Stochastic estimation of soil hydraulic conductivity utilizing self-organizing map method
This study proposes an unsupervised Self-Organizing Map (SOM) approach to enhance saturated hydraulic conductivity (ksat) estimation. Using the extensive FLSOIL database of 6,487 soil samples from Florida, the SOM-based ksat estimation model is optimized based on map size and feature selection, then...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-06-01
|
| Series: | Soils and Foundations |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0038080625000356 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This study proposes an unsupervised Self-Organizing Map (SOM) approach to enhance saturated hydraulic conductivity (ksat) estimation. Using the extensive FLSOIL database of 6,487 soil samples from Florida, the SOM-based ksat estimation model is optimized based on map size and feature selection, then compared with seven empirical equations and three supervised machine learning models. Unlike the other methods, the SOM-based approach provides a probabilistic distribution of ksat, enabling reliability-based design-value determination. Moreover, refining input features particularly by including specific surface area and Kozeny–Carman derived formulas improves accuracy and mitigates bias by the model features, especially in fine-grained soils. |
|---|---|
| ISSN: | 2524-1788 |