Introduction

A tsunami is a series of high-speed moving waves generated by a large impulsive disturbance in the ocean. The word ‘Tsunami’ originates from a Japanese word which refers to waves that amplify their heights and velocities when reaching coastlines (Teh and Koh 2011). It is sometimes referred to as seismic sea wave or tidal wave. Tsunami is a hazardous natural phenomenon that impacts the coastlines. It is generated by several natural events such as seaquake, submarine landslide, and volcanic eruption. Tsunami can also possibly be caused by meteorite impact originated from outer space that survived its path through the earth’s atmosphere. The most common tsunami disasters are generated by earthquakes and created sudden movements on the tectonic plates. Earthquake activities induce series of ocean waves over few meters’ height striking the coastal areas in just several hours post-occurrence of the earthquake.

At the beginning of tsunami formation in the deep ocean, it is small and barely noticed, but the deadly waves become enormous and capable to cause damages when arriving at harbours and coasts. This fact is demonstrated by the 2004 Indian Ocean tsunami which showed that dangerous waves can move at trans-oceanic distances and induced damages on coasts far from the epicentre. The 2004 Indian Ocean tsunami is induced by disastrous severe earthquake at Mw 9.0 located at West Coast of Sumatra. Total of 14 countries surrounding the Indian Ocean experienced the impacts of this disastrous tsunami, including Malaysia during Sunda earthquake. The event created an estimated 800 km fault length, 85 km fault width and initial vertical displacement of 11 m which caused extensive damages and resulted to at least 227,898 fatalities (Loon et al. 2006). The impacts of tsunami are deadly resulting to massive loss of life, injuries and damages to properties. Millions of people’s lives have assuredly changed since this catastrophe. In Malaysia, 68 lives were reported dead due to this deadliest disaster and thousands of lives involved located mostly on the North of West Coast Peninsular Malaysia suffered including Penang, Langkawi, Kedah and Perak (Zahari et al. 2013). Due to unreadiness in confronting tsunami disaster, coastal communities at West Coast Peninsular Malaysia faced life-threatening circumstances and property damages (Colbourne 2005).

Malaysia is surrounded by active earthquake sources located at the regions of Andaman Sea and South China Sea which may trigger at any time. Until now, East Coast of Peninsular Malaysia has not experience any tsunami event, but in the past few years, several studies have found the existence of an earthquake epicentre located at South China Sea which will affect East Coast of Peninsular Malaysia, identified as Manila Trench (Huang et al. 2009; Mardi et al. 2017; Teh and Koh 2011). However, the potential impacts of tsunami induced by movements of seismic activities located off the coast towards coastal areas of East Coast Peninsular Malaysia have not been explored extensively. The active earthquake source from Manila Trench has been identified as high-risk zone that could potentially generate a basin-wide earthquake-induced tsunami within South China Sea (Li et al. 2017). Based on literature review conducted, tsunami waves generated from Manila Trench are expected to reach Malaysia coastal areas within few hours after the earthquake triggered. Unavailability of extensive studies and updated information on impacts of tsunami occurrences from Manila Trench towards Malaysia, especially at the East Coast of Peninsular Malaysia, will affect the country.

The occurrences of the 2018 Sulawesi Tsunami and the 2018 Sunda Strait Tsunami have triggered the urgency to investigate tsunami impacts to Malaysia. In Palu, Indonesia, total of two million people were recorded as victims of the 2018 Sulawesi tsunami that generated the maximum run-up height of 7 m (Hariyanto 2018). Thus, studies related to tsunami impacts are pertinent in order to provide vital information of this disaster, particularly to coastal communities. Therefore, this study is conducted to investigate the impacts of tsunami generated from Manila Trench earthquake source towards East Coast of Peninsular Malaysia. These dangerous waves are likely to adversely affect coastal areas of East Coast Peninsular Malaysia, namely states of Kelantan, Terengganu, Pahang and Johor. Three (3) different earthquake intensities were used in this study namely Mw 7.0, Mw 8.0 and Mw 9.0 in order to generate different scenarios of tsunami event.

Tsunami modelling by using TUNA model was conducted in order to record wave height, arrival time, inundation depth and inundation distance. TUNA model is capable to simulate complete tsunami processes at three (3) main phases, namely generation (GE), propagation (M2), run-up and inundation (RP). TUNA-GE, which applied Okada computation, was used to simulate the initial tsunami wave induced by earthquake events. Tsunami inundation maps were then developed by overlaying TUNA output files on Google Earth maps. Furthermore, major electrical assets identified to be located within the inundated areas of East Coast Peninsular Malaysia consist of power station and substation were demarcated on the inundation maps and submergence water depth at these affected electrical assets were established.

It is vital to analyse the impacts of tsunami towards electrical assets that generate and provide electricity to end users in order to ensure continuous electricity supply. Electricity is very important and has become a part in daily life. Nowadays, there are still lack of information and studies conducted on the vulnerability of important infrastructures including electrical assets (Williams et al. 2019). Hence, the efforts done in this study to analyse major electrical assets located in East Coast Peninsular Malaysia that potentially be impacted by Manila Trench tsunami are crucial and novel. Clear indications can be provided to the owner of power provider on which electrical assets anticipated to be affected and enable them to plan suitable and effective mitigation measures in order to minimise damages.

Besides that, findings from this study will provide information and awareness pertaining to possibilities of tsunami disaster induced from Manila Trench that may affect the coastal areas of Peninsular Malaysia. Coastal communities will be alerted on this disaster, and suitable evacuation time and location can be determined. Development of tsunami-induced inundation maps established in this study for Kelantan, Terengganu, Pahang and Johor is capable to assist various authorities and government agencies on preparation of evacuation plans at the affected areas. Besides that, these inundation maps can be used to establish zoning scheme for future potential coastal developments.

The 2006 United States Geological Survey (USGS) tsunami source workshop has identified three (3) subduction zones, namely North Sulawesi subduction zone, Ryukyu subduction zone and Manila subduction zone, capable to physically harm countries located in the vicinity of South China Sea (Liu et al. 2009). Manila subduction zone, otherwise known as Manila Trench, has been recognised as the largest and most dangerous earthquake-induced tsunami source in South China Sea region as compared to other earthquake sources (Liu et al. 2009). The length of Manila subduction zone is approximately 1000 km, starting from northern tip of Palawan, Philippine, expands to the north along the Western edge of Luzon and ends in Taiwan (Liu et al. 2009). Manila subduction zone formed by two (2) tectonic plates that collided with each other when Sunda Plate sub-ducted underneath the Philippines Sea Plate with depth trench of about 5400 m at approximately 22–25 million years ago (Liu et al. 2007). This giant megathrust rupture is adequate to cause massive earthquake, capable to generate tsunami waves at high speed of approximately 700 km/h (Ren et al. 2015).

