ARAndreas Rademacher

Andreas Rademacher

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Andreas Rademacher has a total of 56 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Improved basal drag of the West Antarctic Ice Sheet from L-curve analysis of inverse models utilizing subglacial hydrology simulations

    by Lea-Sophie Höyns, Thomas Kleiner, Andreas Rademacher, Martin Rückamp, Michael Wolovick, Angelika Humbert

    The West Antarctic Ice Sheet (WAIS) is the focus of current research due to its susceptibility to collapse, which could potentially contribute to rising sea levels. To accurately predict future glacier evolution, precise ice sheet models are essential. The ice discharge of outlet glaciers into the ocean is one key factor here, primarily caused by the basal sliding of ice. Since we cannot directly measure basal properties on a large scale, inverse models can be used to infer the basal drag coefficient by minimizing a cost function that depends on a velocity misfit and a regularization term. We conduct various basal drag inversions to obtain an improved basal drag distribution for the WAIS. Additionally, we perform L-curve analyses to determine the optimal trade-off between the cost function terms that result in smooth L-curves. The domain L-curve is divided into eight subdomains of the study area to assess how well the inverse method performs in different glaciological settings. Pine Island Glacier exhibits the smoothest L-curves, while slow-flowing regions such as Roosevelt Island reveal rather poorly shaped L-curve behavior for the basal drag inversion. This highlights the importance of performing a subdomain L-curve analysis for large-scale inversions to discover potential problematic regions and to establish suitable regularization for different physical conditions. Comprehensive basal drag inversion experiments allow us to test the dependence of both the L-curves and the basal drag results on the nonlinearity of sliding and the inclusion of subglacial effective pressure in the friction law. The analysis suggests that nonlinear friction laws are preferable to linear sliding because of reduced variance in the overall inferred friction coefficient and steeper L-curves leading to a more well-defined corner region. We show that a Budd-type friction law that incorporates effective pressure from a subglacial hydrology model rather than a simple geometry-based approximation achieves improved performance in our inverse model in terms of the total model variance ratio, along with faster convergence and smoother L-curves. Further comparison reveals that the basal drag coefficient field has a less variable spatial structure when an effective pressure from the hydrology model is used instead of a parameterized effective pressure, allowing us to interpret the inverted drag coefficient more precisely in terms of the basal properties rather than the basal hydrology.


  • Goal-Oriented Adaptive Finite Cell Methods

    by Annika Osmers, Andreas Rademacher, Andreas Schröder

  • Goal-Oriented Error Estimates for Adaptive Mixed Finite Element Methods

    by Lothar Banz, Andreas Rademacher, Dominika Thiede

  • Open Access

    Reply on RC2

    by Lea-Sophie Hoeyns, Thomas Kleiner, Andreas Rademacher, Martin Rückamp, Michael Wolovick, Angelika Humbert

    The West Antarctic Ice Sheet (WAIS) is the focus of current research due to its susceptibility to collapse, which could potentially contribute to rising sea levels. To accurately predict future glacier evolution, precise ice sheet models are essential with regard to suitable approximations of physical behavior to the real system and appropriate input values, as well as computing power. The ice discharge of outlet glaciers into the ocean is one key factor here, primarily caused by basal sliding of ice. Since we cannot directly measure basal properties on a large scale, inverse models can be used to infer the basal drag coefficient by minimizing a cost function that depends on a velocity misfit and a regularization term. We conduct basal drag inversions and perform L-curve analyses to find the optimal trade-off between the cost function terms, ending up with smooth L-curves. Additionally, the domain L-curve is break down to eight subdomains of the study area in order to reveal how well the inverse method performs in different glaciological settings. It reveals that Pine Island Glacier being the best area and slow-flowing areas such as Roosevelt Island being among the worst in terms of the L-curve behavior for the basal drag inversion. This highlights the importance of performing a subdomain L-curve analysis, whenever an inversion for a larger domain is calculated to discover problematic regions. Comprehensive basal drag inversion experiments allow us to test the dependence of the L-curve and basal drag results on the non-linearity of sliding as well as on the inclusion of subglacial effective pressure in the friction law. The analysis suggest that non-linear friction laws are preferable to linear sliding because of reduced variance of the overall inferred friction coefficient, faster convergence, as well as steeper L-curves leading to a more accurate choice of weight for the regularization term. We show that a Budd-type friction law that incorporates effective pressure from a subglacial hydrology model rather than a simple geometry-based approximation achieves an improved performance in our inverse model. Further comparison reveals that the effective pressure from the hydrology model accounts for a larger part of spatial structure of the basal drag coefficient than the parameterized effective pressure. Allowing the inverted drag coefficient to more precisely reflect actual variations in basal properties. Finally, a comparison of the inferred basal drag across WAIS with observed locations of subglacial lakes reveals a good match, giving us additional confidence in the spatial basal drag structure revealed by the inverse method.

