Safety Evaluation of Major Nuclear Facilities According to Debris Flow Occurrence Scenarios

Article information

J. Ocean Eng. Technol. 2024;38(6):414-425
Publication date (electronic) : 2024 December 19
doi : https://doi.org/10.26748/KSOE.2024.083
1Researsh fellow, Division for Integrated Water Management, Korea Environment Institute, Sejong, Korea
2Senior Researcher, Research & Development Center, COEN Solution, Seoul, Korea
Corresponding author Seung-min: +82-2-2131-2962, skinsin@nate.com
Received 2024 October 16; Revised 2024 November 11; Accepted 2024 November 22.

Abstract

This study systematically analyzes the impact of sediment disasters, such as those due to debris flows, on key facilities near the Gori nuclear power plant, aiming to identify limitations in the current safety management system. Using 62 years of rainfall data from the Busan weather station, probable rainfall values for return periods of 30, 50, 100, and 200 years were analyzed. These values were applied to various debris flow scenarios to evaluate the flow characteristics and impact forces on structures. The analysis reveals that the impact forces of debris flows increase sharply for return periods of 100 years or more, indicating that existing safety protocols may not adequately account for the topographical characteristics. This study highlights the necessity of a comprehensive safety assessment framework that integrates site-specific risks such as slope and sediment dynamics. By utilizing actual rainfall data, this study provides a safety evaluation that reflects the unique geographical hazards faced by nuclear facilities. The findings offer essential data to guide future policy decisions regarding site selection, design, and operational management of nuclear facilities, ultimately improving the disaster preparedness for high-risk locations in a way that differs from existing studies.

1. Introduction

The Great East Japan Earthquake in 2011 was an accident that clearly showed the tremendous hazards of natural disasters to nuclear facilities, and the damage to the Fukushima nuclear power plant caused by the earthquake and ensuing tsunami had serious social, economic, and environmental impacts not only in Japan but across the world. According to a United Nations (2014) report, the Fukushima nuclear power plant accident had serious impacts on society as it led to a large-scale evacuation with the corresponding mental suffering, produced economic losses, and generated recovery burdens due to the radioactive leaks, which resulted in long-term environmental damage to ecosystems, including the marine environment. This accident has affected the human society, economy, and environment deeply and continuously, and such impacts indicated the need for long-term response and management in various areas. In addition, the residents had to evacuate (Hasegawa, 2013), and the long-term environmental damage caused by radioactive contamination still has effects. This accident served as an opportunity to raise awareness of nuclear safety worldwide, and reminded the urgent need to review the safety evaluation and management methods for nuclear facilities

In the Fukushima nuclear power plant accident (Fig. 1(a)), the absence of safety systems to prepare for such accidents as well as the natural disaster itself were pointed out as main causes. According to a United Nations (2014) report, external power supply was cut off by the earthquake and subsequent tsunami, and the damage to emergency generators and seawater cooling pumps led to the failure of the cooling system, which interrupted reactor cooling. Then, hydrogen accumulated and caused an explosion, resulting in the leakage of radioactive materials into the atmosphere and the sea. Because of the interruption of communication systems and lack of central control, on-site workers responded to the emergency situation without proper equipment or manuals, resulting in significant difficulties in the emergency response, accident response, and recovery work. The flood caused by the unexpected tsunami caused a series of accidents as it cut off power supply for reactor cooling. As shown in Fig. 1(b), power supply to nuclear power plants is usually realized through transmission towers. In nuclear power plants, damage to transmission facilities due to artificial impacts on transmission towers caused by disasters, such as landslides, can be very dangerous.

Fig. 1.

Nnatural disasters that can occur around nuclear power plants

As a result of the increasing frequency of severe rainfall, tsunamis, and high waves accompanied by typhoons, the number of cases involving simultaneous or complex disasters (e.g., landslides and floods) in coastal areas has also increased. To prepare for such situations, various disaster scenarios must be considered in advance. In particular, safety evaluation for disasters such as complex floods and landslides is important. The lessons learned from the Fukushima accident have led to various safety measures, including the selection of nuclear power plant sites considering geological risks, strengthening of emergency response systems, and improvements in physical safeguards to minimize damage in the event of an accident. Areas where nuclear power plants are located, however, are mostly security areas surrounded by mountainous terrains on the coast, and no study has been conducted to predict the damage to major facilities due to the debris flow caused by severe rainfall. The risk of landslides (debris flows) has not been closely evaluated, and research on the impacts of such disasters on nuclear facilities is still insufficient. Landslides can be caused by heavy rainfall, snowmelt, volcanic activities, or earthquakes. They can destroy infrastructure, block access routes, and cause structural damage to facilities. Therefore, thorough research on the effect of debris flows on nuclear plants is required.