Manila subduction zone is capable to induce an earthquake with intensity of Mw 9.0 which is as powerful as the 2004 Indian Ocean tsunami and the 2011 Tohoku-Oki tsunami (Hsu et al. 2016; Megawati et al. 2009). Based on Li et al, (2018), the highest intensity to be generated by Manila Trench is Mw 9.0. This is based on dip angle obtained from earthquake focal mechanisms with similar ranges when compared to other subduction zones known to produced earthquake at Mw 9.0 and above such as South America, Alaska–Aleutians, Sumatra–Java and Japan–Kuril–Kamchatka earthquake sources (Li et al. 2018). The disastrous tsunami waves with leading elevation crest are anticipated to propagate through South China Sea towards China, Taiwan, Vietnam, Thailand, Philippines, Hong Kong, Brunei and Malaysia (Dao et al. 2009).

Distribution of historical earthquake events in South China Sea is recorded by USGS database. Figure 1 shows a total of 74 earthquake events at intensities more than Mw 6.0 occurred at Manila Trench and nearby areas from year 1900 to 2020. Two (2) largest seaquake events occurred in Philippine Islands region and Mindoro, Philippines with intensity reaching to Mw 7.5.

Fig. 1
figure 1

Past earthquake events occurred at Manila Trench

Full size image

Using GPS measurements, it is estimated that convergence between Sunda Plate and Philippines Sea Plate ranges from approximately 50 mm/year in Taiwan to 100 mm/year near North Luzon, while near Zambales and Mindoro Island, the convergence rate is approximately 50 mm/year and 20 mm/year, respectively (Galgana et al. 2007).

Tsunami inundation maps are pertinent in the planning of tsunami early warning system. It provides crucial information on tsunami hazards and vulnerability of the coastal areas. This information helps authorities in making decisions on tsunami evacuation plans and sustainable coastal development plans. Functions of inundation maps are to present the level of tsunami vulnerability for coastal areas in order to authorise communities to act effectively during the catastrophes. Outcomes of inland inundation and run-up heights produced by tsunami modelling are used to develop these inundation maps (Tan et al. 2017). Long computational time is one of the serious limitations in most of tsunami inundation numerical models. Several studies have been conducted to enhance the computational efficiency of tsunami inundation simulation (Musa et al. 2018; Setiyono et al. 2017; Volpe et al. 2018; Zhang and Niu 2020). Other than using numerical modelling, maximum inundation height also can be predicted by using amplification factor approach (Glimsdal et al. 2019). Amplification factors define the relationship between offshore wave height and maximum inundation height, as expressed by linearised plane wave models used for incident waves at different wave characteristics (Glimsdal et al. 2019).

In Malaysia, many studies have been conducted to develop tsunami inundation maps for West Coast Peninsular Malaysia due to earthquake generated from Sunda Trench (Tan et al. 2016; Najihah et al. 2014; Ismail et al. 2012). On the other hand, there was also a study conducted to establish tsunami inundation maps for densely populated areas in Sabah and Sarawak, subjected to submarine slide-induced tsunami from Brunei Slide (Tan et al. 2017). However, there is still no study conducted to establish tsunami inundation maps at coastal areas of Malaysia due to Manila Trench tsunami. Therefore, this study was conducted to develop tsunami inundation maps for whole East Coast of Peninsular Malaysia including Kelantan, Terengganu, Pahang and Johor. This study will benefit coastal communities and authorities in Malaysia since mitigation measures can be planned in order to minimise the loss and damages due to tsunami disaster.

Tsunami impact has been categorised into primary and secondary impact. Primary impact refers to the affect caused by the force of tsunami waves when striking the shores and structures. Secondary impact refers to the affect caused by movement of debris and contaminated flood waters, generated by the primary impact. One of the examples of secondary impact is when explosions and fire triggered in a gas plant after being hit by tsunami debris that lead to the leakage of fuels (Vera San Martín et al. 2018).

During tsunami disaster, buildings and infrastructures were exposed to hazards of inundation. Based on reports and satellite images from Palu City due to the 2018 Sulawesi Tsunami it is revealed that buildings and infrastructure located within 300 m from the coastline experienced severe damages (Paulik et al. 2019). Based on the observation, inundation depths of more than 1.2 m will cause the building to suffer irreparable or complete damage (Paulik et al. 2019). Currently, studies on casualty estimations and buildings vulnerability due to tsunami impacts are well developed, but on the other side, there were still lack of information and studies conducted on the vulnerability of important infrastructures (Williams et al. 2019).

Electricity has become a necessity in life. However, the electrical systems of many facilities and equipment are extremely vulnerable to tsunami inundation (Imai et al. 2019). In many cases, even slight inundation can cause electrical system problems, causing a loss of power sources (Imai et al. 2019; Kisielewicz et al. 2011). After the 2011 Tohoku-Oki tsunami in Japan, it became crucial to build a more resilient power system. Japan experienced destructive damage that led to massive blackouts, affecting 8.5 million people (Aki 2017). Fukushima Daiichi Nuclear Power Station is one of the affected power stations. Damage on these power stations caused nationwide massive blackouts lasted for two (2) weeks. Due to the blackouts, train services were decreased by 30–50% and traffic signals were turned off which caused many students and workers could not reach their schools or offices (Aki 2017).

Many important infrastructures were identified to be located near to coastal areas of South China Sea region that are vulnerable to Manila Trench tsunami (Terry et al. 2017). In Western Luzon, Bataan Nuclear Power Plant is located within the inundation areas if tsunami from Manila Trench occurred. This power plant is facing directly to the subduction zone without any tsunami protection and is situated only 75 km away from Metro Manila that occupied about 12 million populations. At Subic Bay, residential assets and freeport commercial located there are also vulnerable due to its low-lying ground elevation and funnel shape of Subic Bay. Besides that, at western side of Lamma Island located in Hong Kong, a coal-/gas-fired power generation plant is potentially to be hit by dangerous tsunami waves travelling from open sea to the south (Terry et al. 2017). On the other side, Ling Ao and Daya Bay Nuclear Power Plants located in Guangdong province are expected to experience impacts of tsunami generated from the northern section of Manila Trench. The location of Daya Bay Nuclear Power Plant is approximately 30 km from eastern Hong Kong and metropolitan Shenzhen. This has worried many people, especially those living nearby.

In this study, the impacts of tsunami generated from Manila Trench were analysed on major electrical assets located in East Coast Peninsular Malaysia. The affected major electrical assets consist of power station, and substations were demarcated on the developed inundation maps and submergence water depth at these affected electrical assets was established. Electrical power station is an industrial facility for electric power generation. The electrical substations analysed in this study are transmission main intake (TMI), main distribution substation (MDS) and main switching station (MSS) (Hamdan et al. 2018). TMI is the interconnection point of 132 kV or 275 kV transmission system to the distribution system. MDS is the interconnection point between 33 kV distribution system and 11 kV distribution system, while MSS is a place to install high voltage circuit breaker without transformer (Hamdan et al. 2018). All these major electrical assets are very important in order to ensure the generated electricity reaches the consumers. Hence, this study is vital to identify the potential major electrical assets that will be affected by tsunami inundation in order to plan effective mitigation measures that can minimise the damages.

Methodology

Determination of the most significant earthquake fault parameters at Manila Trench

In order to determine the most significant earthquake fault parameters anticipated to affect East Coast of Peninsular Malaysia, fault parameters utilised by previous researchers to investigate tsunami induced by earthquake from Manila Trench had been studied. Next, the selected fault parameters were simulated utilising TUNA-M2 model in order to study the potential impacts of tsunami generated from Manila Trench towards coastal area of Peninsular Malaysia, particularly coastal areas of Terengganu. Fault parameters suggested by Wu and Huang (2009), Liu et al. (2009), Nguyen et al. (2014) and Mardi et al. (2017) were selected to be simulated in this study. This is because these studies considered Manila Trench earthquake source with six (6) fault segments as suggested by USGS. Using the simulated results obtained in this study, an intensive approach is then conducted to determine the most significant set of fault parameters with high potential to create the worst-case tsunami scenario towards offshore areas of Peninsular Malaysia in terms of the highest and fastest first wave arrived. For further analysis, the fault parameters were established as follows:

  1. (a)

    The Manila Trench subduction zone is divided into six (6) fault segments as proposed by USGS (Kirby et al. 2005).

  2. b)

    The latitude and longitude of fault segments, length of fault segments, dip angle and strike angle were set as proposed by Kirby et al. (2005).

  3. (c)

    The rake angle is assumed to be 90° in order to indicate the potential worst-case scenario for tsunami event as it creates the maximum seafloor deformation.

  4. (d)

    The focal depth is assumed to be 15 km as commonly found in many major seismic activities in the region of South China Sea.

  5. (e)

    The width of fault segments is determined by using the empirical formula as suggested by Papazachos (2004).

  6. (f)

    The dislocation (Dis.) is calculated using Eqs. 1 and 2.

    $$M_{ \circ } = \mu \left( {Dis.} \right)\left( L \right)\left( W \right)$$
    (1)
    $$M_{w} = \frac{2}{3}log_{10} M_{ \circ } - 10.7$$
    (2)

Based on all these considerations, the final fault parameters established in this study are tabulated in Table 1, which exhibit the proposed Manila Trench earthquake fault parameters for Mw 7.0, 8.0 and 9.0. In this study, different moment magnitudes (Mw) were used in the simulations in order to create various scenarios of tsunami events.

Table 1 Manila Trench earthquake fault parameters
Full size table

Preparation of study domain

Computational study domain that includes Manila Trench earthquake source and the study area, namely East Coast Peninsular Malaysia, is required to be prepared prior to the commencement of tsunami generation and propagation simulation. The computational domain is necessary to be performed in a rectangular shape to suit the requirements of TUNA-M2 model where Okada model and Shallow Water Equation (SWE) are adopted to generate the initial tsunami wave from Manila Trench source and propagate it throughout the open sea.

Hence, this study applies a computational domain within a rectangle bounded by 0°N to 25°N in latitude and 100°E to 125°E in longitude (2775 km × 2775 km) to be used for simulation of Manila Trench tsunami as shown in Fig. 2. The computational domain is gridded by a grid size of 1500 m with a grid dimension of 1851 × 1851. In order to ensure numerical stability, a corresponding time step of 0.1 s was set for simulation purposes (Aziz et al. 2018a, b). Four (4) numbers of proposed boundaries covering the entire coastline of East Coast Peninsular Malaysia were selected for simulation purposes as shown in Fig. 2. The function of the boundary is to capture tsunami wave height propagated to the study area. East boundary is 50–100 m water depth located along the coast of Kelantan, Terengganu, Pahang and Johor. This is because the propagation of tsunami waves generated using SWE is valid at any water depth (Koh et al. 2009a, b). Hence, based on tsunami simulation carried out in this study, the suggested 50–100 m water depth for east boundary adopted to capture tsunami wave height was found to be sufficient. This suggested water depth was also used by Tan et al. (2016), Tan et al. (2017) and Mardi et al. (2017).

Fig. 2
figure 2

Left: Proposed computational domain. Right: Proposed boundaries of study area

Full size image

Tsunami generation and propagation

In this study, all phases of tsunami are incorporated in the simulation, starting from the generation phase, propagation phase until the run-up and inundation phase. The phases are divided into two (2) main parts. The first part is when the tsunami waves are at sea (wet region); this involves the generation and propagation phases. The second part is when the tsunami waves are entering the land (dry region); hence, the run-up and inundation phase takes place. Tsunami propagation models are established to prepare the coastal communities anticipated to be exposed to tsunami threat and to provide near real-time projection during tsunami propagation through the open sea and before reaching the land (Aziz et al. 2018a, b). In this study, the run-up and inundation model utilises the most recent land topography and sea bathymetric data available to reproduce the desired wave dynamics during the inundation computation.

TUNA-M2 is used in this study, due to its ability to generate and propagate tsunami efficiently. Waves arrival time and waves height will be recorded at the study boundaries during simulation of TUNA-M2 model. Reliable results of wave heights and arrival time are captured at study boundaries, located within the range of 50–100 m ocean depth. Based on Eqs. 4 and 5, it is recommended that the ocean depth, D, can be selected at any values but not too small, else Eqs. 4 and 5 cannot be solved due to infinitely small values of result (Koh et al. 2009a, b). TUNA-M2 adopted Okada model to generate the initial tsunami wave height induced by earthquake and utilises Shallow Water Equation (SWE) to simulate wave propagation.

