Coastal Digital Twins: building knowledge through numerical models and IT tools

SeaDataNet is a pan-European infrastructure that manages and provides access to marine datasets collected by European organisations through research cruises and observational activities in coastal waters, regional seas, and the global ocean. It was founded by National Oceanographic Data Centres (NODCs) and major marine research institutes. The network has expanded through successive EU-funded RTD projects and by contributing to major European initiatives such as EMODnet, Copernicus Marine Service, ENVRI, and the European Open Science Cloud (EOSC).SeaDataNet develops and promotes widely adopted standards, vocabularies, software tools, and services that support FAIR marine data management. Its core service, the CDI (Common Data Index), provides unified online discovery and access to in situ marine observation data managed by more than 115 data centres in 34 countries. The service currently offers access to over 3 million datasets from more than 1,000 European organisations, covering physical, chemical, biological, geological, and geophysical data from European waters and the global ocean. The use of standard metadata, formats, and controlled vocabularies ensures rich, highly FAIR datasets.SeaDataNet also delivers core data services for EMODnet Chemistry, Bathymetry, and Physics, harmonising large volumes of marine data that support the production of thematic data products, including Essential Climate Variable (ECV) and Essential Ocean Variable (EOV) datasets.Environmental science increasingly relies on large, heterogeneous, and rapidly growing data collections that must be efficiently accessed, subsetted, and harmonised for use in models, digital twins, AI workflows, and Virtual Research Environments (VREs). The fully open-source Beacon software, developed by MARIS https://beacon.maris.nl/, addresses these challenges by enabling cloud-native, high-performance data lakes that are fast to deploy and access. Beacon supports parameter harmonisation using metadata annotations based on NERC vocabularies, ECV vocabularies, and the I-ADOPT methodology adopted in ENVRI-HUB Next.To improve the ease of access to subsets of the SeaDataNet CDI data collection, a Beacon instance containing all the open SeaDataNet data was set-up. This now allows users to obtain real-time access to data subsets in multiple data formats (NetCDF, Parquet, Zarr) and flexible querying from Jupyter Notebooks or a newly developed Beacon studio (user interface) for non-technical users. Within ENVRI-HUB Next, this SeaDataNet instance enables on-the-fly access to ECV-related subsets from millions of files via Jupyter Notebooks, ready for use in the Analytical Framework.The presentation focuses on this use case, the technical solution, and its potential applicability for other Research Infrastructures supporting EOSC use cases.

Climate change is affecting the global water, energy and carbon cycle resulting in more severe hydrometeorological events with more societal impact (e.g. precipitation, floods, droughts). Decision support systems for operational or planning purposes are essential to accurately predict and monitor environmental disasters, and optimally manage water and environmental resources now and in the future. The Digital Twin Component (DTC) Hydrology Next project focuses on solutions for monitoring and simulations and forecasting, it requires high-resolution (1 km, 1 hour-day) satellite Earth Observation (EO) data, fully integrated with advanced and spatially distributed modelling systems. Within this scope,  we aim to improve operational reservoir monitoring to obtain reliable estimates  of surface area, water level and volume (i.e. storage). Secondly, we aim to enhance predictions by data assimilation using wflow_sbm (Imhoff et al., 2020, Eilander et al., 2021, van Verseveld et al., 2024, Imhoff et al., 2024). The focus will be on the Rhine catchment focusing on flooding in co-creation with RWS (Dutch ministry of traffic and waterways) and Zambia focusing on reservoir monitoring and flood management working together with WARMA (Water Resources Management Authority). We also consider including  2D hydraulic flood simulations using SFINCS (Leijnse et al., 2021) driven by outputs from wflow_sbm. Eilander, D., van Verseveld, W., Yamazaki, D., Weerts, A., Winsemius, H. C., and Ward, P. J. (2021) A hydrography upscaling method for scale invariant parametrization of distributed hydrological models, Hydrol. Earth Syst. Sci., 25, 5287–5313, https://doi.org/10.5194/hess-25-5287-2021.Imhoff, R. O., van Verseveld, W. J., Osnabrugge, B., and Weerts, A. H. (2020) Scaling Point-Scale (Pedo)transfer Functions to Seamless LargeDomain Parameter Estimates for High-Resolution Distributed Hydrologic Modeling: An Example for the Rhine River, Water Resour. Res., 56, https://doi.org/10.1029/2019WR026807.Imhoff, Ruben and Buitink, Joost and van Verseveld, Willem and Weerts, Albrecht, A fast high resolution distributed hydrological model for forecasting, climate scenarios and digital twin applications using wflow_sbm. Environmental Modelling & Software,179,https://doi.org/10.1016/j.envsoft.2024.106099, 2024Leijnse, T. W. B. et al. (2021). Modeling compound flooding in coastal systems using a computationally efficient reduced-physics solver: Including fluvial, pluvial, tidal, wind- and wave-driven processes. Coastal Engineering, 163, Article 103796. https://doi.org/10.1016/j.coastaleng.2020.103796van Verseveld, W. J., Weerts, A. H., Visser, M., Buitink, J., Imhoff, R. O., Boisgontier, H., Bouaziz, L., Eilander, D., Hegnauer, M., ten Velden, C., and Russell, B.: Wflow_sbm v0.7.3, a spatially distributed hydrological model: from global data to local applications, Geosci. Model Dev., 17, 3199–3234, https://doi.org/10.5194/gmd-17-3199-2024, 2024.       

Coastal observatories integrate operational numerical modelling and data to produce targeted tools for coastal early warning and risk management. They explore the usage of computational services and web-based tools to produce timely and interactive products in support of coastal management, building the foundation for Digital Twins from the ocean to the river basin (Rodrigues et al., 2021). Examples of computational services in this context include OPENCoastS (Oliveira et al., 2020) and SURF (Trotta et al., 2021), for on-demand predictions, and WebGIS platform applications, customized to meet user requirements.In Copernicus Marine Service National Collaboration Programme’s project CONNECT, a multi-purpose coastal service was established for two sites, based on coastal observatories technology (Rodrigues et al., 2021), merging high-resolution model predictions from OPENCoastS deployments and data from the national infrastructure CoastNet (Figure 1). By providing both pre-configured products (such as maps, automatic quality assessments and indicator results) and the capacity for users to build products on-the-fly (such as probing results as time series and building tailored dashboards with user selected information), the CONNECT service is one of the first core coastal Digital Twins applicable for flooding and coastal pollution management (Figure 2). Figure 1 – CONNECT’s coastal service architectureFigure 2 – Sample user services in the Tagus estuary: Virtual sensors and configurable dashboard.Computational services are also applied to the construction of the river predictions (Jesus et al., in review). Particularly suited for dam-controlled river basins, a deep learning service was used to predict river flows at the upstream boundary of CONNECT sites, based on Multilayer Perceptron algorithms, potentially reducing severe phase errors during extreme events in the estuary than conventional persistence approaches. An on-demand platform is being built to allow users to setup their own AI model, with multiple choices for deep learning algorithms, input and output data , and training and validation periods. These tools will be validated in two sites in Africa, addressing co-creation challenges with local communities, and in two sites in Portugal, to address urban-driven contamination and shellfish farm needs.ReferencesRodrigues M., Martins R., Rogeiro J., Fortunato A.B., Oliveira A., Cravo A., Jacob J., Rosa A., Azevedo A., Freire P. , 2021. A Web-Based Observatory for Biogeochemical Assessment in Coastal Regions. Journal of Environmental Informatics, 38(1), 1-15,  https://doi.org/10.3808/jei.202100450Oliveira A., Fortunato A.B.  Rogeiro J., Teixeira J.,  Azevedo A., Lavaud L.,  Bertin X.,  Gomes J., David M.,  Pina J., Rodrigues M.,  Lopes P., 2019. OPENCoastS: An open-access service for the automatic generation of coastal forecast systems, Environmental Modelling & Software,  124, 104585, https://doi.org/10.1016/j.envsoft.2019.104585.Trotta F., Federico I., Pinardi N., Coppini G., Causio S., Jansen E., Iovino D., Masina S., 2021. A Relocatable Ocean Modeling Platform for Downscaling to Shelf-Coastal Areas to Support Disaster Risk Reduction , Frontiers in Marine Science, 8, pp. 317.Jesus, G., Mardani, Z., Alves, E., Oliveira, A. Under review, Deep Learning-Based River Flow Forecasting with MLPs: Comparative exploratory analysis applied to the Tejo and the Mondego rivers, Sensors.

