Skip to content. | Skip to navigation

Personal tools
You are here: Home Theses Development of Generative Models Trained on Decentralised Geo- and Hydrological Data


Prof. Dr. S. Decker
RWTH Aachen
Informatik 5
Ahornstr. 55
D-52056 Aachen
Tel +49/241/8021501
Fax +49/241/8022321

How to find us

Annual Reports





Development of Generative Models Trained on Decentralised Geo- and Hydrological Data

Thesis type
  • Bachelor
Status Finished
Submitted in 2021

In times of increasing weather extremes, weather forecasts and early warning systems for extreme weathers have gained great promise to prevent property damage or to mitigate the risk to the civilian population. Current approaches are based on different data types, such as geo basic data, weather forecasts or other hydrological data, in several data sources (Salas et al., 2020; Kreklow et al., 2020; Hofmann and Schüttrumpf, 2019) However, these approaches rely on data centralisation, which can pose several challenges. First, the centralisation process itself could act as a bottleneck since vast amounts of spatiotemporal data from multiple data sources are transmitted to one single location. Second, depending on the data type, well-known data privacy regulations hinder institutions to share their data. Lastly, data sets suffer from sparsity and incomplete features, which makes data analysis difficult. In particular, weather datasets provide high spatiotemporal data whereas hydrological well data or geobase data are generally incomplete.

One solution to circumvent these immanent problems of data centralisation could be Distributed Analytics (DA) approaches in combination with generative models. DA poses a paradigm shift to address the mentioned challenges by bringing the analysis algorithm to the data instead of vice versa. By design, data stays within institutional borders and the data owner keeps the sovereignty over the data.

This Thesis focuses on the development of suitable generative models in a decentralised setting using decentralised geo- and hydrological data to mitigate the influence of sparse data. In a second step, the developed generator models should be used to train additional models to evaluate their quality (E.g. Prediction modelling or model-based simulation). This Thesis represents a feasibility study showing that decentrally trained generative models could be an alternative to common data centralisation approaches in this domain.

If you are interested in this thesis, a related topic or have additional questions, please do not hesitate to send a message to, or .


  1. Hofmann, J., Schüttrumpf, H., 2019. Risk-Based Early Warning System for Pluvial Flash Floods: Approaches and Foundations. Geosciences 9 (3), 127.
  2. Kreklow, J., Tetzlaff, B., Burkhard, B., Kuhnt, G., 2020. Radar-Based Precipitation Climatology in Germany—Developments, Uncertainties and Potentials. Atmosphere 11 (2), 217.
  3. Salas, D., Liang, X., Navarro, M., Liang, Y., Luna, D., 2020. An open-data open-model framework for hydrological models’ integration, evaluation and application. Environmental Modelling & Software 126 (3 SPEC. ISS), 104622.

For more information, see the following attachment:

Development of Generative Models Trained on Decentralised Geo- and Hydrological Data_final.pdf — PDF document, 402Kb


Foundational Knowledge: Machine Learning, Deep Learning, Data Generation, Synthetic/Artificial Data

Document Actions