During tsunami propagation phase, travelling waves are generated by adopting the theory of SWE for non-dispersive waves with assumption that all waves propagate from deep to shallower ocean at the same speed with the same shape (Liu 2013). TUNA-M2 is a two-dimensional model which solves linear SWE as the governing equation following the proposal prepared by the Intergovernmental Oceanography Commission (IOC) (Teh et al. 2011). Since the wavelength of a tsunami wave is large and the wave height is much smaller than the depth of the ocean, SWE is suitable to be used for the analysis of tsunami waves. This condition is suitable for tsunami propagations in the South China Sea. Explicit finite difference with staggered grids is used to solve SWE in TUNA-M2. The SWE is displayed in Eqs. 3–5.

$$\frac{{\partial\upeta }}{{\partial {\text{t}}}} + \frac{{\partial {\text{M}}}}{{\partial {\text{x}}}} + \frac{{\partial {\text{N}}}}{{\partial {\text{y}}}} = 0$$
(3)
$$\frac{{\partial {\text{M}}}}{{\partial {\text{t}}}} + \frac{\partial }{{\partial {\text{x}}}}\left( {\frac{{{\text{M}}^{2} }}{{\text{D}}}} \right) + \frac{\partial }{{\partial {\text{y}}}}\left( {\frac{{{\text{MN}}}}{{\text{D}}}} \right) + {\text{gD}}\frac{{\partial\upeta }}{{\partial {\text{x}}}} + \frac{{{\text{gn}}^{2} }}{{{\text{D}}^{7/3} }}{\text{M}}\sqrt {{\text{M}}^{2} + {\text{N}}^{2} } = 0$$
(4)
$$\frac{{\partial {\text{N}}}}{{\partial {\text{t}}}} + \frac{\partial }{{\partial {\text{x}}}}\left( {\frac{{{\text{MN}}}}{{\text{D}}}} \right) + \frac{\partial }{{\partial {\text{y}}}}\left( {\frac{{{\text{N}}^{2} }}{{\text{D}}}} \right) + {\text{gD}}\frac{{\partial\upeta }}{{\partial {\text{y}}}} + { }\frac{{{\text{gn}}^{2} }}{{{\text{D}}^{7/3} }}{\text{N}}\sqrt {{\text{M}}^{2} + {\text{N}}^{2} } = 0$$
(5)

g (m/s2) is gravitational acceleration constant, n \((\mathrm{s}/{\mathrm{m}}^{1/3})\) is Manning’s roughness coefficient, M (m2/s) and N (m2/s) are horizontal discharged fluxes in x- and y-direction, respectively, related to velocities u (m/s) and v (m/s) as stated in Eqs. 6 and 7, h (m) represents water depth below mean sea level (MSL), \(\eta\) (m) represents vertical free surface displacement measured from MSL and D (m) =\(h+\eta\) is the total water depth.

$$M = u\left( {h + \eta } \right) = uD$$
(6)
$$N = v\left( {h + \eta } \right) = vD$$
(7)

In this study, the time iterations for all the simulations were set at 450,001 which mean 45,000 s, equivalent to 12 h and 30 min. This is because a corresponding computational time step (DELT) of 0.1 s was set for simulation purposes in order to ensure numerical stability (Aziz et al. 2018a, b). In this study, the decision to choose 12 h and 30 min as the maximum time iterations is based on literature reviews where estimated time taken for tsunami waves to travel from Manila Trench towards East Coast of Peninsular Malaysia is approximately 9 to 11 h after earthquake occurrences (Mardi et al. 2017).

Tsunami run-up and inundation

Based on finite difference method, an in-house model named TUNA-RP has been developed by USM in order to simulate tsunami waves at open ocean onto the shallow beaches. The function of TUNA-RP model is to generate run-up heights of the travelled waves along beaches and to provide the inundation distances on shore. TUNA-RP model has been validated with OAR-PMEL-135 (benchmark problems) established by US National Tsunami Hazard Mitigation Program (Koh et al. 2017). By considering the combinations of various source generation conditions, output from TUNA-M2 simulations is derived as input to TUNA-RP model. Then, the dangerous wave reaching depth at 50–100 m will be led to the coast and seashore. The Nonlinear Shallow Water Equation (NSWE) will be used for incoming tsunami waves at depths of 50–100 m onto the beaches in order to model the wave run-up.

Explicit finite difference method with staggered grids is utilised to solve NSWE. Explicit finite difference method performs well in solving NSWE and is adopted in established tsunami numerical models such as MOST and TUNAMI-N2. Figure 3 shows computational points for staggered scheme where g, u and v are located. Partial derivatives are replaced by finite differences as shown in Eq. 8, while time step Δt is restricted by Courant criterion in Eq. 9 to ensure stability of the numerical scheme. The staggered scheme (Yoon and Liu 1992; Yoon 2002; Koh et al. 2009a, b) is adopted to solve NSWE.

$$\begin{gathered}\upeta _{{{\text{i}},{\text{j}}}}^{{{\text{k}} + 1}} =\upeta _{{{\text{i}},{\text{j}}}}^{{\text{k}}} - \frac{{{\Delta t}}}{{{\Delta x}}}\left[ {{\text{M}}_{{{\text{i}} + 0.5,{\text{j}}}}^{{{\text{k}} + 0.5}} - {\text{M}}_{{{\text{i}} - 0.5,{\text{j}}}}^{{{\text{k}} + 0.5}} } \right] - \frac{{{\Delta t}}}{{{\Delta y}}}\left[ {{\text{N}}_{{{\text{i}},{\text{j}} + 0.5}}^{{{\text{k}} + 0.5}} - {\text{N}}_{{{\text{i}},{\text{j}} - .05}}^{{{\text{k}} + 0.5}} } \right] \hfill \\ {\text{M}}_{{{\text{i}} + 0.5,{\text{j}}}}^{{{\text{k}} + 0.5}} = {\text{ M}}_{{{\text{i}} + 0.5,{\text{j}}}}^{{{\text{k}} - 0.5}} - {\text{gD}}_{{1 + 0.5,{\text{j}}}}^{{\text{k}}} \frac{{{\Delta t}}}{{{\Delta x}}}\left[ {\upeta _{{{\text{i}} + 1,{\text{j}}}}^{{\text{k}}} -\upeta _{{{\text{i}},{\text{j}}}}^{{\text{k}}} } \right] \hfill \\ {\text{D}}_{{{\text{i}} + 0.5,{\text{j}}}}^{{\text{k}}} = {\text{h}}_{{{\text{i}} + 0.5,{\text{j}}}} + 0.5\left[ {\upeta _{{{\text{i}} + 1,{\text{j}}}}^{{\text{k}}} +\upeta _{{{\text{i}},{\text{ j}}}}^{{\text{k}}} } \right] \hfill \\ {\text{N}}_{{{\text{i}} + 0.5,{\text{j}}}}^{{{\text{k}} + 0.5}} = {\text{N}}_{{{\text{i}} + 0.5,{\text{j}}}}^{{{\text{k}} - 0.5}} - {\text{gD}}_{{1 + 0.5,{\text{j}}}}^{{\text{k}}} \frac{{{\Delta t}}}{{{\Delta y}}}\left[ {\upeta _{{{\text{i}} + 1,{\text{j}}}}^{{\text{k}}} -\upeta _{{{\text{i}},{\text{j}}}}^{{\text{k}}} } \right] \hfill \\ {\text{D}}_{{{\text{i}} + 0.5,{\text{ j}}}}^{{\text{k}}} = {\text{h}}_{{{\text{i}} + 0.5,{\text{ j}}}} + 0.5\left[ {\upeta _{{{\text{i}} + 1,{\text{j}}}}^{{\text{k}}} +\upeta _{{{\text{i}},{\text{j}}}}^{{\text{k}}} } \right] \hfill \\ \end{gathered}$$
(8)
$${\Delta t} \le \frac{{{\Delta x}}}{{\sqrt {2{\text{gh}}} }}$$
(9)

where η (m) is the vertical free surface displacement measured from mean sea level, u (m/s) are the fluxes velocity in x-direction and v (m/s) are the fluxes velocity in y-direction. The numerical approximations of derivatives on wave height value at a certain point in staggered scheme are defined by calculating the initial wave height value of a body of water located at an initial point k with coordinates i and j. When the body of water dispersed another point, as an example one point above point k, therefore, the point will be k + 1 while the coordinate turns to i, j + 1.