As a non-engineering measure for flood control, real-time forecasting provides valuable information like the magnitude and occurrence time of flood peak, which is essential for decision-making. In China, many reservoirs are built and operated in major river including ChangJiang River Basin. Operations of reservoirs pose new challenges for real-time forecasting. For example, (1) it’s difficult to calibrate model parameters due to human-impaired streamflow series, (2) the leading time of real-time forecasting is much shorter.To address these challenges, we propose a distributed real-time streamflow forecasting framework using the Xin’anjiang (XAJ) hydrological model. We evaluate different scale of computational units of the XAJ model to better characterize the runoff processes, land surface characteristics, and meteorology factors. We then develop a set of models to calculate model parameters from land surface characteristics, which reduce the calibration requirement. We also develop an algorithm to correct the bias of precipitation forecasts, which is coupled with real-time forecasting framework. This helps to extend the leading time of real-time forecasting.Our proposed framework is tested and validated at Upper Changjiang River Basin and get promising feedbacks.

The paper examines the evolution of innovation evaluation methods, which is directly influenced by technological advancement and changes in the socio-economic environment. Initially, innovation activities were evaluated primarily from a financial perspective, using classic methods such as NPV, IRR, and Payback Period, focusing on economic feasibility but often neglecting the social and environmental impact. Modern approaches, however, are oriented towards a more holistic consideration of various factors, including intangible assets such as knowledge, ecological sustainability, and strategic significance. Digitalization, according to the authors, requires the integration of new methods such as multi-criteria analysis, which accounts for economic, social, environmental, and technological factors, as well as the creation of hybrid models that combine traditional financial tools and innovative approaches. Specifically, real options and digital twins provide flexible evaluation and forecasting, considering potential changes in the market and technology. The paper also highlights the limitations of traditional evaluation methods like NPV and IRR in the context of high uncertainty, typical of innovative projects. The study proposes a methodological approach for comprehensive evaluation of innovative projects based on the integration of financial, social, environmental, and technological factors, utilizing advanced digital technologies. Authors suggest a multi-step evaluation model, which includes multidisciplinary assessment of basic parameters, real options for evaluating strategic flexibility, and digital twins for real-time project implementation forecasting and optimization. These approaches allow for the consideration of dynamic market changes and provide more accurate and adaptive information for strategic decision-making. An interesting aspect of the study is the application of artificial intelligence and digital twins, which significantly enhance the precision of project evaluations and increase their adaptability to external changes. Through case studies of innovative projects, particularly in logistics process management, the paper demonstrates how the proposed methods allow for a comprehensive evaluation, taking into account financial indicators, ecological sustainability, and technological innovation. The results of the study highlight the necessity of integrated approaches to evaluating innovative projects, as digitalization and technological innovations dramatically transform the evaluation processes. These methods enable the creation of more adaptive and accurate models that account for a wide range of factors and can respond promptly to changes in the market environment, which is particularly crucial in the context of digital business transformation.

