Simulation of Hypothetical Radiological Accident at Potential Nuclear Power Plant Sites in Malaysia

After any nuclear incident, the main concern for human well-being is the extent of radiation released beyond the site. This study simulated a similar scenario on three potential nuclear power plant sites in Malaysia, using the Fukushima Daiichi Nuclear Power Plant incident as a reference. The computer model Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) was used to simulate and track the movement of air parcels and the dispersion of radioactive emissions. HYSPLIT analyzed the dispersion profile of radioactive materials, revealing that in S1, S2, and S3, the maximum ground dose was 7.9 mSv, 28.0 mSv, and 7.6 mSv, respectively. The maximum activity deposited on the ground was 62 MBq, 210 MBq, and 14 MBq for S1, S2, and S3, respectively. The analysis of ground deposition indicated that S1 covered an estimated area of 1500 km2, S2 covered 3025 km2, and S3 had the largest coverage of approximately 4537 km2. Overall, this study demonstrates that the hypothetical accident would contaminate the vicinity of the three potential nuclear power plant (NPP) sites.


Introduction
Recently, nuclear sources have emerged as a popular choice for energy production, with most countries worldwide adopting this technology.This alternative energy source is highly efficient, generating energy with minimal wastage.The International Energy Outlook 2013 report estimates that global energy consumption will increase by approximately 56% between 2010 and 2040.Several nations, including France, China, South Korea, Kazakhstan, and the United States, have made strides in developing nuclear technology [1].Considering the global population of 8 billion people, ensuring a stable supply of electricity is of utmost importance, and nuclear power may become a more prominent energy source.
Following the guidelines provided by the Atomic Energy Licensing Board (AELB), Malaysia's regulatory body for nuclear-related matters, construction of the inaugural nuclear power plant (NPP) is anticipated to commence in 2017, with operations set to commence by 2025.Based on the abovementioned guidelines, several potential sites have been identified as suitable for NPP development.Prior research has determined that Tenggaroh and Jemaluang are the most appropriate locations within Johor to construct a nuclear power plant [2,3].
In the wake of a catastrophic nuclear incident, such as the one at Fukushima in 2011, there is a heightened sense of concern.To effectively set up a response system for emergencies that may arise due to nuclear disasters, it is paramount to conduct a comprehensive evaluation of the associated risks.This will enable the relevant authorities to implement appropriate and precise measures, with the choice of the source term being a critical factor in the simulation process.Determining the number of radioactive substances released into the environment, the level of radioactivity, and the extent of exposure to the population mainly depends on the source term.Despite the availability of precise meteorological data, developing a model capable of predicting the distribution and deposition of radionuclides following nuclear energy-related incidents remains a highly intricate undertaking.The outcomes of an investigation into the movement of Chernobyl 137 Cs, utilizing HYSPLIT (Hybrid Single Particle Lagrangian Integrated Trajectory), have indicated that the dissemination of radioactive substances is significantly impacted by both meteorological factors and the characterization of the pollutant source [4].
The HYSPLIT model simulates the deposition of radionuclides resulting from nuclear tests conducted in former test sites without exposing individuals to radiation in the field.This model can be employed locally on a personal computer by downloading and installing it or accessing it online.HYSPLIT employs the log-normal particle distribution, which assigns a proportional share of the overall 137 Cs activity to each particle size bin [5].Mitrakos et al. discovered that the Fukushima Daiichi Nuclear Power Plant catastrophe caused the release of 520 PBq of radioactive materials into the atmosphere from multiple sources, except for noble gases [6].
Within the nuclear accident emergency preparedness framework, a taxonomy is utilized to guide hazard assessment for emergency preparedness and response and develop a protection strategy.The hazard assessment results are then used to inform the protection strategy established before the emergency response [7].In order to enhance emergency preparedness, an assessment of the atmospheric dispersion of 131 I and 137 Cs radionuclides has been conducted to prepare an activity distribution.A previous study utilized data from past nuclear incidents, cumulative effective dose, and the mortality risk among the populace to present an accident scenario in the Southeast Asia region, which was found to be significant [7].The utilization of HYSPLIT can also provide insight regarding the effects on the environment and human health.Researchers conducted a study that generated simulations of scaling factors for air concentrations (measured in Bq.m 3 ) and surface depositions (measured in Bq.m 2 ) of radionuclides that were discharged from a hypothetical NPP.Following the simulation of an NPP, the resulting data concerning the concentration of radionuclides in the air and their surface deposition was analyzed to ascertain the associated environmental and human health impacts [5,8].
The objective of this study is to assess the probability of a Fukushima-like scenario occurring in Malaysia and its potential ramifications on human health.Our primary focus is on analyzing the dispersion of atmospheric and deposition of radionuclides and assessing the risks associated with the radiation emitted from selected potential NPP sites.

