NUS Mechanical Engineering Star Gianmarco Mengaldo's Global Recognition

National University of Singapore (NUS) Assistant Professor Gianmarco Mengaldo has earned a prestigious spot on the World Meteorological Organization's (WMO) Joint Advisory Group on Artificial Intelligence (JAG-AI). This appointment underscores the growing influence of Singapore's higher education institutions in pioneering artificial intelligence applications for critical global challenges like weather prediction and climate modeling. Mengaldo, based in the Department of Mechanical Engineering within NUS's College of Design and Engineering (CDE), brings his cutting-edge expertise in mathematical engineering and explainable AI to the table, positioning NUS as a key player in international meteorological advancements.

Mengaldo's selection highlights how Singapore universities are fostering talent at the intersection of AI and environmental sciences. His work focuses on developing robust data-driven tools that enhance simulations of complex dynamical systems, a vital area as climate variability intensifies worldwide. This milestone not only elevates his profile but also amplifies NUS's contributions to sustainable technologies through interdisciplinary research.

From Imperial College to NUS: Mengaldo's Academic Journey

Gianmarco Mengaldo's path to this appointment began with a PhD from Imperial College London, where he delved into computational mathematics and high-order numerical methods for fluid dynamics. Joining NUS in 2020 as an Assistant Professor, he quickly established himself as a leader in mathematical eXplainable AI (XAI), particularly for scientific applications. His dual appointment in Mechanical Engineering and courtesy in the Department of Mathematics reflects NUS's commitment to cross-disciplinary collaboration.

At NUS, Mengaldo leads the MathEXLab, a hub where researchers blend rigorous mathematics with machine learning to tackle real-world problems. The lab's projects span weather forecasting, climate simulations, and beyond, emphasizing interpretable AI models that maintain physical consistency. This approach addresses a core challenge in AI: ensuring predictions are not just accurate but trustworthy and explainable, crucial for high-stakes fields like meteorology.

Mengaldo's publication record, with over 3,300 citations, includes breakthroughs in hybrid physics-AI models for Earth system digital twins. These innovations enable stable long-term climate projections, a feat traditional models struggle with due to error accumulation over time. His progression from European academia to Singapore exemplifies how NUS attracts global talent to bolster Asia's research ecosystem.

Key Milestones in Mengaldo's Research Career

• PhD at Imperial College London on advanced numerical simulations for aerodynamics.
• Postdoctoral roles advancing AI for geophysical flows.
• NUS tenure since 2020, founding MathEXLab with focus on XAI for dynamical systems.
• Recent publications on AI-stabilized climate emulators in top journals like Nature Communications.

Understanding the WMO Joint Advisory Group on AI

The WMO, a specialized UN agency coordinating global efforts on weather, climate, and water, launched JAG-AI in 2025 to harness artificial intelligence responsibly. Comprising experts from academia, industry, and public sectors, the group advises on integrating AI into meteorological infrastructures. Objectives include accelerating AI adoption for precise forecasting, ethical data use, and bridging gaps between physics-based and data-driven models.

JAG-AI reports to WMO's Infrastructure Commission and Research Board, focusing on standards for AI trustworthiness, interoperability of models, and capacity building in developing nations. Mengaldo's role involves shaping guidelines for AI-enhanced early warning systems, vital as extreme weather events rise. For instance, AI can now predict sub-seasonal patterns with unprecedented speed, aiding disaster preparedness.

This group emerges amid rapid AI progress in weather tech, like Google's GraphCast and ECMWF's AIFS, which rival traditional supercomputer simulations. Yet, challenges persist: AI models often falter in rare events or data-scarce regions like the tropics, where Singapore's expertise shines. Mengaldo's involvement ensures Asian perspectives, especially tropical dynamics, inform global standards. Learn more about WMO's AI strategy here.

Mengaldo's Expertise Aligns Perfectly with WMO Priorities

Mengaldo's research excels in hybrid models fusing physics equations with neural networks, preventing drift in extended forecasts. A recent NUS project led by him developed an AI technique maintaining simulation stability over decades, outperforming conventional methods by preserving energy conservation and physical laws. This is pivotal for climate scenarios under IPCC pathways, where long-horizon accuracy informs policy.