Urban infrastructure, especially smart bridges, is growing rapidly, requiring effective solutions to ensure structural integrity. Conventional Structural Health Monitoring (SHM) systems have limitations in scalability, speed, and accuracy. This paper presents a new edge-enabled digital twin platform for realtime SHM of smart bridges, combining IoT, cloud computing, and edge computing. The architecture offers a high-performance, decentralized scheme of continuous monitoring to facilitate real-time detection and forecasting of structural failure by integrating the sensors on the bridges and the edge devices. The fundamental approach uses an Autoencoder-based anomaly detection, in which Autoencoders are trained to learn to recreate sensor information, and learn to behave normally by modeling the structural behavior of the bridge. In the case of real-time monitoring, the differences between the real sensor values and the reconstructed data are compared, and anomalies are noted, which are indicators of structural problems. This architecture minimizes latency by often processing data at the edge and by improving decision-making by initiating maintenance actions based on identified anomalies. The digital twin model captures the actual behavior of a bridge, providing extensive information on the current condition of the infrastructure. The suggested system is tested in terms of five major performance indicators, namely accuracy, processing time, energy consumption, scalability, and false alarm rate. Indications show that the system delivers better results than traditional SHM systems across a range of key features, including much higher anomaly detection accuracy, shorter processing time, and more efficient energy use. The system can be scaled, and additional bridges with a significantly reduced false alarm rate can be supported, therefore reducing unnecessary maintenance intervention. In general, the edge-enabled digital twin architecture can provide a promising solution to real-time SHM to enhance the safety and efficiency of smart bridges. The next research will involve integrating AI-based predictive analytics into the digital twin system to increase further the capacity of the system to indicate structural failures prior to it happening.

This systematic review and meta-analysis investigates the evolving role of Building Information Modeling (BIM) in managing multidimensional risks associated with sustainable bridge infrastructure projects. Drawing upon the PRISMA methodology, the study rigorously screened and analyzed a total of 87 peer-reviewed journal articles and conference papers published between 2010 and 2024. The selected literature was sourced from five leading academic databases—Scopus, Web of Science, IEEE Xplore, ScienceDirect, and the ASCE Library—ensuring comprehensive coverage of global research on BIM-based risk management practices. The review focuses on how BIM contributes to the identification, analysis, mitigation, and continuous monitoring of technical, financial, environmental, and operational risks throughout the lifecycle of bridge projects. Findings reveal that BIM’s capabilities extend well beyond 3D visualization, offering robust tools for 4D scheduling, 5D cost estimation, and real-time data integration that collectively enhance proactive decision-making. BIM facilitates early risk detection through clash detection, geospatial analysis, and scenario simulations during the planning and design stages, reducing design inconsistencies and constructability issues. Moreover, the study synthesizes evidence demonstrating how BIM supports cost and schedule risk control by enabling dynamic simulation of construction sequences and integration with live pricing data, resulting in reduced cost overruns and schedule delays. The review also highlights the increasing convergence of BIM with Internet of Things (IoT) technologies, digital twins, and Geographic Information Systems (GIS), enabling real-time monitoring of structural health, predictive maintenance, and environmental risk modeling in operational phases. Lifecycle-oriented risk governance through BIM emerged as a key theme, particularly in studies focused on sustainability assessments, material optimization, and long-term asset performance. Institutional adoption patterns, regulatory influences, and policy frameworks were also examined, revealing that countries with mandated BIM protocols exhibit more advanced integration of BIM into public infrastructure risk management. Despite persistent challenges such as fragmented data standards, skill gaps, and limited interoperability, the majority of the reviewed literature—accumulating over 3,400 citations—converges on the conclusion that BIM is an essential enabler of risk-informed, sustainable bridge engineering.

