the interpretability analysis across the four classifiers indicated that the highly influential features are Temperature seasonality, Maximum Temperature of the Warmest Month, and Minimum Temperature of the Coldest Month, suggesting their critical role in the prediction process and their overall importance in interpretability assessments. Additionally, the findings confirmed the reliability of the interpretability techniques used and highlighted the effectiveness of the Global Surrogate approach in addressing the accuracy-interpretability trade-off across all black-box models.
The structural complexity of vegetation may favor the parasitoid diversity of a habitat by increasing the number of potential niches for their hosts and for themselves, but how vegetation structural variables are contributing to diversity is still uncertain. Here we aimed to determine the vegetation variables that influence diversity, abundance, richness and assemblage composition of Darwin wasps (Hymenoptera, Ichneumonidae). We considered the life strategies (idiobiont/koinobiont, and ecto/endoparasitoids) and trophic guilds of Ichneumonidae, and how this relationship is shifted in different temporary seasons. We evaluated several vegetation structural variables (height, cover, species richness, life-forms diversity) of the tree, shrub and herbaceous layers, in five different habitats along a structure complexity gradient of vegetation in a protected area of Central Spain, in two seasons: spring and autumn. Vegetation variables were correlated to Ichneumonidae variables. A non-metric multidimensional scaling (NMDS) was conducted to identify patterns of Ichneumonidae species composition in relation to vegetation variables measured in two seasons. Plant richness negatively affected the abundance of koinobionts and endoparasitoids, whilst a greater cover of the herbaceous layer supported a greater abundance of idiobionts and ectoparasitoids. Ichneumonidae assemblages varied significantly regarding habitats but also differed among the two seasons. The structural complexity of plant communities is not sufficient to predict parasitoid diversity, but some vegetation traits, such as a well-developed herbaceous layer, can be considered when planning conservation management strategies to keep a high parasitoid diversity. Other variables, not only at habitat but also at landscape level, should be considered.
In antagonistic interactions, adaptive responses to reciprocal attacks can drive the cyclical dynamics of the “arms race”. Since attack and defense mechanisms are linked to gene expression, which may also affect mate choice, these interactions can drive speciation on an evolutionary scale. Mathematical and computational models play an important role in the investigation of this type of dynamics, since experiments or observations that uncover such long term effects are not generally possible. Although previous works have explored the roles of spatial scale and phylogenetic association on antagonistic interactions, the role of such ecological interactions on the evolutionary dynamics is still not clear. This article aims to analyze how antagonistic interactions between consumers and resources influence consumer diversity, population dynamics, and speciation. Using an IBM model, we analyze the formation of species and the maintenance of biodiversity across extended temporal scales. We considered that resources can have two phenotypes describing two potential niches for the consumers. We demonstrated that the coexistence of consumers and resources occurs under conditions of low resource mutation and that diversity is generally reduced when the two initial phenotypes of the consumers are similar. Our results also indicate that niche separation is likely when initial niche distance is sufficiently large. In contrast, when the initial phenotypes are similar, niche width tends to expand while separating from each other. These simulations effectively illustrate the impacts of the “arms race” between consumers and resources and their coevolution.
The ladybird Eriopis connexa (Germar, 1824), a voracious aphid predator, faces challenges from insecticide applications, compromising biological control. As a result, the number of studies analysing the resistance and susceptibility of ladybirds has increased. Some studies have found that resistant populations differ in predation and foraging behaviour from susceptible ones. This study modelled the population dynamics of resistant and susceptible E. connexa preying on Aphis gossypii Glover, 1877 and Myzus persicae (Sulzer, 1776). A logistic model with density dependence and type II functional response was constructed to analyse predation dynamics, incorporating bifurcation analysis of predation parameters (attack rate and handling time) and the mortality rate of susceptible ladybirds. This model was used to simulate scenarios that included or excluded insecticide application and aphid resistance. To simulate the effects of insecticide applications, the parameters related to the aphids' intrinsic growth rate (r1 and r2) were changed to reflect the responses of susceptible and resistant populations. The same approach was used for the mortality rate of ladybirds (d2 and d3). The results demonstrated that mortality, attack rate, and handling time were critical in shaping predator-prey interactions. Temporal simulations revealed fluctuating abundances, highlighting the fragility of these interactions under insecticide stress. This study contributed to understanding the ecological implications of insecticides, which disrupt natural predation dynamics, and showed how changes in the rates of behaviours can impact prey control. This research demonstrated the importance of integrated strategies that balance insecticide applications with preserving natural enemies and causing sustainable agricultural practices.
