Digital twin for water management handed over for trial in the Limpopo River Basin

The International Water Management Institute (IWMI) and the Limpopo Watercourse Commission (LIMCOM) have taken a significant step toward enhancing water security in the Limpopo River Basin through a recent demonstration of the Limpopo River Basin Digital Twin. The tool, developed under the CGIAR Initiative on Digital Innovation, was officially handed over to LIMCOM following the signing of a Memorandum of Understanding (MoU) between IWMI and LIMCOM, further solidifying their partnership for sustainable water management.

The historic MoU was signed on 30 October 2024 on the sidelines of the 25th WaterNet/WARFSA/GWPSA Symposium, where a dedicated session was hosted by CGIAR Digital Innovation and the DIWASA project to showcase the functionality of the Digital Twin prototype and how it can transform water management in southern Africa.

During the demonstration, Digital Twin project Angie Garcia Andarcia (IWMI), presented the digital twin alongside a suite of tools and supporting documents that contributed to the prototype that was handed over to LIMCOM. It is anticipated that LIMCOM will trial the Digital Twin for water management while IWMI undertakes further developments.

• Digital Twin Portal: Central hub for accessing and interacting with the digital twin model for the Limpopo River Basin, featuring multiple water resource models, data and tools for simplified interpretation. The foundation of the digital twin is a SWAT hydrological model running in near real time, including a three-month seasonal forecast. A total of 1,408 channels covering 400,000 km² are available for seasonal forecasting, including 305 discharge stations provided by South Africa's Department of Water and Sanitation. The platform also features 303 rainfall stations based on CHIRPS data and incorporates three types of forecasts from ECMWF, supported by a 20-year historical database.
• Visualization application: An interactive tool for visualizing datasets through graphics and maps, facilitating exploration of spatial and temporal trends in water resources.
• River discharge & e-flow Charts: Specialized charts that display natural and present-day river discharge and environmental flows, enabling analysis of the water distribution required to maintain river ecosystems and the beneficiaries of those ecosystems and also providing users with the available supply of water from every river reach. .
• Reservoir volume monitoring: Machine learning-based tool for monitoring reservoir volume, providing near real-time insights into water storage and availability across different reservoirs across the Limpopo Basin. The tool provides daily updates on 1,424 waterbodies and 96 dams.
• Reservoir forecasting: A predictive tool that forecasts future reservoir levels using an experimental machine learning approach, currently available for Middelburg Dam and Loskop Dam. This tool aids in planning for potential water shortages or surpluses, supporting proactive water management strategies.

Environmental Framework Assessment Tool to assess social and environmental risks of water management decisions

Multiple new innovations were developed for the Digital Twin, including:

• A user-friendly Environmental Framework Assessment (EFA) Tool that enables stakeholders to test multiple river management scenarios and assess social and environmental risk outcomes without extensive expertise.
• FISHTRAC, a real-time fish tracking solution where sensors are attached to fish where they gather real-time data on river health and water quality. This data is used to trigger alerts in the Digital Twin to alert managers to environmental problems, particularly related to water pollution, in real time.
• Drone solutions to map river bodies for environmental flow (eflow) assessments. These solutions used high-resolution Lidar and multi-spectral imaging to map the surface of the river while submergible drones were used to map beneath the river surface.
• Virtual Reality Tour: The project produced an immersive VR experience presently only at a single river site on the Olifants River (Balule), allowing users to explore and gain a spatial understanding of key water resources of the area. This tour was developed under the CGIAR Initiative on Diversification in East and Southern Africa in collaboration with CGIAR Digital Innovation.

A user undergoing a experience of the Balule river site at theWaterNet conference. Credit: IWMI

AI for inclusive water management: the Digital Twin virtual assistant

As a core feature of the Digital Twin interface, an AI-based virtual assistant was co-designed with Microsoft Research under the Farm Vibes project to provide water managers, researchers and citizens with guidance on efficient water practices in the basin. This virtual assistant, still in the testing phase and powered by GPT-4, allows users to interact with complex datasets through a natural language chat interface.

By facilitating natural language queries, the assistant enhances accessibility and usability, transforming raw data into actionable information. It can help users draft reports, plot maps and explore various scenarios, making it a crucial tool for anticipating risks and strategizing in developing climate-resilient water management solutions that will improve the sustainable use of natural resources in the basin.

"This chatbot is a first for water resources management, probably anywhere in the world," said Chris Dickens, researher at IWMI. "It is a paradigm shift in what is possible for water resources management, where AI can assist both skilled and unskilled users to understand in their own language, the intricacies of water resources management in a way that is practical and useful to them. This chatbot brings the most complex data into the hands of any user."

