Investigation of sediment thickness in the rammang-rammang maros karst area using microtremor method

The aims of this study were to analyze the value of dominant frequency and amplification, estimate the thickness of the sediment, and to evaluate the seismic vulnerability of the Rammang-Rammang Maros Karst Area. This work was carried out at the Rammang-Rammang Maros Karst Area, Salenrang and Bontolempangan Village, Bontoa District, Maros Regency, South Sulawesi. Microtremor data were analyzed using the Horizontal to Vertical Spectra Ratio (HVSR) method to obtain the H/V curve representing the dominant frequency and amplification values. The dominant frequency values were obtained in the range from 1.389 to 6.251 Hz. The amplification values were in the range from 0.455 to 3.944. The sediment thickness was obtained in the range from 18.521 to 88.029 m, and the seismic vulnerability was in the range from 0.040 to 5.761. High values of sediment thickness were obtained in riverbank areas, which indicated that the weathering that occurred was caused by weathering of limestone and sedimentation by rivers. Meanwhile, low sediment thickness values were obtained in karst hill areas, indicating that the weathering occurred due to the limestone dissolution process. Based on the seismic vulnerability and amplification, the Rammang-Rammang Maros Karst Area is in a relatively safe zone.


Introduction
Karst is a typical landscape from the surface and subsurface formed by the dissolution and deposition of easily soluble rocks, mainly carbonate rocks, by groundwater flow and has a well-developed secondary porosity [1,2,3].Astronomically, the Maros Pangkep Karst area is located between 4°42'49"-5°06'42" S and 119°55'13" E. The total area of Maros-Pangkep Karst is 43.750 ha [2].The uniqueness of the Maros-Pangkep Karst in South Sulawesi that is not owned by other karst areas, is the presence of a unique and distinctive landscape called the karst tower.
The Rammang-Rammang Karst area is part of the Maros Karst cluster located in Salenrang and Bontolempangan Village, Bontoa District, Maros Regency, South Sulawesi.In this karst area, a river flow empties into the sea called the Sungai Pute [3].The activities of tourists visiting the Rammang-Rammang Karst Area and the activities of the local community in the tourism development, such as using boats, causes river erosion.This erosion can be proven by shrinking the river body near the pier.This erosion causes an increase in sedimentation rate and sediment layer thickness, which has an impact on reducing the river capacity and affects the velocity of body wave propagation.
Sediments are fragments of minerals or organic matter transferred from various sources and deposited by the medium of air, wind, ice, or by water [4].The thickness of the sediment layer of an area can be determined from the dominant frequency and amplification value that can be obtained from microtremor measurements.Microtremor is a natural harmonic vibration over a short period arising from small vibrations below the ground surface or due to artificial sources.Microtremor method is widely used in research on soil structure to obtain information about the characteristics of soil sediment layers [5].
Data from microtremor measurements can be analyzed using the HVSR method, which estimates the characteristics of subsurface sediment layers [6].The parameters obtained from data processing with the HVSR method are the dominant frequency f 0 and amplification A 0 which interprets the geological conditions of an area [7].The sediment thickness can be related to the dominant frequency of soil vibration and the velocity of the shear wave ( ).
where is the thickness of the sediment layer (in m), V s is the velocity of the shear wave (in m/s), and f is the dominant frequency (in Hz) [8].
The dominant frequency value obtained from microtremor measurements can be utilized to analyze the soil type of the measurement area.One of the classifications of soil types used in classifying dominant frequency values is as follows.Sediment thickness plays a role when an earthquake occurs, namely the amplification effect/ strengthening earthquake [10].Amplification is the magnification of seismic waves that occur due to significant differences between layers.In other words, seismic waves will experience magnification if they propagate in one medium to another medium that is softer than the initial medium through which it passes [11].The soil amplification value obtained in microtremor measurements is divided into 4 zones, namely: (a) low amplification zones A 0 < 3; (b) medium amplification zone 3 ≤ A 0 < 6; (c) high amplification zone 6 ≤ A 0 < 9; and (d) very high amplification zones A 0 ≥ 9 [ 2].
Information about the dominant frequency and amplification of an area can also be used to analyze the seismic vulnerability of an area to earthquakes vibration.The seismic vulnerability index (Kg) is an index that describes the level of vulnerability of the surface soil layer to deformation during an earthquake [13].The seismic vulnerability index relates to the geomorphological conditions of an area.If an area with a thick sedimentary layer is accompanied by amplification of seismic wave vibrations, it will produce a large seismic vulnerability index value [14].The seismic vulnerability index can be calculated using the following equation.
where g, A , and f are the seismic vulnerability index (in s), the amplification value, and the ground dominant frequency (in Hz), respectively [15].The seismic vulnerability index values are classified into three classes, namely (a) low seismic vulnerability index Kg < 3; (b) medium seismic vulnerability index 3 ≤ g < 6; and (c) high ei mic vulnerability index Kg > 6 [16].

