Abstract
A geo-spatial data integration method for three-dimensional subsurface stratification is proposed in this study. The proposed method integrates the boring data modified with the cross-validation based outlier detection method and the geophysical testing results using indicator kriging to offer the appropriate criteria of P-wave velocity, which are derived site specifically to classify the local geomaterials for dam site. Cross-validation for the outlier analysis of boring data is a test to evaluate the susceptibility of variogram models or kriging models and to reduce the statistical uncertainty of the boring data, and indicator kriging, the integration method, is characterized by geostatistical non-linear procedures to model the variability of spatial attributes. Using the integration method, the site-specific criteria of geomaterials are determined. The computer software is developed for the proposed method with ArcGIS developer tool and GSLIB. The results show that this proposed method presents more reliable stratification results than the conventional classification criteria.
| Original language | English |
|---|---|
| Article number | 168 |
| Pages (from-to) | 1-10 |
| Number of pages | 10 |
| Journal | Environmental Earth Sciences |
| Volume | 75 |
| Issue number | 2 |
| DOIs | |
| State | Published - 1 Jan 2016 |
Keywords
- Dam site
- Geo-spatial data integration
- Indicator kriging
- Outlier analyses
- Subsurface stratification