Professor, Jamia Millia Islamia, India
Environmental degradation has profound, interconnected global effects that affect almost every aspect of life on Earth, from human health and economic stability to the very survival of countless species, biodiversity and ecosystems. When considering acute environmental pressures, air and water pollution are particularly important, as they cause millions of premature deaths each year, along with water scarcity, forest degradation, and industrial emissions. This research assesses AI and Big Data applications for environmental monitoring globally and in India, with a focus on air quality management and water quality monitoring. We identified India, which witnessed extreme weather events recorded on over 255 days in late 2024 and 64% of Indian companies actively deploying AI for sustainability, significantly above the global average of 44%. India represents a critical case for evaluating scalable, data-driven environmental solutions.
To start with, an exhaustive bibliometric analysis is being conducted and reported to identify key solutions and strategies for addressing and monitoring environmental degradation on a global scale, enabled by AI and big data adoption. Additionally, have tried to explore these solutions and strategies specifically from the perspective of India. The latest research database, such as Scopus, has been considered. Furthermore, we have examined the main challenges posed by environmental degradation, particularly concerning air and water quality, and how they are being addressed using AI and big data, comparing the global landscape with that of India. The selected keywords comprehensively represent the subject domain, ensuring broad and relevant coverage of the research area. This study systematically analyses publications from the last 25 years, identifying the most significant and highly relevant contributions in the field. Water pollution seems more serious than air pollution, as more papers are reported on it. The top three research contributors are China, the US, and the UK, in that order. Internationally, best practices include satellite-based AI platforms such as Climate TRACE (tracking 660 million polluters) and eDNA-integrated water assessments.
India, however, leads in hyperlocal innovations. The CHETNA project uses convolutional neural networks on Sentinel-2 imagery to map 44,000 brick kilns nationwide, a sector responsible for 15% of PM 2.5 and 6% of CO2 emissions. The UNDP Gurugram project integrates 50 IoT sensors with the VAYU AI platform. For water, an AI-driven DSS for the Ganga River employs LSTM networks to forecast pH, Dissolved Oxygen, and Biochemical Oxygen Demand up to 5 days in advance with 96.86% accuracy, and it incorporates an automated alert mechanism. When comparing Global with Indian challenges, we see that both contexts address data sparsity. Still, India uniquely prioritises low-cost sensor calibration, model interpretability, and resource-optimised alert systems, as demonstrated by the Ganga DSS. Indian research excels in hyperlocal source apportionment. While global water models achieve 94% accuracy, they lack the monsoon-adaptive interpretability and operational early warning infrastructure of India. There are innovative Indian solutions, such as biodegradable, sunlight-activated AI water and an Air quality filter. Two detailed case studies illustrate the implementation of an AI-driven vehicular emission intervention: an AI-Driven DSS platform integrating IoT sensors, satellite data, and hydrodynamic models to provide real-time, actionable insights into the River’s water quality. The analysis validates that these India-derived strategies align with advanced environmental AI practices already implemented in other high-density urban centres such as Hong Kong, where hyperlocal sensing, satellite-terrestrial integration, interpretable water-quality AI, predictive sewer maintenance, IoT-GIS multimodal fusion, and corporate sustainability mandates are well established.
Professor, Jinan University, China
Microbial degradation has gained significant attention as an environmentally friendly and cost-effective approach for pollution control and ecological remediation. However, traditional methods for screening functional microbes and investigating the molecular mechanisms of pollutant degradation often rely on phenotypic observation and empirical trial-and-error, which suffer from low throughput and lack molecular-level information. Due to the wide variety and structural complexity of organic pollutants, their degradation processes typically involve multi-enzyme synergistic reactions, participation of electron transfer chains, and dynamic reprogramming of metabolic pathways. To address these challenges, our group focused on typical emerging contaminants such as endocrine-disrupting chemicals, pesticides, and antibiotics. By integrating biological databases with bioinformatics and multi-omics technologies, a high-throughput screening strategy for identifying degrading microbes was developed. The regulatory relationships among key enzymes, coenzymes, electron transfer chains, and metabolites during pollutant degradation were systematically analyzed, revealing the molecular regulatory network that couples degradation enzyme catalysis with carbon metabolic flux. These findings provide scientific insights into the common metabolic pathways and regulatory principles underlying the biodegradation of typical organic pollutants, and also establish a theoretical and methodological foundation for the precision bioremediation of contaminated environments.
