Learn and participate
Abstract submission
We welcome submissions to both symposia and general sessions. Please submit your abstracts to either a symposium or general session on the conference topics listed below and, on our website, using the online abstract submission system S’ouvre dans une nouvelle fenêtre.
Abstract submission deadline: 2 May 2025
Conference topics
1. Dynamic ecosystem models
2. Uncertainty analysis
3. Ensemble modelling
4. Data assimilation and optimization techniques
5. Machine learning and (big) data
6. Model integration, metamodels
7. Individual-based modelling
8. Software and tools
9. Bioenergetics: Dynamic energy budget models
10. Network modelling
11. Models of socio-ecological systems
12. Models of global, climate and land-use change
13. Sustainability and resilience
14. Ecosystem services
15. Biodiversity and conservation
16. Community models
17. Marine ecology and fisheries
18. Forests
19. Freshwaters (lakes and rivers)
20. Models of epidemics
21. Other
You can submit as many abstracts to the conference for review as you would like. If, after the review by the committee, you have more than one paper accepted for the conference, you will need to register to attend and pay an additional paper fee for each additional paper (i.e., for the 2nd, 3rd, 4th papers – not the 1st). Please note this is for papers that you are the presenting author of, not papers that you are co-author of.
Successfully submitted abstracts will be acknowledged with an electronic receipt including an abstract reference number, which should be quoted in all correspondence. Allow at least 2 hours for your receipt to be returned to you.
For revisions or queries regarding papers already submitted:
For revisions or queries regarding abstracts already submitted please contact the Conference Content Executive S’ouvre dans une nouvelle fenêtre, (please do not email credit card information under any circumstances). If you do not receive acknowledgement for your abstract submission or you wish to make any essential revisions to an abstract already submitted, please DO NOT RESUBMIT your abstract, as this may lead to duplication. Please email the Conference Content Executive S’ouvre dans une nouvelle fenêtre, (please do not email credit card information under any circumstances) with details of any revisions or queries. Please quote your reference number if you have one. A condition of submission is that if accepted one of the authors will present at the conference.
Symposia
4 | Enhancing coastal ecosystem resilience to climate and land use change |
6 | Marine ecology and fisheries by marine eco-environmental informatics |
11 | Social-ecological systems modelling |
13 | Predictive modelling and real-time surveillance: Strengthening Ebola response in Liberia with epidemic intelligence systems |
15 | Innovative approaches to ecosystem-based fisheries management: integrating climate resilience and socio-economic factors |
16 | Reservoir and riverine water quality modeling workshop with 1D and 2D software |
17 | Advancing ecological modelling for sustainable ecosystem management |
20 | Artificial intelligence and machine learning to advance process modelling in ecological and environmental sciences |
21 | Integrating machine learning and genome data digitalisation for biodiversity conservation and dynamic ecosystem modelling |
23 | Integration of integrated assessment models (IAM) and land change models (LCM) for sustainable ecosystem management |
27 | Sustainable and resilient ecosystems modelling: innovations for future challenges |
28 | Oneness water quality with neural network |
29 | Marine social ecological system modelling for the human ocean coupling |
31 | Towards better understanding of plastic transport and accumulation processes in inland waters |
32 | Harness the power of AI and process understanding for predictive ecology |
34 | Climate change, REDD+, and gendered benefit sharing in forest-dependent communities of Africa |
35 | Modeling relationships between species diversity and ecosystem services and functions |
37 | Towards improved understanding of ecosystem dynamics in arid and semi-arid regions with limited water resources |
40 | Modelling forest ecosystems: integration through multiple scales |
41 | Advancing predictive models of forest greenhouse gas exchange: from biogeochemical processes to global impacts |
42 | Ecosystem modeling for environmental impact assessment of human activities |
46 | Modeling of watersheds, lake, ocean and ecosystems related to climate change impacts |
49 | Transforming data into action: advancing ecological modeling for sustainable ecosystem management in Africa |
52 | Socio-ecological system modeling for natural resources management and ecological conservation |
53 | Advancements in machine learning algorithms for ecological big data processing |
57 | Advances in Data Assimilation (DA) and optimization techniques |
59 | Dynamic energy budget modeling |
60 | Ecosystem services and community interplay in mountain areas |
61 | Challenges and advances in integrating ecological and earth system models across multiple sectors |
63 | Land change modeling for sustainable land management |
64 | Modeling water quality and ecosystems in estuarine and coastal areas: Japan and other countries under human activities and climate change |
65 | Data-driven ecosystems: Modeling for improved health outcomes and sustainable development |
66 | Modeling species distributions and interactions in ecological communities |
67 | Marine plastic modelling: Biological interactions, transport dynamics, and ecosystem impacts |
68 | Advances in signal processing for bioacoustics |
69 | Marine biogeochemical/ecological modelling: advancements and best practices |
4 - Enhancing coastal ecosystem resilience to climate and land use change
Sathaporn Monprapussorn Srinakharinwirot University, Thailand
Abstract
Coastline is the area where land meets the sea. It provides place for various activities such as marine transportation, fisheries, tourism, aquaculture, mangrove habitat and integral part to nation economy that provides many benefits to local communities and ecosystem. Intense climate change is likely to increase physical threats to both human and ecological systems along the coasts e.g. floods, sea level rise, biodiversity degradation, health and well-being of socio-ecological systems which some important coastal habitats such as mangrove forest, coral reefs, estuaries may become so rapidly degraded and reduced in area that they would be considered to have a significant role in enhancing the climate resilience of coastal resources and communities. In the meantime, effects of human dimension and interventions on coastal ecosystem are very difficult to separate from the effects of ecological dimension, limiting our understanding of human-environment interaction as well as potential solutions to climate change. The Socio-ecological systems provide useful guidance of how to assess human-environmental interactions to increasing coastal resilience. For example, shrimp aquaculture or tourism infrastructure development have consumed coastal resources and habitats declines while climate change has also driven widespread collapse of coral reef ecosystems. The aim of this session focuses on sharing the past and recent impacts of physical and socio-ecological change on coastal sustainability in term of climate and land use change, including varieties of decision-support tools that have been emerged to support more systematic analysis of coastal ecosystem services. The role of ecosystem services in protecting biodiversity and habitats, providing essential resources, buffering land from disaster are very important to increase resilience of coastal communities and sustainability. Enhancing coastal ecosystem resilience to land use and climate change through highlighting the importance of integrative systems modelling will be crucial in improving our understanding and achieving coastal ecosystem resilience.
6 - Marine ecology and fisheries by marine eco-environmental informatics
Meilin Wu1, Hui Zhao2
1South China Sea Institute of Oceanology Chinese Academy of Sciences, China. 2Guangdong Ocean University, China
Abstract
Ocean represent some of the most dynamic and productive ecosystems on the planet, serving as critical interfaces between terrestrial and marine environments. These areas are characterized by a complex interplay of biotic and abiotic factors, which are increasingly influenced by both anthropogenic activities and natural changes. Long-term and short-term ecological monitoring networks have been established in sea to evaluate environmental problems such as this measurement of hydro-chemical variables, biological indicators, and fish in the sea environment will aid better understanding of aquatic environment. These monitoring programs produce huge datasets, and it becomes really difficult to extract latent meaningful information from these datasets. To extract the latent meaningful information, multivariate statistical analysis and different biotic indices for biodiversity data are used. It may include factor analysis, cluster analysis, discriminant analysis, self-organizing maps, artificial neural network, canonical correspondence analysis, redundancy analysis and many biotic indices. Marine eco-environmental informatics identify different patterns in the datasets and provides meaningful underlying information which would be rather difficult just seeing the raw data.
