Teaching GPR Data Processing Methods Using Jupyter Notebooks (from A-Scan to C-Scan)

Authors:
Publisher website
Google Scholar
Abstract
Ground Penetrating Radar (GPR) data processing is often performed using proprietary, GUI-based software that lacks flexibility for automation, customization, specialization, and integration into narratives, particularly for teaching purposes. To address this limitation, we developed interactive processing suite based on the Python Scientific Software Stack and the Jupyter Notebook environment, designed to enhance teaching and learning experiences in classrooms, field courses, and student internships. Our processing suite includes essential GPR data processing techniques such as frequency filtering, mean trace removal, various gain methods, singular value decomposition, hyperbola analysis, and Stolt migration. Additionally, it enables data visualization through A-scan, B-scan, and C-scan representations, with amplitude maps for enhanced spatial analysis. The integration of interactive widgets further demonstrates the versatility of Jupyter Notebooks, allowing students to explore GPR data dynamically. To bridge theory and practice, we incorporated Jupyter notebooks that seamlessly interact with the numerical modeling software gprMax, offering students hands-on experience in data simulation and interpretation. This approach has been successfully applied in student-led projects, particularly in archaeological studies. Three examples are given in this paper. Student feedback has been positive, highlighting the effectiveness of this method in improving their understanding of GPR data processing. Several students reported gaining proficiency in using the software independently, reinforcing the value of interactive and open-source tools in geophysics education.

This publication has 11 references indexed in Scilit: