31-03-2025
G42 and NVIDIA collaborate to introduce a hyper-local weather forecasting model
Abu Dhabi's G42 is exploring the expansion of a hyper-local AI-powered weather forecasting system that it developed in collaboration with NVIDIA.
The generative downscaling model offers faster, more cost-effective simulations, as well as the ability to predict extreme events by learning the small-scale physics of the region.
G42 successfully carried out an end-to-end fog simulation across the UAE, with the progression of fog across the country from morning to evening within a single day, showcasing the platform's advanced visualisation capabilities and its integration of real, high-resolution fog forecast data.
G42 and NVIDIA AI-powered weather forecasting
This breakthrough enables enhanced fog forecasts, impacting the productivity and safety of people across sectors like transportation, logistics, public safety, insurance, renewable energy, and utilities. By improving safety, efficiency, and risk management capabilities, the model has achieved enhancements in speed and cost-effectiveness compared to traditional forecasting systems.
The model was built by the Earth-2 Climate Tech Lab in Abu Dhabi, which was established as part of G42's partnership with NVIDIA.
Inception, a G42 company, customised CorrDiff (part of the NVIDIA's Earth-2 platform), to provide detailed 200-meter resolution forecasts for Abu Dhabi and other urban areas. The much-improved level of granularity sets a benchmark in micro-weather simulations and high-resolution forecasting tailored to specific urban environments.
This enables governments, industries, and communities to anticipate the impact of extreme weather with timely warnings and thus initiate effective response measures. Core42, G42's digital infrastructure company, will host the platform and provide accelerated compute and inference services leveraging NVIDIA hardware.
Andrew Jackson, Chief Executive Officer of Inception, commented: 'For AI to be truly transformative, it should be an accessible tool for governments and industries worldwide. Through our collaboration with NVIDIA, we are bringing cutting-edge forecasting capabilities within reach of those who need them most.
'Because CorrDiff is designed to adapt to local weather behaviours, this technology is not only improving forecasting for the UAE but can also be tailored for regions worldwide facing climate volatility. The ability to generate hyper-local predictions gives decision-makers the confidence to act faster, plan better, and build more resilient systems.
'Beyond its impact on weather forecasting, this breakthrough has far-reaching applications in aviation, urban mobility, energy grid optimisation, and environmental planning. Hyper-local insights can reduce flight delays, improve road safety, optimise renewable energy distribution, and support climate-adaptive urban development.'
Enhancing safety with AI
G42 is now planning to take the technology beyond the UAE, particularly in climate-vulnerable regions across the Global South, such as Africa, South Asia, and Southeast Asia, where accurate forecasting can be a critical tool for disaster preparedness and infrastructure resilience.
Dr Abdulla Al Mandous, Director General of the National Center of Meteorology UAE, added: 'AI-powered weather forecasting has the potential to revolutionise high-quality, high-resolution weather and disaster management solutions, particularly in this accelerating phase of climate change.
'By enhancing prediction accuracy and enabling hyper-local, real-time forecasts, this technology empowers better decision-making and strengthens resilience against climate challenges. We are excited to see how these innovations will shape the future of sustainability and drive more effective, data-driven solutions.'
Dion Harris, Senior Director of HPC and AI Factory Solutions at NVIDIA, said: 'Hyper-local weather forecasting is becoming crucial to determining patterns and people positioning as climates around the world undergo change. G42 is innovating using the NVIDIA Earth-2 platform to drive access to actionable information in areas that need it most.
'Weather forecasting has always required significant computational power, but AI is redefining what's possible. With physics-informed neural networks and high-resolution climate modelling, we can now generate faster, more detailed, and more adaptive predictions. The scalability of accelerated computing allows us to process vast amounts of data efficiently, helping industries and governments better understand and respond to extreme weather patterns.'
