MiGIS: micromorphological soil and sediment thin section analysis using an open-source GIS and machine learning approach

Zickel, Mirijam; Gröbner, Marie; Röpke, Astrid; Kehl, Martin

doi:https://doi.org/10.5194/egqsj-73-69-2024

Articles | Volume 73, issue 1

https://doi.org/10.5194/egqsj-73-69-2024

© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/egqsj-73-69-2024

© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 73, issue 1

Research article

|

26 Jan 2024

Research article |

| 26 Jan 2024

MiGIS: micromorphological soil and sediment thin section analysis using an open-source GIS and machine learning approach

Mirijam Zickel, Marie Gröbner, Astrid Röpke, and Martin Kehl

Model code and software

MiGIS toolbox for QGIS 3 Mirijam Zickel and Marie Gröbner https://doi.org/10.5281/zenodo.10527165

GDAL, version 3.5.1 Even Rouault et al. https://doi.org/10.5281/zenodo.6801315

lennepkade/dzetsaka: Fix bug in processing provider with vector files (Dzetsaka QGIS classification plugin), version 3.5.1 Nicolas Karasiak https://doi.org/10.5281/zenodo.3463523

qgis/QGIS: 3.22.10 (final-3_22_10), version 3.22 qgis-bot. https://doi.org/10.5281/zenodo.7986774

scikit-learn/scikit-learn: scikit-learn 1.0.1, version 1.0.0, Olivier Grisel et al. https://doi.org/10.5281/zenodo.5596244

Short summary

With our open-source toolbox, MiGIS for QGIS 3, we intend to advance digital micromorphological analysis. This approach focuses on the classification of micromorphological constituents based on their distinct colour values (multi-RGB signatures), acquired using flatbed scanning of thin sections in different modes (transmitted, cross-polarised, and reflected light). The resulting thin section maps enable feature quantification, visualisation of spatial patterns, and reproducibility.