Introduction Plastic pollution in the Pasig River is not only a waste management problem but also an information and coordination challenge. Multiple government agencies contribute expertise and carry out complementary responsibilities across the river system. However, because each organization focuses on different aspects of river management, developing a comprehensive understanding of pollution sources, movement patterns, and impacts requires integrating diverse datasets and perspectives. Responsibility for the river is shared among national, regional, and local government bodies, each with its own mandate, data resources, and operational systems. While coordination among these agencies is well established, the information that supports planning and decision-making is often maintained on separate platforms. Since the river functions as a single interconnected system, creating a shared, integrated view of river conditions and pollution dynamics is valuable. A digital twin can provide this common operating picture, enabling agencies to align interventions, set priorities, and make more informed decisions across jurisdictions. Stretching 27 kilometers from Laguna de Bay to Manila Bay, the Pasig River drains one of the Philippines’ most densely populated urban areas. Metro Manila generates thousands of tons of waste each day, exceeding landfill capacity, and some of this waste leaks into waterways. The river is estimated to carry 63,000 tons of plastic into the ocean annually. To help address this challenge, the Asian Development Bank (ADB) launched an innovation challenge with support from Japan’s High-Level Technology Fund, through the regional technical assistance project Promoting Action on Plastic Pollution from Source to Sea in Asia and the Pacific, and the ADB–Korea Climate Technology Hub. Arup and RiverRecycle were selected through the challenge. Working with government partners, ADB and Arup then developed a digital twin prototype for the Pasig River—a virtual representation that integrates satellite imagery, hydrological and tidal data, infrastructure information, and waste surveys to support planning and coordinated decision-making. Why has plastic waste in the Pasig River been so difficult to manage? Plastic waste continues to persist in the Pasig River because of dense urbanization, complex hydrodynamics, and the vast, interconnected nature of a river system that crosses many jurisdictions. More than 48 tributaries carry waste from across Metro Manila into the river, while rainstorms wash garbage from streets, markets, and informal settlements that lack adequate waste and sanitation services. Cleanup operations are further complicated by seasonal tidal reversals, which cause waste to circulate through tributaries and esteros before flowing back toward Laguna de Bay. The scale and composition of the problem are equally challenging. An AI analysis of a one-month camera survey at Buhangin Creek, a major tributary in the San Juan River system, detected more than 300,000 pieces of plastic. More than half consisted of film plastics, single-use plastics, and foam materials. Alongside the physical challenge is an informational one. Much of the data already exists, generated and maintained by agencies for their own operations. The Department of Environment and Natural Resources (DENR) tracks estero interventions and manages trash traps. The Metro Manila Development Authority (MMDA) oversees cleanup operations and pumping stations. The Philippine Atmospheric, Geophysical and Astronomical Services Administration (PAGASA) monitors rainfall and water levels. Local governments maintain waste management records. Coordination mechanisms across these agencies are in place. DENR works with local governments through regular meetings on 10-year solid waste management plans, participation in local boards, joint monitoring of facilities, and circular economy programs in Pasig, Caloocan, and Quezon City. Its Environmental Management Bureau has installed nearly 200 trash traps across Metro Manila. These efforts align with national policy. DENR leads implementation of the Ecological Solid Waste Management Act, strengthened by the Extended Producer Responsibility Act of 2022, which requires large enterprises to recover plastic packaging waste. It also drives the National Plan of Action on Marine Litter, targeting zero waste to Philippine waters by 2040. What remains missing is a single technical platform to consolidate datasets. Without it, building a system-wide picture—where plastic originates, where it accumulates, and which interventions are most effective—requires repeated, ad hoc coordination across agencies. How does a digital twin give government agencies a shared view of the river? Answering these questions required giving agencies a common picture of the river. A digital twin is a virtual representation of the Pasig River that combines live and historical data into a single view, allowing agencies to see the river as a whole system alongside the data they maintain for their own mandates. Built on the open-source TerriaJS platform, it integrates datasets from different agencies while using tiered access to let them contribute data without relinquishing control. The platform was co-designed with its users, starting from the decisions they needed to make rather than the data that happened to be available. How does the platform turn scattered data into decisions agencies can act on? At its core, the platform brings together data from multiple sources into a single, navigable catalogue. More than 40 datasets are available for public access, with about 200 more in development. It presents information in 2D, 3D, and 4D while connecting to real-time feeds such as PAGASA rainfall and water-level data. The value lies not in the data itself but in what it allows planners to see. A hydraulic model predicts water levels and river discharge from rainfall and tidal conditions, generating velocity heat maps that help agencies identify strategic locations for trash traps because faster flows carry more waste. Satellite layers complement these insights by tracking turbidity, floating debris, and plastic accumulation along the river. Screenshot of the open-source, TerriaJS-based visualization platform showing how river waste hotspots overlap with flood-prone areas, giving planners a common view for coordinated decision-making.Source: Authors. At the end of the six-month proof of concept, the codebase was handed over to DENR for long-term hosting, future enhancement, and scaling. Moreover, the project delivered a digital twin ecosystem built on stakeholder collaboration, shared knowledge, and real-world operational insights. How are artificial intelligence and machine learning being used to tackle plastic waste along the river? Time-lapse cameras installed along tributary rivers captured months of footage, providing the data needed to train a plastic-detection application. The team manually tagged thousands of images to train an AI model that classifies three types of waste: rigid plastics, flexible film or soft plastics, and foam. The model also estimates the speed and direction of floating debris. Early testing achieved an accuracy of 60%–70%, demonstrating