Groundwater Productivity Potential Mapping Using Logistic Regression and Boosted Tree Models: The Case of Okcheon City in Korea
This study analyzed Groundwater Productivity Potential (GPP) using different models in a geographic information system (GIS) in Okcheon area, Korea. These models used the relationship between groundwater-productivity data, including specific capacity (SPC) and transmissivity (T), and its related hydrogeological factors. Data about related factors, including topography, lineament, geology, forest and soil were constructed to a spatial database. Additionally, T and SPC data were collected from 86 well locations. Then, GPP were mapped using the Logistic Regression (LR) and Boosted Tree Regression (BT) models. The resulting GPP maps were validated using Area Under Curve (AUC) analysis with the well data. The GPP maps using the LR and BT models had accuracies of 85.04 and 81.66% with T value, respectively. And the GPP maps using the LR and BT models had accuracies of 82.22 and 81.53% with SPC value, respectively. These results indicate that LR and BT models can be useful for GPP mapping.
KeywordsGroundwater potential GIS Logistic regression Boosted tree Korea
- 2.Lee, S., Hong, S.-M., & Jung, H.-S.: GIS-based groundwater potential mapping using artificial neural network and support vector machine models: the case of Boryeong city in Korea. Geocarto. Int., 1–15 (2017)Google Scholar