Micro-Zonation of hot spring manifestation site at Umbul Niti, South Lampung based on microtremor

In 2021, an intriguing event unfolded at Umbul Niti village when a resident’s well drilling endeavor unexpectedly yielded to a hot spring, thus transforming the site into a small local tourist attraction that captivates visitors and enhances local economies. However, it is crucial to prioritize the safety of visitors and infrastructure, considering the geological implications of hot springs are related to unstable weak zones that would be severely damaged when an earthquake strikes. Weak zones often exhibit heightened susceptibility to seismic events. Using Horizontal to Vertical Spectral Ratio (HVSR) analysis of micro-tremor data that was recorded for 40 minutes at 25 locations, the Amplification (A 0) and Natural Frequency (f 0) were obtained; then the Seismic Vulnerability Index (Kg ) were calculated. The study area has a dominant frequency of around 4.74 - 9.7 Hz indicating the dominance of hard Tertiary Type I soil from Lampung Formation (QTI). Kg value ranges from 0.08 - 3.57 interpreted as mostly low hazard level. However, in the NW - SE direction the value indicates a high Kg value which means a relatively high hazard level, passing through the hot spring location. Thus, the area at the NW-SE zone might require more specific treatment.


Introduction
The development of tourist attractions often hinges upon the discovery of unique natural features that captivate visitors and enhance local economies.In 2021, an intriguing event unfolded at Umbul Niti village, South Lampung when a resident's well drilling endeavor unexpectedly yielded hot water [1], thus transforming the site into a small local tourist attraction [2].The emergence of this natural hot spring presents an opportunity for the area to be developed into a professional tourist destination.However, it is crucial to prioritize the safety of visitors and infrastructure, considering the geological implications associated with hot springs and their connection to potentially unstable weak zones [3].Weak zones often exhibit a high seismic vulnerability index (  ), therefore seismic vulnerability index mapping and micro-zonation analysis of these regions becomes imperative for effective hazard mitigation urban planning, and engineering design [4].
Various methods exist for micro-zonation, including analyzing recordings from earthquakes, explosions, or numerical simulations.However, these approaches are challenging to apply and come with high costs [5].Efficient application of earthquake recordings is hindered in regions with low seismic activity due to insufficient data.In such scenarios, an alternative approach involves capturing 1314 (2024) 012083 IOP Publishing doi:10.1088/1755-1315/1314/1/012083 2 micro-tremor signals (ambient seismic noise) generated by various natural or anthropogenic sources, such as wind, traffic, and ocean waves.This method can be used to estimate the natural frequency and the amplification factor using the Horizontal-to-Vertical spectral ratio (HVSR).
Certain seismologists have drawn a connection between the HVSR of micro-tremors and the HVSR of Rayleigh waves [6].This correlation is based on the resemblance in characteristics, as ambient noise predominantly comprises surface waves.[7].However, Nogoshi and Igarashi [8] and Ohta [9] pointed out that the peak frequency of the horizontal-to-vertical spectral ratio (HVSR) does not contain Rayleigh wave energy.As a result, this frequency peak can be attributed to the multiple reflections of SH waves.In the presence of a soft sediment layer, the horizontal movement of the sediment is influenced by the multiple reflections of SH waves, leading to greater horizontal displacement compared to vertical motion.Conversely, in the case of a solid bedrock layer, the horizontal movement matches the vertical motion [10].This study aimed to map the seismic vulnerability index and identify potential seismic hazards associated with the geological features present in the area.
The region is situated within the collision zone between the Eurasian Plate and the Indo-Australian Plate.Subduction of the Indo-Australian Plate beneath the Sunda Plate has resulted in the formation of the Sunda Arc, which includes active volcanic arcs and trench-forearc systems.The region's convergence between these plates has led to intense deformation and seismic activity.The major fault systems, such as the Semangko Fault, have influenced the deformation pattern and controlled the distribution of lithological units.Folds, both tight and open, are observed in several areas, reflecting compressional forces.Volcanic domes, associated with volcanic activity, are common in the region [11].The lithological units in Umbul Niti Village comprise a diverse range of rock types, including volcanic rocks, sedimentary rocks, and metamorphic rocks.Volcanic rocks, such as basalt, andesite, and rhyolite, are abundant in the region due to volcanic activities associated with the subduction zone.Sedimentary rocks, including sandstone, mudstone, and limestone, are present in various formations.Metamorphic rocks, such as schist and gneiss, can be observed in localized areas.
The stratigraphic sequences in Umbul Niti Village consist of several formations, representing different geological periods.The oldest unit is the Pre-Tertiary basement complex, composed of metamorphic and igneous rocks.Overlying this complex is the Tertiary volcanic sequence, characterized by extensive lava flows, pyroclastic deposits, and volcanic breccias.The Neogene sedimentary units, including sandstones and mudstones, unconformably overlie the volcanic rocks.Quaternary deposits, such as alluvium and volcaniclastic deposits, are present in river valleys and low-lying areas.