As nuclear power plants are mostly located near the coast, studies have been actively conducted to investigate the direct causes of the damage caused by storm surges, high waves, and tsunamis. In particular, according to the results of a study, the exposure of nuclear power plants to a combination of two or more natural disasters may cause serious damage. The disasters that may occur around nuclear power plants are classified as shown in Fig. 2 (Choi et al., 2021). Kang and Hahm (2018) analyzed the interaction between earthquakes and landslides from a perspective of multiple disasters. Based on this, they comprehensively evaluated the risk of landslides by expanding the probabilistic safety assessment (PSA). In the PSA, they analyzed the landslide probability and travel distance based on the earthquake intensity and slope vulnerability, and re-quantified the key damage frequency. However, they pointed out the difficulties in collecting basic data for quantitative numerical analysis of the complex interactions between multiple disasters. Therefore, they evaluated the risk of earthquake-induced landslides by expanding the PSA framework, but underscored the need for a practical evaluation of the actual flow of sediment. Hwang et al. (2024), based on Typhoon Bolaven, emphasized the increasing uncertainty and vulnerability of coastal disaster prediction, and highlighted the need to improve the accuracy and speed of typhoon surge prediction amid the increasing climate crisis. In addition, wave predictions for each sea area were obtained through machine learning (Park and Kang, 2024), and the major factors determining the wave shear rate near low-crested and submerged structures (LCS) were evaluated using artificial neural networks (Kim et al., 2022). In addition, studies on the wave reduction effects of submerged breakwaters and seawalls have been conducted to predict damage to inland structures located along the coast and reduce high waves from the sea. The performance of flood mitigation measures against the land flow caused by tsunami-like waves was analyzed through numerical analysis (Dang et al., 2022), and the damage to buildings caused by typhoon-induced wind in coastal areas was predicted and evaluated (Kang et al., 2022).

Fig. 2.

Conceptual illustration of the effects of various natural hazards on a virtual nuclear power plant

Erosion control dams are mainly used to reduce the fluid force of debris flows. Kim et al. (2018) evaluated the static stability (sliding, overturning, and bearing capacity) and dynamic stability (member force) of dam structures, and the impact force acting on erosion control dams was found to range from 50 to 900 kPa depending on various natural environments (Proske et al., 2009). In addition, Lee et al. (2019) analyzed the changes in behavioral characteristics of a debris flow, such as its volume, thickness, speed, and diffusion range, using the DAN3D model. Accurate evaluation of the impact force acting on a structure by external factors is very important, and the reliability of a stability assessment can be ensured only through an accurate evaluation. However, few studies have been conducted on the evaluation of the impact force of debris flows that may occur in mountainous terrains on the coast. In an era of climate crisis, examinations based on multiple scenarios involving not only single disasters but also the simultaneous occurrence of various disasters are required, and preparing countermeasures based on them is urgently needed.

The purpose of this study is to identify the limitations of the current safety management system by systematically analyzing the impacts of sediment disasters (e.g., debris flows) on nuclear facilities, evaluating their impact forces on structures, and investigating their inundation characteristics. Specifically, the probability of rainfall duration was analyzed based on the 62-year (1961 to 2022) rainfall data from the Busan weather station near the Gori nuclear power plant, and the impact assessment was performed according to debris flow scenarios resulting from the probable rainfall for each return period. The analysis results showed that the change in impact force was insignificant until a return period of 30 or 50 years, but the impact force tended to rapidly increase as it increased from 100 to 200 years depending on the debris flow occurrence point. This indicates that the existing system does not fully consider the topographical characteristics, and a comprehensive safety evaluation system that reflects them is required. In addition, the effect of debris flows on structures and the risk they pose were analyzed specifically according to disaster occurrence scenarios to provide practical safety evaluation data for nuclear facilities. The data can be utilized to prepare guidelines for policy decision-making regarding the site selection, design, and operation of nuclear facilities located in high-risk areas, and they are expected to further strengthen the sediment disaster response capabilities.

2. External Force in Debris Flows

2.1 Landslide Impact Assessment Model

Various numerical analysis techniques (Table 1) have been developed in many studies, and research has been conducted to optimize the parameters required for each numerical analysis. To achieve the purpose of this study, an optimal numerical analysis model had to be adopted. This study required a model that could analyze the impact of the movement of sediment (a debris flow) by calculating the runoff sediment volume according to the maximum amount of rainfall that could occur during a typhoon, taking into account the rainfall characteristics and climate change. In particular, it was important to select a model able to reproduce the terrain around a nuclear power plant and evaluate the risk of major facilities inside the nuclear power plant. Two-dimensional (2D) models can analyze debris flow inundation in alluvial fans based on the research results of Flo-2D (Wu et al. 2012), Debris-2D (Chae et al., 2010), Rapid Mass Movement Simulation (RAMMS, Chang et al., 2024), Kanako 2D (Kim et al., 2016), and Debris Flow Urban-2D (DFU-2D). The Flo-2D model, however, cannot consider the inflow of erosive materials and rainfall. When the reproducibility of an actual debris flow event (Mt. Umyeon) was examined, it was found that the flow velocity was underestimated (Kim et al., 2013). Entrainment plays a very important role in analyzing debris flows. It changes the size and intensity of a debris flow, and the entrained materials may accumulate 10 to 50 times the initial soil moisture (Lee et al., 2020). Takahashi and Nakagawa (1991) developed an entrainment equation, which is used in the Kanko-2D, DFU-2D, and iRIC (Zhu et al., 2018) models. In particular, the DFU-2D model enables an inundation analysis that takes into account the structures (buildings) in alluvial fans.