Fig. 3
figure 3

Computational points for a staggered scheme

Full size image

A corresponding time step of 0.1 s is set for both TUNA-M2 and TUNA-RP models to avoid additional interpolation on the time steps (Samarasinghe et al. 2022). Time step of 0.1 s has been widely used in tsunami simulation studies since it fulfilled the Courant criterion stability (Musa et al. 2018; Oishi et al. 2015). However, in this study, the grid size used in TUNA-M2 is different from TUNA-RP. In TUNA-M2, the grid size is set to be 1500 m, while in TUNA-RP, the grid size is set at 30 m in order to generate higher precision results in the development of high-resolution inundation map. Hence, in this study, interpolation process for TUNA-M2 output files is required to be created by using linear interpolation.

In addition, another required input files for TUNA-RP simulation are integrated topographic and bathymetric data. For purpose of simulating generation and propagation, ETOPO1 topography and bathymetric data obtained from NOAA are sufficient due to its occurrences in deep ocean. Since run-up and inundation phase occurred on land where valuable lives and properties existed, accurate and precise ground elevation data are pertinent to be prepared for simulation purposes. Initiatives were taken to obtain high precision and updated resolution of land topography data from Jabatan Ukur dan Pemetaan Malaysia (JUPEM).

The resolution of land topography data obtained from JUPEM is 30 m × 30 m grid size which was found to be excellent for tsunami simulation (Zolghadr et al. 2022). Study conducted by Montoya et al. in 2017 found that grids finer than 30 m resolution are not necessary when computing the outcomes of inundation depths, inundation distances and flow velocities at study areas (Montoya et al. 2017). For bathymetric data, the data used in this study are derived from ETOPO1. These two (2) types of data have to be integrated into a single combined file at 30 m × 30 m resolution. Both topographic and bathymetric data are interpolated by using Kriging method. Kriging is one of the spatial interpolation methods that apply advance mathematical formulas to evaluate unknown values at unknown points based on data obtained at known points.

Development of inundation maps and demarcation of affected major electrical assets

In this study, tsunami inundation map is developed to identify the area anticipated to be inundated by the tsunami event. The attempt to establish tsunami-induced inundation maps is pertinent, since it provides critical information on tsunami threats and vulnerability at coastal areas and hence helps in preparation of mitigation measures at the affected areas. Results obtained from simulation of tsunami run-up and inundation are used to develop the inundation maps.

Based on this information, inundation maps for state of Kelantan, Terengganu, Pahang and Johor were developed by graphical plotting in a mapping tool named Golden Surfer Software (Surfer). Surfer is used to plot raw data into coloured graphical map that displays information of inundated height and land inundation distance. These graphical maps were then converted and exported into Keyhole Markup Language (KML) format file in order to be overlaid on Google Earth platform. KML is an Extensible Markup Language notation to express geographic annotation and visualisation within Internet-based, two-dimensional maps and three-dimensional Earth browsers. It is specifically developed to be compatible with Google Earth. It became an international standard of Open Geospatial Consortium in 2008. The position of KML file overlaid on Google Earth platform is adjusted based on World Geodetic System (WGS84) UTM Zone 48N as geo-reference coordinate system.

This study further continued to demarcate the affected major electrical assets located within the anticipated inundation areas. These major electrical assets consist of electrical power station and substation. Information and locations of these assets, in terms of longitude and latitude, are obtained from the designated power provider. The locations obtained are overlaid on the inundation maps, and the affected electrical assets were analysed.

Inundated height at the affected major electrical assets is important since this information will benefit owner of the power provider to prepare suitable mitigation measures in the case of tsunami occurrences. In order to establish the submerged water depth at each affected major electrical asset, ground elevation data of that particular area are required. These data are obtained from the designated power provider or via Google Map using elevation finder. Upon obtaining the elevation data at each affected major electrical asset, it can then be overlaid on the inundation maps where the submerged water depth is calculated.

Results and discussion

Tsunami generation and propagation modelling

Simulation on generation and propagation phase of tsunami generated from Manila Trench towards East Coast of Peninsular Malaysia has successfully been conducted in this study. Tsunami wave propagation recorded from earthquake epicentre across South China Sea at Mw 9.0 is presented in Fig. 4. Red-coloured wave represents the elevation waves while blue-coloured wave is the depression waves. Most of the countries surrounding the vicinity South China Sea including Vietnam, Taiwan, China, Hong Kong, Philippines, Brunei, Thailand and Malaysia are anticipated to experience tsunami catastrophe at several hours after the generation phase.

Fig. 4
figure 4

Simulated tsunami wave propagation from Manila Trench earthquake source

Full size image

After an earthquake occurred at the epicentre, the energy transferred to the sea water surface and generates elevation and depression waves of tsunami. Based on Fig. 4, at zero (0) hour, the elevation waves are facing the South China, while depression waves are facing Philippines. After generation phase, the elevation tsunami waves travelled through open ocean towards countries located within South China Sea region. At 2 h after earthquake event, tsunami waves already arrived Taiwan, Vietnam and Palawan in Philippines. After 3 h, tsunami waves are anticipated to reach China, Hong Kong and Sabah. Sarawak is expected to experience tsunami from Manila Trench at 4.5 h after earthquake occurrences. These dangerous waves continued to propagate at south-west direction, and it appears that tsunami waves propagate towards offshore of East Coast Peninsular Malaysia at approximately after 9 h post-earthquake occurrences. This is due to the distance between Manila Trench earthquake source and Peninsular Malaysia which is about 2000 km away (Aziz et al. 2018a, b).

Table 2 summarises the simulated arrival time of tsunami waves from Manila Trench earthquake source towards the study area boundaries located at East Coast of Peninsular Malaysia at Mw 7.0, 8.0 and 9.0. The arrival time is when the first peak of leading elevation wave arrives at each node. Each study area boundary consists of several nodes. Hence, the results of tsunami arrival time for each study area were presented in range. Results recorded from the tsunami simulations in this study showed that tsunami waves generated by earthquake at Mw 9.0 arrived at the boundary earlier than other moment magnitudes (Mw). Tsunami arrival time slightly depends on the size of earthquake intensity. Larger earthquake intensity causes faster tsunami arrival time at coastal areas (Saleh et al. 2012). It can be concluded that tsunami waves generated from Manila Trench earthquake source are anticipated to arrive at East Coast of Peninsular Malaysia at approximately 9 h after earthquake occurrences.

