arXiv subjects

Luca Schmidt

Publications and source records attributed to Luca Schmidt.

6 recordsLinked to original sources

arXiv2026

How Can Machine Learning Emulators Best Support Climate Science?

For decades, physics-based climate models have been used to provide insights for climate decision-making. Their application is, however, constrained by significant computational and technical demands. Machine learning (ML) emulators offer a way to reduce these high computational costs; yet, it remains challenging to use ML emulators effectively in climate research. In practice, climate scientists often bypass emulators altogether, and machine learning researchers frequently develop them as methodological showcases without proving their practical utility. The reasons are diverse, ranging from limited accessibility and a lack of specialized knowledge to broader concerns about the physical grounding of ML methods. Here, we discuss limitations and introduce a framework for guiding emulator development, considering both climate science and machine learning perspectives. We argue that designing easy-to-adopt emulators that address clearly defined tasks and demonstrate their reliability is essential. This offers a promising path towards making machine-learning approaches more relevant and usable for applied climate research.

cs.LG↗

arXiv2025

RainShift: A Benchmark for Precipitation Downscaling Across Geographies

Earth System Models (ESM) are our main tool for projecting the impacts of climate change. However, running these models at sufficient resolution for local-scale risk-assessments is not computationally feasible. Deep learning-based super-resolution models offer a promising solution to downscale ESM outputs to higher resolutions by learning from data. Yet, due to regional variations in climatic processes, these models typically require retraining for each geographical area-demanding high-resolution observational data, which is unevenly available across the globe. This highlights the need to assess how well these models generalize across geographic regions. To address this, we introduce RainShift, a dataset and benchmark for evaluating downscaling under geographic distribution shifts. We evaluate state-of-the-art downscaling approaches including GANs and diffusion models in generalizing across data gaps between the Global North and Global South. Our findings reveal substantial performance drops in out-of-distribution regions, depending on model and geographic area. While expanding the training domain generally improves generalization, it is insufficient to overcome shifts between geographically distinct regions. We show that addressing these shifts through, for example, data alignment can improve spatial generalization. Our work advances the global applicability of downscaling methods and represents a step toward reducing inequities in access to high-resolution climate information.

cs.CV↗

arXiv2025

A novel method for quantifying enzyme immobilization in porous carriers using simple NMR relaxometry

Enzyme immobilization plays a crucial role in enhancing the stability and recyclability of enzymes for industrial applications. However, traditional methods for quantifying enzyme loading within porous carriers are limited by time-consuming workflows, cumulative errors, and the inability to probe enzymes adsorbed inside the pores. In this study, we introduce Time-Domain Nuclear Magnetic Resonance (TD-NMR) relaxometry as a novel, non-invasive technique for directly quantifying enzyme adsorption within porous carriers. Focusing on epoxy methyl acrylate carriers, commonly used in biocatalysis, we correlate changes in T2 relaxation times with enzyme concentration, leading to the development of an NMR-based pore-filling ratio that quantifies enzyme loading. Validation experiments demonstrate that TD-NMR-derived adsorption curves align closely with traditional photometric measurements, offering a reliable and reproducible alternative for enzyme quantification. The accessibility of tabletop TD-NMR spectrometers makes this technique a practical and cost-effective tool for optimizing biocatalytic processes. Furthermore, the method holds promise for real-time monitoring of adsorption dynamics and could be adapted for a wider range of carrier materials and enzymes.

cond-mat.soft↗

arXiv2024

A Generative Framework for Probabilistic, Spatiotemporally Coherent Downscaling of Climate Simulation

Local climate information is crucial for impact assessment and decision-making, yet coarse global climate simulations cannot capture small-scale phenomena. Current statistical downscaling methods infer these phenomena as temporally decoupled spatial patches. However, to preserve physical properties, estimating spatio-temporally coherent high-resolution weather dynamics for multiple variables across long time horizons is crucial. We present a novel generative framework that uses a score-based diffusion model trained on high-resolution reanalysis data to capture the statistical properties of local weather dynamics. After training, we condition on coarse climate model data to generate weather patterns consistent with the aggregate information. As this predictive task is inherently uncertain, we leverage the probabilistic nature of diffusion models and sample multiple trajectories. We evaluate our approach with high-resolution reanalysis information before applying it to the climate model downscaling task. We then demonstrate that the model generates spatially and temporally coherent weather dynamics that align with global climate output.

cs.LG↗

arXiv2024

Climate data selection for multi-decadal wind power forecasts

Reliable wind speed data is crucial for applications such as estimating local (future) wind power. Global Climate Models (GCMs) and Regional Climate Models (RCMs) provide forecasts over multi-decadal periods. However, their outputs vary substantially, and higher-resolution models come with increased computational demands. In this study, we analyze how the spatial resolution of different GCMs and RCMs affects the reliability of simulated wind speeds and wind power, using ERA5 data as a reference. We present a systematic procedure for model evaluation for wind resource assessment as a downstream task. Our results show that higher-resolution GCMs and RCMs do not necessarily preserve wind speeds more accurately. Instead, the choice of model, both for GCMs and RCMs, is more important than the resolution or GCM boundary conditions. The IPSL model preserves the wind speed distribution particularly well in Europe, producing the most accurate wind power forecasts relative to ERA5 data.

stat.AP↗

arXiv2024

Wind Power Assessment based on Super-Resolution and Downscaling -- A Comparison of Deep Learning Methods

The efficient placement of wind turbines relies on accurate local wind speed forecasts. Climate projections provide valuable insight into long-term wind speed conditions, yet their spatial data resolution is typically insufficient for precise wind power forecasts. Deep learning methods, particularly models developed for image super-resolution, offer a promising solution to bridge this scale gap by increasing the spatial resolution of climate models. In this paper, we compare the performance of various deep learning models on two distinct tasks: super-resolution, where we map artificially coarsened ERA5 data to its native resolution, and downscaling, where we map native ERA5 to high-resolution COSMO-REA6 data. We evaluate the models on their downstream application in forecasting long-term wind power, emphasizing the impact of spatial wind speed resolution on wind power estimates. Our findings highlight the importance of aligning models and evaluation metrics with their specific downstream applications. We show that a diffusion model outperforms other models for estimating the wind power potential by better preserving the wind speeds' distributional and physical properties.

stat.AP↗