Method
The study was conducted using HYSPLIT version 5, which was released in April 2021.This software is versatile in creating air parcel trajectories, simulating local, regional, and long-range transport, and predicting air contaminants' dispersion and deposition on various geographical and temporal scales [7].HYSPLIT has also been utilized for predicting the dispersion of wildfire smoke, wind-blown dust, and volcanic ash, as well as air dispersion products in the event of chemical and nuclear mishaps [9,10].
The HYSPLIT model employs a combination of Lagrangian and Eulerian methods to simulate particle transport, including turbulent transport, scavenging, decay, and air concentration calculation.The model uses the Eulerian modeling technique to integrate pollutant fluxes caused by diffusion and advection at each grid cell interface to calculate air concentrations for each cell.This approach enables accurate and efficient prediction of air quality and pollutant dispersion.The Lagrangian modelling technique is utilized to determine air concentrations by incorporating the contributions of individual pollutant puffs that are transported across the grid cell in accordance with their trajectory [11].
The calculation of particle advection involves determining the average three-dimensional velocity vectors from the initial position X(t) to the estimated final position X'(t+(t)).Advection refers to the movement of fluid, typically in a horizontal direction in air or water, for the purpose of transferring heat or substances.The Langragian dispersion model is utilized to compute trajectories in simulations involving intricate dispersion and deposition processes [12].Linear interpolation is applied to the velocity vectors in both spatial and temporal domains.The underlying principle employed to determine the location is based on Equation (1).
This study mainly focuses on designing a HYSPLIT simulation model of hypothet-cal NPPs based on FDNPP incidents on selected potential sites in Malaysia as well as computing a radiological assessment, hazard evaluation and spatial distribution at affected areas.This research study endeavors to examine potential sites in Malaysia for the installation of a nuclear power plant (NPP).The sites that have been nominated have been carefully selected based on several essential requirements, namely geology and seismology, atmospheric extremes and dispersion, safety zones population considerations, emergency planning, hydrology, and industrial, military and transportation facilities.These criteria have been identified as essential components for establishing a secure and effective NPP [13].The beginning of any accident sequence can be traced back to a pre-anticipated occurrence, such as an earthquake, flood, or tsunami [14].Furthermore, internal factors such as the loss of off-site power, a coolant leakage, or human-caused errors can also serve as initiator events.Any such triggering events can adversely affect the nuclear reactor control and safety systems, releasing a substantial amount of radioactive fission into the environment.The locations mentioned above are a product of a regional investigation and are identified as S1) Rungkup, Hilir Perak, Perak; S2) Jugra, Kuala Langat, Selangor; and S3) Tenggaroh, Mersing, Johor [15].Despite the occurrence of multiple radionuclide releases during the FDNPP incident, this simulation has limited its focus to ten specific ones.These radionuclides include Niobium-95, Silver-110, Tellurium-132, Iodine-131, Iodine-132, Xenon-133, Caesium-137, Caesium-134, Barium-140, and Lanthanum-140, all of which were detected in Takasaki, Japan during the FDNPP incident [16].Emitting gases and particles into the atmosphere following catastrophic events such as those in Fukushima and Chernobyl can have far-reaching and long-lasting effects.The issue at hand typically spans the spatial and temporal scales of the Earth's radius and a few months, respectively.These scales of air dispersion and transmission can significantly impact the behavior of the released material, including radioactive decay, gas-particle conversion, and dry and wet deposition.It is generally assumed that the release and dispersion of radionuclides occur over three days or 108 hours following the start of the incident.The estimated release height of these radionuclides is between 0 and 200 meters, with a selected height of 100 meters [17].Table 2 provides details on the parameters used to calculate this simulation.Dose intakes through inhalation and external exposure pathways were calculated using Equation (2).
where e‫ܧ‬ is the dose rate coefficient, A is the activity concentration and t is the duration of a person being exposed to the pollutant.‫ܧ݁‬ is dependant to the altitude of the radionuclide analysis whether it is focused on ground or air.Value of A is directly obtained from the HYSPLIT system and multiplied by the External Exposure to Radionuclides in air, water and soil (EPA-402/R19/002 report from EPA United States Environmental Protection Agency).After the amount of total effective dose is calculated for each distance, it was recorded in a table for each location and the results are compared to the allowed dose among public people.After calculating the dose conversion factor and activity concentration, the result is the rate of the equivalent dose received by adult people near the area.Thus, by multiplying it with 3600s to determine accumulated dose in a day.This is to determine the total effective dose received by adults on the first day of the accident due to the ground deposition value.
Using the general method approved by the ICRP organization (equation ( 3)), cancer risks associated with exposure to a specific amount of annual effective dose equivalent from the radionuclides are calculated (equation ( 4)).The total effective dose from the NPP in a nominated site can be calculated and recorded for different locations and directions.The cancer fatality risk in the population is calculated using equation (3), where the dose-to-risk coefficient and E(t) is the total dose are multiplied to obtain the risk [17] whereas the risk of mortality and morbidity is shown in equation ( 4) [18]. 1