Mark Graham from GroundTruth presenting the results of work to empower citizen scientists with digital tools. Credit: IWMI

Amplifying community voices through citizen science

During the WaterNet session, GroundTruth showcased the impact of digital tools developed in partnership with CGIAR Digital Innovation. The session highlighted the importance of partnerships at the civic level, emphasizing citizen science, capacity sharing and youth involvement. The Enviro-Champs group, comprising approximately 1,000 citizen scientists in KwaZulu-Natal province, South Africa, has collected over 1,200 water quality measurements using custom data collection apps such as MiniSASS. Through a partnership with the YOMA-UNICEF platform, young people are receiving training in citizen science methods that is recognized in the form of an online, block-chain based CV along with other incentives.

The goal is to leverage the learnings from this experience and apply them to the Limpopo River Basin, using these tools to submit data that will be accessible to decision-makers through the digital twin. This will be made possible with the support of the Enabel project in 2025. By integrating citizen science data, the digital twin empowers local communities to contribute valuable information, bridging data gaps and enhancing water management insights across the region.

Driving change through multi-stakeholder collaboration

The Digital Twin project embodies a multistakeholder co-design approach., collaborating with local agencies and leveraging country-specific resources to ensure adaptability, scalability, and alignment with regional needs. The digital twin project brought together over 20 experts, including researchers, hydrologists, environmental flow specialists, remote sensing and IoT experts, software developers, data scientists, data engineers, modelers, and IT professionals. Key contributors included GroundTruth, Hidromod, Rivers of Life, AWARD, YOMA-UNICEF, Pegasys Global Consulting, Central University of Technology, Free State, and technological partners such as Microsoft Research (Farm Vibes project) and Amazon Web Services. Regional support from LIMCOM, which integrated the digital twin into its strategy, was pivotal. This collaboration was part of the "Integrated Transboundary River Basin Management for the Sustainable Development of the Limpopo River Basin" (LIMCOM-UNDP/GEF) project, culminating in the first digital twin for river basin management in the Limpopo basin

Democratizing FAIR DATA Interoperability across the river basin

The Open Data Cube (ODC) is a robust open-source software package that supports the digital twin by providing structured access to satellite and geospatial data for analysis. It is built on the foundation of Digital Earth Africa under a cloud-based AWS architecture, adhering to FAIR data principles (findability, accessibility, interoperability, and reuse of digital assets) within the ODC for the digital twin. Within the ODC for the digital twin project, seven products have been incorporated, with two key products highlighted:

• Drought Monitoring: The Digital Twin Drought Monitoring Tool provides real-time analysis of drought conditions using key indexes such as the Standardized Precipitation Index (SPI), which measures rainfall deficiency over various time scales, and the Vegetation Condition Index (VCI), which assesses vegetation health relative to long-term observations. These insights help decision-makers identify drought severity and plan effective responses.
• Irrigating mapping areas: The Irrigated Mapping Tool within the digital twin project uses advanced machine learning algorithms to identify and map irrigated areas across the Limpopo River Basin. By analyzing satellite imagery and integrating various data points, the tool can accurately differentiate between irrigated and non-irrigated land. This approach enhances the precision of land-use classification, enabling stakeholders to monitor irrigation patterns, optimize water allocation, and plan for sustainable agricultural practices.

Henry Roman, Country Representative, South Africa, IWMI (left) and Sergio Sitoe, Executive Secretary, LIMCOM (right)

The future for water management in the Limpopo River Basin

The collaboration to create the Limpopo River Basin Digital Twin, extended through the MoU signed between IWMI and LIMCOM, represents a paradigm shift in water management across the Limpopo River Basin and potentially a model for other contexts. On the technological side, it was shown that the integration of AI and real-time data can create powerful platforms for decision-makers, researchers, and communities to better understand water dynamics, plan for future scenarios, and adapt to changing conditions. The result is to break down the divide between complex data models and practical user-friendly management of water resources, making water management more responsive and inclusive.

"The development of the digital twin prototype for the Limpopo Basin, running alongside LIMCOM's own management information systems provides a big leap forward for the basin and the region, to allow innovative management of river basin resources, testing and making use of new technologies at the cutting edge," said Eddie Riddell, Regional Coordinator, LIMCOM UNDP-GEF7 project

***

The LIMCOM-UNDP/GEF project, implemented by LIMCOM in partnership with the Global Water Partnership Southern Africa (GWPSA), supported by UNDP and funded through the Global Environment Facility (GEF), aims to uplift the living standards of the basin's population and conserve its resources and ecosystem services through various community level interventions. Leveraging CGIAR's digital innovation, Digital Earth Africa and IWMI's extensive experience in water research and near real-time basin hydrological modeling, supported by The Leona M. and Harry B. Helmsley Charitable Trust, we aims to harness state-of-the-art technologies to enhance water management in the Limpopo River Basin in a collaborative effort of actions.

If you are interested to learn more about the project: [email protected]

Public link

Please use the above public link if you want to share this noodl on another website.