MAMichael Aich

Michael Aich

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Michael Aich has a total of 17 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Generating realistic global precipitation fields from modelled atmospheric circulation

    by Michael Aich, Sebastian Bathiany, Philipp Hess, Yu Huang, Niklas Boers

    Improving the representation of precipitation in Earth system models (ESMs) is critical for assessing the impacts of climate change and especially of extreme events like floods and droughts. In existing ESMs, precipitation is not resolved explicitly, but represented by parameterizations. These typically rely on resolving approximated but computationally expensive column-based physics, not accounting for interactions between locations. They struggle to capture fine-scale precipitation processes and introduce systematic biases. We present a novel approach, based on generative machine learning, which combines a conditional diffusion model with a regression model to generate accurate, high-resolution ( $$0.25^\circ $$ ) global daily precipitation fields from a small set of prognostic atmospheric variables. Unlike traditional parameterizations, our framework efficiently produces ensemble predictions, capturing uncertainties in precipitation, and does not require fine-tuning by hand. We train our model on the ERA5 reanalysis and present a method that allows us to apply it to unseen ESM data across different climate scenarios, enabling fast probabilistic generation of precipitation fields. By leveraging interactions between global prognostic variables, our approach provides an alternative parameterization approach that mitigates biases present in the ESM precipitation while maintaining consistency with its large-scale (annual) trends. This work demonstrates that complex precipitation patterns can be learned directly from large-scale atmospheric variables, offering a computationally efficient method to obtain high-resolution precipitation without the cost of running the dynamical model at such high resolution.

  • Open Access

    Better serving impact assessments via AI: Reconstructing daily extremes from spatiotemporal downscaling of monthly fields

    by Yu Huang, Sebastian Bathiany, Shangshang Yang, Michael Aich, Philipp Hess, Niklas Boers

    Climate impact assessment studies strongly depend on fine representations of meteorological fields. Downscaling addresses the trade-off between data requirements and storage capacity, yet the faithful replication of extreme-value statistics and spatiotemporal consistency presents a persistent issue. We present an efficient generative AI model for spatiotemporal downscaling. Using coarse-resolution monthly fields as inputs, the model reconstructs sequences of daily fields with the enhanced spatial resolution. The AI-generated daily fields accurately reproduce spatial coherence, temporal persistence, and extreme-value characteristics, showing strong agreement with ground-truth daily observations. We look forward to applying this framework more effectively to future studies on the impacts of extreme events.

  • Top Cited Papers
    Open Access

    Conditional diffusion models for downscaling and bias correction of Earth system model precipitation

    by Michael Aich, Philipp Hess, Baoxiang Pan, Sebastian Bathiany, Yu Huang, Niklas Boers

    Climate change exacerbates extreme weather events like heavy rainfall and flooding. As these events cause severe socioeconomic damage, accurate high-resolution simulation of precipitation is imperative. However, existing Earth System Models (ESMs) struggle to resolve small-scale dynamics and suffer from biases. Traditional statistical bias correction and downscaling methods fall short in improving spatial structure, while recent deep learning methods lack controllability and suffer from unstable training. Here, we propose a machine learning framework for simultaneous bias correction and downscaling. We first map observational and ESM data to a shared embedding space, where both are unbiased towards each other, and then train a conditional diffusion model to reverse the mapping. Only observational data is used for the training, so that the diffusion model can be employed to correct and downscale any ESM field without need for retraining. Our approach ensures statistical fidelity and preserves spatial patterns larger than a chosen spatial correction scale. We demonstrate that our approach outperforms existing statistical and deep learning methods especially regarding extreme events.


  • Open Access

    WIND: Weather Inverse Diffusion for Zero-Shot Atmospheric Modeling

    by Michael Aich, Andreas Fürst, Florian Sestak, Carlos Ruiz-Gonzalez, Niklas Boers, Johannes Brandstetter

    Deep learning has revolutionized weather and climate modeling, yet the current landscape remains fragmented: highly specialized models are typically trained individually for distinct tasks. To unify this landscape, we introduce WIND, a single pre-trained foundation model capable of replacing specialized baselines across a vast array of tasks. Crucially, in contrast to previous atmospheric foundation models, we achieve this without any task-specific fine-tuning. To learn a robust, task-agnostic prior of the atmosphere, we pre-train WIND with a self-supervised video reconstruction objective, utilizing an unconditional video diffusion model to iteratively reconstruct atmospheric dynamics from a noisy state. At inference, we frame diverse domain-specific problems strictly as inverse problems and solve them via posterior sampling. This unified approach allows us to tackle highly relevant weather and climate problems, including probabilistic forecasting, spatial and temporal downscaling, sparse reconstruction and enforcing conservation laws purely with our pre-trained model. We further demonstrate the model's capacity to generate physically consistent counterfactual storylines of extreme weather events under global warming scenarios. By combining generative video modeling with inverse problem solving, WIND offers a computationally efficient paradigm shift in AI-based atmospheric modeling.

  • Open Access

    Reply on RC2

    by Michael Aich, Philipp Hess, Baoxiang Pan, Sebastian Bathiany, Yu Huang, Niklas Boers

    Climate change exacerbates extreme weather events like heavy rainfall and flooding. As these events cause severe socioeconomic damages, accurate high-resolution simulation of precipitation is imperative. However, existing Earth System Models (ESMs) struggle resolving small-scale dynamics and suffer from biases. Traditional statistical bias correction and downscaling methods fall short in improving spatial structure, while recent deep learning methods lack controllability and suffer from unstable training. Here, we propose a machine learning framework for simultaneous bias correction and downscaling. We train a generative diffusion model purely on observational data. We map observational and ESM data to a shared embedding space, where both are unbiased towards each other and train a conditional diffusion model to reverse the mapping. Our method can correct any ESM field, as the training is independent of the ESM. Our approach ensures statistical fidelity, preserves large-scale spatial patterns and outperforms existing methods especially regarding extreme events.