11 - Social-ecological systems modelling
Brian Fath Towson University, USA. International Institute for Applied Systems Analysis, Austria. Masaryk University, Czech Republic
Abstract
Social-ecological systems (SES) consider the interactions, influences, linkages, and dependencies between humans and nature. Investigations of SES often use a complex adaptive systems framework to explore the bio-geo-physical resources and the social and institutional actors. This session invites presentations that study SES models in the context of Sustainable Development using resilience, robustness, metabolic, or other process-based, dynamical techniques. Models that include the role of local or traditional knowledge is encouraged.
13 - Predictive modelling and real-time surveillance: Strengthening Ebola response in Liberia with epidemic intelligence systems
Saeed Ahmad1,2, Fahmeeda Idrees1 1Tampere University, Finland. 2The Task Force for Global Health (TFGH), USA
Abstract
This symposium will explore the integration of predictive modelling and real-time Epidemic Intelligence Surveillance (EIS) systems, highlighting their pivotal role in strengthening the Ebola response in Liberia. The 2014–2016 West African Ebola outbreak revealed significant gaps in early detection, outbreak forecasting, and resource optimization. By employing advanced epidemic modelling frameworks, such as SEIR (Susceptible-Exposed-Infectious-Recovered) and SEIR-QD (quarantine dynamics), alongside real-time surveillance systems, Liberia made substantial strides in outbreak management.
The session will explore how predictive models, informed by surveillance data streams—including community event-based surveillance (CEBS), contact tracing systems, and health facility reporting—helped identify transmission hotspots, forecast epidemic trends, and prioritize interventions. These models supported critical decisions such as isolation of cases, safe burial practices, and strategic deployment of resources to high-risk areas, reducing the overall impact of the epidemic.
Key discussions will include: • The development and application of epidemic models tailored to the Liberian context. • Real-time integration of surveillance data to enhance model accuracy and forecasting capabilities. • Overcoming challenges related to data quality, underreporting, and delayed reporting. • Strengthening the feedback loop between modelling outputs and ground-level response strategies.
The symposium will also highlight lessons learned from the Liberian Ebola response, focusing on scalability and sustainability for future public health emergencies. Emphasis will be placed on the importance of interdisciplinary collaboration between epidemiologists, data scientists, and field surveillance teams to create agile and resilient epidemic response systems.
By combining predictive analytics with real-time surveillance, Liberia’s experience offers a blueprint for managing high-burden epidemics in low-resource settings. This session aims to inspire the global public health community to adopt data-driven strategies that enhance epidemic preparedness, response, and resilience while advancing global health security.
15 - Innovative approaches to ecosystem-based fisheries management: integrating climate resilience and socio-economic factors
Salma Aboussalam Cadi Ayyad University, Morocco
Abstract
Ecosystem-based fisheries management (EBFM) is essential for ensuring the sustainability of marine resources while balancing ecological health and socio-economic needs. This symposium, titled "Innovative Approaches to Ecosystem-Based Fisheries Management: Integrating Climate Resilience and Socio-Economic Factors," aims to explore cutting-edge methodologies and frameworks that enhance EBFM practices in the context of a rapidly changing environment.We will bring together researchers, practitioners, and policymakers to discuss innovative strategies that incorporate climate resilience into fisheries management. Topics will include adaptive management practices that respond effectively to climate variability, the development of multi-species harvest strategies that consider ecological interactions, and the integration of socio-economic modeling to assess the impacts of fishing on local communities.Participants will also engage with decision support systems that enable scenario analysis and facilitate informed management choices. By highlighting successful case studies and collaborative approaches, this symposium will showcase how stakeholder involvement can lead to more effective EBFM outcomes.The goal is to foster interdisciplinary dialogue and share best practices that empower fisheries managers to navigate the complexities of marine ecosystems while promoting sustainable fishing practices. Ultimately, this symposium seeks to contribute to the development of resilient fisheries management frameworks that align ecological integrity with socio-economic viability in the face of global challenges such as climate change and overfishing. Join us in advancing the conversation on innovative EBFM approaches that ensure the long-term health of our oceans and the communities that depend on them.
16 - Reservoir and riverine water quality modeling workshop with 1D and 2D software
Zhonglong Zhang Portland State University, USA
Abstract
CE-QUAL-W2 (W2) is a comprehensive 2D longitudinal-vertical hydrodynamic and water quality model. It simulates hydrodynamics, water temperature, nutrient cycles, eutrophication, and various other water quality constituents in river and reservoir systems. This model has proven invaluable for reservoir management, conducting thermal and water quality assessments, evaluating impacts from diverse stressors, updating reservoir operation manuals, and supporting environmental impact statement studies. HEC-ResSim (ResSim), on the other hand, is designed to simulate the behavior and performance of a network of reservoirs under varying hydrologic and operational scenarios. As a unique reservoir management model, it allows users the flexibility to define operating rules to meet multiple objectives across one or more reservoirs. Additionally, ResSim seamlessly integrates a 1D water quality engine, enabling the simulation of water temperature and nutrient cycles along a longitudinal axis for rivers and a vertical axis for reservoirs.
This workshop aims to provide participants with an in-depth understanding of the latest versions of CE-QUAL-W2 and HEC-ResSim, their water quality capabilities, input requirements, model development processes, and result analysis techniques. By the end of the workshop, attendees will have acquired practical skills in utilizing both 2D and 1D reservoir and riverine water quality models. Through a combination of lectures, demonstrations, and hands-on exercises, participants will learn to set up water quality simulations, execute model runs, visualize results, and effectively conduct alternative analyses.
17 - Advancing ecological modelling for sustainable ecosystem management
Prerna Mehta GD Rungta Educational Society, India
Abstract
The special symposium “Advancing Ecological Modelling for Sustainable Ecosystem Management” is aimed at attracting scholars, students, and professionals who are interested in the importance of ecological modelling in sustainable ecosystem management. When species in ecosystems are pressed against adaptive challenges including climate alteration, habitat destruction, and biological diversity loss, sound procedures that use powerful information analyses is crucial.
This event will focus on new advanced quantitative and qualitative approaches to modeling of large data to foster ecosystem resilience and sustainability. Participants are going to discuss and present state-of-art methodologies ranging from the machine learning techniques to the agent based models of the management of the variety of ecological systems.
This also implies that the symposium will also focus on the qualities of interdisciplinarity which combines technology with ecology for the improvement of management approaches. In this way, our workshops try to establish what are the current difficulties of the use of this approach in the real world (integration of data sets and model checking, for instance) and present what recent advancements have been achieved to overcome them.
Through promoting a sharing of knowledge, this symposium shall provide potential benefits in terms of professional networking and development, as well as the promotion of theoretical ecological knowledge. We would like to invite you to explore with us on how some of the most modern eco- simulation techniques can be applied to creating long-lasting preservation policies for our environment to benefit our future generations. Altogether, it is possible to create the international platform for generation of new ideas and actions that will serve as a basis for solving ecological problems.