Model selection

Therefore, this study adopted the DFU-2D model, which uses the same governing equations as the iRIC model and can utilize optimized parameters, as a model for simulating a debris flow, evaluating the effects of erosion control structures, and reproducing the urban topography so as to evaluate the impact of sediment disasters in areas around a nuclear power plant (Kim et al., 2020).

2.2 Calculation of External Forces

The peak flow rate of a debris flow was calculated based on the runoff sediment volume. The runoff sediment volume that may occur in the target area can be calculated using Eq. (1), which was presented by the National Institute for Land and Infrastructure Management (NILIM, 2016) of Japan. As for the peak flow rate of a debris flow, the external force is calculated using Eq. (2).

(1) Pv=103RTA1λ(Cd1Cd)fr
(2) Qsp=C*C*CdQp
(3) fr=0.05(logA2.0)2+0.05
(4) Cd=ρtanθ(σρ)(tanφtanθ)
where RT is the 24-h cumulative rainfall, A is the watershed area (km2), λ is the porosity, Cd is the volumetric sediment concentration of the debris flow (calculated as the debris flow concentration in the area where the debris flow occurred), fr is the runoff correction rate, C* (= 0.65) is the maximum volumetric sediment concentration of the debris flow, and Qp is the outflow, which is calculated using the synthetic rational method. Different design frequencies are applied to major facilities in Korea, and rainfall data analysis is required for each frequency suitable for use. Therefore, in this study, the probable rainfall by rainfall duration was analyzed using the past rainfall data (Busan), and the results are presented in Table 2. As it is difficult to investigate the characteristics of the sediment that constitutes the topography of the Gori nuclear power plant, the physical properties of the surrounding mountainous terrain were assumed.

Results of the analysis of probable rainfall by rainfall duration (Busan)

The parameters of the model for conducting a numerical analysis of debris flows in the mountainous area of the Gori nuclear power plant were set, and they are listed in Table 3. The amounts of sediment that may occur by rainfall scenario (probable rainfall frequency: 30, 50, 100, and 200 years) were calculated (Eq. (1)) and used as the boundary conditions (input) of the model. The computation time was set to 10 min (when the debris flow reaches the stationary state). Table 3 presents the analytical conditions used in this study according to debris flow occurrence scenarios.

Parameters for the model

2.3 Study Area

The Gori nuclear power plant is the first nuclear power plant in Korea, and its commercial operation started in April 1978 with of Gori Unit 1 (Fig. 3). As this facility is designated as a national security facility, the disclosure of many data is limited, including the structure’s specifications, aerial photographs, digital elevation model (DEM) data, and ground conditions. Therefore, based on the published data, the input data of the model were constructed by referring to the approximate location and form of the structure using Google Maps. In spite of the limitations for accurately reproducing the actual situation, it was considered possible to quantitatively analyze the flow field of debris flows.

Fig. 3.

Study area for the Gori nuclear power plant

In this study, if a landslide (debris flow) occurs in mountainous areas around the nuclear power plant site, the landslide risk map provided by the Korea Forest Service is not sufficient to closely examine its impact on major facilities in downstream areas. Therefore, debris flow inundation in downstream areas was evaluated by rainfall scenario, and a numerical debris flow analysis was conducted to analyze the impact forces acting on structures. In this process, for the numerical analysis of debris flows, various data are required on the major parameters and factors that cause landslides (nine factors). As most of the data are not disclosed, the model (Fig. 4) was constructed using major material properties of the past and surrounding areas.

Fig. 4.

Environment around the Gori nuclear power plant

3. Numerical Model

The flow characteristics and equilibrium equation of the modeled debris flow may be applied to various environments with steep slopes and potential debris flows. The inundation prediction of a debris flow was simulated by adopting an equation (Takahashi, 1991) that can be applied to steep slopes similar to the topographical characteristics of the Gori nuclear power plant. This method was proposed through on-site observation and hydraulic model testing of debris flows in mountainous terrains. Nakagawa et al. (2003) evaluated the spread and effect of debris flows on potential sediment erosion and deposition conditions for terrains similar to that of the Gori nuclear power plant. In addition, the parameters used in the model, such as the sediment concentration (C*) and internal fiction angle (ψ), were evaluated based on the soil characteristics. These values are essential for calculating the impact force and predicting the debris flow movement patterns around the target to be examined

3.1 Debris Flow Urban-2D (DFU-2D)

The DFU-2D model of Kim et al. (2015) was used to analyze 2D debris flows on steep slopes and evaluate impermeable erosion-control dams, and the reliability of the model (Kim et al., 2020) was evaluated through a comparison with the results of the hydraulic model experiment of Shrestha (2009). The configuration of the model is as follows.