Horizontal-to-vertical spectral ratio (HVSR)
Nakamura defines micro-tremor as consisting of Rayleigh waves and body waves.If the typical geological structure of a sedimentary basin is shown in Figure 1, then the horizontal () and vertical spectra () on the surface ground of the basin are written as follows The amplification factor of horizontal  ℎ and vertical   motion of sedimentary ground the ratio between the frequency spectra of the surface and the basement rock.Here in equations ( 1) and (2); ℎ and  represent horizontal and vertical amplification factors for incident body waves,  and  denote horizontal and vertical spectra of the basement ground, and  and  correspond to the spectra of horizontal and vertical directions of Rayleigh waves.Shear wave velocity and weathering layer thickness affect the natural frequency [12].In the sedimentary layer, when there is no effect of the Rayleigh wave, the horizontal motion can receive a large amplification while the vertical motion stays unamplified.This is caused by the multiple reflections of SH waves.In this case, the value of  becomes equal to .When Rayleigh wave affects the motion at the surface, then  will be much larger than , and the amplification factor at the surface can be written as equation 5 [13].
The parameter related to the vulnerability level of an area caused by an earthquake is called the seismic vulnerability index (  ).The bigger the value of the seismic vulnerability index, the higher the alert of that area.The index could be calculated using equation (6). 0 is the amplification factor, while  0 is the natural frequency [14].

Micro-zonation and soil classification
Amplification could be used to determine the seismic hazard rank as proposed by Midorikawa.It measures how much the ground motion will be amplified when an earthquake strikes.The site is classified into four categories from very low to relatively high hazard [15].Kanai [16] classified soil types based on their natural frequency and predominant period from very soft to hard; and based on their seismic vulnerability index from low to high.If the earthquake's frequency is the same as the site's, the ground will resonate and produce higher amplitude.The predominant period indicates subsurface lithology characteristics.The vulnerability index calculates the risk level when an earthquake takes place.A high seismic vulnerability index is associated with soft sediment which prone to ground movement.

Methods
25 locations of the three components (North -South, East -West, and Vertical/ Up -Down) microtremor survey was recorded in 2023 near the Umbul Niti hot spring, covering about 4 km 2 area as seen in Figure 2.Each data has a 30-40 minute length of recording and was separated around 500 m.The data recorded meet the five reliability and clarity criteria by SESAME European Research Project 2004 [19].The hot spring is in alignment with the existing fault (red line in Figure 2) which also aligns with The Way Belerang Hot Spring.The research follows steps as shown in Figure 3.After data acquisition, data processing was done using Geopsy, an open-source software.Butterworth filter from 1 -10 Hz was applied to eliminate highfrequency signal.Seismograms after filtering from the MSJTM03 point are given in Figure 4 (a).The uppermost is the vertical component, followed by the North-South component in the middle, and the East-West component in the lower one.Data filtering was done to eliminate some high-frequency signals.The micro-tremor wave is characterized by stable non-spiky amplitude.Short Time Average over Long Time Average (STA/ LTA) technique was deployed to determine the waves accurately (colored region in Figure 4 (b)), and window the signal so that only the micro-tremor signal would be analyzed further.The window length met the  0 > 10/( ℎ) criteria.Each signal chunk was then Fast Fourier transformed so that the spectral analysis could be conducted to determine natural frequency and amplification.The peak of the H/V curve indicates the natural frequency value and the level of the correspondence peak is the Amplification.Both natural frequency and amplification were utilized to calculate the seismic vulnerability index.Three maps:  0 ,  0 ,    were created to visualize the distribution of each parameter.Interpretation of these three parameters was conducted based on table from Kanai and Midorikawa.Natural frequency and amplification are determined from the peaks exhibited from each point.The result is listed in Table 1.Natural frequency ( 0 ) is rather high ranging from 4.74 -9.70 Hz, giving a predominant period range of 0.103 -0.211 s while the range for amplification is 0.88 -5.53.The calculation gave   from 0.08 -3.57.The interpretation of each point is given in Table 1.Due to its predominant period, the study area mostly consists of Type 1 (hard) soil from the Tertiary Lampung Formation (QTI) which could consist of Tuff.Several areas (MSJTM03 and MSJTM10) are considered Type II (medium) soil with thinner layers.Two locations are interpreted as having relatively high hazard levels due to their amplification and   : MSJTM09 and MSJTM19.MSJTM19 is a survey location with the closest distance from the Umbul Niti hot spring.One location is considered to have a moderate hazard level (MSJTM15).The spatial distribution of each parameter ( 0 ,  0 ,   ) is presented in Figure 6.Measurement points are indicated by a black triangle.The Umbul Niti hot spring itself is labeled as 0 (zero).Due to natural frequency, Type II soil is distributed in the direction of North East (NE) from Umbul Niti hot Spring (reddish area in Figure 6 (a)).Amplification informs the relatively high hazard levels are also located in the NE direction while to the South East (SE) of Umbul Niti hot spring, the hazard level is considered moderate (reddish area in Figure 6 (b)).The two locations (NE and SE) are also categorized as mediumrisk zones by the seismic vulnerability index.Despite considered as Type II (medium) soil, around Umbul Niti hot spring itself is considered as very low to low seismic hazard zone.However, some safety measures remain compulsory because the subsurface condition might change over the time due to the hot spring.

Figure 2 .
Figure 2. Map of study area in Umbul Niti, South Lampung.25 microtremor measurements yellow dot) around the Umbul Niti hot spring (green triangle).The red line indicates the presence of a fault.

Figure 3 .
Figure 3. Research flow chat to determine the seismic vulnerability index (  )

Figure 6 .
Figure 6. 0 (a),  0 (b), and   (c) distribution within the study area.The red arrows point to a relatively high hazard levels zone.The measurement points (MSJTM10 -MSJTM02) were depicted in every picture by a black triangle.

Table 1 .
Parameters obtained and calculated with its interpretation within the study area