An Operational Deep Learning System for Satellite-Based High-Resolution Global Nowcasting

2025-10-14

AI Summary

Traditional numerical weather prediction (NWP) suffers from high latency, coarse spatiotemporal resolution, and globally uneven accuracy for precipitation nowcasting (0–12 hours); meanwhile, existing machine learning approaches rely heavily on dense radar observations, limiting applicability in radar-sparse regions—particularly across the Global South. To address these limitations, we propose Global MetNet: the first end-to-end deep learning nowcasting system that fuses multi-source satellite data—including geostationary meteorological imagery and the GPM CORRA precipitation product—with global NWP inputs. Its key innovation is the first satellite-driven, globally scalable precipitation forecasting system delivering minute-level updates at high spatiotemporal resolution (~1 km, 2-min intervals). Experiments demonstrate substantial improvements over state-of-the-art hourly forecasting systems across critical metrics—including Critical Success Index (CSI) and Fractions Skill Score (FSS)—with particularly pronounced gains in radar-absent regions, where it outperforms high-resolution NWP models from Europe and North America. Global MetNet is now operational in Google Search, serving millions of users worldwide.

Technology Category

• Machine Learning: Scalability of ML Systems
• Search and Optimization: Metareasoning and Metaheuristics
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Precipitation nowcasting, which predicts rainfall up to a few hours ahead, is a critical tool for vulnerable communities in the Global South frequently exposed to intense, rapidly developing storms. Timely forecasts provide a crucial window to protect lives and livelihoods. Traditional numerical weather prediction (NWP) methods suffer from high latency, low spatial and temporal resolution, and significant gaps in accuracy across the world. Recent machine learning-based nowcasting methods, common in the Global North, cannot be extended to the Global South due to extremely sparse radar coverage. We present Global MetNet, an operational global machine learning nowcasting model. It leverages the Global Precipitation Mission's CORRA dataset, geostationary satellite data, and global NWP data to predict precipitation for the next 12 hours. The model operates at a high resolution of approximately 0.05° (~5km) spatially and 15 minutes temporally. Global MetNet significantly outperforms industry-standard hourly forecasts and achieves significantly higher skill, making forecasts useful over a much larger area of the world than previously available. Our model demonstrates better skill in data-sparse regions than even the best high-resolution NWP models achieve in the US. Validated using ground radar and satellite data, it shows significant improvements across key metrics like the critical success index and fractions skill score for all precipitation rates and lead times. Crucially, our model generates forecasts in under a minute, making it readily deployable for real-time applications. It is already deployed for millions of users on Google Search. This work represents a key step in reducing global disparities in forecast quality and integrating sparse, high-resolution satellite observations into weather forecasting.

Problem

• Addresses high latency and low resolution of traditional weather prediction
• Overcomes sparse radar coverage limitations in Global South forecasting
• Provides real-time global precipitation nowcasting with satellite data

Innovation

• Uses deep learning for global precipitation nowcasting
• Leverages satellite and global NWP data inputs
• Generates high-resolution forecasts in under a minute