20 - Artificial intelligence and machine learning to advance process modelling in ecological and environmental sciences
Clement Sohoulande USDA-ARS Coastal Plains Soil, Water, and Plant Research Center, Florence, SC, USA
Abstract
Research-driven knowledge accumulated over decades in ecological and environmental sciences has led to significant progress in the development of process models. In general, the assumptions used in these process models are based on the scientific understanding of factors affecting biophysical phenomena. Hence, process models have been widely used to simulate and upscale various ecological and environmental responses. Today there is an increasing research interest in artificial intelligence (AI) and machine learning (ML) to solve ecological and environmental problems. Indeed, several literatures reported the use of AI and ML models to address ecological and environmental phenomena. However, these AI and ML models do not necessarily offer a comprehensive structure to relate the biophysical mechanisms and interactions driving ecological and environmental behaviors. Furthermore, AI and ML models learn from data and their applicability is limited by data availability. In contrast, process models are less data-demanding, and they use established biophysical concepts to simulate ecological and environmental responses. Nevertheless, the capitalization on the scientific advances in AI and ML could be an avenue to improve the performance of process models. This symposium aims to provide an expert understanding of the state of the art in the use of AI and ML in ecological and environmental modelling. The symposium will help elucidate potential avenues to improve process models with AI and ML. Two well-established speakers will be invited to deliver a half-hour talk each. Following, the two talks, a half-hour open-floor discussion will be animated by a panel. The panel will include the two speakers and the organizer/chair (if needed additional panelists may be added). Participants will be allowed to question and comment on the subject. We expect this symposium to drive the attention of scientists, scholars, and professionals interested in ecological and environmental modelling.
21 - Integrating machine learning and genome data digitalisation for biodiversity conservation and dynamic ecosystem modelling
Hajar Fauzan Ahmad University of Malaysia Pahang Al-Sultan Abdullah Faculty of Industrial Sciences and Technology, Malaysia
Abstract
This symposium will include topic focusing on the integration of artificial intelligence (AI), machine learning, and genome data digitalisation to advance biodiversity conservation and ecosystem modelling. With the rapid growth of genomic data, there is a pressing need to develop innovative approaches that leverage big data analytics, machine learning, and optimization techniques to model complex ecological systems or resolving cryptic species.
The session will highlight cutting-edge research and applications in biodiversity modelling, such as genomic data generation approach, genome-based species distribution models, habitat suitability prediction, and conservation prioritization. It will explore dynamic ecosystem modelling techniques that incorporate genomic insights into understanding population dynamics, adaptive responses, and ecosystem resilience. Participants will discuss approaches to uncertainty analysis, ensemble modelling, and data assimilation, with a focus on genomic datasets. Emphasis will also be placed on developing user-friendly tools and software for genomic data processing and integration into broader socio-ecological and ecosystem service models. The session will encourage cross-disciplinary discussions, connecting AI specialists, ecologists, and conservation biologists to address critical challenges in biodiversity conservation under changing global conditions.
This symposium aligns with key conference topics, including machine learning and (big) data, biodiversity and conservation, dynamic ecosystem models, and sustainability. It aims to foster collaboration and share innovative solutions to enhance the effectiveness of conservation efforts through genome data digitalisation and modelling advancements. Target Participants:
1. Ecologists and conservation biologists 2. Data scientists and machine learning experts 3. Genomics and bioinformatics researchers 4. Model developers and software engineers 5. Policy makers and conservation organizations 6. Academics and students 7. Multidisciplinary collaborators
23 - Integration of integrated assessment models (IAM) and land change models (LCM) for sustainable ecosystem management
Peichao Gao, Yifan Gao Beijing Normal University, China
Abstract
Sustainable ecosystem management is a critical approach to addressing global environmental challenges, including climate change, biodiversity loss, and resource degradation. Sustainable ecosystem management focuses on balancing ecological, social, and economic demands to ensure the long-term health and resilience of ecosystems. Integrating models with diverse specialties enhances our understanding of the complex interplay of factors in sustainable ecosystem management, providing valuable insights into future dynamics. Among the numerous models, IAM and LCM are typical representatives widely used in ecosystem management and climate policy research. IAMs excel at simulating policy-driven climate scenarios across multiple sectors and factors, while LCMs offer spatially explicit insights into land use dynamics. Integrating IAMs and LCMs enables researchers and policymakers to explore the cascading effects of climate policies on land change, predict future scenarios of land use and biodiversity, and evaluate trade-offs among competing land demands, including agriculture, urban development, and conservation. We welcome contributions on theoretical and methodological advancements in IAM and LCM integration, as well as case studies showcasing their practical applications. Key objectives include:
1. Developing frameworks to evaluate the cascading effects of climate policies on land changes. 2. Utilizing multi-source data, including remote sensing and socioeconomic datasets, to improve input quality and model accuracy. 3. Performing scenario-based analyses to evaluate future land use patterns, ecosystem services, and management strategies. 4. Creating decision-support systems for policymakers to evaluate sustainable land management strategies.
27 - Sustainable and resilient ecosystems modelling: innovations for future challenges Delin Fang, Cuncun Duan
Beijing Normal University, China
Abstract
Ecosystems worldwide are increasingly threatened by climate change, urbanization, and anthropogenic pressures, requiring innovative modelling approaches to enhance resilience and sustainability. This symposium aims to bring together researchers and practitioners from diverse disciplines to advance the frontiers of ecosystem modelling, bridging theoretical insights with practical applications across urban, agricultural, and natural ecosystems. The symposium will explore state-of-the-art modelling techniques to assess ecosystem responses to environmental stressors, emphasizing predictive models for ecosystem vulnerability, adaptation strategies, and resilience-building mechanisms. Key topics include ecological network analyses, machine learning applications, and stochastic modelling frameworks that inform sustainable decision-making. By fostering a cross-disciplinary dialogue, this symposium seeks to integrate ecological, technological, and socio-economic perspectives, encouraging innovative solutions for ecosystem sustainability. We invite contributions from researchers working on advanced modelling methodologies, case studies demonstrating real-world applications, and interdisciplinary approaches that connect ecosystem science with policy and management. This session will serve as a platform for knowledge exchange, promoting collaborative research and actionable strategies to enhance ecosystem resilience in an era of environmental uncertainty.
28 - Oneness water quality with neural network
Yao-Te Wang National Taipei University of Technology, Taiwan
Abstract
In the past, limited monitoring technology and weak data management systems made it nearly impossible to compare water quality across countries accurately. This often led to fragmented policies and inefficient use of resources. Although multinational cooperation and standardization efforts have improved the situation, the lack of real-time data sharing still poses major challenges. Looking ahead, achieving global consistency in water quality will rely on the robust integration of big data and cloud platforms, with artificial intelligence at the forefront. Machine learning can handle the cleaning and calibration of large, complex datasets, enhancing accuracy, while deep learning techniques can quickly detect anomalies, predict pollution trends, and deliver real-time insights to decision-makers. This approach unlocks new possibilities for more effective water management, supporting sustainable development for both human communities and ecosystems. By bringing together fragmented historical data into a smart, cohesive analytical framework, researchers and policymakers can close existing gaps, encourage interdisciplinary collaboration, and build long-term ecological resilience—all within a unified, data-driven system.