In weather contexts, his work on explainable emulators reveals why AI makes certain predictions, demystifying black-box limitations. For Singapore, prone to thunderstorms and haze, such tools could refine nowcasting. MathEXLab collaborations with international partners, including ECMWF, have already yielded open-source tools like PyTorch-based climate emulators, democratizing access.

Statistics underscore the impact: AI weather models now achieve 90-95% accuracy for medium-range forecasts, up from 80% in physics-only systems, per recent benchmarks. Mengaldo's contributions position NUS researchers to influence WMO's push for equitable AI deployment globally.

Singapore's Higher Education Leading AI-Climate Nexus

NUS's appointment of Mengaldo mirrors broader Singapore university thrusts in AI for sustainability. The Climate and Weather Research Alliance Singapore (CAWRAS), uniting NUS, NTU Singapore, A*STAR, and NEA, channels $25 million via the Weather Science Research Programme (WSRP). Launched in 2025, it targets tropical urban weather prediction using AI foundation models and urban climatology.

NUS strengths in hydroclimatology and AI modeling complement NTU's Earth Observatory collaborations. For students, this means enriched curricula: NUS offers modules in computational climate science, attracting global talent. Singapore's National AI Strategy 2.0 allocates resources for 100,000 AI upskilled workers by 2025, with universities central.

Details on CAWRAS initiatives are available here, showcasing multi-institutional synergy.

Benefits of National Alliances for Singapore Universities

• Shared supercomputing for AI training on tropical datasets.
• Interdisciplinary PhD programs blending AI, engineering, and earth sciences.
• Industry partnerships for real-time weather apps in smart nation initiatives.
• Global visibility elevating Singapore HE rankings.

AI Revolutionizing Meteorology: Opportunities and Hurdles

AI transforms meteorology by processing petabytes of satellite and radar data instantaneously. Models like FourCastNet, inspired by Mengaldo's methods, forecast 15-day weather at 20km resolution in minutes. In climate, AI downscales global models to city scales, predicting heatwaves with 85% skill.

For Singapore, AI aids haze forecasting from Indonesian fires and urban flooding risks. NUS research integrates ensemble learning for uncertainty quantification, essential for aviation and agriculture. Challenges include data biases in underrepresented tropics and ensuring model generalizability across seasons.

Mengaldo advocates hybrid approaches: AI learns physics constraints, reducing hallucinations. This aligns with WMO's ethical AI framework, emphasizing transparency and inclusivity.

Implications for NUS Students and Researchers

Mengaldo's role opens doors for NUS students. MathEXLab hosts undergraduates in AI-weather projects, many publishing in NeurIPS Climate tracks. Graduate programs in Mechanical Engineering now emphasize XAI electives, preparing for roles in MSS or private firms like IBM Weather.

Singapore's 30% R&D GDP target by 2030 fuels HE investments: NUS received $60M for marine-climate centers. Peers like NTU's AI Research Centre develop similar tools, fostering competition and collaboration. Students gain from hackathons simulating WMO scenarios, building portfolios for global jobs.

Future Outlook: Singapore's Role in Global AI-Meteorology

As climate threats escalate—IPCC projects 2-4x tropical cyclones by 2100—AI's role amplifies. Mengaldo's JAG-AI tenure could steer WMO towards tropical-focused benchmarks, benefiting SEA nations. NUS aims to lead AI4Climate hubs, with spin-offs commercializing emulators.

Challenges: talent retention amid US/China poaching, ethical AI governance. Solutions lie in Singapore's agile policies, like AI Verify framework. For higher ed, this means curricula evolving faster, with 40% AI-embedded courses by 2030 at NTU/NUS.

Stakeholders—from policymakers to farmers—stand to gain from precise predictions, underscoring HE's societal impact. Explore NUS's full announcement here.

Photo by Musemind UX Agency on Unsplash

Broadening Horizons: Collaborations and Actionable Insights

NUS's global ties, via Mengaldo's ECMWF links, enable student exchanges and joint grants. Researchers can apply hybrid AI to local issues like sea-level rise modeling for Tuas mega-port. Actionable steps: enroll in NUS XAI modules, join MathEXLab internships, contribute to open-source climate repos.

Singapore HE's edge: multicultural teams yielding robust models. Future: quantum-AI hybrids for exascale simulations. This appointment signals NUS's ascent in AI-driven sciences, inspiring next-gen scholars.