3.1.1 Basic equations

As the basic equations in the Cartesian coordinate system of a debris flow consisting of a mixture of sediment and water, the momentum equations (Eqs. (5) and (6)), the continuity equation for the entire volume (Eq. (7)), and the continuity equations for coarse particles (Eq. (8)) and fine particles (Eq. (9)) are as follows:

(5) Mt+β(uM)x+β(vN)y=ghsinθbxoghcosθbxo(zb+h)xτbxρt
(6) Nt+β(uM)x+β(vN)y=ghsinθbyoghcosθbyo(zb+h)yτbyρt
(7) ht+Mx+Ny=ib
(8) (CLh)t+(CLM)x+(CLN)y={ibC*L(ib0)ibC*DL(ib<0)
(9) {(1CL)CFh}t+{(1CL)CFM}x+{(1CL)CFN}y={ib(1C*L)C*F(ib0)ib(1C*DL)C*F(ib<0)
(10) zbt+ib=0
where M = uh, and N = vh are the flow rate fluxes in the x and y directions, u and v are the average flow velocities, h is the water level, β is the momentum correction factor (=1.25), g is the gravitational acceleration, ρt = σCL + ρm (1 −CL) is the density of the mixture of particles and water (≃apparent density of the debris flow), σ is the density of particles, ρm is the density of water, Tbx and Tby are the bottom shear stresses in the x and y directions, ib is the deposition (ib <0), or erosion (ib ≥0) rate, CL and CF are the volumetric concentrations of coarse and fine particle components of the total particle components in the debris flow, C*L and C*F are the coarse and fine particle concentrations in the sedimentary layer at the bottom, and C*DL is the concentration in the sedimentary layer when particles are separated from the flow to be deposited. zb is the erosion or deposition depth for which Eq. (12) is used.

3.1.2 Bottom shear stress

As for the total shear stress of the debris flow composition equation (Eq. (11)), the first term can be expressed with C0 (adhesion), the second term with Ty (shear stress), and the third term with μ (coefficient of viscosity) and du/dz (velocity gradient). The fourth term can be expressed with the stress caused by the collision between particles and stress caused by turbulent mixing.

(11) τ=C0+τy+μ(dudz)+A(1e2)σ1b|dudz|(dudz)ρuv¯

The following approximate expressions were used to apply the debris flow composition equation to the model.

(12) τ=τy+aisina{(C*CL)1/31}2σdm2(uz)2
(13) τy=pstanφ
(14) ps=f(CL)(σρ)CLg(hz)cosθ
(15) f(CL)={CLC3C*C3;CL>C30;CLC3
where ai is the experimental value, a is the particle collision angle (aisina = 0.02), ps is the static pressure, C* is the maximum sediment concentration (= 0.65), and C3 is the limited sediment concentration (= 0.48). The bottom shear stress according to the sediment concentration is as follows:
  • (1) For a fully developed stony debris flow; CL >0.04 C*

    (16) τbx=uu2+v2τyx+ρfbuu2+v2
    (17) τby=vu2+v2τyy+ρfbvu2+v2
    (18) τyx=f(CL)(σρ)CLghcosθxtanφ
    (19) τyy=f(CL)(σρ)CLghcosθytanφ

  • (2) For an immature debris flow; 0.02 ≤CL ≤0.04C*

    (20) τbx=uu2+v2τyx+ρfbuu2+v2
    (21) τby=vu2+v2τyy+ρfbvu2+v2

  • (3) For a turbulent flow; CL <0.02

    (22) τbx=uu2+v2τyx+ρfbuu2+v2
    (23) τby=vu2+v2τyy+ρfbvu2+v2

3.1.3 Erosion and deposition velocity

The erosion and deposition of a debris flow are dominated by the saturation of the bottom surface, the sediment concentration in the flow, and the parameters of the equilibrium concentration, and their properties also have very complex structures. In this study, the following approximate expressions presented by Takahashi (1991) to reproduce the flow characteristics of a debris flow were adopted:

  • (1) Erosion velocity; CLC

    (24) ib=δeCCLC*Cu2+v2dm

  • (2) Deposition velocity

    (25) ib=δd(1u2+v2pUe)CCLC*DLu2+v2
    where δe (= 0.007) and δd (= 1.0) are the erosion and deposition coefficients, respectively, and p (=2/3) is the experimental value. Ue is the equilibrium velocity, which is expressed as follows:
    (26) Ue=25dm[gsinθeasinα{CL+(1CL)ρmσ}]1/2{(C*DLCL)1}h3/2
    (27) tanθe=CL(σρm)tanφCL(σρm)+ρm
    where θe is the bottom slope and dm is the representative particle size of the coarse particle components resulting from the collision between particles.

3.1.4 Equilibrium of sediment concentration

The debris flow concentration equations in the equilibrium state can be divided into stony debris flow, immature debris flow, and turbulent water flow according to the bottom gradient. The following equations were applied (Nakagawa et al., 2003):

  • (1) Stony debris flow; θw >0.138

    (28) C=tanθw(σρm)(tanφtanθw)

  • (2) Immature debris flow; 0.03 <θw ≤0.138

    (29) C=6.7{tanθw(σρm)(tanφtanθw)}2

  • (3) Turbulent water flow with bed load transport; θw ≤0.03

    (30) C=(1+5tanθw)tanθwσρm(1α02τ*cτ*)(1α02τ*cτ*)
    (31) α02=2{0.425(σ/ρt)tanθw/(σ/ρt1)}1(σ/ρt)tanθw/(σ/ρt1)
    (32) τ*c=0.04×101.72tanθw
    where ϕ is the internal friction angle, T*c is the dimensionless critical shear stress, and T*c is the dimensionless shear stress. In addition, Takahashi (1991) divided the debris flows into stony debris flows (C′ >0.4C*), immature debris flows (0.1≤C ≤0.4C*) and turbulent water flows (C <0.1)according to the sediment concentration.