Risk of Mortality Bq h Mortality Air concentration Breathing Rate Morbidity
Morbidity Coefficient

Release and dispersion of radioactive materials in the air
The HYSPLIT model was used to simulate the release and dispersion of radionuclides in the air from S1, S2 and S3.The model is configured to display the total accumulated dose in mSv in intervals of every 3 hours.Figure .2 shows the dispersed radionuclides in the air after release starting from 1500 H, 14 March 2011.For S1, it is shown that in Figure 2.a), the radionuclides move in the direction of northwestern of the site of release.In the first 3 hours, the radionuclides have moved across roughly 20 km in the airspace of 100 m and have covered 60 km in another 3 hours.9 hours after release, the maximum dose in air concentration dropped to 4.6 × 10 -1 mSv and could not be detected in the following hours.The further away the radionuclides traveled, lesser dose was detected in the air.In the 6th hour frame, a more significant difference was observed in contour colours at around 0.1 mSv -0.2 mSv.It can also be deduced that most of the fallout have been deposited onto the Strait of Malacca.For the release at S2 and S3, the model was adjusted to simulate air concentration between 0m and 500m to ensure all radionuclides were accounted for.The simulation for S2 can be seen in Figure 2.b), where the direction of the radionuclides in the air after release is towards the west.Within the first 3 hours, the radionuclides appeared to be clustering within close proximity of the release site before stretching to the west for about 27 km as it approached the 6 th -hour display.Before completely depositing on the ground, the radionuclides released from S2 was able to shift northwest for approximately 27 km with the maximum dose recorded being 5.5 × 10 -1 mSv.Similar to the model for S1, most of the radionuclides are expected to be deposited onto the sea.

c) Maximum Dose in air concentration
Based on Figure 2.c), which shows the accumulated dose in air concentration for S3 as site of release, the radionuclides were headed towards the south.Three hours after the dispersion, it can be seen that most of the radionuclides were located 14 km away from the site of release.Between 6 hours until 9 hours, the HYSPLIT model simulated the radionuclides covered a distance of 27 km to the south from the site of release.After 12 hours of release, all sites have the same outcome where the radionuclides have completely deposited on the ground.However, for the S3 model, the radionuclides had moved inland risking a lot of potential human health risk.