  • Open Access

    Reply on RC1

    by Lea-Sophie Hoeyns, Thomas Kleiner, Andreas Rademacher, Martin Rückamp, Michael Wolovick, Angelika Humbert

    The West Antarctic Ice Sheet (WAIS) is the focus of current research due to its susceptibility to collapse, which could potentially contribute to rising sea levels. To accurately predict future glacier evolution, precise ice sheet models are essential with regard to suitable approximations of physical behavior to the real system and appropriate input values, as well as computing power. The ice discharge of outlet glaciers into the ocean is one key factor here, primarily caused by basal sliding of ice. Since we cannot directly measure basal properties on a large scale, inverse models can be used to infer the basal drag coefficient by minimizing a cost function that depends on a velocity misfit and a regularization term. We conduct basal drag inversions and perform L-curve analyses to find the optimal trade-off between the cost function terms, ending up with smooth L-curves. Additionally, the domain L-curve is break down to eight subdomains of the study area in order to reveal how well the inverse method performs in different glaciological settings. It reveals that Pine Island Glacier being the best area and slow-flowing areas such as Roosevelt Island being among the worst in terms of the L-curve behavior for the basal drag inversion. This highlights the importance of performing a subdomain L-curve analysis, whenever an inversion for a larger domain is calculated to discover problematic regions. Comprehensive basal drag inversion experiments allow us to test the dependence of the L-curve and basal drag results on the non-linearity of sliding as well as on the inclusion of subglacial effective pressure in the friction law. The analysis suggest that non-linear friction laws are preferable to linear sliding because of reduced variance of the overall inferred friction coefficient, faster convergence, as well as steeper L-curves leading to a more accurate choice of weight for the regularization term. We show that a Budd-type friction law that incorporates effective pressure from a subglacial hydrology model rather than a simple geometry-based approximation achieves an improved performance in our inverse model. Further comparison reveals that the effective pressure from the hydrology model accounts for a larger part of spatial structure of the basal drag coefficient than the parameterized effective pressure. Allowing the inverted drag coefficient to more precisely reflect actual variations in basal properties. Finally, a comparison of the inferred basal drag across WAIS with observed locations of subglacial lakes reveals a good match, giving us additional confidence in the spatial basal drag structure revealed by the inverse method.

  • Open Access

    Comment on egusphere-2024-1251

    by Lea-Sophie Höyns, Thomas Kleiner, Andreas Rademacher, Martin Rückamp, Michael Wolovick, Angelika Humbert