Urban wastewater microbiomes are complex and temporally dynamic, offering valuable insight into community-scale microbial ecology and potential public health trends. However, existing wastewater-based studies often remain descriptive, lacking tools for predictive modeling. In this study, we introduce a digital twin framework that forecasts microbial abundance trajectories in urban wastewater using an interpretable generative model, Q-net. Trained on a 30-week longitudinal metagenomic dataset from seven wastewater treatment plants, the model captures temporal microbial dynamics with high fidelity ( for key taxa; at the final timepoint). Beyond accurate forecasting, Q-net provides transparent model structure through conditional inference trees and enables simulation of realistic microbial trends under hypothetical scenarios. This work demonstrates the potential of digital twins to move wastewater microbiome studies from static snapshots to dynamic, predictive systems, with broad implications for environmental monitoring and microbial ecosystem modeling.

The prediction of photovoltaic (PV) system performance has been intensively studied as it plays an important role in the context of sustainability and renewable energy generation. In this paper, a digital twin (DT) model based on a domain-matched transformer is proposed using convolutional neural network (CNN) for domain-invariant feature extraction, transformer for PV performance prediction, and domain adaptation neural network (DANN) for domain adaptation. The effectiveness of the proposed framework is validated using a PV power prediction dataset. The results indicate an accuracy improvement of up to 39.99% in model performance. Additionally, experiments with varying numbers of timestamps demonstrate enhanced PV power prediction performance as parameters are continuously updated within the DT framework, offering a reliable solution for real-time and adaptive PV power forecasting.

In environmental modelling, estimating the confidence level in conceptual model parameters is necessary but difficult. Having a realistic estimation of the uncertainties related to the parameters is necessary i) to assess the possible origin of the calibration difficulties (correlation between model parameters for instance), and ii) to evaluate the prediction confidence limits of the calibrated model. In this paper, an application of the Metropolis algorithm, a general Monte Carlo Markov chain sampling method, for the calibration of a four-parameter lumped urban stormwater quality model is presented. Unlike traditional optimisation approaches, the Metropolis algorithm identifies not only a "best parameter set", but a probability distribution of parameters according to measured data. The studied model includes classical formulations for the pollutant accumulation during dry weather period and their washoff during a rainfall event. Results indicate mathematical shortcomings in the pollutant accumulation formulation used.

Autogenous and allogenous control of the phytoplankton standing crop and biomass was proved in Mex Bay, west of Alexandria (Egypt). Among the clear signs of eutrophication are the heavy visible blooms, which became regular events, at intermittent periods during the warm seasons. The causative organisms progressed differently, the dinoflagellate, Gymnodinium catenatum, is well known toxic species. The Environmental Impact Assessment model (EIA) was applied to Mex Bay. It is a three-dimensional water quality and flow model, based on the three interacted dimensional: longitude, latitude and depth, calculations of time and space variations. The model deals with processes that affecting the water quality. It was initiated with measured field data and load input. The model was used to simulate phytoplankton biomass transport and ambient nutrient concentrations (nitrate, phosphate & ammonia). The model results and the field measurements are compared in order to fit the model. Generally, the calculated flow and concentrations corresponded well with observations. Deviations are attributed to several reasons. A notable influence of the main outfall on the concentrations, particularly in the near shore area, as well as, on the seawater quality was detected. The spatial distribution of the phytoplankton standing crop coincided clearly with the accompanied chlorophyll a concentrations calculated by the model. The results stressed the need to reduce load input into Mex Bay, in order to counteract eutrophication in Alexandria waters. Nitrogen reduction is more important for the whole area, while phosphorous reduction is more effective in the bay.