Associate Professor, Lingnan University, China
The growing availability of open government data (CSDI Portal), supported by Hong Kong's open data policy formalized in 2018, has extended the opportunities for environmental monitoring, risk assessment, and predictive analytics in urban settings. This presentation explores how open data can be integrated with Artificial Intelligence of Things (AIoT) and Geospatial Artificial Intelligence (GeoAI) to support near real-time urban environmental monitoring and sustainability assessment.
By combining static and dynamic datasets through APIs and applying geospatial analysis tools and GeoAI, these approaches enable more responsive, scalable, and evidence-based environmental intelligence. The presentation will showcase innovative projects that demonstrate how AIoT and GeoAI can be used to detect spatial patterns, identify environmental risks, and support decision-making related to urban resilience, public health, and sustainable city management.
The sharing will also highlight interactive map-based dashboards that communicate analytical outputs to stakeholders and the public in near real time. In addition, it will reflect on practical challenges in data integration, model development, interpretability, and implementation. Overall, the presentation illustrates how open data and geospatial technologies can support transparent, collaborative, and actionable approaches to urban environmental sustainability.
CEO/CTO, Bio-FD&C Plant Cell Research Institute, Korea
Plant cell culture is emerging as a next-generation cellular agriculture platform that bridges biodiversity conservation and sustainable bioresource production. BIO-FD&C, a KOSDAQ-listed Korean biotechnology company (KOSDAQ: 251120), has established an advanced plant cell platform based on proprietary cell line design and cultivation technologies, enabling the stable, scalable, and sustainable production of high-value biomaterials for cosmetic and food applications. Beyond industrial utilization, this platform offers a promising strategy for conserving endangered, endemic, and climate-sensitive plant resources while reducing pressure on natural habitats caused by overharvesting. Representative efforts include conservation-oriented studies on rare plant resources, including Arctic plants and the native flora of Ulleung Island and Jeju Island. Ongoing research in space farming further expands the future scope of plant cell-based biomanufacturing, highlighting plant cell culture as an innovative solution for resilient, low-impact, and sustainable bioresource systems.
Teaching Fellow, The Hong Kong Polytechnic University, China
Urban parks play an important role in supporting environmental sustainability in high-density cities by improving local thermal conditions, enhancing urban livability, and contributing to climate resilience. This presentation uses Hong Kong as a case study to explore how LiDAR and 3D models can help us better understand the environmental functions of urban parks. By combining field observation, UAV imagery, LiDAR-based reconstruction, and digital 3D modelling, the study compares three parks with different sizes, vegetation structures, and surrounding urban conditions. The findings show that urban parks can create measurable local cooling effects, although their performance varies according to park size, vegetation structure, and roadside exposure. Beyond environmental measurement, the presentation also highlights how LiDAR and 3D models make complex urban ecological patterns more visible, interpretable, and easier to communicate. In particular, immersive approaches such as AR and VR offer new ways to represent how green spaces function within dense city environments. The presentation argues that environmental sustainability is not only about increasing green space, but also about improving how we analyse, represent, and communicate the role of urban parks in practice.
Assistant Professor, Electronic Engineering and Computer Science, Hong Kong Metropolitan University
Associate Professor, School of Electronic, Information and Electrical Engineering, Shanghai Jiao Tong University
Courtesy Appointment, Oregon State University
Associate Professor, Director of Occupational and Environmental Institute, School of Public Health, Lanzhou University
Senior research fellow, Shenzhen Customs Information Center
Senior engineer, Shenzhen Langhe Tech Limited
Assistant Professor, University of Birmingham, U.K.
Chemical pollution presents a critical global challenge, with over 350,000 chemicals registered for use and thousands more entering the environment annually. Traditional reductionist models often fail to capture the reality of chemical mixtures and their non-linear interactions with shifting environmental factors, leaving significant gaps in regulatory knowledge and ecological protection.
Associate Professor, San Raffaele University of Rome, Italy
The transition toward environmentally sustainable food systems requires innovative strategies to reduce waste, optimize resource utilization, and generate value from underexploited agro-industrial side streams. Mass spectrometry–based omics technologies provide powerful analytical platforms to decode the complexity of food matrices and to guide sustainable processing strategies, effectively transforming food waste into high-value molecular insight.