    The West Antarctic Ice Sheet (WAIS) is the focus of current research due to its susceptibility to collapse, which could potentially contribute to rising sea levels. To accurately predict future glacier evolution, precise ice sheet models are essential with regard to suitable approximations of physical behavior to the real system and appropriate input values, as well as computing power. The ice discharge of outlet glaciers into the ocean is one key factor here, primarily caused by basal sliding of ice. Since we cannot directly measure basal properties on a large scale, inverse models can be used to infer the basal drag coefficient by minimizing a cost function that depends on a velocity misfit and a regularization term. We conduct basal drag inversions and perform L-curve analyses to find the optimal trade-off between the cost function terms, ending up with smooth L-curves. Additionally, the domain L-curve is break down to eight subdomains of the study area in order to reveal how well the inverse method performs in different glaciological settings. It reveals that Pine Island Glacier being the best area and slow-flowing areas such as Roosevelt Island being among the worst in terms of the L-curve behavior for the basal drag inversion. This highlights the importance of performing a subdomain L-curve analysis, whenever an inversion for a larger domain is calculated to discover problematic regions. Comprehensive basal drag inversion experiments allow us to test the dependence of the L-curve and basal drag results on the non-linearity of sliding as well as on the inclusion of subglacial effective pressure in the friction law. The analysis suggest that non-linear friction laws are preferable to linear sliding because of reduced variance of the overall inferred friction coefficient, faster convergence, as well as steeper L-curves leading to a more accurate choice of weight for the regularization term. We show that a Budd-type friction law that incorporates effective pressure from a subglacial hydrology model rather than a simple geometry-based approximation achieves an improved performance in our inverse model. Further comparison reveals that the effective pressure from the hydrology model accounts for a larger part of spatial structure of the basal drag coefficient than the parameterized effective pressure. Allowing the inverted drag coefficient to more precisely reflect actual variations in basal properties. Finally, a comparison of the inferred basal drag across WAIS with observed locations of subglacial lakes reveals a good match, giving us additional confidence in the spatial basal drag structure revealed by the inverse method.

  • Open Access

    Comment on egusphere-2024-1251

    by Lea-Sophie Höyns, Thomas Kleiner, Andreas Rademacher, Martin Rückamp, Michael Wolovick, Angelika Humbert

    The West Antarctic Ice Sheet (WAIS) is the focus of current research due to its susceptibility to collapse, which could potentially contribute to rising sea levels. To accurately predict future glacier evolution, precise ice sheet models are essential with regard to suitable approximations of physical behavior to the real system and appropriate input values, as well as computing power. The ice discharge of outlet glaciers into the ocean is one key factor here, primarily caused by basal sliding of ice. Since we cannot directly measure basal properties on a large scale, inverse models can be used to infer the basal drag coefficient by minimizing a cost function that depends on a velocity misfit and a regularization term. We conduct basal drag inversions and perform L-curve analyses to find the optimal trade-off between the cost function terms, ending up with smooth L-curves. Additionally, the domain L-curve is break down to eight subdomains of the study area in order to reveal how well the inverse method performs in different glaciological settings. It reveals that Pine Island Glacier being the best area and slow-flowing areas such as Roosevelt Island being among the worst in terms of the L-curve behavior for the basal drag inversion. This highlights the importance of performing a subdomain L-curve analysis, whenever an inversion for a larger domain is calculated to discover problematic regions. Comprehensive basal drag inversion experiments allow us to test the dependence of the L-curve and basal drag results on the non-linearity of sliding as well as on the inclusion of subglacial effective pressure in the friction law. The analysis suggest that non-linear friction laws are preferable to linear sliding because of reduced variance of the overall inferred friction coefficient, faster convergence, as well as steeper L-curves leading to a more accurate choice of weight for the regularization term. We show that a Budd-type friction law that incorporates effective pressure from a subglacial hydrology model rather than a simple geometry-based approximation achieves an improved performance in our inverse model. Further comparison reveals that the effective pressure from the hydrology model accounts for a larger part of spatial structure of the basal drag coefficient than the parameterized effective pressure. Allowing the inverted drag coefficient to more precisely reflect actual variations in basal properties. Finally, a comparison of the inferred basal drag across WAIS with observed locations of subglacial lakes reveals a good match, giving us additional confidence in the spatial basal drag structure revealed by the inverse method.