  • Open Access

    Reply on RC1

    by Michael Aich, Philipp Hess, Baoxiang Pan, Sebastian Bathiany, Yu Huang, Niklas Boers

    Climate change exacerbates extreme weather events like heavy rainfall and flooding. As these events cause severe socioeconomic damages, accurate high-resolution simulation of precipitation is imperative. However, existing Earth System Models (ESMs) struggle resolving small-scale dynamics and suffer from biases. Traditional statistical bias correction and downscaling methods fall short in improving spatial structure, while recent deep learning methods lack controllability and suffer from unstable training. Here, we propose a machine learning framework for simultaneous bias correction and downscaling. We train a generative diffusion model purely on observational data. We map observational and ESM data to a shared embedding space, where both are unbiased towards each other and train a conditional diffusion model to reverse the mapping. Our method can correct any ESM field, as the training is independent of the ESM. Our approach ensures statistical fidelity, preserves large-scale spatial patterns and outperforms existing methods especially regarding extreme events.

  • Open Access

    Comment on egusphere-2025-2646

    by Michael Aich, Philipp Hess, Baoxiang Pan, Sebastian Bathiany, Yu Huang, Niklas Boers

    Climate change exacerbates extreme weather events like heavy rainfall and flooding. As these events cause severe socioeconomic damages, accurate high-resolution simulation of precipitation is imperative. However, existing Earth System Models (ESMs) struggle resolving small-scale dynamics and suffer from biases. Traditional statistical bias correction and downscaling methods fall short in improving spatial structure, while recent deep learning methods lack controllability and suffer from unstable training. Here, we propose a machine learning framework for simultaneous bias correction and downscaling. We train a generative diffusion model purely on observational data. We map observational and ESM data to a shared embedding space, where both are unbiased towards each other and train a conditional diffusion model to reverse the mapping. Our method can correct any ESM field, as the training is independent of the ESM. Our approach ensures statistical fidelity, preserves large-scale spatial patterns and outperforms existing methods especially regarding extreme events.

  • Open Access

    Comment on egusphere-2025-2646

    by Michael Aich, Philipp Hess, Baoxiang Pan, Sebastian Bathiany, Yu Huang, Niklas Boers

    Climate change exacerbates extreme weather events like heavy rainfall and flooding. As these events cause severe socioeconomic damages, accurate high-resolution simulation of precipitation is imperative. However, existing Earth System Models (ESMs) struggle resolving small-scale dynamics and suffer from biases. Traditional statistical bias correction and downscaling methods fall short in improving spatial structure, while recent deep learning methods lack controllability and suffer from unstable training. Here, we propose a machine learning framework for simultaneous bias correction and downscaling. We train a generative diffusion model purely on observational data. We map observational and ESM data to a shared embedding space, where both are unbiased towards each other and train a conditional diffusion model to reverse the mapping. Our method can correct any ESM field, as the training is independent of the ESM. Our approach ensures statistical fidelity, preserves large-scale spatial patterns and outperforms existing methods especially regarding extreme events.

  • Open Access

    Reply on CEC2

    by Michael Aich, Philipp Hess, Baoxiang Pan, Sebastian Bathiany, Yu Huang, Niklas Boers

    Climate change exacerbates extreme weather events like heavy rainfall and flooding. As these events cause severe socioeconomic damages, accurate high-resolution simulation of precipitation is imperative. However, existing Earth System Models (ESMs) struggle resolving small-scale dynamics and suffer from biases. Traditional statistical bias correction and downscaling methods fall short in improving spatial structure, while recent deep learning methods lack controllability and suffer from unstable training. Here, we propose a machine learning framework for simultaneous bias correction and downscaling. We train a generative diffusion model purely on observational data. We map observational and ESM data to a shared embedding space, where both are unbiased towards each other and train a conditional diffusion model to reverse the mapping. Our method can correct any ESM field, as the training is independent of the ESM. Our approach ensures statistical fidelity, preserves large-scale spatial patterns and outperforms existing methods especially regarding extreme events.

  • Open Access

    Reply on CEC1

    by Michael Aich, Philipp Hess, Baoxiang Pan, Sebastian Bathiany, Yu Huang, Niklas Boers

    Climate change exacerbates extreme weather events like heavy rainfall and flooding. As these events cause severe socioeconomic damages, accurate high-resolution simulation of precipitation is imperative. However, existing Earth System Models (ESMs) struggle resolving small-scale dynamics and suffer from biases. Traditional statistical bias correction and downscaling methods fall short in improving spatial structure, while recent deep learning methods lack controllability and suffer from unstable training. Here, we propose a machine learning framework for simultaneous bias correction and downscaling. We train a generative diffusion model purely on observational data. We map observational and ESM data to a shared embedding space, where both are unbiased towards each other and train a conditional diffusion model to reverse the mapping. Our method can correct any ESM field, as the training is independent of the ESM. Our approach ensures statistical fidelity, preserves large-scale spatial patterns and outperforms existing methods especially regarding extreme events.