The major parameters required for the numerical analysis of the debris flow (Table 4) and the topographic map constructed using the DEM data of the target area (Fig. 5) are as follows:

Parameters for the numerical analysis

Fig. 5.

Ground level around Gori Nuclear Power Plant

4. Results

In this study, the possible amount of sediment according to the rainfall scenario in the mountainous terrain around the Gori nuclear power plant was estimated, and the debris flow field at the nuclear power plant site in the downstream area was analyzed through debris flow inundation analysis (Fig. 6). The flow velocity was evaluated when the debris flow generated in the upstream area reached the downstream area. Based on the flow velocity, the impact forces acting on structures were evaluated. The main results of this study are as follows.

Fig. 6.

Catchment areas in the study for the Gori nuclear power plant

4.1 Debris Flow Flooding Characteristics

The developed DFU-2D model was verified based on a hydraulic model experiment carried out by Kim et al. (2020), and the results of the total runoff sediment volume in the debris flow, sediment outflow, and sediment concentration were compared and verified. In addition, the results of the hydraulic model experiment of Shrestha (2009) were used to verify the model assuming the installation of an erosion control dam. It was found that the model satisfactorily reproduced the deposition patterns in front of the erosion control dam over time. Based on the above results, it was determined that the model adequately reproduced a debris flow.

Fig. 7 shows the debris flow occurrence points (Points A, B, and C) to examine the effects of the impact forces of a debris flow on major facilities located to the east and south of the Gori nuclear power plant. The flow velocity at the front end of the debris flow and its arrival time were evaluated to examine the debris flow effects on major facilities in the downstream area and evaluate the impact forces acting on these facilities. This was performed by calculating the amount of sediment generated according to the rainfall frequency (Table 3).

Fig. 7.

Study catchments in Gori

For the debris flow originated when the sediment in the upper part of the mountainous area collapsed and flowed to the downstream area with surface water, the arrival time and flow velocity at the front end rapidly changed according to the topographical characteristics (slope). The inundation patterns in the downstream area were also different from those of a general fluid (clear water) depending on the location of the structure and the bottom slope.

The time it took for the generated debris flow to reach major facilities located in the downstream area was examined according to the debris flow occurrence scenario (Point A, B, and C). It was found that the arrival time tended to gradually decrease as the probable rainfall frequency increased because the amount of rainfall and the size of the generated debris flow increased. The arrival time varied depending on the location and terrain where the debris flow occurred. While it took approximately 1 to 2 min for the generated debris flow to reach the downstream area at points A and B, it took 30 to 40 s to reach the transmission facility located in the downstream area at point C. Fig. 8 shows the inundation patterns of the debris flow that reached the downstream area according to the structure and topographical conditions. Important data, such as the maximum dispersion distribution and the deposition depth of sediment, are provided.

Fig. 8.

Debris flow calculation results for each scenario (return period: 200 years)

The flow velocity of the debris flow at the front end is a very important parameter when evaluating its impact force on structures, and accurate evaluation is very important. In this study, a computation time of 0.1 s was applied to evaluate the flow velocity at the front end immediately before the debris flow reached the structures, and the results are presented in Table 5.

Results of velocity and arrival time for the debris flow

4.2 Impact Pressure of the Debris Flow

Unlike a general fluid, it is significantly difficult to evaluate the accurate impact force of a debris flow as it is a mixture of sediment and fluid. Previous studies proposed evaluation formulas through various hydraulic model experiments according to the hydrodynamic model (Bugnion et al., 2011; Yamamoto et al., 1998; Kim, 2013) and solid collision model (Mizuyama, 1979; Kim, 2013). In this study, the impact pressure of the debris flow was calculated using the following equation:

(33) Ppeak=kpρdv2(kp=1.0,ρd=1.33t/m3)
where Ppeak is the maximum debris flow impact pressure, kp is an empirical factors, ρd is the density of debris flow, v is the velocity of debris flow. Due to the limitations for accurately calculating the geological condition (D50) of the Gori nuclear power plant, the coefficient (kp = 1.0) presented by Yamamoto et al. (1998) was applied for kp. Fig. 9 shows the results for the impact pressure of the debris flow obtained using the flow velocity at its front end.

Fig. 9.

Debris flow calculation results for each scenario (return period: 200 years).

The fastest debris flow occurred at Point A owing to the higher bottom slope compared to that of other areas. At this point, the impact force acting on major facilities located in the downstream area was also evaluated to be the largest. However, because slope protection is installed at Point A, the impact force is expected to be lower than the evaluated impact force if the current conditions are accurately considered.

5. Conclusions

In this study, a numerical analysis of a debris flow was conducted to estimate the inundation characteristics of a debris flow in downstream areas and its impact force on structures. Debris flow occurrence scenarios according to rainfall conditions were created because the landslide risk map provided by the Korea Forest Service is insufficient to closely examine the effects of landslides (debris flows) in mountainous areas near a nuclear power plant site on major facilities in downstream areas.