The Mumbuca stream, which flows through the urban area of Monte Carmelo, Minas Gerais, in the Triângulo Mineiro region, is the water body receiving raw sewage; the deficiency of a sewage system hampers the maintenance of the equilibrium of the aquatic ecosystem and brings social and economic losses to the local population. The objective of this study is to diagnose and forecast the water quality in an urban lotic environment, specifically the Mumbuca Stream in the city of Monte Carmelo (MG), with the main impacting variables being land use and occupation as well as urban sewage and stormwater drainage infrastructure. With a bimonthly seasonality during drought periods, the parameters pH, water temperature, total suspended solids, turbidity, dissolved oxygen, biochemical oxygen demand, total phosphorus, organic nitrogen, ammonia, and nitrate were characterized at four monitoring stations along the Mumbuca Stream. Regarding the mathematical modeling of self-purification using the WASP computational tool, after calibrating the kinetic terms inputted into the simulations of the parameters dissolved oxygen, biochemical oxygen demand, ammonia, nitrate, and total phosphorus, the maximum pollutant loads of organic matter, nitrogen, and phosphorus that the stream can support without compromising the limit concentrations established in CONAMA Resolution 357 (2005) were evaluated. The results obtained from the calibration process of the water quality model indicate that the Mumbuca Stream, during drought periods, does not have self-purification capacity for the pollutant loads from the clandestine connections of raw domestic sewage, with the maximum admissible loads obtained in the study being 7.0 kg BOD/day, 1.0 kg Ammonia/day, and 0.02 kg Total Phosphorus/day.

Geographic Information System (GIS) is quickly becoming a critical component to develop and sustain asset management for today’s wastewater utilities as most of their data is geographically referenced. This technology offers sophisticated data management and spatial analysis capabilities that can greatly improve and facilitate urban drainage infrastructure modeling and analysis applications. This paper presents a comprehensive GIS-based decision support system that integrates several technologies for use in the effective management of urban stormwater collection systems. It explicitly integrates ESRI ArcGIS geospatial model with advanced hydrologic, hydraulic, and water quality simulation algorithms, nature-based global optimization techniques including genetic algorithms for design and calibration of stormwater management models, automated dry weather flow generation and allocation, and automated subcatchment delineation and parameter extraction tools to address every facet of urban drainage infrastructure management. The geocentric interface allows seamless communication among the various modules. The resulting decision support system effortlessly reads GIS datasets, extracts necessary modeling information, and automatically constructs, loads, designs, calibrates, analyzes and optimizes a representative urban drainage model considering hydrologic and hydraulic performance requirements. It also makes it easy to run, simulate and compare various modeling scenarios, identify system deficiencies, and determine cost-effective physical and operational improvements to achieve optimum performance and regulatory compliance. These combined capabilities provide favorable geospatial environment to assist wastewater utilities in planning, designing, and operating reliable systems and in optimizing their capital improvement programs.

Bayesian bootstrapping in real-time probabilistic photovoltaic power forecasting

Development of innovative tools for real-time monitoring and forecast of environmental health impacts is central to effective public health interventions and resource allocation strategies. Though a need for such generic tools has been previously echoed by public health planners and regional authorities responsible for issuing anticipatory alerts, a comprehensive, robust and scalable real-time operational framework for predicting temperature-related excess deaths at local scale has not been developed yet. Filling this gap, we propose a flexible conceptual framework for coupling publicly available operational weather forecasts with temperature-mortality risk functions specific to small census-based zones, the latter derived using state-of-the-art environmental epidemiological models. Utilising high-resolution temperature data forecast by a leading European meteorological centre, we demonstrate a real-time application to forecast the excess mortality during the July 2022 heatwave over England and Wales. The output by way of expected temperature-related excess deaths at small geographic areas on different lead times, can be automated to generate maps at various spatio-temporal scales, thus facilitating preventive action and allocation of public-health resources in advance. While the real-case example discussed here demonstrates an application for predicting (expected) heat-related excess deaths, the framework can also be adapted to other weather-related health risks and to different geographical areas, provided data on both meteorological exposure and the underlying health outcomes are available to calibrate the associated risk functions. The proposed framework addresses an urgent need for predicting the short-term environmental health burden on public health systems globally, especially in low- and middle-income regions, where rapid response to mitigate adverse exposures and impacts to extreme temperatures are often constrained by available resources.