Deposition of radioactive materials on the ground
The deposition of radionuclides on the ground was also simulated for S1, S2 and S3.The exact configuration was used to display the total accumulated dose.However, it was done in intervals of every 6 hours to allow better observation in the change of contours.Figure 4.a) shows the accumulated dose at ground level at S1 with the first sampling at 1800 H, 14 March 2011.Assuming all of the radionuclides have been deposited as in Figure 2, the 12 th hour frame in Figure 4. b) shows the total area that has been contaminated which is estimated to be 1500 km 2 .Within the first 6 hours, the maximum dose recorded is 6.2 × 10 -1 mSv however, the reading rose exponentially to 3.1 mSv in the next 6 hours.This data showed similar significance in radiological risk as the study done by Shamsuddin et.al. [3].Fortunately, due to careful planning and standard criterias of NPP selection, most radionuclides were deposited far away from land and settlements [19].It is also observed that 18 hours after release, the reading recorded around S1 was more than 1 mSv which is the annual dose limit for the general public [20].The simulation of S2 can be referred in Figure 4.b) where the first 6 hours have a greater dose reading compared to S1.It is observed that the maximum reading is 2.9 mSv after 6 hours of release compared to S1 which is only 0.62 mSv.The radionuclides have covered a larger area estimated to be 3025 km 2 , doubling the model in S1.The data also suggested that even though 100% deposition occurred after 12 hours, maximum area contamination was discovered to happen after 18 hours.This is most likely to occur because the radionuclides are slightly moving in the wind direction even after reaching ground level.The maximum reading increased significantly at the 12 th hour frame from 11 mSv, peaking at 28 mSv at 36 th frame accumulatively.The annual dose limit threshold for workers in radiation related fields is 20mSv per year.Using that as a guideline, workers involved this postulated accident can only remain around S2 until 24 hours after the release.The 24th hour frame for S2 shows that the annual dose limit for worker was reached with the proximity reading ranging from 1 mSv -20 mSv.This tells us that any action to evacuate or to decontaminate the area must be done with higher priority as the dose distributed around S2 is much higher.
Figure 4. c) shows the deposition of radionuclides with the release site in S3, Tenggaroh, Mersing, Johor.Starting in the 3rd hour frame, the depostion is directed towards the southwest with maximum area contamination taking place in the 24th hour frame.The maximum area covered is evaluated at around 4537km 2 , making it the largest contamination among all three potential sites.The model also demonstrated that the accident in S3 has to be dealt with as the highest priority as the fallout involves not only the population of Johor but the population of Singapore as well.The accident is also to be treated as a transboundary accident as it has affected areas outside its initial location.

Activity in air concentration and ground deposition
Table 4 shows the radionuclide activity in air concentration within 12 hours after release.The activity seems to increase steadily within the first 9 hours and finally deposited to the ground, shown in the 12th hour column to be 0 in activity reading.This pattern is somewhat expected to be similar to the reading of the dose distribution.The initial reading of activity is the highest at S1 with 1.1 × 10 6 Bq, followed by S2 with 7.7 × 10 5 Bq and S3 with 5.2 × 10 5 respectively.Table 4. shows the radionuclide activity deposited on the ground within 36 hours after release.The activity seems to be increasing steadily for all 3 sites.Based on Table 4., initial activity reading after 3 hours of deposition at S1 is the highest among all three sites with 7.3 × 10 5 Bq.However, Figure 7.b) shows that after 36 hours, the activity reading at S2 is highest with a reading of 2.1 × 10 8 Bq.The difference in the increasing rate can be seen as S2 trendline is more significant compared to S1 and S3.

Mortality risk from inhalation and ground deposition
Table 4 shows the mortality risk probability from inhalation of radionuclides in the air at all three potential sites.Based on the data obtained from the calculation of mortality risk, the risk of people dying are only viable within the first 12 hours, before all the radionuclides are deposited on the ground.After the 12 th hour mark, the mortality risk drops to null.
Mortality risk from exposure of radionuclides deposited on the ground is very different compared to inhalation from the air.Table 5. depicts the mortality risk probability from radionuclides deposited on the ground.It can be observed that starting from release up until 36th hours mark, the probability of people dying increases steadily at all three sites.The highest reading for S1, S2 and S3 after 36 hours are 5.82; 19.7 and 1.31 deaths per 1000 people.