Wetland conversion to agricultural use is widespread globally, particularly in semi-arid regions. As key indicators of wetland responses to global change, plant diversity and biomass are fundamentally influenced by management practices and plant functional strategies. While existing research has largely focused on aboveground traits, the role of root functional traits in mediating plant diversity and biomass remains poorly understood. This study investigated the predictive value of root functional traits for plant species diversity and biomass across a range of agriculturally managed wetlands in the Songnen Plain of semi-arid China. The management practices included natural wetlands (NW), moderately grazed wetlands (MG), mowed and grazed wetlands (MSG), mowed wetlands (TM), and heavily grazed wetlands (HG). The results demonstrated NW exhibiting the highest biomass but lowest diversity, whereas HG sites displayed the opposite pattern. Root traits also varied significantly across wetland sites. Stepwise regression analysis revealed that root phosphorus content (RPC), root carbon-to-phosphorus ratio (RCPR), and root length (RL) were key predictors of diversity (R2 = 0.957), while root carbon content (RCC) and specific root area (SRA) are robust predictors of biomass (R2 = 0.762). The partial least squares structural equation model (PLS-SEM) further elucidated two primary pathways by which root functional traits influence diversity and biomass: (1) a direct effect of morphological traits (RL) exerted on diversity, and (2) an indirect effect of morphological traits (SRA, RL) on both diversity and biomass via modulation of root nutrient acquisition traits (RCC, RPC, RCPR). Notably, trade-offs between RL and RPC, together with synergies between SRA and RPC, were associated with reduced diversity but increased biomass. In contrast, synergies between RL and RCC and trade-offs between SRA and RCC exerted negative effects on biomass. In conclusion, integrated root trait combinations and their interactions serve as robust predictors of plant diversity and biomass in agriculturally managed wetlands. These findings advance the theoretical framework for understanding wetland ecological dynamics and provide valuable insights for sustainable resource management in semi-arid regions.
This study presents a four-dimensional mathematical model with time-varying parameters to analyse the diverse effects of global warming on marine ecology over the next 100 years. Key environmental factors, including rising sea surface temperature and decreasing dissolved oxygen concentrations, are evaluated in relation to their influence on plankton species. The model's predictions are validated through a case study, comparing results with prior research. Findings indicate that rising temperatures accelerate the dilution of dissolved oxygen, significantly affecting plankton densities, with zooplankton being more susceptible to temperature changes than phytoplankton. This reduction in zooplankton and oxygen levels is anticipated to impact overall ocean productivity. The study also proposes a threshold for annual temperature increments aligned with global environmental targets. Additionally, a second model incorporating a time delay examines the period required for phytoplankton-released toxins to impact zooplankton populations. Results suggest that the time delay has minimal long-term effect on marine ecology within the study time frame. Overall, this research provides insights into the impact of atmospheric changes due to global warming on oceanic ecosystems.
The marine protected area (MPA) of the Arrabida Natural Park is a mid-latitude hotspot for biodiversity. To understand its trophic structure, a highly defined food web network was assembled for this ecosystem, consisting of 884 taxa. Network analysis showed that humans are the top predators, as well as various seabirds, dolphins and sharks. This web is dominated by intermediate species, and its general organization follows previously reported patterns for other marine and coastal ecosystems. Two swimming crabs, Polybius navigator and Polybius henslowii, assume important roles as mid-trophic level consumers and prey, due to their high connectivity in the network. The cuttlefish, Sepia officinalis, a cephalopod of high commercial value, assumes the most pivotal role in the network, as it is the species with the highest number of prey and is among the top 10 most highly connected species (with more links to other species). Additionally, the cuttlefish is among the species with shortest path length, that is the lowest number of links connecting it to any other species. Since, this cephalopod is highly mobile and extends its territory outside the MPA, into the Sado estuary, where it is the main target of local fisheries, and is exposed to various pollution sources, close monitoring the local population of cuttlefish is of the utmost importance, not only in the Arrabida MPA but also in the adjacent Sado estuary.
Rhea pennata populations play a vital ecological role in the Andean highland ecosystems. This study aims to critically assess the evolution, scope, and focus of global scientific research on Rhea pennata, in order to identify trends, gaps, and opportunities that can guide future conservation and ecological studies on the species. For this purpose, we realize a bibliometric analysis of 73 articles published from 1974 to 2024. For this purpose, the bibliometric tools Biblioshiny in R and the VOSviewer were utilized. Results reveal an urgent need to expand scientific research on this species, given its projected 50-year extinction risk in Peru. Furthermore, a low publication rate and several emerging research areas with potential for future investigation were identified. Key authors, significant keywords, influential sources, and high-impact publications in this field were also highlighted. Argentina stands out, particularly through the Applied Zoology Centre of the University of Cordoba, for its major scientific contributions and international collaborations. Although there are reports of captive breeding of the Rhea pennata, no successful reintroduction cases in the wild have been documented. In contrast, population increases have been observed in protected natural areas without direct human intervention and in semi-captive conditions, suggesting a shift in conservation strategies for the Rhea pennata compared to current approaches.
Studying plankton systems encompasses different interests, including understanding ecological cycles and developing sustainable strategies in aquaculture research regarding food security. Zooplankton farming is economically valuable, and its production may depend primarily on the availability of phytoplankton and other external food sources. However, diverse factors may affect overall phytoplankton-zooplankton interactions. For example, phytoplankton's defense mechanisms, such as finding refuge and releasing toxins or low phytoplankton's sustainable environments, can decrease zooplankton populations. Another critical factor is the adverse effects of pollution on plankton systems, which are more frequently present in water bodies. Still, zooplankton may survive harsh conditions if present pollutants are in low concentrations and external sources, including animal waste, are available. The partial understanding of these trophic interactions depends on initial assumptions, and using stochastic approaches may reduce the gap between deterministic mathematical outcomes and reality. In this work, we have mathematically described a planktonic system under the above assumptions using a deterministic model as well as its stochastic version. Our findings suggest that zooplankton growth is possible under polluted environments by providing them with external food sources, complementing phytoplankton availability. However, in these circumstances, random external environmental factors may cause the phytoplankton population to collapse. Through stochastic numerical experiments, we estimate which possible scenarios are more likely to induce phytoplankton extinction in these plankton systems.