Given that most of the major parameters required for the numerical analysis of the debris flow in the target area of the study are set as undisclosed data due to its special characteristics, data for the situation before the construction of the nuclear power plant and the material properties of the surrounding area were applied to the model. To respond to extreme rainfall, the amounts of sediment that can be generated for each scenario (30, 50, 100, and 200 years) were calculated and applied as boundary conditions of the debris flow. In addition, the inundation characteristics of the debris flow and its impact force on structures were evaluated. In particular, scenarios with relatively high risks (slope disasters and transmission towers) were selected and an impact assessment was performed considering the topographical characteristics of the Gori nuclear power plant.

In the numerical analysis results, the largest impact force occurred at the most extreme rainfall intensity (200 years). In particular, the impact force of the debris flow linearly increased according to its frequency at occurrence points A and C, but it tended to slightly increase from the frequency of 100 years at Point C. These results indicate that, through the constructed numerical model, it is possible to evaluate the flow characteristics of a debris flow according to the topographical characteristics of the area around the Gori nuclear power plant and the effect of this debris flow on major facilities.

This study modeled the effect of debris flows on the Gori nuclear power plant under various rainfall scenarios, and found that the speed and impact force of a debris flow significantly increased in the event of extreme rainfall, especially on steep slopes and areas directly exposed to sediment. This is a pioneering study that modeled the effect of debris flows on nuclear facilities, rather than on the general coastal infrastructure, and it provides important data with respect to safety assessment and disaster preparedness planning for nuclear power plants in sediment-rich areas. Based on these results, nuclear facilities located in similar terrains should introduce safety protocols according to their topographical characteristics in preparation for the risk of landslides. Specifically, the importance of reinforced barrier installation, rapid evacuation planning, and design adjustment was confirmed. In addition, this study confirmed the need for additional experimental and observational data to verify debris flow models at nuclear power plants. In the future, research should be conducted on debris flow mitigation technologies for high-risk power plant locations.

Notes

Yeonjoong Kim serves on the journal publication committee of the Journal of Ocean Engineering and Technology and had no role in the decision to publish this article. No potential conflicts of interest relevant to this article were reported.

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (Ministry of Science and ICT) (No. RS-2022-00144325).

