WMO is making a strategic shift to integrate Artificial Intelligence (AI) to advance Earth system science.
WMO is making a strategic shift to integrate Artificial Intelligence (AI) to advance Earth system science in accordance with a decision of the World Meteorological Congress in 2023. This move aims to improve the understanding of Earth systems towards offering better environmental services globally. The Organization and its partners have already started exploring how AI can enhance early warning systems and foster collaboration across sectors. The partners in the Early Warnings for All initiative are also seeking to integrate AI to improve disaster management and climate adaptation. This paradigm shift in the WMO approach is apparent in many areas.
WMO participates in a United Nations focus group that standardizes data collection, storage, processing, and AI model development. It will also contribute to a forthcoming meeting at NASA's Goddard Space Flight Center, which aims to provide equitable access to data and promote inclusive learning towards a larger Global Initiative on resilience to natural hazards through AI solutions (RESOLUTION).
As part of the Early Warnings for All initiative, WMO and the United Nations Office for Disaster Risk Reduction (UNDRR)are using AI to enhance the financing tracking mechanisms to ensure coherence, alignment and financial efficiency. This approach benefited from an earlier collaboration between WMO and WomenInData on a datathon, with female students from around the world, to pilot AI applications to that finance and to better understand the impacts of heat wave early warnings. Learnings from the pilot led to efforts to automate the Early Warnings for All finance tracking effort, which several multilateral banks have joined. WMO is currently exploring the use of AI for the tagging and tracking of investments.
n addition, The Horizon Europe project MedEWSa (Mediterranean and Pan-European Forecast and Early Warning System against Natural Hazards), a European Union-Horizon funded project led by WMO, exemplifies how AI and emerging technologies join forces to strengthen early warning systems by improving forecasts, detection and monitoring of natural hazards and extreme weather events, and to make informed decisions.
The agility and speed of AI has already made astonishing advances. Companies, such as Google's DeepMind, Huawei and Nvidia, have released AI-based medium-range weather forecasting models in the past year, which have been incorporated into "traditional" ECMWF (European Centre for Medium-Range Weather Forecasts) models. AI is also likely to deliver better forecasting and risk monitoring for "hyper local" or kilometre-scale events - such as thunderstorms, which can give rise to extreme rainfall, tornados or damaging hail - as well as for improved hurricane track forecasting.
Over the last years, significant strides have been made towards making Earth system science more accessible through AI. Capability for processing different data modalities, such as time series forecasting, image processing or text processing, have been enhanced and AI and natural language processing models for verifying scientific claims have been developed. These point to AI's potential to refine the Earth system science approach by offering more accurate and unbiased information. A better understanding of the limitations of AI in Disaster Risk Reduction (DRR) is required to realize its benefits, which will demand interdisciplinarity, multistakeholder involvement and international collaboration.
The WMO engagement with AI in Earth system science includes technology but also embraces and fosters a more well-informed global community confronted by climate challenges.