This presentation introduces an integrated omics framework applied to marine and plant by-products. Case studies include the valorization of Atlantic herring side streams through high-resolution peptidomics, the ultrasound-assisted functionalization of soybean okara explored by combined peptidomic and metabolomic profiling, and the molecular characterization of thinned apple residues as underexploited agricultural resources. These examples demonstrate how untargeted LC–MS/MS workflows enable the identification of bioactive peptides and metabolites, revealing how green processing technologies reshape compositional patterns, enhance compound release, and unlock functional value within circular economy models.
Beyond compositional profiling, lipidomics emerges as a systems-level strategy to assess the metabolic impact of recovered bioactive compounds, emphasizing the shift from molecule-centered identification to pathway-driven biological interpretation. The integration of omics datasets with data-driven and artificial intelligence–assisted analytical approaches further enables the extraction of meaningful biological patterns from complex datasets, supporting predictive and systems-level interpretations.
High-resolution analytical platforms generate multidimensional datasets. By integrating omics-driven insights with advanced computational and AI-based approaches, analytical science acts as a bridge between food technology, environmental sustainability, and data-informed innovation, supporting smarter and more sustainable resource management across the food chain.
Professor, Federal University of Technology and Environmental Sciences, Nigeria
Environmental sustainability challenges, ranging from pollution and biodiversity loss to climate change and ecosystem degradation are increasingly complex, interconnected, and data-intensive. Addressing them requires more than traditional environmental monitoring; it demands intelligent systems capable of learning from vast, heterogeneous datasets and supporting evidence-based decision-making.
This presentation will explore how AI and Big Data analytics are reshaping environmental sustainability through integrated monitoring, predictive modelling, and intelligent decision systems. It will discuss how machine learning, remote sensing, IoT-enabled sensing platforms, and digital twin technologies provide real-time analysis of soil health, marine ecosystems, air quality, land-use change, and carbon dynamics.
Some specific case study problems in this domain such as predictive risk assessment, anomaly detection in pollution systems, climate impact forecasting, and optimization of restoration strategies will be highlighted, with particular attention on how multi-source environmental data can be fused into intelligent ecosystem decision platforms.
In essence, it will encapsulate how computational Intelligence complements ecological science with data-driven AI systems to support adaptive management, enhance resilience, and advance the Sustainable Development Goals.
Prorector for AI; Associate Professor, University of Insubria, Italy
Artificial Intelligence and Big Data are rapidly transforming how humanity understands and addresses environmental sustainability challenges. This keynote explores the evolving role of data-driven intelligence in supporting climate action, natural resource management, and ecosystem preservation. By integrating massive heterogeneous datasets—from satellite observations and IoT environmental sensors to citizen science and open data platforms—AI is enabling unprecedented capabilities in environmental monitoring, prediction, and decision-making.
The talk provides a structured overview of the most influential AI methodologies and technical infrastructure currently applied to sustainability domains, including machine learning for weather modeling, deep learning for environmental pattern recognition, and advanced analytics for resource optimization. Through selected real-world case studies, the keynote illustrates how these technologies are being successfully deployed across multiple sectors, with particular emphasis on agriculture and food systems. Examples include precision agriculture enabled by remote sensing and predictive analytics, AI-driven crop monitoring and yield optimization, smart irrigation and soil health management, food quality evaluation, and nowcasting through intelligent forecasting.
Senior Lecturer, University of Toronto, Canada
This study proposes a Behavioral Nudge AI system designed to embed sustainability into the daily workflows of traders and salespeople in capital markets. By analyzing client communications, market context, and product inventory, the system delivers real-time, contextual suggestions, such as highlighting a transition fund to a client with a renewed net-zero commitment, without mandating action. The concept operationalizes sustainability at the point of distribution, accelerating the adoption of sustainable products while preserving professional autonomy. Two hypotheses are proposed for empirical testing: first, that the system will measurably increase both sustainability-focused client conversations and transaction conversions without adding to client contact time; and second, that its long-term effectiveness depends on user perceptions of nudge relevance and autonomy preservation, with overly prescriptive or irrelevant systems risking friction and fatigue. This research offers a path for financial institutions to move from sustainability aspiration to frontline execution in a risk-aware, data-driven manner.