29 - Marine social ecological system modelling for the human ocean coupling Mingbao Chen1, Jiaxue Wu2
1Macau University of Science and Technology Foundation, Macao. 2Sun Yat-Sen University, China
Abstract
Marine social-ecological systems are facing increasing pressures, including climate change, overfishing, coastal urbanization, and transboundary pollution. Traditional single disciplinary models have difficulty capturing the nonlinear interactions between human behavior, ecological processes, and institutional design. These issues include:（1）Interdisciplinary/ Transdisciplinary modeling methods. Integrating ecological dynamics, socioeconomic data (such as fishery policies, community livelihoods) and advanced technologies (AI, network analysis) to develop dynamic models of coupled human-ocean systems.（2）Data fusion and scale challenges，Exploring how to unify multi-source heterogeneous data (satellite remote sensing, traditional knowledge, policy texts) to solve multi-scale modeling problems from local to global.（3）Policy coordination and practical application. Demonstrating how models support dynamic marine protected area (MPA) design, fishery conflict mitigation, and climate change adaptation strategies through case studies, emphasizing the importance of "science-policy-community" coordination. This session will focus on:（1）AI-enhanced hybrid modeling: the application of meta-learning frameworks in optimizing marine SES models, such as predicting the cascading effects of sea level rise on coastal community livelihoods.（2）Social-ecological network analysis: Quantify the spatial mismatch of ocean connectivity and governance fragmentation using higher-order networks, and reveal the potential mechanisms of institutional failure.（3）Participatory modeling practice: Demonstrate cases of embedding indigenous knowledge into model development (such as community-led fisheries management in Pacific islands) to enhance the cultural inclusiveness and practicality of the model.
Through the discussion, we hope to: (1) form a technical roadmap for marine SES modeling and identify key bottlenecks for interdisciplinary cooperation. (2) incubate a transboundary marine protection and cooperation network and promote the development of open source modeling tools (such as Python-based SES modeling libraries) and visualization platforms. (3) draft a "Guide to Modeling Marine Coupled Systems" to provide methodological support for the United Nations "Decade of Ocean Science" and the Global Biodiversity Framework.
31 - Towards better understanding of plastic transport and accumulation processes in inland waters
Tadanobu Nakayama1, Batdulam Battulga2, Masayuki Kawahigashi3 1National Institute for Environmental Studies, Japan. 2Tohoku University, Japan. 3Tokyo Metropolitan University, Japan
Abstract
Plastic pollution is considered to be one of the recent environmental problems, and such pollutants in streams, rivers, and oceans pose potential risks to human health and the environment. Once plastic is released into the environment, it is gradually degraded by physical, chemical, and biological processes, and therefore, is impossible to remove and remain there indefinitely. Furthermore, existing studies on plastic waste in freshwater systems have suggested land-derived plastics are one of the main sources of marine plastic pollution. However, the fate of plastics is dependent on various factors such as their size, density, shape, and polymer type, etc. Distribution and quantities across environmental compartments, and transport mechanisms remain mostly unknown. These knowledge gaps should be addressed to quantify the dispersal and eventual fate of plastics in the environment, towards better assessment of potential risks and effective mitigation measures. This session aims to explore innovative modeling, theory, and monitoring approaches on macro-, micro- and nanoplastics across freshwater and coastal ecosystems from small- to large-scales. Sink-source relationships of plastics across different compartments (rivers, lakes, groundwater, urban water, floodplains, estuaries, and freshwater biota) are also welcome. The application of cutting-edge technologies such as integrating artificial intelligence and deep learning to detect plastics will be showcased.
32 - Harness the power of AI and process understanding for predictive ecology
Yuanyuan Huang1, Yiqi Luo2 1Institute of Geographical Sciences and Natural Resource Research, Chinese Academy of Sciences, China. 2Soil and Crop Sciences Section, School of Integrative Plant Science, Cornell University, USA
Abstract
The increasing availability of diverse ecological datasets presents both challenges and opportunities for ecological modeling. While process-based models remain essential for understanding fundamental ecological dynamics, artificial intelligence (AI) techniques offer powerful tools for improving predictive accuracy, reducing computational burdens, and uncovering hidden patterns. This symposium will explore the synergies between AI and process-based models, discussing how hybrid approaches can enhance ecological theory, prediction, management, and policy-making. It will focus on leveraging AI to improve the calibration, validation, and forecasting capabilities of process-based models, incorporating knowledge-based insights to improve AI predictions, and enabling more accurate, high-resolution forecasts of ecological dynamics and ecosystem responses to environmental changes in a timely manner.
Through a combination of theoretical insights and practical case studies, this session will focus on how AI-driven tools can automate and optimize model development by boosting data collection capability, enhancing computation efficiency, integrating multimodal information, uncovering hidden patterns, quantifying uncertainty, identifying key drivers of ecological processes and discovering novel connections (e.g., through Agent AI or/and foundation models). We will examine the challenges and opportunities of merging mechanistic, process-oriented models with data-driven AI approaches, emphasizing the importance of maintaining model interpretability while embracing the flexibility and adaptability of AI techniques.
By bringing together experts in ecological modeling and AI, this symposium aims to foster interdisciplinary collaboration and accelerate the application of advanced modeling strategies in ecosystem management and sustainability. Discussions will cover key methodological advancements, future research directions, and the ethical considerations of AI in ecological modeling. The ultimate goal is to establish a new paradigm for predictive ecology, where AI and process understanding work synergistically to improve environmental decision-making and management strategies in the face of a rapidly changing world.
34 - Climate change, REDD+, and gendered benefit sharing in forest-dependent communities of Africa
Rachel Yeboah Nketiah University of Energy and Natural Resources, Ghana
Abstract
At the international level, REDD+ projects are overseen by various political entities, with varying levels of negotiation and execution. African governments must incorporate gender into forest management at the management level and advocate for gender equality in environmental policy. Do REDD+ policies help to fight for gender equity in Africa’s forest-dependent communities, or do they contradict one another when it comes to tackling the impact of climate change? Can they be controlled and applied to enhance Africa’s forest governance? The chapter provides answers focusing on the concepts of non-discrimination and equality as well as participation, benefit sharing, and inclusion in discussing the connection between climate change, REDD+, and gender in reducing vulnerabilities of forest fringe communities in Africa. Here it explores the concept of effective participation and benefits sharing in REDD+ agreements, focusing particularly on the policy frameworks capturing how climate change, REDD+, and gender interplay, especially concerning the rights of women in these policies in reducing vulnerabilities among forest fringe communities in Africa (Ghana, Congo, Mozambique, and Tanzania). This chapter follows a structured and standard process by filling the knowledge gap on climate change, REDD+, and gender justice in forest-dependent communities of Africa through a literature review by examining REDD+ policy in action and practice frameworks in Africa and its climate sensitivity policies to gender and women’s rights in practice.
