Applications Quantum Machine Learning Innovation Center Hamburg, Innovation Center Ulm, Oberpfaffenhofen
Industry partner


We are improving climate models using quantum machine learning for robust technology assessment and mitigation recommendations.

To this end, we are exploiting the potential of quantum computing for the improvement of climate models. We will also make use of machine learning methods and their efficient and extensive evaluation of Earth observation data to address the goals of DLR in cooperation with industrial partners. To achieve this, we are developing a prototype of a climate model based on Quantum Machine Learning (QML). This can reduce uncertainties in climate predictions that are used for robust technology assessment and mitigation recommendations across diverse applications in the aerospace, transport and energy sectors.


Earth System Models (ESMs) simulate the physical climate and biogeochemical cycles under a wide range of forcings. However, given their complexity, there are continuing uncertainties in their predictions – particularly with respect to differences in the representation of processes that occur at scales smaller than the resolution of the model grid. These processes need to be approximated by parameterisations. One solution is to develop high-resolution versions of the ICOsahedral Nonhydrostatic (ICON) model. Here, clouds and convection are explicitly resolved.

However, cloud-resolving models are extremely computationally intensive and can therefore only be used to cover very short periods of time and/or only for a small region. Relying solely on supercomputing performance improvements is therefore not a solution. The latest developments in machine learning have great potential – they can eliminate systematic errors in climate models and contribute to their considerable improvement.


The Klim-QML staff will work closely with the experts at the DLR innovation centres in Ulm and Hamburg, the Earth System Model Evaluation and Analysis (PA-EVA) department at the DLR Institute of Atmospheric Physics, their cooperation partners and with industrial partners on innovative concepts and algorithms. All Klim-QML work areas benefit from the expertise available in the department, especially in the evaluation of climate models with Earth observation data and in the development of deep learning methods for the parameterisation of climate models within the framework of the Understanding and Modelling the Earth System with Machine Learning (USMILE) project. As a result, we expect an improvement and accelerated development of climate models and their simulations, which we will demonstrate using the developed prototype of a QML climate model as an example of a specific use case.

All News