References

Bugnion L, McArdell B. W, Bartelt P, Wendeler C. 2011;Measurements of hillslope debris flow impact pressure on obstacles. Landsildes 9:179–187. https://doi.org/10.1007/s10346-011-0294-4.
Chae B.-G, Liu K.-F, Kim M.-I. 2010;A case study for simulation of a debris flow with DEBRIS-2D at Inje, Korea. The Journal of Engineering Geology 20(3):231–242.
Chang H.-J, Lee H.-J, Kim S.-G. 2024;A simulation of a small mountainous chachment in Gyeoungbuk using the RAMMS model. Journal of Korean Society of Disaster and Security 17(1):1–8.
Choi E, Ha J. G, Hahm D, Kim M. 2021;A review of multihazard risk assessment: Progress, potential, and challenges in the application to nuclear power plants. International Journal of Disaster Risk Reduction 53:101933. 1–13. https://doi.org/10.1016/j.ijdrr.2020.101933.
Dang H. V, Shin S, Lee E, Park H, Park J.-N. 2022;Numerical investigation of countermeasure effects on overland flow hydrodynamic and force mitigation in coastal communities. Journal of Ocean Engineering and Technology 36(6):364–379. https://doi.org/10.26748/KSOE.2022.036.
Hasegawa R. 2013. Disaster Evacuation from Japan’s 2011 Tsunami Disaster and the Fukushima Nuclear Accident (Studies No.05/13). IDDRI. https://www.iddri.org/sites/default/files/import/publications/study0513_rh_devast-report.pdf.
Hwang T, Seo S.-C, Jin H, Oh H, Lee W.-D. 2024;Effects of storm waves caused by Typhoon Bolaven (1215) on Korean coast: A comparative analysis with deepwater design waves. Journal of Ocean Engineering and Technology 38(4):149–163. https://doi.org/10.26748/KSOE.2024.044.
Kang B, Kim J, Kwon Y, Choi J, Jang Y, Kwon S. 2022;An analysis of the impact of building wind by field observation in Haeundae LCT area, South Korea: Typhoon Omais in 2021. Journal of Ocean Engineering and Technology 36(6):380–389. https://doi.org/10.26748/KSOE.2022.027.
Kang S, Hahm D. 2018;Development of an earthquake-induced landslide risk assessment approach for nuclear power plants. Nuclear Engineering and Technology 50(8):1372–1386. https://doi.org/10.1016/j.net.2018.07.016.
Kim K, Kim D, Seo J, Lee C, Woo C, Kang M, Jeong S, Lee D. 2018;Evaluating stability and functionality of hybrid erosion control dam for reducing debris flow damage in forested catchment nearby urban area. Journal of Korean Society of Forest Science 107(1):59–70. https://doi.org/10.14578/jkfs.2018.107.1.59.
Kim S, Paik J, Kim K. S. 2013;Run-out modeling of debris flows in Mt. Umyeon using FLO-2D. Journal of the Korean Society of Civil Engineers 33(3):965–974. https://doi.org/10.12652/Ksce.2013.33.3.965.
Kim T. G. March. 16. 2022;일본 동북지역 쓰나미 임박...후쿠시마 M7.3 지진 [Fukushima, Japan hit by a 7.3 magnitude earthquake, tsunami warning issued for the northeastern region]. Seoul Shinmun https://www.seoul.co.kr/news/international/japan/2022/03/16/20220316500208.
Kim T, Lee W.-D, Kwon Y, Kim J, Kang B, Kwon S. 2022;Prediction of wave transmission characteristics of low crested structures using artificial neural network. Journal of Ocean Engineering and Technology 36(5):313–325. https://doi.org/10.26748/KSOE.2022.024.
Kim Y. J. 2013. Study on Hydraulic Characteristics of Debris Flow Breakers and Sabo Dams with a Flap [Doctoral dissertation,. Kyoto University.
Kim Y. J, Yoon J.-S, Tanaka K, Hur D.-S. 2015;Prediction of a debris flow flooding caused by probable maximum precipitation. Journal of Korea Water Resources Association 48(2):115–126. https://doi.org/10.3741/JKWRA.2015.48.2.115.
Kim Y, Jun K, Jun B. 2016;Analysis of debris flow disaster area using terrestrial LiDAR and Kanako-2D. Journal of the Korean Society of Hazard Mitigation 16(5):113–118. https://doi.org/10.9798/KOSHAM.2016.16.5.113.
Kim Y, Kim T, Kim D, Yoon J. 2020;Debris flow characteristics and sabo dam function in urban steep slopes. Journal of Korea Water Resources Association 53(8):627–636. https://doi.org/10.3741/JKWRA.2020.53.8.627.
Lee D. W, Lee S.-R, Park J.-Y. 2019;Numerical simulation of debris flow behavior at Mt. Umyeon using the DAN3D model, Korea. Journal of the Korean Society of Hazard Mitigation 19(3):195–202. https://doi.org/10.9798/KOSHAM.2019.19.3.195.
Lee S, An H, Kim M, Lee H. 2020;Analysis of debris flow simulation parameters with entrainment effect: a case study in the Mt. Umyeon. Journal of Korea Water Resources Association 53(9):637–646. https://doi.org/10.3741/JKWRA.2020.53.9.637.
Mizuyama T. 1979;Computational method and some considerations on impulsive force of debris flow acting on sabo dams. Journal of the Japan Society of Erosion Control Engineering 112:40–43.
Nakagawa H, Satofuka Y, Kawaike K. 2003;Numerical simulation of sediment disasters caused by heavy rainfall in Camuri Grande basin, Venezuela 1999. Proceedings of the 3rd Conference on Debris-flow Hazards Mitigation: Mechanics, Prediction and Assessment, Switzerland Rotterdam :671–682.
National Institute for Land and Infrastructure Management (NILIM). 2016;Manual of technical standard for establishing sabo master plan for debris flow and driftwood 364. Technical note of national institute for land infrastructure management :25–32. https://www.nilim.go.jp/lab/bcg/siryou/tnn/tnn0364pdf/ks0364.pdf.
Park H. J, Kang Y. J. 2024;Comparison of wave prediction and performance evaluation in Korea waters based on machine learning. Journal of Ocean Engineering and Technology 38(1):18–29. https://doi.org/10.26748/KSOE.2023.040.
Park J. C. August. 6. 2020;벌채⋅송전탑 설치로 ‘와르르’…산사태 피해 주민들 분통 [Residents affected by landslides express ‘outrage’ over logging and installation of transmission towers]. Yonhap News https://www.yna.co.kr/view/AKR20200806134200064.
Proske D, Suna J, Hübl J. 2009;Estimation of design impact forces of debris flows. Proceeding of Safety Reliability and Risk of Structures, Infrastructures and Engineering System :1323–1330.
Shrestha B. B. 2009. Study on mitigation measures against debris flow disasters with driftwood [Doctoral dissertation,. Kyoto University;
Takahashi T. 1991. Debris flow: Monograph Series of IAHR Balkema. p. 1–165.
Takahashi T, Nakagawa H. 1991;Estimation of a debris flow and the area of hazard for a given rainfall, Study of streams and watersheds of high hydraulic irregularity. Proceedings of 24th IAHR Congress :161–168.
United Nations. 2014. UNSCEAR 2013 Report to the General Assembly with Scientific Annexes United Nations. p. 1–311.
Wu Y.-H, Kiu K.-F, Chen Y.-C. 2012;Comparison between FLO-2D and Debris-2D on the application of assessment of granular debris flow. Journal of Mountain Science 10(2):293–304. https://doi.org/10.1007/s11629-013-2511-1.
Yamamoto A, Yamamoto S, Toriihara M, Hirama K. 1998;Impact load on sabo dam due to debris flow. Journal of the Japan Society of Erosion Control Engineering 51(2):22–30. https://doi.org/10.11475/sabo1973.51.2_22.
Yamamoto A, Yamamoto S, Toriihara M, Hirama K. 1998;Impact Load on Sabo Dam due to Debris flow. Journal of the Japan Society of Erosion Control Engineering 51(2):22–30. https://doi.org/10.11475/sabo1973.51.2_22.
Zhu Y, Ishikawa T, Shimizu Y. 2018;Surface flow analysis of typhoon 10 in-duced slope failure based on digital elevation modeling. Proceedings of the 9th Symposium on Sediment-Related Disasters, :121–126.

