Wave-Based Seismic Characterization of the 2025 Myanmar Earthquakes (Mw 7.7 and Mw 6.7)
RajibDolai
0 views
15 slides
Sep 30, 2025
Slide 1 of 15
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
About This Presentation
This study conducts a comprehensive waveform analysis of two significant earthquakes that struck Myanmar on March 28, 2025, Mw 7.7 and Mw 6.7, using advanced signal processing techniques. Seismic records from the Seismological Facility for the Advancement of Geoscience (SAGE) were analyzed across th...
Slide Content
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 1
Wave-Based Seismic Characterization of the 2025 Myanmar
Earthquakes (Mw 7.7 and Mw 6.7)
Debasri Samanta
1
*, Rajib Kumar Dolai
2
1
Research Scholar, Department of Physics, Vidyasagar University, Midnapore, West Bengal,
India
2
Assistant Professor, Department of Economics, Tamralipta Mahavidyalaya, Tamluk, West
Bengal, India
*Corresponding Author
E-Mail Id: [email protected]
ABSTRACT
This study conducts a comprehensive waveform analysis of two significant earthquakes that
struck Myanmar on March 28, 2025, Mw 7.7 and Mw 6.7, using advanced signal processing
techniques. Seismic records from the Seismological Facility for the Advancement of
Geoscience (SAGE) were analyzed across three components (BH1, BH2, BHZ) through Fast
Fourier Transform (FFT), Power Spectral Density (PSD), Continuous Wavelet Transform
(CWT), and Probability Density Function (PDF) methods. These approaches enabled
detailed characterization of the spectral, temporal, and statistical properties of the ground
motion. The Mw 7.7 earthquake exhibited dominant low-frequency energy (0–0.5 Hz),
prolonged shaking, and higher amplitudes, particularly in vertical and horizontal
components, indicating strong resonance potential and elevated structural risk. In contrast,
the Mw 6.7 event produced shorter-duration, lower-amplitude bursts with greater high-
frequency attenuation and reduced hazard implications. FFT and PSD analyses quantified
the stark energy contrast, while CWT results revealed the localized evolution of seismic
energy. PDF analysis further highlighted differences in amplitude distribution, distinguishing
the dynamic behaviours of the two events.
Keywords: Seismic signal processing, Myanmar earthquake 2025, fast fourier transform
(FFT), power spectral density (PSD), wavelet transform, probability density function (PDF)
INTRODUCTION
Earthquakes are among the most
devastating natural hazards, capable of
causing widespread destruction to
infrastructure and significant loss of life.
On March 28, 2025, Myanmar experienced
two major earthquakes within a short time
span: the first occurred at 06:20:52 UTC
with a magnitude of Mw 7.7 at coordinates
22.001°N, 95.925°E, followed by a second
event at 06:32:04 UTC with a magnitude
of Mw 6.7 at 21.698°N, 95.969°E. Both
events generated strong seismic waves that
were captured by multiple seismological
stations, revealing complex rupture
processes and substantial energy release.
Myanmar is located in a seismically active
zone due to the oblique convergence
between the Indian and Eurasian tectonic
plates. In the region of the Sagaing Fault,
the Burma microplate and the Eurasian
plate move in opposite directions—while
the Indian plate is shifting northward, the
Eurasian plate is pushing southward. This
tectonic interaction accumulates stress
along fault lines and eventually releases
energy in the form of seismic waves that
propagate through the Earth's crust.
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 1
Wave-Based Seismic Characterization of the 2025 Myanmar
Earthquakes (Mw 7.7 and Mw 6.7)
Debasri Samanta
1
*, Rajib Kumar Dolai
2
1
Research Scholar, Department of Physics, Vidyasagar University, Midnapore, West Bengal,
India
2
Assistant Professor, Department of Economics, Tamralipta Mahavidyalaya, Tamluk, West
Bengal, India
*Corresponding Author
E-Mail Id: [email protected]
ABSTRACT
This study conducts a comprehensive waveform analysis of two significant earthquakes that
struck Myanmar on March 28, 2025, Mw 7.7 and Mw 6.7, using advanced signal processing
techniques. Seismic records from the Seismological Facility for the Advancement of
Geoscience (SAGE) were analyzed across three components (BH1, BH2, BHZ) through Fast
Fourier Transform (FFT), Power Spectral Density (PSD), Continuous Wavelet Transform
(CWT), and Probability Density Function (PDF) methods. These approaches enabled
detailed characterization of the spectral, temporal, and statistical properties of the ground
motion. The Mw 7.7 earthquake exhibited dominant low-frequency energy (0–0.5 Hz),
prolonged shaking, and higher amplitudes, particularly in vertical and horizontal
components, indicating strong resonance potential and elevated structural risk. In contrast,
the Mw 6.7 event produced shorter-duration, lower-amplitude bursts with greater high-
frequency attenuation and reduced hazard implications. FFT and PSD analyses quantified
the stark energy contrast, while CWT results revealed the localized evolution of seismic
energy. PDF analysis further highlighted differences in amplitude distribution, distinguishing
the dynamic behaviours of the two events.