35 - Modeling relationships between species diversity and ecosystem services and functions
TianXiang Yue1, Herzog Felix2, ZeMen Fan1 1Institute of Geographic Sciences and Natural Resources Research, China. 2Agroscope Division of Agroecology and Environment, Switzerland
Abstract
Species diversity has two components: richness, also called species density, based on the total number of species present, and evenness, based on the relative abundance of species and the degree of its dominance thereof. The relationships between species diversity and ecosystem services and functions have emerged as central issues in ecological and environmental sciences since 1970s. Its major questions include short-term effects of species diversity on ecosystem processes such as productivity and long-term effects such as ecosystem stability. There is evidence from case-studies showing that ecosystem services and functions were inversely related to species diversity – and other case studies show the contrary, with increasing productivity due to diversification of species and niche complementarity. Those relationships may differ between managed (agro and forestry) ecosystems and natural ecosystems. This Symposium will focus on models and machine learning methods for understanding relationships between species diversity and ecosystem services and functions, grounded in empirical evidence.
37 - Towards improved understanding of ecosystem dynamics in arid and semi-arid regions with limited water resources
Qinxue Wang, Tadanobu Nakayama National Institute for Environmental Studies, Japan
Abstract
Groundwater, the most abundant freshwater on Earth, is undergoing significant depletion worldwide. As a crucial buffer during periods of insufficient surface water and precipitation, groundwater remains an essential resource in arid and semi-arid regions, where its availability is increasingly vital to meeting rising water demands. In Mongolia, for instance, groundwater overuse and degradation have become pressing concerns, particularly in the urban and economic center of Ulaanbaatar and the Southern Gobi mining region. These challenges are exacerbated not only by human activities, such as livestock grazing, urban expansion, and mining activities, but also by climatic disturbances that severely impact local ecosystems. This session will explore advances in theoretical frameworks and approaches for understanding ecosystem dynamics in arid and semi-arid regions with limited water resources, including those that rely heavily on groundwater. Key topics include the development of methodologies to assess spatial-temporal variations of water availability, particularly in data-scarce regions, to enhance water resource management. Additionally, it is critical to refine models at the plot scale to better capture water-limiting factors and improve assessments of pasture carrying capacity and ecosystem vulnerability at both local and regional scales. Furthermore, we welcome discussions on the application of emerging technologies, such as integrating artificial intelligence and machine learning, to enhance the estimation and prediction of ecosystem degradation. By fostering interdisciplinary collaboration and methodological advancements, this session aims to contribute to a deeper understanding of ecosystem resilience and sustainable water management in fragile dryland environments.
40 - Modelling forest ecosystems: integration through multiple scales
Juan A. Blanco Public University of Navarre, Spain
Abstract
Following the sharp increase in data availability and massive data analyses, modelling is now being applied in forest ecosystems at multiple scales, from continental and regional studies on forest and species distribution to leaf scales, able to model plague and pathogen infestation in individual branches or to model carbohydrates flows through the tree. Similarly, models are being applied to understand short-time processes such as nutrient transport in cells, to extremely long scales such as genetic drift in tree populations. From the application point of view, models are also being applied in purely ecological issues such as coexistence and resource competition among tree species to socio-economic-ecological issues such as design of tourist routes inside natural reserves, estimating timber flow to sawmills under climate change scenarios, or engineering concerns such as designing forest harvesting machinery with lower ecological footprint (in energy, rut formation, noise, etc.). All these multi-scale topics are interconnected, and therefore the objective of this symposium is to show side by side the cutting edge research on models applied to forests will encourage the advancement of more interdisciplinary and comprehensive modelling approaches.
41 - Advancing predictive models of forest greenhouse gas exchange: from biogeochemical processes to global impacts
Weifeng Wang Nanjing Forestry University Collaborative Innovation Center of Sustainable Forestry in Southern China of Jiangsu Province, China
Abstract
Forest ecosystems play a key role in the global carbon cycle, acting as significant carbon sinks and sources through processes such as carbon sequestration and trace gas emissions. The dynamic changes in forest gas exchange, including fluxes of carbon dioxide (CO2) and methane (CH4), not only influence the carbon storage capacity of these ecosystems but also have profound impacts on global climate patterns. Accurate modeling of these exchanges is essential for developing effective forest management strategies and climate policies. This symposium aims to advance the field by exploring cutting-edge advancements in modeling forest greenhouse gas dynamics, addressing key challenges, and fostering collaborative opportunities. It will bridge scales from microbial processes to global climate impacts, emphasizing innovation in data integration, computational tools, and policy applications. Key themes will cover biological processes such as plant-microbe interactions and soil biogeochemistry, as well as modeling innovations involving process-based models and machine learning. The symposium will also address data integration from various sources, including remote sensing and in-situ measurements. Additionally, it will highlight the importance of policy and management applications, such as validating carbon offsets and understanding biodiversity conservation benefits, and discuss emerging challenges like representing extreme events in global models. By bringing together ecologists, modelers, atmospheric scientists, and policymakers, this event seeks to catalyze progress in forest greenhouse gas modeling, ensuring that scientific advancements directly inform strategies for climate resilience and sustainability.
42 - Ecosystem modeling for environmental impact assessment of human activities
Daisuke Kitazawa, Jinxin Zhou The University of Tokyo Institute of Industrial Science, Japan
Abstract
Various activities such as energy resource development and food production are carried out in the ocean. It is necessary to predict the impacts of these human activities on the surrounding environment and to conserve the marine environment. For this purpose, predictions using numerical simulations are useful in addition to environmental monitoring. Ecosystem models are becoming an important tool for predicting the impacts of human activities on material cycles and changes in ecosystems. This session will focus on ecosystem modeling that contributes to the environmental impact assessment of human activities.
46 - Modeling of watersheds, lake, ocean and ecosystems related to climate change impacts
Eiji Komatsu Meiji University, Japan. Laboratory for Ecological Reconstruction Science inc., Japan
Abstract
Climate change impacts our society in many different ways. Drought can harm food production and terrestrial/aquatic ecosystems, and accelerate wildfire. Flooding can lead to spread of damage ecosystems and infrastructure. Droughts, floods, and other weather-related phenomena have significant direct and indirect socioeconomic impacts as well as environmental and ecological impacts. Earth warming has been progressing faster than originally predicted in recent years, with the average global temperature rising above 1.5°C for the first time last year. Therefore there will be great possibility that the impact of climate change will be enormous in the future.
Ecological modeling play an essential tool to understand these impacts, predict future projections and environmental and ecological transitions after climate change disaster, and select rational adaptation measures. Thus studies on modeling methodology and their application is one of the most important issues.
On the other hand, estimation of socio-economic impacts and losses due to natural disasters caused by climate change that may occur in the future and socio-economic evaluation of avoidance and adaptation measures are also important. Consequently ecological modeling should be integrated with socio-economic aspect and modeling to evaluate and predict the whole extent of the impact of climate change.
This session will address these studies and discuss the future of modeling for climate change impacts.
49 - Transforming data into action: advancing ecological modeling for sustainable ecosystem management in Africa
Adigla Appolinaire Wédjangnon University of Parakou, Benin
Abstract
This symposium will explore how ecological modeling can significantly enhance sustainable ecosystem management in Africa, particularly for biodiversity conservation. The continent faces critical challenges, such as climate change, deforestation, and habitat loss, which severely threaten its rich biodiversity. To tackle these urgent issues, we need effective strategies that utilize ecological modeling. These models help us understand complex ecosystems, predict how they respond to various environmental factors, and make well-informed decisions.
