• Korean Journal of Remote Sensing
• /
• v.38 no.6_2
• /
• pp.1723-1735
• /
• 2022

Landslides are one of the most prevalent natural disasters, threating both humans and property. Also landslides can cause damage at the national level, so effective prediction and prevention are essential. Research to produce a landslide susceptibility map with high accuracy is steadily being conducted, and various models have been applied to landslide susceptibility analysis. Pixel-based machine learning models such as frequency ratio models, logistic regression models, ensembles models, and Artificial Neural Networks have been mainly applied. Recent studies have shown that the kernel-based convolutional neural network (CNN) technique is effective and that the spatial characteristics of input data have a significant effect on the accuracy of landslide susceptibility mapping. For this reason, the purpose of this study is to analyze landslide vulnerability using a pixel-based deep neural network model and a patch-based convolutional neural network model. The research area was set up in Gangwon-do, including Inje, Gangneung, and Pyeongchang, where landslides occurred frequently and damaged. Landslide-related factors include slope, curvature, stream power index (SPI), topographic wetness index (TWI), topographic position index (TPI), timber diameter, timber age, lithology, land use, soil depth, soil parent material, lineament density, fault density, normalized difference vegetation index (NDVI) and normalized difference water index (NDWI) were used. Landslide-related factors were built into a spatial database through data preprocessing, and landslide susceptibility map was predicted using deep neural network (DNN) and CNN models. The model and landslide susceptibility map were verified through average precision (AP) and root mean square errors (RMSE), and as a result of the verification, the patch-based CNN model showed 3.4% improved performance compared to the pixel-based DNN model. The results of this study can be used to predict landslides and are expected to serve as a scientific basis for establishing land use policies and landslide management policies.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.14 no.2
• /
• pp.102-117
• /
• 2009

This study was carried out to investigate the extent to which benthic environment of Ulsan Bay was disturbed by organic materials and trace metals from the megacity and industrial complex. Field survey for benthic environment and macroinvertebrate community was seasonally conducted from February to November 2006 at nine stations covering the inside and outside of the bay. TOC was on average 1.7% while four (As, Cu, Pb, Zn) out of seven trace metals measured exceeded the Effects Range Low (ERL) in most of the stations. Total number of species sampled was 199 and mean density was 4,578 ind./$m^2$, both of which were greatly dominated by the polychaetes. Dominant species were Aphelochaeta monilaris (22.6%), Ruditapes philippinarum (17.1%), Magelona japonica (12.2%), Lumbrineris longifolia (9.9%) and their distribution was ruled by the difference in the benthic environmental condition of each station. From the multivariate analyses, four stational groups were identified: northern part of the bay, middle and lower part of the bay, the intersection of Taewha River and Gosa stream and outside of the bay. As a result, the community heterogeneity of inner bay was much more greater than that of outer bay. SIMPER analysis showed that four groups were represented by R. philippinarum-Capitella capitata, A. monilaris-Balanoglossus carnosus, Sinocorophium sinensis-Cyathura higoensis and M. japonica-Ampharete arctica, respectively. Spatio-temporal changes of macroinvertebrate communities in Ulsan Bay were closely related to those of depth, mean grain size and organic content, and Zn was also a meaningful factor in that context.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.28 no.3
• /
• pp.279-293
• /
• 1995

The most obvious and easily recognizable sources of potential water pollution are point sources such as domestic and industrial wastes. But recently, the potential effects of nonpoint sources on water quality have been increased apparently. In order to evaluate the characteristics and the effects of nonpoint sources on water quality, this study was performed in Suyeong Bay from May, 1992 to July, 1992. The depth-averaged 2-dimensional numerical model, which consists of the hydrodynamic model and the diffusion model was applied to simulate the water quality in Suyeong Bay. When flowrate was $65.736m^3/s,$ the concentration of pollutants (COD, TSS and VSS) at Oncheon stream (Sebeong bridge) during second flush were very high as much as 121.4mg/l of COD, 1148.0mg/l of TSS and 262.0mg/1 of VSS. When flowrate was 4.686m^3/s, the concentration of pollutants $(TIN,\;NH_4\;^+-\;N,\;NO_2\;^--N\;and\;PO_4\;^{3-}-P)$ during the first flush were very high as much as 20.306mg/1 of TIN, 14.154mg/1 of $NH_4\;^+-N$, 9.571mg/l of $NO_2\;^--N$ and l.785mg/l of $PO_2\;^{3-}-P$ As results of the hydrodynamic model simulation, the computed maximum velocity of tidal currents in Suyeong Bay was 0.3m/s and their direction was clockwise flow for ebb tide and counter clockwise flow for Hood tide. Four different methods were applied for the diffusion simulation in Suyeong Bay. There were the effects for the water quality due to point loads, annual nonpoint loads and nonpoint loads during the wet weather and the investigation period, respectively. The efforts of annual nonpoint loads and nonpoint loads during the wet weather seem to be slightly deteriorated in comparison with the effects of point loads. However, the bay was significantly polluted by the nonpoint loads during the investigation period. In this case, COD and SS concentrations ranged 2.0-30.0mg/l, 7.0- 200.0mg/l in ebb tide, respectively. From these results, it can be emphasized that the large amount of pollutants caused by nonpoint sources during the wet weather were discharged into the bay, and affected significantly to both the water quality and the marine ecosystem. Therefore, it is necessary to consider the loadings of nonpoint pollutants to plan wastewater treatment plant.