Morbidity risk from inhalation
The risk of contracting an illness or other health issue due to exposure to ionizing radiation is known as the morbidity risk from exposure to ionizing radiation.The type of radiation known as ionizing radiation has sufficient energy to free securely bonded electrons from atoms, which can harm DNA and cellular structures.
Table 6 shows the morbidity risk probability from inhalation of radionuclides in the air at all three potential sites.Based on the data obtained from the calculation of mortality risk, the risk of people suffering from both cancerous and non-cancerous illnesses is also only viable within the first 12 hours, before all the radionuclides are deposited on the ground.After the 12 th hour mark, the morbidity risk drops to null.
The likelihood of morbidity resulting from exposure from ground deposition at each of the three prospective sites is shown in Table 7.According to information gleaned from the calculation of morbidity risk, people are in danger of developing cancer and non-cancerous disorders where the morbidity risk only gradually increases as time passes.After 36 hours, the number of people getting radiation induced disease at S1, S2 and S3 are 6.72; 22.7 and 1.52 per 1000 people.

Conclusion
This study investigated the simulation of radioactive releases from three potential nuclear power plant (NPP) locations in Malaysia, focusing on atmospheric release, deposition, dispersion, and transport.A hypothetical accident scenario was simulated using the HYSPLIT model and drawing insights from the 2011 FDNPP incident.The results revealed that for sites S1 and S2, the released radionuclides followed a specific direction before completely depositing on the ground within 12 hours.Due to their proximity to the western coasts of Peninsular Malaysia, most of the radionuclides were deposited in the sea.However, a portion persisted and settled near the release site.The ground deposition analysis showed contamination areas of 1500 km 2 and 3025 km 2 for S1 and S2, respectively, with peak coverage occurring at different time frames.In contrast, S3 exhibited distinct behavior, with scattered radionuclides traveling south before settling after 12 hours.Within 9-12 hours, measurements in Singapore indicated that the radioactive ground deposition constituted a transboundary accident, covering the largest region of approximately 4537 km 2 .Consequently, if a similar disaster to the FDNPP incident were to occur in Malaysia, S1 and S2 would be the most suitable NPP locations due to their minimal damage compared to S3. Regarding the potential health impact, fatality rates increased progressively from discharge to the 36-hour mark across all three sites.At 36 hours, the highest fatality rates per 1000 persons were observed as 5.82 for S1, 19.7 for S2, and 1.31 for S3.Similarly, radiationinduced disease cases per 1000 people were reported as 6.72 for S1, 22.7 for S2, and 1.52 for S3 after 36 hours.

Figure 1 .
Figure 1.Nominated potential site for study.

Figure 3 .
Figure 3. Accumulated Dose at ground level at S1, S2 and S3 after 36 hours of release.

Figure 4 .
Accumulated Dose at ground level at S1, S2 and S3 after 36 hours of release.

Figure 5 .
Figure 5. Rate of change of Maximum dose deposited on the ground at S1, S2 and S3.

Figure 6 .
Figure 6.Contour map for S3 with multiple vital locations.

Figure 7 .
Figure 7. Maximum activity in air concentration and ground deposition at S1, S2 and S3.

Table 1 .
Potential NPP sites for hypothetical accident.

Table 2 .
HYSPLIT model configuration of the release of radionuclides.

Table 3 .
Radioactive materials released to the environment.

Table 4 .
Mortality risk probability from inhalation of radionuclides in the air.

Table 5 .
Mortality risk probability from radionuclides deposited on the ground.

Table 6 .
Morbidity risk probability from inhalation of radionuclides in the air concentration.

Table 7 .
Morbidity risk probability from radionuclides deposited on the ground.