Table 2 Arrival time of tsunami waves from Manila Trench earthquake source
Full size table

Tsunami run-up and inundation modelling

In this study, run-up and inundation modelling was conducted for East Coast of Peninsular Malaysia in order to simulate actual inundation event occurred inland during tsunami catastrophe. Table 3 summarises results obtained in this study for maximum inundated depth and inundation distance at different moment magnitudes (Mw) for overall areas of Kelantan, Terengganu, Pahang and Johor. The inundation depth is obtained from the ground level until the surface of flood, while inundation distance is the maximum distance calculated perpendicularly from the shoreline.

Table 3 Maximum inundation depth and inundation distance at different moment magnitudes
Full size table

Based on results obtained in this study, it was found that tsunami at Mw 9.0 is anticipated to incur the most impact towards East Coast of Peninsular Malaysia. Terengganu is expected to experience the most disastrous tsunami waves with maximum inundated depth at 6.0 m height and maximum inland inundation distance up to 6 km. This is followed by Kelantan with maximum inundated depth anticipated to reach 4.0 m and maximum inland inundation distance of 5.5 km. For Pahang, the highest inundated depth is anticipated at 4.5 m while the maximum inland inundation distance is up to 3.3 km. Johor which is located at the south of East Coast Peninsular Malaysia is expected to experience the lowest impacts from Mw 9.0 Manila Trench tsunami at maximum inundated depth of 1.2 m height and up to 0.4 km inland inundation distance. Based on these depth and distances, it can be concluded that Manila Trench earthquake generated tsunami waves of Mw 9.0 will incur devastating damages at East Coast of Peninsular Malaysia.

In this study, at Mw 8.0, the maximum depth of inundation is found at 0.14 m in Terengganu, followed by Pahang and Kelantan at 0.1 m and 0.09 m height, respectively. In Johor, the inundated depth is very insignificant at only 0.03 m. While for Mw 7.0, the maximum inundation depth is recorded at 0.04 m in Terengganu. On the other hand, the inundation distances at East Coast of Peninsular Malaysia for Mw 8.0 are found to range between 5 and 60 m, while the range for Mw 7.0 is from 1 to 3 m. Hence, there is no impact of tsunami waves generated from Mw 7.0 of Manila Trench earthquake tsunami at East Coast of Peninsular Malaysia. This is due to the smaller fault dislocation created by Mw 7.0 that generated smaller amount of energy in tsunami waves as compared to higher moment magnitudes (Mw). In addition, the distance of Peninsular Malaysia that is located further away from Manila Trench earthquake source contributed to the loss of energy in tsunami waves during propagation phase, causing the waves to dissipate before arriving at East Coast of Peninsular Malaysia.

Development of tsunami inundation map

In this study, tsunami inundation maps were developed and overlaid on Google Earth. The inundation maps were developed for four (4) states located at East Coast of Peninsular Malaysia namely Kelantan, Terengganu, Pahang and Johor. These tsunami inundation maps were developed for three (3) different earthquake intensities namely Mw 9.0, 8.0 and 7.0. Results obtained in this study showed that Manila Trench earthquake at Mw 9.0 is anticipated to produce disastrous tsunami impacts towards East Coast of Peninsular Malaysia, while at Mw 8.0 and 7.0, low tsunami impacts were identified towards the study areas.

Kelantan

Kelantan is located at the north of East Coast Peninsular Malaysia. It is an agricultural state with vast paddy fields, rural fishing villages and casuarina-lined coast. For simulation purposes, the entire coastline of Kelantan is divided into three (3) parcels namely K1, K2 and K3 as illustrated in Fig. 5. The districts anticipated to be impacted by tsunami generated from Manila Trench are Tumpat, Kota Bharu, Bachok and Pasir Puteh.

Fig. 5
figure 5

Parcels (K1–K3) of inundation map in Kelantan

Full size image

At Mw 9.0 earthquake, Kelantan is anticipated to be tremendously impacted by tsunami generated from Manila Trench. Table 4 tabulates overall inundation depths and inland inundation distances for Kelantan at Mw 9.0. The results showed that the highest inundation depth anticipated to be experienced by Kelantan is up to 4.0 m height, while the maximum inland inundation distance is expected to reach 5.5 km.

Table 4 Inundation depths and inundation distances in Kelantan at Mw 9.0
Full size table

As shown in Fig. 6(K1), the inundation area for parcel K1 is located from Taman Sri Dalam Rhu in Tumpat to Kubang Golok in Bachok where three (3) districts, namely Tumpat, Kota Bharu and Bachok, are situated. The maximum inundation depth is anticipated at 2.9 m and the maximum inland inundation distance is up to 4.5 km. Kubang Golok is expected to experience the highest inundation depth due to its short distance location to the nearby river mouth. Tsunami waves that mixed together with nearby river water resulted to increment of water level at nearby areas. Location of nearby river mouth at Pulau Gajah, Pengkalan Chepa, caused the incoming waves to reach further inland. Chances of Sultan Ismail Petra Airport in Pengkalan Chepa to be inundated are reduced due to its location further from the coastline. All villages located nearby the coastal areas are found to be inundated by this anticipated tsunami event.

Fig. 6
figure 6

Inundation map in Kelantan at Mw 9.0

Full size image

As shown in Fig. 6(K2), inundation map for parcel K2 includes areas in Bachok from Kubang Golok until Kampong Telong. The highest inundation depth was found to be at 3.5 m between Kubang Golok and Pantai Irama Bachok. The farthest inland inundation distance is 5.5 km due to location of nearby river known as Sungai Kemasin. Beginning from Kubang Golok, Sungai Kemasin flows south to Jelawat where the tsunami waves mixed with river water of Sungai Kemasin to move further inland. This is anticipated to cause Jelawat to experience the highest inland inundation distance.

Figure 6(K3) shows the inundation map established for parcel K3 where it includes the areas from Kampong Telong in Bachok to Kampung Pachakan in Besut located at the border of Kelantan and Terengganu states. Based on simulation conducted in this study, these areas are expected to experience significant impact with the highest inundation depth of up to 4.0 m and maximum inland inundation distance of up to 5.0 km between Kampung Nelayan Kandis and Tok Bali. Two (2) river mouths of Sungai Semerak at Tok Bali and Cherang Ruku caused nearby areas to experience higher inundation depth and further inland inundation distance. Coastal areas of Kelantan are found to be densely packed with fishing villages. Based on tsunami simulation carried out in this study, it is anticipated that all these villages will experience severe submergence by the incoming waves.