Methodology of Research
The microtremor measurement was conducted in the Rammang-Rammang Maros Karst Area, Salenrang and Bontolempangan Village, Bontoa District, Maros Regency, South Sulawesi.. Astronomically, it is located between 4°54'51.61"-4°55'50.51"S and 119°35'30.38"-119°37'22.58"E. The microtremor measurements are conducted according to the standards made by SESAME European Research Project.Microtremor measurements were carried out on an area of about 2.62 km 2 consisting of 11 measurement points.Microtremor data recording was taken for ± 60 minutes with an additional time of 5 minutes for measurement points with relatively much noise.The measurement points should be located on solid and flat ground, away from buildings, and large trees that will cause noise and in areas that are relatively quiet from community activities.The choice of measurement points is conducted in a spread way, including near river bodies, around residential areas, and on karst hills to indicate differences in the thickness of the sediment layer in these three areas.
Microtremor data processing was analyzed using Horizontal to Vertical Spectra Ratio (HVSR) method.The measurement results were recorded in three component vibration signals, namely horizontal (East-West), horizontal (North-South), and vertical (Up-Down).Data in mini seed format can be processed using Geopsy software to obtain a horizontal to vertical (H/V) spectrum ratio.Each measurable signal component in the measurement is analyzed using the Fast Fourier Transform (FFT) algorithm.Fourier transform processing aims to convert signal components from the time domain in the initial microtremor data into frequency domains so that the spectrum of each signal component is produced [17].
where ( ) is the Fourier transform of f(t).
The subsequent data processing is filtering and windowing signals with large noise, which is performed with Geopsy.Then, the process of combining signal components into the HVSR analysis is carried out.The ratio of the spectrum between the total horizontal component to the vertical component is calculated using the following equation.
where is the ratio spectrum HVSR, (f) is the horizontal component spectrum east-west, (f) is the horizontal component spectrum north-south, and (f) is the vertical component spectrum [6].The result of combining signals in HVSR analysis is an HVSR curve that present information about dominant frequency values and amplification (H/V).

Result and Discussion
The microtremor measurement results obtained for each measurement point are processed using Geopsy software to obtain dominant frequency and amplification values.Then, by adding soil shear wave velocity data obtained through the USGS website, the three data can be used to determine sediment layer thickness values and seismic vulnerability indices, as shown in Table 2.According to [9,17], soil types obtained from the research location can be classified by dominant frequency and the results are listed in Table 3.The dominant frequency value obtained in Bontolempangan Village varies despite the small space between measurement points.The dominant frequency values for points T1 and T2 are 5.49386 Hz and 6.25119 Hz, while for points T3, T4, and T5 range from 1.38898 -2.69978 Hz.The difference in values between these points is due to geological conditions at different research locations.For points T1 and T2 are in karst hills while points T3, T4, and T5 are close to the river so that the dominant frequency values obtained at points T1 and T2 are greater than the values frequencies obtained at points T3, T4, and T5.
Based on the classification by John et al. and the Building Seismic Safety Council, it was found that rocks and hard soils dominated the soil type in the research area.This result is in line with previous research on the physical and mechanical properties of Maros Karst rocks [19].The previous research states that the physical properties of Maros Karst rock have a density of 2-3 gr/cm 3 and a porosity value of 4-30 %, which indicates that the type of rock in the Taman Batu Maros is a type of karstified dolomite and limestone.The mechanical properties of Maros Karst rock have an average compressive strength value of 26.67 MPa.According to International Social Rock Minerals, rocks with compressive strength worth 25-50 MPa are categorized as medium-hard rocks.
Based on the results of processing HVSR curve data, a map of the distribution of soil amplification values in the Rammang-Rammang Maros Karst Area was obtained (see Figure 3).The soil amplification value obtained in the range from 0.455115 to 3.94393.Measurement point 2 has the smallest amplification value of 0.455115 while measurement point 11 has the largest amplification value of 3.94393.Based on the determination of the amplification zone [12], the soil amplification zone of this study can be obtained and listed in Table 4.