promising potential while highlighting the need for further refinement. The digital twin’s AI module detects and classifies river plastic into hard, soft, and foam types, which can inform measures for upstream waste reduction. Source: Authors. Designed for non-technical users, the platform integrates these outputs into the Digital Twin's Plastic Waste Module, allowing everyone from city officials to barangay waste collectors to upload footage and generate classified reports. A future phase will combine this capability with the Digital Twin's Hydrology/Hydraulic Module to predict where plastic is likely to accumulate under different conditions, helping agencies intervene before it spreads. What insights did the prototype reveal about plastic leakage pathways and pollution hotspots? Estero de Tutuban had the highest proportion of plastic waste, accounting for about 53% of sampled waste. Its mix of dense housing, commercial activity, and small-scale industry illustrates how difficult plastic waste is to manage in high-density urban areas. The prototype identified four leakage pathways: urban stormwater runoff, street-sweeping discharge, illegal dumping into esteros, and industrial wastewater discharge. Each requires a different response, ranging from community enforcement to regulatory and policy measures. Single-use plastics dominated the waste stream, particularly sachets for everyday goods. This points to an upstream challenge: plastic pollution is as much a packaging and production issue as it is a waste management problem. A more detailed hydraulic model is needed to better understand how floating plastics move with changing water velocities. Consultations also identified a practical constraint: trash traps cannot be placed where currents are strongest because they may be damaged and allow plastic to continue downstream. Future work should therefore consider siting and design together. A community survey reinforced these findings. Ninety percent of residents identified garbage as a serious problem, while 75% described nearby esteros as dirty. The results highlight the importance of sustained service delivery on the ground—support that data and digital tools can enhance but never replace. How can Philippine local governments and agencies use these insights to design more effective interventions? The prototype gives Philippine agencies a common platform for planning and operational decisions. It helps identify where waste is most likely to accumulate, supporting better decisions on the placement of floodgates, drainage systems, flood management measures, and waste collection infrastructure. A shared platform for river intelligence strengthens coordination across agencies and promotes greater public accountability. More than a visualization tool, it provides a foundation for integrated urban environmental management. In practice, the platform can help agencies complement cleanup operations with more anticipatory and better-targeted action. It enables them to prioritize investments in waste collection, public education, infrastructure, and in-river cleanup, while informing local development, climate, and disaster risk management plans. The platform also has clear limits. Data alone cannot replace service delivery, enforcement, and accountability on the ground, but it can help make those interventions better designed and better targeted. Why did this approach work, and what made Philippine government agencies willing to share their data? It started with the use case, not the technology. The platform was designed around the decisions users needed to make rather than the data that happened to be available. It treated engagement as collaboration. More than 18 government agencies were involved in co-defining, co-designing, and co-developing the platform, combining top-down mandates with bottom-up input from the people who collect waste. It addressed the real barriers to data sharing: transparency and trust. Sharing data across institutions requires clarity on how information will be stored, used, and attributed. Tiered access, data-sharing agreements, and clear lines of accountability provided that assurance. It started small and remained open. A lightweight, open-source, and modular prototype allowed agencies to see value quickly and adopt the features that mattered most. Handing the codebase to the DENR under an open license also strengthened local ownership. What can other cities and countries learn from the Pasig River experience? Start with a “digital twin lite.” A low-cost prototype built in weeks to months lets governments understand the platform’s value before scaling. Put governance before technology. Data-sharing agreements, tiered access, and quality controls are what make a shared platform effective. Without them, even the best technology will struggle. Build readiness in steps. Moving from paper records to digitized information, then to connected systems, and finally to digital transformation is a gradual process rather than a single leap. Design for adaptability and replication. A modular, open-source architecture allows the platform to be adapted and scaled across different urban waterways and cities instead of being tied to a single vendor. Ask the Experts Christine Po King Chan Principal Urban Development Specialist (Finance and Investment), Sectors Department 3, Asian Development Bank Christine Po King Chan has more than 24 years of international experience in structured and project finance, private equity, capital markets, and sustainable investing. She serves as Principal Urban Development Specialist at ADB, leading private capital structuring for urban and water infrastructure. She previously served as Senior Advisor to ADB’s Vice President (Market Solutions), providing guidance on strategic and transactional matters. Before joining ADB, she held investment roles in New York, Hong Kong, and Manila. She holds an MBA from the Wharton School and a BA from UC Berkeley. Rudolph Peralta Digital Automation Leader, Arup Manila Rudolph Peralta is a geomatics engineer at Arup with more than 12 years of experience in geographic information systems (GIS), remote sensing, LiDAR, and geospatial analytics. He specializes in applying digital technologies and spatial data to climate resilience, environmental management, and urban planning. His work focuses on developing data-driven solutions that support evidence-based decision-making for governments and development partners. Christian Rieza Knowledge Management Specialist (Consultant), Asian Development Bank Christian Rieza is a development communication and knowledge management consultant with experience in climate change, governance, urban resilience, and public health in Asia and the Pacific. He supports knowledge management, communications, and coordination for water and urban development initiatives at the Asian Development Bank, including the Rejuvenating Pasig River for a Livable Manila initiative. He previously supported COVID-19 response efforts with the Philippine Department of Health and USAID. He holds a BS in Development Communication, cum laude, from the University of the Philippines. Leave your question or comment in the section below: View the discussion thread.