• Journal of the Korean Geotechnical Society
• /
• v.37 no.5
• /
• pp.47-63
• /
• 2021

Recently, the demand for three-dimensional (3D) underground maps from the perspective of digital twins and the demand for linkage utilization are increasing. However, the vastness of national geotechnical survey data and the uncertainty in applying geostatistical techniques pose challenges in modeling underground regional geotechnical characteristics. In this study, an optimal learning model based on multi-layer perceptron (MLP) was constructed for 3D subsurface lithological and geotechnical classification in Seoul, South Korea. First, the geotechnical layer and 3D spatial coordinates of each borehole dataset in the Seoul area were constructed as a geotechnical database according to a standardized format, and data pre-processing such as correction and normalization of missing values for machine learning was performed. An optimal fitting model was designed through hyperparameter optimization of the MLP model and model performance evaluation, such as precision and accuracy tests. Then, a 3D grid network locally assigning geotechnical layer classification was constructed by applying an MLP-based bet-fitting model for each unit lattice. The constructed 3D geotechnical layer map was evaluated by comparing the results of a geostatistical interpolation technique and the topsoil properties of the geological map.

• Journal of the Korean Geotechnical Society
• /
• v.38 no.4
• /
• pp.59-70
• /
• 2022

There is a growing interest in evaluating earthquake damage and determining disaster prevention measures due to the magnitude 5.8 earthquake in Pohang, Korea. Since the liquefaction phenomena occurred extensively in the residential area as a result of the earthquake, there was a demand for research on liquefaction phenomenon evaluation and liquefaction disaster prediction. Liquefaction is defined as a phenomenon where the strength of the ground is completely lost due to a sudden increase in excess pore water pressure caused due to large dynamic stress, such as an earthquake, acting on loose sand particles in a short period of time. The liquefaction potential index, which can identify the occurrence of liquefaction and predict the risk of liquefaction in a targeted area, can be used to create a liquefaction hazard map. However, since liquefaction assessment using existing field testing is predicated on a single borehole liquefaction assessment, there has been a representative issue for the whole targeted area. Spatial interpolation and geographic information systems can help to solve this issue to some extent. Therefore, in order to solve the representative problem of geotechnical information, this research uses the kriging method, one of the geostatistical spatial interpolation techniques, and constructs a geotechnical information database for liquefaction and spatial interpolation. Additionally, the liquefaction hazard map was created for each return period using the constructed geotechnical information database. Cross validation was used to confirm the accuracy of this liquefaction hazard map.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1035-1046
• /
• 2022

This study evaluated the applicability of multi-resolution data fusion for land-cover classification. In the applicability evaluation, a spatial time-series geostatistical deconvolution/fusion model (STGDFM) was applied as a multi-resolution data fusion model. The study area was selected as some agricultural lands in Iowa State, United States. As input data for multi-resolution data fusion, Moderate Resolution Imaging Spectroradiometer (MODIS) and Landsat satellite images were used considering the landscape of study area. Based on this, synthetic Landsat images were generated at the missing date of Landsat images by applying STGDFM. Then, land-cover classification was performed using both the acquired Landsat images and the STGDFM fusion results as input data. In particular, to evaluate the applicability of multi-resolution data fusion, two classification results using only Landsat images and using both Landsat images and fusion results were compared and evaluated. As a result, in the classification result using only Landsat images, the mixed patterns were prominent in the corn and soybean cultivation areas, which are the main land-cover type in study area. In addition, the mixed patterns between land-cover types of vegetation such as hay and grain areas and grass areas were presented to be large. On the other hand, in the classification result using both Landsat images and fusion results, these mixed patterns between land-cover types of vegetation as well as corn and soybean were greatly alleviated. Due to this, the classification accuracy was improved by about 20%p in the classification result using both Landsat images and fusion results. It was considered that the missing of the Landsat images could be compensated for by reflecting the time-series spectral information of the MODIS images in the fusion results through STGDFM. This study confirmed that multi-resolution data fusion can be effectively applied to land-cover classification.

• Economic and Environmental Geology
• /
• v.56 no.3
• /
• pp.331-341
• /
• 2023

When creating a soil contamination map using geostatistical techniques, there are various sources that can affect prediction errors. In this study, a grid-based soil contamination map was created from the sampling data of heavy metal concentrations in soil in abandoned mine areas using Ordinary Kriging. Five factors that were judged to affect the prediction error of the soil contamination map were selected, and the variation of the root mean squared error (RMSE) between the predicted value and the actual value was analyzed based on the Leave-one-out technique. Then, using a machine learning algorithm, derived the top three factors affecting the RMSE. As a result, it was analyzed that Variogram Model, Minimum Neighbors, and Anisotropy factors have the largest impact on RMSE in the Standard interpolation. For the variogram models, the Spherical model showed the lowest RMSE, while the Minimum Neighbors had the lowest value at 3 and then increased as the value increased. In the case of Anisotropy, it was found to be more appropriate not to consider anisotropy. In this study, through the combined use of geostatistics and machine learning, it was possible to create a highly reliable soil contamination map at the local scale, and to identify which factors have a significant impact when interpolating a small amount of soil heavy metal data.