Article information Continued

Fig. 1.

Nnatural disasters that can occur around nuclear power plants

Fig. 2.

Conceptual illustration of the effects of various natural hazards on a virtual nuclear power plant

Fig. 3.

Study area for the Gori nuclear power plant

Fig. 4.

Environment around the Gori nuclear power plant

Fig. 5.

Ground level around Gori Nuclear Power Plant

Fig. 6.

Catchment areas in the study for the Gori nuclear power plant

Fig. 7.

Study catchments in Gori

Fig. 8.

Debris flow calculation results for each scenario (return period: 200 years)

Fig. 9.

Debris flow calculation results for each scenario (return period: 200 years).

Table 1.

Model selection

Dimension Program Entrainment process1) Inondation Sabo dam Structure GUI Adoption
2D Flo-2D
Debris-2D
RAMMS
Kanako 2D
DFU-2D
iRIC (Morpho2DH)
1)

Entrainment process: In the process of flowing along a mountain stream, debris flows undergo entrainment that reproduces the erosion of the bed and deposition at the bank.

Table 2.

Results of the analysis of probable rainfall by rainfall duration (Busan)

Return period Type Rainfall duration (min)

10 60 120 180 240 360 540 720 900 1080 1440
30 28.8 98.1 137.3 163.8 189.6 225.5 253.4 275.4 297.0 310.2 333.5
26.8 89.5 127.2 152.2 171.5 201.1 233.4 257.5 276.6 292.1 315.8
28.8 96.4 131.5 159.6 186.6 216.4 - 277.9 - - 339.8
50 31.0 106.8 149.6 178.4 206.5 242.2 275.6 299.4 322.9 337.2 362.9
28.9 97.4 138.5 165.7 186.7 218.8 253.9 280.2 300.9 317.7 343.6
31.1 104.7 142.9 173.5 202.9 235.1 - 302.3 - - 370.3
100 34.1 118.5 166.3 198.0 229.3 268.8 305.6 331.7 357.8 373.8 402.4
31.6 108.2 153.9 184.0 207.2 242.9 282.0 311.4 334.6 353.4 382.2
34.1 115.9 158.4 192.3 224.9 260.3 - 335.2 - - 411.4
200 37.1 130.2 182.9 217.6 252.1 295.4 335.4 364.0 392.6 410.1 441.8
34.4 118.9 169.2 202.2 227.5 266.3 309.1 341.3 366.9 387.8 402.2
37.0 127.1 173.8 211.1 246.9 285.4 - 367.9 - - 452.3

➀ Probable rainfall calculated this time (Busan Observatory, 62 years: 1961–2022)

➁ Improvement and supplementary research on probable rainfall

➂ Busan City sewerage maintenance basic plan (change)

Note: Probable rainfall (unit: mm)

Table 3.

Parameters for the model

Parameter Unit Case1 Case2 Case3 Case4
Return period yr 30 50 100 200
Rainfall (24 h) mm/d 333.5 362.9 402.4 441.8
Basein arae km2 0.171 0.171 0.171 0.171
Input discharge m3/s 8.79 9.74 11.04 12.36
Grid size (Δx, Δy) m 5 5 5 5
Δt s 0.01 0.01 0.01 0.01
Runoff sediment volume m3 14,402 15,672 17,337 19,079
D50 m 0.1 0.1 0.1 0.1
tanϕ - 0.7 0.7 0.7 0.7
C* - 0.65 0.65 0.65 0.65
C3 - 0.48 0.48 0.48 0.48

Note: C* = Sediment concentration in the bed; C3 = Limitative sediment concentration

Table 4.

Parameters for the numerical analysis

Parameter Unit Value
Rainfall (24 h) mm/d -
Runoff sediment volume m3 -
Basin area km2 -
Δx, Δy m -
Δ t s -
Density of a sediment particle (σ) g/cm3 2.65
Density of a water (ρ) g/cm3 1.0
Tangent of the internal friction angle of sediment (tanϕ) - 0.7
Limitative sediment concent (C3) - 0.48
Erosion coefficient (δe) - 0.0018
Deposition coefficient (δd) - 0.045

Table 5.

Results of velocity and arrival time for the debris flow

Parameter Point Return period

30 50 100 200
Max. velocity (m/s) Point-A-A 3.04 3.14 4.40 8.58
Point-B-A 4.74 4.95 5.73 6.08
Point-C-A 3.89 4.09 4.82 5.98
Arrival time (s) Point-A-A 78.1 73.2 66.5 61.8
Point-B-A 130.7 120.7 109.3 103.8
Point-C-A 40.6 37.9 34.2 31.5