We will begin by discussing the specific environmental challenges impacting biodiversity in various African regions. These challenges require customized management approaches that address the unique needs of each area. The symposium will highlight different types of ecological models, such as species population models, habitat models, and landscape models, and will showcase their practical applications in conservation efforts. We will present case studies that demonstrate how these models have successfully guided conservation initiatives, improved land management, and protected endangered species. Additionally, we will address the challenges related to data availability and quality in Africa. Participants will engage in discussions about the importance of local knowledge in developing effective models, emphasizing how insights from local communities can enhance the modeling process. We will focus on strategies for collecting reliable data and making the best use of existing resources efficiently.
Furthermore, the symposium will share best practices for employing these models in policy-making and resource management. By bringing together researchers, policymakers, and practitioners, we hope to foster collaboration and facilitate the exchange of ideas.
In conclusion, this symposium will underscore the vital role of ecological modeling in promoting biodiversity and sustainable practices that benefit both people and the environment. By working together, we can transform data into actionable strategies, ensuring a healthier and more resilient future for Africa’s diverse ecosystems.
52 - Socio-ecological system modeling for natural resources management and ecological conservation
Qian Zhang1, Feng Wu2 1College of Land Science and Technology, China Agricultural University, China. 2Institute of Geographic Sciences and Natural Resources Research Chinese Academy of Sciences Library, China
Abstract
Symposium Overview:
The management of natural resources and ecological conservation efforts are increasingly complex challenges that require an integrated understanding of both ecological and social systems. Socio-ecological system (SES) modeling has emerged as a powerful tool to address these challenges by capturing the dynamic interactions between human activities and ecological processes. This symposium aims to bring together researchers, practitioners, and policymakers to explore the latest advancements in SES modeling, with a focus on its application in natural resources management and ecological conservation.
The symposium will feature presentations and discussions on innovative modeling approaches, case studies, and methodologies that integrate social and ecological data to inform sustainable natural resource management and ecological conservation strategies. By fostering interdisciplinary dialogue, this symposium will contribute to the development of more effective and equitable solutions for managing natural resources and conserving ecosystems in the face of global environmental change.
Key Themes:
Integrated Modeling Approaches: Presentations on the development and application of integrated models that combine social and ecological data to simulate complex socio-ecological systems.
Case Studies in Resource Management: Case studies demonstrating the use of SES modeling to address challenges in forestry, fisheries, water resources, agriculture, and other natural resource sectors.
Conservation Strategies: Discussions on how SES modeling can inform the design and implementation of conservation strategies, including protected area management, species recovery, and habitat restoration.
Policy and Decision Support: Exploration of how SES models can be used to support policy-making and decision-making processes, with a focus on stakeholder engagement and participatory modeling.
Challenges and Future Directions: Critical reflections on the limitations of current SES modeling approaches and identification of emerging research frontiers, including the role of big data, machine learning, and scenario-based analysis.
53 - Advancements in machine learning algorithms for ecological big data processing
Manish Kumar Pandey1, Prashant Kumar Srivastava2,3, Vikas Dugesar4, Sanjeev Kumar Srivastava3, Ram Avtar5, Amit Srivastava4 1Birla Institute of Technology, India. 2Banaras Hindu University Institute of Environment & Sustainable Development, India. 3University of the Sunshine Coast, Australia. 4IRRI South Asia Regional Centre, India. 5Hokkaido University Graduate School of Environmental Science, Japan
Abstract
The rapid growth of ecological data from remote sensing, IoT-enabled environmental sensors, and citizen science initiatives has led to the emergence of Big Data for ecosystem modelling. Traditional modeling approaches struggle to handle the complexity, scale, and heterogeneity of such data. Recent advancements in machine learning (ML) algorithms have revolutionized ecological modeling by enabling the extraction of meaningful patterns, enhancing predictive accuracy, and improving decision-making in conservation and resource management.
The Symposium on Advancements in Machine Learning Algorithms for Big Data Analysis for Ecosystem Modeling aims to bring together researchers, practitioners, and policymakers to discuss cutting-edge developments in ML-driven tools for ecological modeling. The symposium will focus on novel methodologies, real-world applications, and interdisciplinary collaborations to tackle global ecological challenges.
Topics of Interest
Machine Learning Innovations for Big Data Analysis for Ecosystem Modeling • Deep learning (CNNs, RNNs, transformers) for ecological prediction • Graph-based learning for ecosystem network modeling • Bayesian deep learning and probabilistic ML for uncertainty estimation • Reinforcement learning for adaptive environmental management • Hybrid AI-physics-informed models for ecosystem dynamics Handling Large-Scale and Heterogeneous Ecological Data • Federated learning and decentralized ML for distributed ecological data • Spatiotemporal modeling for high-resolution ecological forecasting • Explainable AI (XAI) for interpretable ecological models • Transfer learning and domain adaptation for ecological datasets • High-performance computing (HPC) for scalable ML in ecology Applications of ML in Ecological Modeling • Species distribution modeling and biodiversity conservation • Climate change impact assessment and mitigation strategies • Ecosystem service valuation using ML-driven analytics • Monitoring and controlling invasive species with AI-driven models • ML-powered precision conservation and sustainable land-use planning
57 - Advances in Data Assimilation (DA) and optimization techniques
Manish Kumar Pandey1, Prashant Kumar Srivastava2,3, Vikas Dugesar4, Ram Avtar5, Sanjeev Kumar Srivastava3 1Birla Institute of Technology, India. 2Banaras Hindu University Institute of Environment & Sustainable Development, India. 3University of the Sunshine Coast, Australia. 4IRRI South Asia Regional Centre, India. 5Hokkaido University Graduate School of Environmental Science, Japan
Abstract
The rapid evolution of data assimilation (DA) and optimization techniques has significantly improved predictive modeling, decision-making, and control across diverse scientific and engineering domains. Data assimilation, which integrates observational data with numerical models, has seen major advancements through machine learning-enhanced methodologies, hybrid ensemble-variational approaches, and the utilization of high-resolution remote sensing data. Simultaneously, optimization techniques have advanced to address high-dimensional, non-convex problems, leveraging stochastic gradient-based methods, evolutionary algorithms, Bayesian optimization, and reinforcement learning. The synergy between AI-driven optimization and traditional numerical techniques has led to more efficient parameter estimation, hyperparameter tuning, and real-time decision-making in many applications.
Considering the application in the multidisciplinary domains, this symposium is dedicated to exploring cutting-edge methodologies that enhance our understanding, prediction, and management of ecological systems. As ecological modeling becomes increasingly data-driven, the integration of advanced DA and optimization techniques is crucial for improving ecosystem forecasting, biodiversity conservation, and resource management. This symposium provides a platform to discuss recent developments, challenges, and interdisciplinary applications of these techniques in ecological research.
