<p>Hydrogeophysics is interdisciplinary by definition. As researchers strive to gain quantitative information on process-relevant subsurface parameters while integrating non-geophysical measurements, multi-physical geoscientific models are often developed that simulate the dynamic process and its geophysical response. Such endeavors are associated with considerable technical challenges due to coupling of different numerical models, which represents an initial hurdle for students and many practitioners. Even technically versatile users often end up with individually tailored solutions at the cost of scientific reproducibility.</p><p>We argue that the reproducibility of studies in computational hydrogeophysics, and therefore the advancement of the field itself, needs versatile open-source software. One example is pyGIMLi - a flexible and computationally efficient framework for modeling and inversion in geophysics. The library provides management for structured and unstructured 2D and 3D meshes, finite-element and finite-volume solvers, various geophysical forward operators, as well as a generalized Gauss-Newton based inversion framework.</p><p>In this contribution, we highlight some of the recent advances and use cases in research and education since its 1.0 release in 2017 (R&#252;cker et al., 2017) including:</p><ul><li>generalized modeling and inversion frameworks for conventional, joint, time-lapse and process-based inversion</li>
<li>geostatistical regularization operators for unstructured meshes (Jordi et al., 2018)</li>
<li>improved constraints in the presence of petrophysical parameter transformations demonstrated by an estimation of water, ice, and air in partially frozen systems (Wagner et al., 2019)</li>
<li>3D visualization leveraging upon PyVista (Sullivan and Kaszynski, 2019)</li>
<li>simulation of electrical streaming potentials</li>
<li>complex-valued forward modeling and inversion of induced polarization</li>
<li>forward modeling with anisotropic parameters</li>
<li>availability for Mac OS</li>
<li>improved API and documentation</li>
</ul><p>Since the library is freely available and platform-compatible, it is also well suited for teaching. We demonstrate examples from Master level university courses and public outreach, where learners can interactively change model and acquisition parameters to study their influence on a hydrogeophysical process simulation. Finally, we would like to use this opportunity to discuss future developments with the community.</p><p><strong>References</strong></p><p><span>Jordi, C., Doetsch, J., G&#252;nther, T., Schmelzbach, C., & Robertsson, J. O. (2018). Geostatistical regularization operators for geophysical inverse problems on irregular meshes. <em>Geophysical Journal International</em>, 213(2), 1374&#8211;1386. </span><span></span></p><p><span>R&#252;cker, C., G&#252;nther, T., Wagner, F.M., 2017. pyGIMLi: An open-source library for modelling and inversion in geophysics, <em>Computers and Geosciences</em>, 109, 106-123. </span><span></span></p><p><span>Sullivan, C., & Kaszynski, A. (2019). PyVista: 3D plotting and mesh analysis through a streamlined interface for the Visualization Toolkit (VTK). <em>Journal of Open Source Software</em>, 4(37), 1450. </span><span></span></p><p><span>Wagner, F. M., Mollaret, C., G&#252;nther, T., Kemna, A., & Hauck, C. (2019). Quantitative imaging of water, ice and air in permafrost systems through petrophysical joint inversion of seismic refraction and electrical resistivity data. <em>Geophysical Journal International</em>, 219(3), 1866&#8211;1875. </span><span></span></p>
Gym-ANM: Open-source software to leverage reinforcement learning for power system management in research and education
