Physics-informed Multi-task Neural Networks for Joint Mapping of Hydraulic Conductivity and Preferential Flow Paths in Fractured Aquifers: PI-XNET

This repository contains the data and code used in the research work titled "Physics-informed Multi-task Neural Networks for Joint Mapping of Hydraulic Conductivity and Preferential Flow Paths in Fractured Aquifers: PI-XNET". It is advisable to install the necessary Python libraries before using the code. Please find the download link for the Python-based MODFLOW groundwater modeling suite (https://files.pythonhosted.org/packages/cd/db/25c68f8f945e381a8e406f41324c3a7815bcc9d4399b98fb5b7c1680c86f/modflow_setup-0.6.1.tar.gz). Additionally, the binary executable files for MODFLOW (mf6) are available for both Windows and Linux operating systems in the repository. It is required to install the Python package "FloPy" for MODFLOW to simulate steady-state groundwater heads. Use either "pip install flopy" or "conda forge" to install the FloPy package. Please ensure that the binary executable files are saved in your working directory when executing the code. If any parts of the code are used or reproduced, the original source and author must be cited.