Topics of Interest
Innovations in Data Assimilation for Ecology • Machine learning-enhanced data assimilation for ecological forecasting • Multi-source data fusion (remote sensing, in situ observations, citizen science) • Uncertainty quantification and error correction in ecological models • Bayesian and probabilistic approaches for ecological state estimation Optimization Techniques for Ecological Systems • Stochastic and evolutionary algorithms for ecosystem management • Reinforcement learning for adaptive conservation strategies • Spatial-temporal optimization for biodiversity and habitat modeling • High-performance computing for large-scale ecological optimization Applications in Ecological Modeling • Climate change impact assessment and mitigation strategies • Invasive species control and biodiversity conservation • Sustainable agriculture and land-use planning • Ecosystem services valuation and resource allocation Future Directions and Challenges • Integration of physics-informed AI and ecological modeling • Scalable and interpretable AI-driven ecological models • Data scarcity and uncertainty in ecological predictions
59 - Dynamic energy budget modeling
Romain Lavaud1, Nina Marn2, Marko Jusup3, Konstadia Lika4 1Louisiana State University Agricultural Center, USA. 2Ruđer Bošković Institute, Croatia. 3Japan Fisheries Research and Education Agency, Japan. 4University of Crete Department of Biology, Greece
Abstract
DEB theory provides a framework for constructing families of related models, rooted in a mechanistic description of the individual metabolic processes. This mechanistic approach is essential for linking functional traits to predictive variables, ensuring that models remain adaptable to shifting environmental conditions. A key challenge in model applications is the assessment of parameter values. Many applications of DEB theory have demonstrated the capacity to extract meaningful parameters from data, enabling models to inform effective decision-making. This symposium will showcase the latest advancements in DEB theory and its applications, fostering cutting-edge, cross-disciplinary developments to address the pressing ecological challenges of our time.
60 - Ecosystem services and community interplay in mountain areas
Santatanu Patnaik Rajiv Gandhi University, India
Abstract
Mountain areas are rich in biotic endowment, have high topographic variations and have sparse but clustered population. These areas are most vulnerable to disaster and extreme events, essentially being fragile in nature. Despite the complex interplay between topography, climate extremes, biological hotspots, dynamic and uncertain environmental setup, the indigenous communities have been surviving with a delicate balance between nature and demand for resource to meet their ends. The recent understanding of dynamism of mountain ecosystem and looking into it through the prism of services rendered by it, have offered a better way to systematically analyze and prescribe appropriate space and time framework for development of native communities sustainably. One of the ideal regions for comprehension of such concepts is the Himalayas. Many agencies and scholars are engaged in different parts of the world to explore nuances of man-nature interactions in mountainous areas. As there is no end-of-tunnel of investigation, to reach at a steady state of development paradigm and optimal use of services offered by mountain ecosystem, a larger platform for academic discussion and scientific assessment will provide pathway for governance and policy prescription to keep mountain areas beautiful and charming.
61 - Challenges and advances in integrating ecological and earth system models across multiple sectors
John Kim1, Alla Golub2, Yueyang Jiang3, Fang Li4, Erwan Monier5, Christopher Reyer6, Leonardo Salas7, Brent Sohngen8 1USDA Forest Service Western Wildland Environmental Threat Assessment Center, USA. 2Purdue University, USA. 3Rice University, USA. 4Chinese Academy of Sciences, China. 5University of California Davis, USA. 6Potsdam Institute for Climate Impact Research (PIK) e V, Germany. 7Point Blue Conservation Science, USA. 8The Ohio State University, USA
Abstract
A wide variety of models are used to simulate and study ecological and earth system processes. Global models generate global emissions scenarios, socio-economic scenarios, land-use change scenarios and combine them to create projections of future climate. Future climate data are in turn ingested by various impact models across many sectors. In this arena of modeling, models are often coupled together to create a new model. For example, a land surface model may be integrated into a climate system model; a wildfire model may be integrated into an ecosystem model; or a global timber market model may be linked with a global vegetation model. Integrating two or more models faces significant challenges, as the component models employ different techniques, cover different spatio-temporal domains, or use different spatial or temporal resolutions, and span sectors with different conceptual frameworks. Model integration may be loose or asynchronous, where one model’s outputs are used as inputs of another; or the models may be tightly integrated, running synchronously or combined into a single simulation. In this session, we invite presentations describing thorny challenges, novel solutions, and success stories in model integration across multiple sectors at any spatio-temporal scale. We aim to share experiences, stimulate innovative integration approaches, and learn from our mutual successes in simulating natural systems and coupled social-ecological systems.
63 - Land change modeling for sustainable land management
Ronald C. Estoque1, Peter H. Verburg2, Yuji Murayama3 1Center for Biodiversity and Climate Change, Forestry and Forest Products Research Institute, Japan. 2Environmental Geography Group, IVM Institute for Environmental Studies, Vrije Universiteit Amsterdam, The Netherlands. 3Faculty of Life and Environmental Sciences, University of Tsukuba, Japan
Abstract
Land change—encompassing land use and land cover change (LUCC)—is a key driver of ecosystem alterations, significantly impacting the provision and sustainability of ecosystem services. Understanding LUCC’s underlying drivers and anticipating future land transformations through spatially explicit modeling are essential for effective land management. How land is used and managed has profound implications for resilience, sustainability, and well-being. Thus, land change modeling plays a critical role in informing transformative planning and governance.
By integrating geospatial technologies and scenario analysis, land change models provide spatially explicit insights into past, present, and future land transitions. Hence, they enable decision-makers to assess trade-offs, predict outcomes, and support informed planning that balances ecological, economic, and social priorities. Simulating land changes under socio-economic, environmental, and policy scenarios helps identify sustainable pathways.
This session aims to highlight the role of land change modeling in advancing sustainable land management by showcasing diverse modeling approaches and their applications across ecosystems, scales, and contexts. It will also demonstrate how land change models inform conservation, climate change mitigation and adaptation, and resilience-building efforts, fostering interdisciplinary collaboration and innovative solutions for sustainable land systems.
Submissions are welcome on the development and advancement of land change models (or land use models), their applications, methodological innovations, integration with other modeling frameworks, and broader perspectives on land change modeling and land system science.
64 - Modeling water quality and ecosystems in estuarine and coastal areas: Japan and other countries under human activities and climate change
Masayasu Irie1, Akio Sohma2 1Osaka University, Japan. 2Osaka Metropolitan University, Japan
Abstract
In Japan, some coastal areas that once suffered from water quality deterioration due to excessive nutrient loading are now experiencing improvements as a result of continued load reduction. However, the primary goal of load reduction—namely, the recovery of biological resources and biodiversity—has not yet been achieved.
This symposium will provide an opportunity to discuss modeling that contributes to the elucidation, prediction, evaluation, and communication of water quality and ecosystems in estuarine and coastal areas, which are significantly affected by rapid changes in human activities in terrestrial areas and climatic conditions.
The symposium will introduce recent efforts in water quality and ecosystem modeling and present initiatives aimed at the recovery of ecosystems and biodiversity. Additionally, presentations on efforts in Japan and other countries are welcome and will be made, fostering an exchange of opinions on ecosystem modeling.
65 - Data-driven ecosystems: Modeling for improved health outcomes and sustainable development
Adaoyibo Okpala University College London, UK
Abstract
This paper investigates the critical relationship between urban green spaces and community health outcomes, emphasizing the role of data-driven modeling in ecosystem management. As urbanization intensifies, the need for sustainable practices that enhance public health and biodiversity becomes increasingly urgent. Our analysis draws on health metrics, including air quality and public health indicators, to illustrate how urban green spaces contribute to improved community well-being.
Key findings indicate that effective ecosystem management leads to cleaner air and water, thereby mitigating health risks associated with pollution. By integrating sustainable development principles, we propose actionable recommendations for increasing urban green spaces, which not only foster biodiversity but also promote healthier communities. The paper highlights the importance of transitioning from data to decision-making, showcasing case studies that exemplify the profound implications of modeling in managing ecosystems. For instance, a dynamic modeling study in Dominica revealed the interconnectedness of ecological, economic, and societal factors in tourism development, while a Systemic Spatial Decision Support System in Karachi demonstrated the value of community engagement in addressing environmental challenges.
