Filters

2018 - 201912020 - 20212

More filters

2022 - 20233
Reset filters

Settings

Author: Boers, Niklas × Author: Gelbrecht, Maximilian × Author: Kurths, Jürgen × End date: 2023 ×

Search Results

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Analysis of a bistable climate toy model with physics-based machine learning methods
    (Berlin ; Heidelberg : Springer, 2021) Gelbrecht, Maximilian; Lucarini, Valerio; Boers, Niklas; Kurths, Jürgen
    We propose a comprehensive framework able to address both the predictability of the first and of the second kind for high-dimensional chaotic models. For this purpose, we analyse the properties of a newly introduced multistable climate toy model constructed by coupling the Lorenz ’96 model with a zero-dimensional energy balance model. First, the attractors of the system are identified with Monte Carlo Basin Bifurcation Analysis. Additionally, we are able to detect the Melancholia state separating the two attractors. Then, Neural Ordinary Differential Equations are applied to predict the future state of the system in both of the identified attractors.
  • Thumbnail Image
    Item
    Neural partial differential equations for chaotic systems
    ([London] : IOP, 2021) Gelbrecht, Maximilian; Boers, Niklas; Kurths, Jürgen
    When predicting complex systems one typically relies on differential equation which can often be incomplete, missing unknown influences or higher order effects. By augmenting the equations with artificial neural networks we can compensate these deficiencies. We show that this can be used to predict paradigmatic, high-dimensional chaotic partial differential equations even when only short and incomplete datasets are available. The forecast horizon for these high dimensional systems is about an order of magnitude larger than the length of the training data.
  • Thumbnail Image
    Item
    Phase coherence between precipitation in South America and Rossby waves
    (Washington, DC [u.a.] : Assoc., 2018) Gelbrecht, Maximilian; Boers, Niklas; Kurths, Jürgen
    The dominant mode of intraseasonal precipitation variability during the South American monsoon is the so-called precipitation dipole between the South Atlantic convergence zone (SACZ) and southeastern South America (SESA). It affects highly populated areas that are of substantial importance for the regional food supplies. Previous studies using principal components analysis or complex networks were able to describe and characterize this variability pattern, but crucial questions regarding the responsible physical mechanism remain open. Here, we use phase synchronization techniques to study the relation between precipitation in the SACZ and SESA on the one hand and southern hemisphere Rossby wave trains on the other hand. In combination with a conceptual model, this approach demonstrates that the dipolar precipitation pattern is caused by the southern hemisphere Rossby waves. Our results thus show that Rossby waves are the main driver of the monsoon season variability in South America, a finding that has important implications for synoptic-scale weather forecasts.