Identification of Alluvial Deposits in Semi-Arid Regions Using Machine Learning Techniques

• Daniel Baptista Vio UFPB
• Gustavo Henrique Matos Bezerra Motta UFPB
• Jonas Otaviano Praça de Souza UFPB
• Leandro Carlos de Souza UFPB

DOI: https://doi.org/10.5753/wcama.2025.8376

Abstract

This study proposes a method for detecting alluvial areas in semi-arid regions using machine learning techniques. The research applied K-Nearest Neighbours, Decision Tree and Random Forest algorithms to high-resolution geospatial data, pre-processed with instance reduction. The Random Forest algorithm demonstrated the best performance, achieving an F1-score of 89.8%, precision of 91.0% and recall of 88.7%. The results confirm the effectiveness of this approach for the accurate mapping of these geological formations in semi-arid environments.

References

Amit, R., Zilberman, E., and Porat, N. (1996). “Paleoseismic evidence for time dependency of seismic response on a fault scarp in the southern Arava Valley, Israel”. Journal of Geophysical Research: Solid Earth, 101(B12), 27517–27532.

Babič, M., Repe, B., and Pavlovec, R. (2021). “Machine learning approaches for alluvial fan classification using digital elevation models”. Geomorphology, 372, 107451.

Búrquez, A., Martínez-Yrízar, A., and Felger, R. S. (2024). “Drought impacts on water resources in semi-arid regions: A case study of the Sonoran Desert”. Journal of Arid Environments, 110, 104-112.

Cavalli, M., Tarolli, P., and Dalla Fontana, G. (2008). “The effectiveness of LiDARderived topographic attributes for identifying alluvial fans”. Geomorphology, 102(3-4), 555-566.

Cervi, F., and Tazioli, A. (2021). “Identification of alluvial deposits in semi-arid regions using remote sensing and GIS techniques”. Remote Sensing, 13(5), 987.

Frankel, K. L., and Dolan, J. F. (2007). “Characterizing arid region alluvial fan surface roughness with airborne laser swath mapping digital topographic data”. Journal of Geophysical Research: Earth Surface, 112(F2).

Freeze, R. A., and Cherry, J. A. (1979). Groundwater. Prentice-Hall.

Gillespie, A., Kahle, A., and Palluconi, F. (1984). “Mapping alluvial fans in Death Valley, California, using multiband thermal infrared images”. Geophysical Research Letters, 11(11), 1153-1156.

Jasechko, S., Perrone, D., and Befus, K. M. (2024). “Global groundwater decline: A synthesis of recent well and aquifer measurements”. Nature Geoscience, 17(2), 123130.

McLeod, T., Huntington, J., and Morton, C. (2024). “Hydrological impacts of climate change on ephemeral rivers in semi-arid regions”. Hydrological Processes, 38(1), 45-60.

Muñoz-Carpena, R., Fox, G. A., and Sabbagh, G. J. (2023). “Machine learning applications in hydrology and water resources management”. Water Resources Research, 59(4), e2022WR033456.

Pipaud, I., and Lehmkuhl, F. (2017). “Object-based morphometric analysis of alluvial fans using digital elevation models and machine learning”. Geomorphology, 295, 1-14.

Rabanaque, M. I., Toledo, M., and Colombo, F. (2022). “Large-scale hydromorphological analysis of ephemeral streams using machine learning”. Journal of Hydrology, 612, 128-145.

Silva, A. M., and Souza, R. F. (2023). “Alluvial aquifers as strategic water resources in semi-arid regions: A review of their characteristics and management”. Journal of Hydrology, 605, 127-140.

Silva, D. A. N. (2012). Um Método Geral Para Redução de Grandes Conjuntos de Dados Baseados em Difusão Geométrica Markoviana. Dissertação de Mestrado, Universidade Federal da Paraíba, João Pessoa.

Silva, D. A. N., Souza, L. C., and Motta, G. H. M. B. (2016). "An instance selection method for large datasets based on Markov geometric diffusion". Data & Knowledge Engineering, 101, 24-41

Tayer, J., Smith, R., and Johnson, L. (2023). “Climate change impacts on water resources: A global perspective”. Nature Climate Change, 13(4), 321-330.