Moreover, we pose thought-provoking questions regarding the collaboration between data scientists and ecologists, and the role of community involvement in implementing data-driven decisions. This essay underscores the necessity of leveraging data and modeling to cultivate resilient ecosystems that benefit both human health and sustainable development. As we confront pressing environmental issues, embracing data-centric approaches will be essential for fostering a healthier and more sustainable future.
Keywords: Urban Green Spaces, Community Health, Sustainable Development, Ecosystem Management, Data-Driven Decision Making.
66 - Modeling species distributions and interactions in ecological communities
Shinji Fukuda1, Young-Seuk Park2 1Tokyo University of Agriculture and Technology, Japan. 2Kyung Hee University, Republic of Korea
Abstract
Species distribution models (SDMs) have evolved significantly with the integration of data-driven techniques, enhancing their predictive accuracy and applicability across diverse ecological landscapes. However, many SDMs still focus on individual species, overlooking the critical roles of species interactions in shaping distributions within ecological communities. This symposium highlights recent innovations in SDMs, with a focus on deep learning, machine learning, statistical modeling, remote sensing, and big data.
We invite discussions on novel methodologies that enhance both single-species and multi-species SDMs through the integration of high-resolution spatiotemporal datasets, ensemble modeling strategies, and AI-driven predictions to assess species occurrence under dynamic environmental conditions. Special attention will be given to approaches that incorporate biotic interactions such as competition, predation, mutualism and facilitation into predictive models. The symposium will also address critical challenges, including data quality and quantity, model uncertainty, generalizability, and the interpretability of complex models.
By bringing together researchers from ecology, data science, and modeling disciplines, this symposium aims to foster interdisciplinary collaboration, bridge methodological gaps, and drive the development of robust, scalable, and transparent SDMs. Ultimately, this exchange will contribute to improving biodiversity conservation strategies and ecosystem management by advancing SDMs to capture both biotic and abiotic complexities.
67 - Marine plastic modelling: Biological interactions, transport dynamics, and ecosystem impacts
Shin-ichi Ito1, Nikolai Maximenko2, Déborah Benkort3, Haodong Xu1 1The University of Tokyo, Japan. 2University of Hawaii at Manoa, USA. 3Université du Québec à Rimouski, Canada
Abstract
Marine macro/microplastics pollution has been a growing environmental issue for the last decades. Previous studies have focused on the plastic concentration in seawater and sediments, their global distribution and transport mechanisms. However, the fate and behaviors of plastics are still unclear.
Some of the plastics with higher density than seawater and are expected to sink to the seafloor once they lose their buoyancy. However, these heavy plastics are observed at the ocean surface and in the water column, raising questions about processes governing their movement. Conversely, plastics with lower density than seawater, should normally float at the sea surface, but have been found in the seafloor sediments, suggesting that biofouling and bio-aggregation might change their sinking behavior in the seawater. However, the precise role of biological effects in plastics transport is still undetermined, making it difficult to estimate and predict their long-term fate and distribution in the ocean.
On the other hand, macro/microplastics can serve as habitats for various marine biotas including viruses, bacteria, and periphyton. These attached marine biotas could facilitate the long-distance transport of invasive species. In addition, the plastics could act as vectors for chemical contaminants, by releasing additives from their own or by carrying attached pollutants. These transport processes of invasive biotas and chemical components may have consequences on marine ecosystems. In addition, macro/microplastics are accidentally ingested by many marine organisms, representing another major source of concern. The understanding of the ecological impact of those macro/microplastics on the marine ecosystem are difficult to assess and remains insufficiently studied.
This session will focus on marine macro/microplastics modelling approaches to investigate biological interactions influencing plastic fate and transport and their impacts on marine ecosystems. We will also discuss issues in plastic modelling and future strategies to improve our understanding of these complex dynamics.
68 - Advances in signal processing for bioacoustics
Shinji Fukuda1, Kohei Yatabe1, Tsuyoshi Shimmura1, Naohisa Nakashima2, Hiroki Kobayashi3 1Tokyo University of Agriculture and Technology, Japan. 2Obihiro University of Agriculture and Veterinary Medicine, Japan. 3The University of Tokyo, Japan
Abstract
Bioacoustics has become an essential tool for biodiversity monitoring and the study of animal behavior. However, effective and efficient analysis of both large-scale and small, sparse acoustic datasets remains a significant challenge. This symposium highlights advances in signal processing techniques for ecological modelling, including time-frequency analysis, sound source separation, species classification, and sound source localization, to enhance the extraction of ecological insights from bioacoustic recordings. We welcome contributions exploring automated species identification, species distribution modelling, behavioral analysis, and ecosystem monitoring frameworks using bioacoustic data. Special emphasis will be placed on innovative approaches that leverage AI, deep learning, and computational acoustics to improve detection accuracy and ecological interpretation. Additionally, we invite discussions on methodologies for overcoming data scarcity, such as transfer learning, data augmentation, and few-shot learning, which are critical for improving the model performance in data-limited scenarios. This symposium aims to foster interdisciplinary collaboration and discussions on emerging methodologies by bridging the fields of acoustics, ecology, and data science. The insights shared among the participants will contribute to advancing ecological models that incorporate bioacoustics, enhancing their applications in biodiversity conservation and environmental assessments across both small and large datasets.
69 - Marine biogeochemical/ecological modelling: advancements and best practices
Yasuhiro Hoshiba1, Sachihiko Itoh2, S. Lan Smith3 1Japan Agency for Marine-Earth Science and Technology, Japan. 2The University of Tokyo, Japan3
Abstract
Marine biogeochemical and ecological models have increasingly been applied to understand, assess and project marine ecosystem services such as climate regulation, water quality control and fisheries production. While most such models commonly considered interactions of nutrients and plankton under the influence of hydrodynamic circulation, there currently is a broad spectrum of complexity in the models, in terms of key elements considered, functional types, hydrodynamic representations, and the interaction with human activities, etc. Furthermore, capturing biodiversity and key ecological processes in large scale models such as Earth system models remains a pressing challenge. This session aims to share recent technological advancements and best practices in such modelling in the open ocean (including those targeting regional areas, marginal seas and the global ocean) along these various axes. We welcome modelling studies of any complexities for any issues, including carbon transport, ocean acidification, ecological regime shifts and future predictions, as well as intercomparisons and interdisciplinary studies linking natural and social sciences.
8th Biennial best young researcher paper award
Win your registration fee and $200 USD to attend the conference.
Any author, age 35 or younger, who published a paper in Ecological Modelling S’ouvre dans une nouvelle fenêtre in 2024–2025 or has a paper accepted before the deadline is eligible.
Forward a PDF of the paper with indication for consideration of the prize to Brian Fath S’ouvre dans une nouvelle fenêtre. Self-nominations are accepted.
Papers will be evaluated by a small group of ISEM officers for overall impact, originality, and quality of work. Contact Brian Fath S’ouvre dans une nouvelle fenêtre for further questions as needed.
Deadline: 30 March 2025
Please do not email credit card information under any circumstances.