Terengganu

Findings obtained from simulation conducted in this study indicated that tsunami generated from Manila Trench is expected to produce significant impacts towards coastal areas of Terengganu. Terengganu is a state with a long coastline, with several offshore islands, popular of its tourism industries that captivated beachgoers and snorkelers due to its picture-perfect beaches. In addition, fishing and agriculture were also pertinent industries in Terengganu. In this study, the coastline of Terengganu is divided into 12 parcels (T1–T12) consists of districts namely Besut, Setiu, Kuala Terengganu, Marang, Dungun dan Kemaman as shown in Fig. 7.

Fig. 7
figure 7

Parcels (T1–T12) of inundation map in Terengganu

Full size image

Table 5 summarises results obtained on inundation depths and inundation distances for Terengganu at Mw 9.0. Inland intrusion of tsunami waves is expected at maximum distance of 6.0 km and up to 6.0 m inundation depth.

Table 5 Inundation depths and inundation distances in Terengganu at Mw 9.0
Full size table

Tsunami inundation event in Besut located at Parcel T1 was found to be slightly less as compared to other places in Terengganu with inundation depth of up to 1.5 m and maximum inland inundation distance at 0.5 km. Setiu located in parcel T2 is anticipated to confront inland intrusion of tsunami waves up to 2.9 km with a maximum inundation depth at 3.3 m. On the other hand, Kuala Terengganu located in parcels T3, T4 and T5 is found to be the most affected area in East Coast of Peninsular Malaysia with maximum inundation depth of 6.0 m and inundation distance of up to 6.0 km. Coastal areas of Kuala Terengganu are estimated to be completely submerged due its close proximity to the shoreline and its low-lying topography. Several fishing villages located near the shoreline are also anticipated to be severely inundated at Mw 9.0.

District of Marang located in parcel T6 and T7, is anticipated to experience tsunami inundation distance at a maximum of 1.7 km and maximum inundation depth at 5.9 m. Dungun, Terengganu, located in parcel T8–T11 is anticipated to experience very high inundation depth of up to 5.7 m, at the area between Kampung Kerta Pulas to Kampung Kongsi. The inland inundation distance is estimated to be of up to 4.6 km. This is because of nearby river mouth existence known as Sungai Dungun. Last but not least, Kemaman located in parcel T12 is anticipated to experience maximum inundation depth at 4.8 m and maximum inundation distance at 2.0 km. In this study, tsunami inundation maps of parcel T1–T12 were developed for Terengganu at Mw 9.0. The inundation maps established for parcel T4, T5 and T6 are illustrated in Fig. 8.

Fig. 8
figure 8

Inundation map in Terengganu at Mw 9.0

Full size image

Pahang

Pahang is the largest state in Peninsular Malaysia. It is acknowledged for its mountains, rainforests and beaches. Similar to Terengganu, Pahang also has long coastline. Three (3) districts are located at coastline of Pahang, namely Kuantan, Pekan and Rompin. These districts have the possibilities to face the tsunami waves originated from Manila Trench, particularly at the possible worst-case scenario of Mw 9.0. In this study, the coastline of Pahang is divided into 12 parcels for tsunami simulation purposes, namely P1–P12, as shown in Fig. 9.

Fig. 9
figure 9

Parcels (P1–P12) of inundation map in Pahang

Full size image

At Manila Trench earthquake of Mw 9.0, Pahang is anticipated to experience significant impacts of tsunami disaster. Table 6 summarises results of inundation depths and inundation distance at coastline of Pahang which have been divided into 12 parcels, P1–P12. These parcels comprise of coastal areas in Pahang from Kuantan to Rompin. Coastal areas of Kuantan are located in parcel P1 until half of parcel P5. Based on results obtained in this study, in Kuantan, the highest inundation depth is at Cherating in parcel P1 with 3.5 m height. Parcel P1 is illustrated in Fig. 10(P1). This is due to low-lying terrain of this area and nearby river mouth. Besides that, the maximum inland inundation distance in Kuantan reached up to 1.6 km at Pantai Teluk Cempedak, known for its beautiful beaches.

Table 6 Inundation depths and inundation distances in Pahang at Mw 9.0
Full size table
Fig. 10
figure 10

Inundation map in Pahang at Mw 9.0

Full size image

District of Pekan which is located at half of parcel P5 until half of parcel P10 is anticipated to experience maximum tsunami waves inland intrusion of up to 3.3 km and inundated depth is reaching to 4.5 m height. The maximum inland inundation distance appeared to be at parcel P6 where the river mouth of Sungai Pahang is located as illustrated in Fig. 10(P6). Theoretically, tsunami waves can intrude further inland with the existence of this river mouth. The highest inundation depth in Pekan is anticipated to be experienced at parcel P7 due to the low-lying terrain of this area as illustrated in Fig. 10(P7).

Coastal areas of Rompin are located from half of parcel P10–P12. The inundation depths and distances in Rompin were found to be quite low as compared to other locations in Pahang. The maximum inundated depth is found to be at about 1.5 m and the maximum inland inundation distance is 0.5 km. This is because Rompin is located further down at south of Pahang thus, further from Manila Trench earthquake source. In theory, energy in tsunami waves dissipated due to the further distance travelled. In this study, tsunami-induced inundation maps of parcels P1–P12 were developed for Pahang at Mw 9.0. The inundation maps established for parcels P1, P6 and P7 are illustrated in Fig. 10.

Johor

Johor is located south of Peninsular Malaysia, known for its beaches and rainforest. Its location is further from Manila Trench earthquake source as compared to Kelantan, Terengganu and Pahang. Hence, it is expected Johor will experience the lowest impact of tsunami induced by earthquake from Manila Trench as compared to the three (3) states located at East Coast of Peninsular Malaysia. In this study, coastline of Johor is divided into eight (8) parcels, J1–J8 comprising of two (2) districts, namely Mersing and Kota Tinggi, shown in Fig. 11. Mersing is located in parcel J1 until half of parcel J5, while Kota Tinggi is located in parcel half of J5 until parcel J8.

Fig. 11
figure 11

Parcels (J1–J8) of inundation map in Johor

Full size image

Based on the tsunami modelling conducted in this study at Mw 9.0, Johor is anticipated to experience impact of tsunami generated from Manila Trench earthquake source with inundation depths and inundation distances summarised in Table 7. Mersing located in parcel J1 until half of parcel J5 is anticipated to experience tsunami waves of up to 1.2 m inundation depth and up to 0.4 km inland inundation distance. Both the highest inundation depth and inland inundation distance are identified to be experience at parcel J1 since it is located at the uppermost of Johor, near to the border of Pahang and Johor.

Table 7 Inundation depths and inundation distances in Johor at Mw 9.0
Full size table

Hence, it can be concluded that the distance of the affected area and the earthquake source is an important factor in identifying the impact of tsunami event. The nearer the affected area to the earthquake source, the greater the impact received. Other factors affecting the degree of tsunami impact received are bathymetric of the ocean, topography of the coastal areas and existence of river mouth.

