• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.164-164
• /
• 2023

This study investigates the importance of flood-influencing factors on the accuracy of flood risk mapping using the integration of remote sensing-based and machine learning techniques. Here, the Extreme Gradient Boosting (XGBoost) and Random Forest (RF) algorithms integrated with GIS-based techniques were considered to develop and generate flood risk maps. For the study area of NAPA County in the United States, rainfall data from the 12 stations, Sentinel-1 SAR, and Sentinel-2 optical images were applied to extract 13 flood-influencing factors including altitude, aspect, slope, topographic wetness index, normalized difference vegetation index, stream power index, sediment transport index, land use/land cover, terrain roughness index, distance from the river, soil, rainfall, and geology. These 13 raster maps were used as input data for the XGBoost and RF algorithms for modeling flood-prone areas using ArcGIS, Python, and R. As results, it indicates that XGBoost showed better performance than RF in modeling flood-prone areas with an ROC of 97.45%, Kappa of 93.65%, and accuracy score of 96.83% compared to RF's 82.21%, 70.54%, and 88%, respectively. In conclusion, XGBoost is more efficient than RF for flood risk mapping and can be potentially utilized for flood mitigation strategies. It should be noted that all flood influencing factors had a positive effect, but altitude, slope, and rainfall were the most influential features in modeling flood risk maps using XGBoost.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.3 no.3
• /
• pp.158-164
• /
• 1998

We introduce a statistical scheme to classify seabed from acoustic profiling data acquired using Chirp sonar system. The classification is based on grouping of signal traces by similarity index, which is computed using the K-L (Karhunen-Lo$\grave{e}$ve) transform of the Chirp profiling data. The similarity index represents the degree of coherence of bottom-reflected signals in consecutive traces, hence indicating the acoustic roughness of the seabed. The results of this study show that similarity index is a function of homogeneity, grain size of sediments and bottom hardness. The similarity index ranges from 0 to 1 for various types of seabed material. It increases in accordance with the homogeneity and softness of bottom sediments, whereas it is inversely proportional to the grain size of sediments. As a real data example, we classified the seabed off Cheju Island, Korea based on the similarity index and compared the result with side-scan sonar data and sediment samples. The comparison shows that the classification of seabed by the similarity index is in good agreement with the real sedimentary facies and can delineate acoustic response of the seabed in more detail. Therefore, this study presents an effective method for geoacoustic modeling to classify the seafloor directly from acoustic data.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.33 no.3
• /
• pp.965-974
• /
• 2013

Multiple debris flows occurred on July 27, 2012 in Mt. Umyeon, which resulted in 16 casualties and severe property demage. Accurate reproducing of the propagation and deposition of debris flow is essential for mitigating these disasters. Through applying FLO-2D model to these debris flows and comparing the results with field observations, we seek to evaluate the performance of the model and to analyse the rheological model parameters. Representative yield stress and dynamic viscosity back-calculated for the debris flows in the northern side of Mt. Umyeon are 1022 Pa and 652 $Pa{\cdot}s$, respectively. Numerical results obtained using these parameters reveal that deposition areas of debris flows in Raemian and Shindong-A regions are well reproduced in 63-85% agreement with the field observations. However, the propagation velocities of the flows are significantly underestimated, which is attributable to the inherent limitations of the model that can't take the entrainment of bed material and surface water into account. The debris flow deposition computed in Hyeongchon region where the entrainment is not significant appears to be in very good agreement with the field observation. The sensitivity study of the numerical results on model parameters shows that both sediment volume concentration and roughness coefficient significantly affect the flow thickness and velocity, which underscores the importance of careful selection of these model parameters in FLO-2D modeling.