Keywords:
- soil texture
- artificial intelligence models
- remote sensing
- soil analysis
- modeling
Abstract
This article explores the use of artificial intelligence (AI) technologies for analyzing soil texture based on machine learning models and remote sensing data. Four datasets, combining archival and recent data with spectral indices, were used to evaluate model performance. The AdaBoost algorithm achieved the highest accuracy (68.2%) for soil texture prediction. The results demonstrate that AI models, especially those incorporating both historical and recent data, provide accurate and scalable solutions for soil texture analysis, contributing significantly to precision agriculture and sustainable land management.
References
McBratney, A.B., Santos, M.M., & Minasny, B. (2003). On digital soil mapping. Geoderma, 117(1-2), 3-52. https://doi.org/10.1016/S0016-7061(03)00223-4.
Minasny, B., McBratney, A.B., Malone, B.P., & Wheeler, I. (2013). Digital mapping of soil carbon. In: Sparks, D.L. (Ed.), Advances in Agronomy, 118, 1-47. Academic Press. https://doi.org/ 10.1016/B978-0-12-405942-9.00001-7.
Wälder, K., Wälder, O., Rinklebe, J., & Menz, J. (2008). Estimation of soil properties with geostatistical methods in floodplains. Archives of Agronomy and Soil Science, 54(3), 275-295. https://doi.org/10.1080/03650340801900617.
Hengl, T., et al. (2015). Mapping soil properties of Africa at 250 m resolution. PLoS One, 10(6), e0125814.
Taghizadeh-Mehrjardi, R., et al. (2016). Digital mapping of soil organic carbon at multiple depths. Geoderma, 266, 98-110.
McBratney, A.B., et al. (2003). On digital soil mapping. Geoderma, 117 (1-2), 3-52.