District of Kota Tinggi located in half of parcel J5 until parcel J8 is anticipated to experience maximum inundation depth at 0.8 m while the inland inundation distance is up to 0.20 km. Therefore, it can be concluded that most coastal areas in Kota Tinggi are expected to survive from this anticipated catastrophe. In this study, tsunami-induced inundation maps of parcels J1–J8 were developed for Johor at Mw 9.0.

Mw 8.0 and Mw 7.0 earthquake for Kelantan, Terengganu, Pahang and Johor

In this study, it is found that the aftermath effects of Mw 8.0 tsunami for Kelantan are very low since the highest inundation depth obtained in this study is 0.09 m and the maximum inland inundation distance is 0.6 m. Therefore, at Mw 8.0 tsunami, it can be concluded that, the impacts are insignificant to coastal areas of Kelantan. On the other hand, the results of inundation depth and distance for Mw 7.0 tsunami are found to be too small and insignificant since the inundation depth is found to be less than 1 cm and the maximum inland inundation distance is 2 m. Hence, it can be concluded that the state of Kelantan will not experience tsunami at Mw 7.0.

The maximum inundation depth, for Terengganu at Mw 8.0, is anticipated at only 0.14 m which is considered low. The maximum inundation distance is anticipated at 0.7 km. It can therefore be concluded that Terengganu is expected not to experience significant impact from tsunami originated from Manila Trench at Mw 8.0. Hence, Terengganu can be safe from danger of tsunami disaster from Manila Trench at Mw 8.0 and below. For tsunami at Mw 7.0, it is anticipated that Terengganu will not experience inundation since values on inundation depths and inundation distances were found to be small and insignificant. The highest inundation depths obtained at Terengganu coastline are 0.04 m, and the maximum inundation distance is found to be at 0.003 km. Therefore, it can be concluded that Terengganu is safe from impact of tsunami disaster generated from Manila Trench at Mw 7.0.

For Pahang at Mw 8.0 and Mw 7.0, it is anticipated that coastal areas of Pahang will not experience dangerous waves of tsunami since values obtained for inundation depths and inundation distances were found to be small and insignificant. For tsunami at Mw 8.0, the highest inundation depth is 0.1 m, and the maximum inland inundation distance was found to be 0.04 km. On the other hand, for tsunami at Mw 7.0, the highest inundation depths recorded for coastline of Pahang are 0.02 m and the maximum inland inundation distance is 0.001 km. Therefore, it can be concluded that Pahang is safe from the impact of tsunami disaster generated from Manila Trench at Mw 8.0 and Mw 7.0.

For Johor at Mw 8.0 and Mw 7.0, it is anticipated that Johor will not experience the foreseen disaster from Manila Trench since values of inundation depths and inundation distances were found to be small and insignificant. For tsunami at Mw 8.0, the highest inundation depths on the coastline of Johor are 0.03 m, while the maximum inland inundation distance was found to be 0.005 km. On the other hand, for tsunami at Mw 7.0, the highest inundation depths obtained for coastline of Johor are 0.01 m and the maximum inland inundation distance was found to be 0.001 km. Therefore, it can be concluded that Johor is safe from the impact of tsunami disaster generated from Manila Trench at Mw 8.0 and Mw 7.0.

Demarcation of affected major electrical assets and establishment of submergence depth

Based on the demarcation of major electrical assets at Mw 9.0, there are a total of 29 assets identified to be inundated by the anticipated tsunami event. On the other hand, no major electrical assets are anticipated to be affected by tsunami inundation event at Mw 8.0 and 7.0. The inland intrusion of tsunami waves at Mw 8.0 and 7.0 is anticipated not to reach any locations of the electrical assets at East Coast Peninsular Malaysia. Hence, all the major electrical assets are anticipated to be safe from the dangerous tsunami waves at Mw 8.0 and 7.0. In this study, the anticipated submergence water depth for each of the affected asset is categorised based on colour code. These colour codes are summarised in Table 8. Figure 12 shows sample from anticipated major electrical assets to be submerged by tsunami at Mw 9.0 for the four study areas.

Table 8 Colour code of submergence depth
Full size table
Fig. 12
figure 12

anticipated major electrical assets to be submerged by tsunami at Mw 9.0 for the four study areas

Full size image

Conclusion and recommendations

Simulations on Manila Trench tsunami were performed starting from generation, propagation to run-up and inundation on three different earthquake intensities, namely Mw 7.0, Mw 8.0 and Mw 9.0. In this study, the coastline of East Coast of Peninsular Malaysia is divided into four study areas which include all states located at East Coast of Peninsular Malaysia, namely Kelantan, Terengganu, Pahang and Johor. At Mw 7.0, wave arrival times at study area boundaries of East Coast of Peninsular Malaysia were found to be at range of 9.0–11.0 h, while simulation at Mw 8.0 indicated the range of wave arrival time at 8.89–10.85 h. Apart from that, tsunami waves at Mw 9.0 arrived at the boundaries between 8.72 and 10.7 h. Hence, it can be concluded that tsunami waves generated by earthquake from Manila Trench will reach East Coast of Peninsular Malaysia approximately nine hours’ post-earthquake occurrences. Results achieved from tsunami simulations in this study indicated that the highest possible Mw 9.0 will result to tremendous tsunami impacts towards the coastal areas. At Mw 8.0, the impacts were found to be lesser, while at Mw 7.0, no significant impacts from tsunami were identified to be experienced at the coastal areas.

From tsunami inundation maps, it was found that In Kelantan, at Mw 9.0, the maximum inundation distance is expected to reach up to 5.5 km and the maximum inundation depth is anticipated to rise up to 4.0 m. In Terengganu, tsunami generated from Manila Trench at Mw 9.0 is anticipated to reach the coastal areas at maximum inundation depth of 6.0 m and maximum inland inundation distance of 6.0 km. Apart from that, tsunami generated from Manila Trench at Mw 9.0 is anticipated to hit Pahang with maximum inundation depth of 4.5 m and maximum inland inundation distance at 3.3 km. The southern state of East Coast Peninsular Malaysia, Johor, is also anticipated to be impacted by Manila Trench tsunami at Mw 9.0 with maximum inundation depth of 1.2 km and maximum inland inundation distance at 0.4 km.

In this study, run-up and inundation simulations were conducted using bathymetric data obtained from ETOPO1 at unsimilar resolution with topography data obtained from JUPEM. Hence, it requires additional work to interpolate and integrate both types of data to become an acceptable computational domain. Therefore, for future studies, it is highly recommended to obtain bathymetric data from local government agency, namely Pusat Hidrografi Nasional Malaysia (PHN), where finer resolution of bathymetric data can be acquired from PHN as compared to ETOPO1 data.