Table 4. Classification of Amplification Zones based on Amplification
Value.

Zone
Points Amplification Value Zone 1 T1, T2, T3, T4, T7, T8, T9, T10 As seen in Table 4, there are eight measurement points included in zone 1 which is a low amplification zone and there are three measurement points included in zone 2 which is a medium amplification zone.The amplification value obtained in this study is included in the safe zone which means that when vibration occurs, the amplification caused is not large and does not cause damage.The area with a high soil amplification value is the measurement area near the river while the area with a low amplification value is in the karst hills area.The reason is because the amplification factor depends on the wave velocity, which is related to the rock density.A decrease in rock density will increase the amplification value.Limestone as a rock that dominates the Rammang-Rammang Maros Karst Area is included in medium hard-rock based on its compressive strength.Therefore, the amplification value in the karst hills area is low due to the low density of the limestone rock.

2 Sediment Layer Thickness of Rammang-Rammang Maros Karst Area
The sediment thickness of the Rammang-Rammang Maros Karst Area can be calculated using equation ( 1) and the map is shown in Figure 4. Based on the results of overlaying contour maps with measurement location acquisition maps, it was found that the thickness of sediment layers are thicker was near rivers with a low dominant frequency, while the thickness of sediment layers was thinner near residential areas and karst hills with a high dominant frequency.This result indicates that the thicker sediment layers found close to the river result from limestone weathering and sedimentation by the river.The thinner sediment layers found in residential areas and karst hills interpret that sedimentation is caused only by limestone weathering.
The data processing results show that sediment thickness is related to the dominant frequency, so if the dominant frequency is higher, the sediment thickness is thicker.Conversely, the lower the dominant frequency, the thinner the sediment thickness.Based on these results, it is known that the dominant frequency and thickness of sediments have an inversely proportional relationship.The sediment layer's thickness will affect the buildings' response above the surface when a vibration occurs (earthquake) because the magnitude of the earthquake will be amplified.

3 Seismic Vulnerability of Rammang-Rammang Maros Karst Area
Based on the calculation results using Equation ( 2) where the dominant frequency and amplification values was obtained from the HVSR curve analysis, the seismic vulnerability map of the Rammang-Rammang Maros Karst Area is obtained as shown in Figure 5.The seismic vulnerability index value obtained ranged from 0.039626 to 5.761426.Measurement point 2 has the smallest seismic vulnerability index of 0.039626, while measurement point 11 has the largest seismic vulnerability index of 5.761426.Based on the classification by [15], the seismic vulnerability index obtained in this study can be classified in the following table.The seismic vulnerability index value is obtained by analyzing dominant frequency values and amplification from the HVSR curve for each measurement point.The dominant frequency and amplification factors influenced the variation in seismic vulnerability index values obtained in this study.The seismic vulnerability index is directly proportional to the magnitude of amplification and inversely proportional to the dominant frequency of the soil.The dominant frequency and amplification factor obtained for each measurement point are different because of the different geological conditions at each measurement point.The measurement point near the river is an area of soft and medium soil, so it has a higher seismic vulnerability.The karst hills area is hard soil with a lower seismic vulnerability index.Based on the parameters used in its calculations, a high seismic vulnerability index is obtained at points with low dominant frequency values and high amplification.
The seismic vulnerability index can be correlated to the geomorphological conditions of an area.If an area has a higher seismic vulnerability index, the level of earthquake risk will also be higher.A high level of seismic vulnerability index is generally found in areas with a low dominant frequency.This implies that relatively thick sedimentary layers that cover the bedrock have a high seismic vulnerability index.In thick sedimentary layers, if followed by the amplification of seismic wave vibrations, it will produce a high seismic vulnerability index value as well.
The previous research about the seismic vulnerability index has been conducted by Sababurrohman in the mining area dominated by limestone [20].Sababurrohman reported that the low seismic vulnerability index in the research area was due to geological conditions in the surface layer consisting of limestone which has more compact rock properties.