  • Open Access

    Improved basal drag of the West Antarctic Ice Sheet from L-curve analysis of inverse models utilizing subglacial hydrology simulations

    by Lea-Sophie Höyns, Thomas Kleiner, Andreas Rademacher, Martin Rückamp, Michael Wolovick, Angelika Humbert

    The West Antarctic Ice Sheet (WAIS) is the focus of current research due to its susceptibility to collapse, which could potentially contribute to rising sea levels. To accurately predict future glacier evolution, precise ice sheet models are essential with regard to suitable approximations of physical behavior to the real system and appropriate input values, as well as computing power. The ice discharge of outlet glaciers into the ocean is one key factor here, primarily caused by basal sliding of ice. Since we cannot directly measure basal properties on a large scale, inverse models can be used to infer the basal drag coefficient by minimizing a cost function that depends on a velocity misfit and a regularization term. We conduct basal drag inversions and perform L-curve analyses to find the optimal trade-off between the cost function terms, ending up with smooth L-curves. Additionally, the domain L-curve is break down to eight subdomains of the study area in order to reveal how well the inverse method performs in different glaciological settings. It reveals that Pine Island Glacier being the best area and slow-flowing areas such as Roosevelt Island being among the worst in terms of the L-curve behavior for the basal drag inversion. This highlights the importance of performing a subdomain L-curve analysis, whenever an inversion for a larger domain is calculated to discover problematic regions. Comprehensive basal drag inversion experiments allow us to test the dependence of the L-curve and basal drag results on the non-linearity of sliding as well as on the inclusion of subglacial effective pressure in the friction law. The analysis suggest that non-linear friction laws are preferable to linear sliding because of reduced variance of the overall inferred friction coefficient, faster convergence, as well as steeper L-curves leading to a more accurate choice of weight for the regularization term. We show that a Budd-type friction law that incorporates effective pressure from a subglacial hydrology model rather than a simple geometry-based approximation achieves an improved performance in our inverse model. Further comparison reveals that the effective pressure from the hydrology model accounts for a larger part of spatial structure of the basal drag coefficient than the parameterized effective pressure. Allowing the inverted drag coefficient to more precisely reflect actual variations in basal properties. Finally, a comparison of the inferred basal drag across WAIS with observed locations of subglacial lakes reveals a good match, giving us additional confidence in the spatial basal drag structure revealed by the inverse method.

  • Top Cited Papers
    Open Access

    Deep learning methods for partial differential equations and related parameter identification problems

    by Derick Nganyu Tanyu, Jianfeng Ning, Tom Freudenberg, Nick Heilenkötter, Andreas Rademacher, Uwe Iben, Peter Maass

    Recent years have witnessed a growth in mathematics for deep learning—which seeks a deeper understanding of the concepts of deep learning with mathematics and explores how to make it more robust—and deep learning for mathematics, where deep learning algorithms are used to solve problems in mathematics. The latter has popularised the field of scientific machine learning where deep learning is applied to problems in scientific computing. Specifically, more and more neural network (NN) architectures have been developed to solve specific classes of partial differential equations (PDEs). Such methods exploit properties that are inherent to PDEs and thus solve the PDEs better than standard feed-forward NNs, recurrent NNs, or convolutional neural networks. This has had a great impact in the area of mathematical modelling where parametric PDEs are widely used to model most natural and physical processes arising in science and engineering. In this work, we review such methods as well as their extensions for parametric studies and for solving the related inverse problems. We also show their relevance in various industrial applications.


  • Open Access

    Dual weighted residual error estimation for the finite cell method

    by Paolo Di Stolfo, Andreas Rademacher, Andreas Schröder

    The paper presents a goal-oriented error control based on the dual weighted residual method (DWR) for the finite cell method (FCM), which is characterized by an enclosing domain covering the domain of the problem. The error identity derived by the DWR method allows for a combined treatment of the discretization and quadrature error introduced by the FCM. We present an adaptive strategy with the aim to balance these two error contributions. Its performance is demonstrated for several two-dimensional examples.