| Title: | Modeling hydrological functioning of karst aquifer systems in Slovenia using geomorphological features and random forest algorithm |
|---|
| Authors: | ID Janža, Mitja (Author)
ID Hudovernik, Valter (Author)
ID Serianz, Luka (Author)
ID Stroj, Andrej (Author) |
|---|
| Files: |  PDF - Presentation file, download (7,40 MB)
MD5: A3BAFC58CF092CB1B3DFD1E770598E0A
 URL - Software, visit https://doi.org/10.1016/j.ejrh.2025.102774
URL - Supplement, visit https://www.sciencedirect.com/science/article/pii/S2214581825006032?via%3Dihub#sec0140
|
|---|
| Language: | English |
|---|
| Typology: | 1.01 - Original Scientific Article |
|---|
| Organization: |  GeoZS - Geological Survey of Slovenia
|
|---|
| Abstract: | Study region: Slovenia
Study focus: This study investigates the relationship between the hydrological functioning of karst aquifer systems and the geomorphological characteristics of their catchments. It is based on the analysis of discharge time series from 15 karst springs. Hydrograph analysis of these time series was used to estimate 11 hydrological parameters that characterize aquifer system functioning. A spatial analysis of morphological, geological, and hydrological data was carried out to assess 7 lumped geomorphological features of the catchments. These features (independent variables) and hydrological parameters (dependent variables) were used to develop random forest models for predicting the hydrological functioning of karst springs.
New hydrological insights for the region: The developed methodological approach provides a basis for improved characterization and prediction of the hydrological functioning of ungauged karst systems. Groundwater availability in these systems is largely controlled by aquifer retention capacity and spring discharge variability. These characteristics can be inferred from hydrological parameters that can be predicted using the developed random forest models. Feature importance analysis indicated that catchment area, cave density, and slope gradient are the most important geomorphological features for predicting the hydrological characteristics of spring discharge. |
|---|
| Keywords: | karst aquifer, random forest, machine learning, ungauged catchment, spring discharge, recession curve analysis |
|---|
| Publication status: | Published |
|---|
| Publication version: | Version of Record |
|---|
| Publication date: | 19.09.2025 |
|---|
| Publisher: | Elsevier |
|---|
| Year of publishing: | 2025 |
|---|
| Number of pages: | 16 str. |
|---|
| Numbering: | vol. 62, [article no.] 102774 |
|---|
| PID: | 20.500.12556/DiRROS-23666  |
|---|
| UDC: | 556.3 |
|---|
| ISSN on article: | 2214-5818 |
|---|
| DOI: | 10.1016/j.ejrh.2025.102774 |
|---|
| COBISS.SI-ID: | 249995267 |
|---|
| Note: |
|
|---|
| Publication date in DiRROS: | 22.10.2025 |
|---|
| Views: | 168 |
|---|
| Downloads: | 106 |
|---|
| Metadata: |    |
|---|
|
:
|
Copy citation |
|---|
| | | | Share: |  |
|---|
Hover the mouse pointer over a document title to show the abstract or click
on the title to get all document metadata. |