• Journal of the Korean Geographical Society
• /
• v.47 no.3
• /
• pp.390-406
• /
• 2012

Recently, stone walls in Jeju Island have gained more attention as their importance in cultural value and tourism source got recognized. In addition, there are several projects to restore the walls underway at local governments' level. This study was conducted with the idea that it is also necessary to find out the background to build stone walls and its implication. The study mainly investigated the height of 'Uldam' or a stone wall surrounding the property including the house by using wind maps, geological maps and interview with local residents. The study found out that stone walls are more likely to be higher than others when they were built to protect the property against wind and stood against maximum wind and easily catched stone surround house.

• Journal of Environmental Impact Assessment
• /
• v.20 no.6
• /
• pp.879-890
• /
• 2011

The GOSAT (Greenhouse gases Observing SATellite) data provide new opportunities the most regionally complete and up-to-date assessment of $CO_2$. However, in practice, GOSAT records often suffer from missing data values mainly due to unfavorable meteorological condition in specific time periods of data acquisition. The aim of this research was to identify optimal spatial interpolation techniques to ensure the continuity of $CO_2$ from samples taken in the North East Asia. The accuracy among ordinary kriging (OK), universal kriging (UK) and simple kriging (SK) was compared based on the combined consideration of $R^2$ values, Root Mean Square Error (RMSE), Mean Error (ME) for variogram models. Cross validation for 1312 random sampling points indicate that the (UK) kriging is the best geostatistical method for spatial predictions of $CO_2$ in the East Asia region. The results from this study can be useful for selecting optimal kriging algorithm to produce $CO_2$ map of various landscapes. Also, data users may benefit from a statistical approach that would allow them to better understand the uncertainty and limitations of the GOSAT sample data.

• Tunnel and Underground Space
• /
• v.5 no.4
• /
• pp.336-346
• /
• 1995

Geotechnical Information System (GTIS) for efficient management of three dimensional borehole data has been developed. Some problems were raised during the input process of borehole data, and alternative solutions were sought. According to the previous geotechnical reports, there is no unified weathering classification scheme. A criterion, 100 times/30cm from SPT, was turned out inappropriate to the discrimination of weathered rock from weathered soil. It has also been suggested that weathered soil, weathered rock, soft rock, and hard rock should be defined as CW, HW, MW, and SW~fresh condition. For better comparison of RQD, the use of NX size coring is recommended for the whole area although BX size coring has been used in excavated area. The limit of drilling depth up to 1 m from the top of surface of hard rock should be extended to avoid possible wrong interpretation of rock head due to the existence of corestone. The input data were analysed by geostatistical methods. It is found that the range in semivariogram is about 300m, and the variance of gneiss is greater than that of granite. It is because the granite data analysed came from almost single uniform rock mass(i.e.Seoul granite), but gneiss data came from the rock mass(i.e. Gyeonggi gneiss complex experienced several metamorphic metamorphic processes.

• Economic and Environmental Geology
• /
• v.30 no.5
• /
• pp.493-498
• /
• 1997

The purpose of this study is to develop geostatistical model for evaluating the abundance of deep-sea manganese nodule. The abundance data used in this study were obtained from the KODOS (Korea Deep Ocean Study) area. The variation of nodule abundance was very high within short distance, while sampling methods was very limited. As the distribution of nodule abundance showed non-gaussian, indicator simulation method was used instead of conditional simulation method and/or ordinary kriging. The abundance data were encoded into a series of indicators with 6 cutoff values. They were used to estimate the conditional probability distribution function (cpdf) of the nodule abundance at any unsampled location. The standardized indicator variogram models were obtained according to variogram analysis. This SIS method had the advantage over other traditional techniques such as the turning bands method and ordinary kriging. The estimating values by indicator conditional simulation near high abundance area were more detailed than by ordinary kriging and indicator kriging. They also showed better spatial characteristics of distribution of nodule abundance.

• The Korean Journal of Applied Statistics
• /
• v.18 no.2
• /
• pp.271-280
• /
• 2005

Kriging methods as traditional spatial data analysis methods and geographical weighted regression models as statistical analysis methods are compared. In this paper, we apply data from the Ministry of Environment to spatial analysis for practical study. We compare these methods to performance with monthly carbon monoxide observations taken at 116 measuring area of air pollution in 1999.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.20 no.3
• /
• pp.171-180
• /
• 2016

Earthquake event keeps increasing every year, and the recent cases of earthquake hazards invoke the necessity of seismic study in Korea, as geotechnical earthquake hazards, such as strong ground motion, liquefaction and landslides, are a significant threat to structures in industrial hub areas including coastal facilities. In this study, systemized framework of integrated assessment of earthquake-induced geotechnical hazard was established using advanced geospatial database. And a visible simulation of the framework was specifically conducted at two coastal facility areas in Incheon. First, the geospatial-grid information in the 3D domain were constructed with geostatistical interpolation method composed of multiple geospatial coverage mapping and 3D integration of geo-layer construction considering spatial outliers and geotechnical uncertainty. Second, the behavior of site-specific seismic responses were assessed by incorporating the depth to bedrock, mean shear wave velocity of the upper 30 m, and characteristic site period based on the geospatial-grid. Third, the normalized correlations between rock-outcrop accelerations and the maximum accelerations of each grid were determined considering the site-specific seismic response characteristics. Fourth, the potential damage due to liquefaction was estimated by combining the geospatial-grid and accelerations correlation grid based on the simplified liquefaction potential index evaluation method.