Conclusion
From the experimental results, several conclusions can be drawn as follows.The distribution of dominant frequency values obtained in the ranges from 1.389 Hz to 6.251 Hz and is classified according to the NEHRP.The distribution of amplification value in the Rammang-Rammang is classified according to the NEHRP.The Rammang Maros Karst Area is included in a safe zone with soil amplification values in the rang from 0.455 to 3.944.This means that when there is a vibration, the wave amplification factor that occurs in the region is not large.It was found that the Rammang-Rammang Maros Karst Area is an area dominated by rocks and hard soil, but some of it is soft soil.
The high sediment layer thickness in the Rammang-Rammang Maros Karst Area was obtained in the riverside area which indicates that the weathering that occurred was caused by limestone weathering and sedimentation by the river.Low sediment layer thickness values were obtained in karst hills areas and residential areas indicating that weathering was caused only by limestone weathering.The value of the thickness of the sediment layer obtained ranges from 18.521 m to 88.029 m.
The seismic vulnerability index values were in the range from 0.040 to 5.761.The seismic vulnerability index value obtained is below 6, meaning that the Rammang-Rammang Maros Karst Area has a small level of damage risk when a vibration (earthquake) occurs.The Rammang-Rammang Maros Karst area is cathegorized as an area with a low to medium seismic vulnerability index value.

10thFigure 2 .
Figure 2. Map of the Distribution of Dominant Frequency Values of the Rammang-Rammang Maros Karst Area.The dominant frequency value obtained in this study ranged from 1.38898 Hz to 6.25119 Hz.Measurement point 3 has the smallest dominant frequency value of 1.38898 Hz, while measurement point 2 has the largest dominant frequency value of 6.25119 Hz.Low dominant frequency values are obtained at measurement points around rivers with a value range of 1-3 Hz, dominant frequency values that are being obtained at measurement points around residential areas with a range of values 3-4 Hz, while the dominant high frequency value is at the measurement point in the karst hilly area with a value range of 4-7 Hz.According to[9,17], soil types obtained from the research location can be classified by dominant frequency and the results are listed in Table3.

Figure 3 .
Figure 3. Map of the distribution of soil amplification values of the Rammang-Rammang Maros karst area.

Figure 4 .
Figure 4. Map of the thickness of the sediment layer of the Rammang-Rammang Maros karst area.The value of the thickness of the sediment layer obtained ranges from 18.52082 m to 88.02863 m.Measurement point 4 has the smallest sediment layer thickness of 18.52082 m, while measurement point 8 has the largest sediment layer thickness of 88.02863 m.The sediment layer thickness value is obtained from the dominant frequency analysis of the HVSR curve and soil shear wave velocity.In this study, the value of the shear wave velocity used is the value of Vs 30 , which is the average value of the shear wave velocity at depths up to 30 m obtained through the USGS website with velocity values ranging from 310.3822193 m/s to 496.1420403 m/s.Based on the results of overlaying contour maps with measurement location acquisition maps, it was found that the thickness of sediment layers are thicker was near rivers with a low dominant frequency, while the thickness of sediment layers was thinner near residential areas and karst hills with a high dominant frequency.This result indicates that the thicker sediment layers found close to the river result from limestone weathering and sedimentation by the river.The thinner sediment layers found in residential areas and karst hills interpret that sedimentation is caused only by limestone weathering.The data processing results show that sediment thickness is related to the dominant frequency, so if the dominant frequency is higher, the sediment thickness is thicker.Conversely, the lower the dominant frequency, the thinner the sediment thickness.Based on these results, it is known that the

Table 2 .
The results of microtremor data processing at coordinates 4°54'57.3"-4°55'45.5"Sand119°35'58.1"-119°36'52.0"E.Distribution of Dominant Frequency Value and Amplification Value of Rammang-Rammang Maros Karst Area Based on the results of processing HVSR curve data, a map of the distribution of dominant frequency values in the Rammang-Rammang Maros Karst Area was obtained, as shown in Figure2.

Table 3 .
Classification of soil types by dominant frequency.

Table 5 .
Seismic Vulnerability Index Value Classification