Neural Network

AI for early warnings addressing floods and droughts

OnlineDiscovery - AI for Earth and Sustainability Science

  • Register

    Watch

    * Register (or log in) to the Neural Network to add this session to your agenda or watch the replay

    • Date
    22 September 2025
    Timeframe
    16:00 - 17:00 CEST Geneva
    Duration
    60 minutes
    Share this session

    While there is a general expectation that climate change will further alter the patterns and frequencies of extreme weather and Climate events, it remains challenging to accurately localize and predict where the impacts of extreme events could occur in the future. Had we localized and predicted such impacts earlier, we would have been able to prepare informed decisions that could lead to better Adaptation strategies. This is an example of an adaptation and mitigation Action that can reduce vulnerability and promote climate resilience. This talk explores applications of artificial intelligence for early warning and forecasting of extreme impacts, such as extreme floods and agricultural droughts. We will discuss three topics. First, we will present RiverMamba, a deep learning forecasting system that is capable of forecasting floods up to seven days in advance with high spatial resolution and on a global scale. In the second part, we will present Focal-TSMP, a deep learning model that forecasts long-term agricultural droughts from climate simulations. Finally, we will introduce a novel deep learning method to identify drivers of extreme agricultural droughts from reanalysis data.

     

    Learning Objectives:

    Describe the challenges in localizing and predicting extreme climate events and their relevance for adaptation strategies.
    Explain how AI and deep learning methods, such as RiverMamba and Focal-TSMP, are used to forecast floods and agricultural droughts.
    Interpret outputs from AI-based climate forecasting systems to identify drivers of extreme events.
    Assess the potential of AI-driven early warning systems to inform decision-making and enhance climate resilience.
    Propose novel approaches for integrating AI forecasts into adaptation and mitigation planning

     

    Recommended Mastery Level / Prerequisites:

    Recommended Mastery Level: Intermediate – participants should have a basic understanding of climate science, extreme events, or AI/machine learning concepts.

     

    Prerequisites:

    • Familiarity with climate data, including temperature, precipitation, and drought/flood indices.
    • Basic understanding of AI/ML concepts, especially deep learning.
    • Interest in climate adaptation, resilience, or early warning systems.
    Share this session
      Add to Calendar
      Discover more from programme stream
      Discovery - AI for Earth and Sustainability Science
      Do AI models know what they know? How explainable AI...

      Do AI models know what they know? How explainable AI...

      24 November 2025 - 17:00 to 18:00

      Risks and solutions when using non-stationary training data in earth...

      Risks and solutions when using non-stationary training data in earth...

      8 December 2025 - 17:00 to 18:00

      Related stories

      Unlocking clean energy transparency with Enerlink

      Around the world, environmental degradation continues to threaten ecosystems, economies, and communities, underscoring the urgency of accelerating climate action. That...
      Climate | Innovation Factory

      Bringing AI into human rights: why dialogue matters

      At the 2025 AI for Good Global Summit, the workshop titled “AI for Human Rights: Smarter, Faster, Fairer Monitoring”, organized...
      Global Summit

      Ending data disparity: Snowflake’s vision for AI and tri-sector partnerships

      My first job was at a summer camp in the mountains, where I worked as a camp counselor. The camp...
      Global Summit

      Are you sure you want to remove this speaker?