Keywords: Seismic signal processing, Myanmar earthquake 2025, fast fourier transform
(FFT), power spectral density (PSD), wavelet transform, probability density function (PDF)
INTRODUCTION
Earthquakes are among the most
devastating natural hazards, capable of
causing widespread destruction to
infrastructure and significant loss of life.
On March 28, 2025, Myanmar experienced
two major earthquakes within a short time
span: the first occurred at 06:20:52 UTC
with a magnitude of Mw 7.7 at coordinates
22.001°N, 95.925°E, followed by a second
event at 06:32:04 UTC with a magnitude
of Mw 6.7 at 21.698°N, 95.969°E. Both
events generated strong seismic waves that
were captured by multiple seismological
stations, revealing complex rupture
processes and substantial energy release.
Myanmar is located in a seismically active
zone due to the oblique convergence
between the Indian and Eurasian tectonic
plates. In the region of the Sagaing Fault,
the Burma microplate and the Eurasian
plate move in opposite directions—while
the Indian plate is shifting northward, the
Eurasian plate is pushing southward. This
tectonic interaction accumulates stress
along fault lines and eventually releases
energy in the form of seismic waves that
propagate through the Earth's crust.
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 2
Source: United States Geological Survey
Fig. 1: Macroseismic Intensity Map.
Seismic waves are typically recorded in
three directions: BH1 (Broadband
Horizontal 1), which captures ground
motion in the north-south direction; BH2
(Broadband Horizontal 2), which records
east-west motion; and BHZ (Broadband
Vertical), which measures vertical ground
movement. To better understand the
dynamic behaviour of these waves, signal
processing techniques are essential. Fast
Fourier Transform (FFT) reveals the
dominant frequency components of the
seismic signal, while Power Spectral
Density (PSD) shows how seismic energy
is distributed across different frequencies.
Wavelet Transform offers a time-
frequency representation of the event,
helping to identify the evolution of
frequency components over time.
Additionally, Probability Density Function
(PDF) analysis provides statistical insight
into the amplitude distribution of ground
motion, indicating the nature of seismic
shaking during the event.
A wide range of analytical techniques have
been developed to simulate and analyse
earthquake ground motions and their
effects on structures, using various signal
processing and modelling approaches to
better understand seismic behaviour.
Boore (2003) [1] introduced the stochastic
method to simulate high-frequency ground
motions using amplitude and random
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 2
Source: United States Geological Survey
Fig. 1: Macroseismic Intensity Map.
Seismic waves are typically recorded in
three directions: BH1 (Broadband
Horizontal 1), which captures ground
motion in the north-south direction; BH2
(Broadband Horizontal 2), which records
east-west motion; and BHZ (Broadband
Vertical), which measures vertical ground
movement. To better understand the
dynamic behaviour of these waves, signal
processing techniques are essential. Fast
Fourier Transform (FFT) reveals the
dominant frequency components of the
seismic signal, while Power Spectral
Density (PSD) shows how seismic energy
is distributed across different frequencies.
Wavelet Transform offers a time-
frequency representation of the event,
helping to identify the evolution of
frequency components over time.
Additionally, Probability Density Function
(PDF) analysis provides statistical insight
into the amplitude distribution of ground
motion, indicating the nature of seismic
shaking during the event.
A wide range of analytical techniques have
been developed to simulate and analyse
earthquake ground motions and their
effects on structures, using various signal
processing and modelling approaches to
better understand seismic behaviour.
Boore (2003) [1] introduced the stochastic
method to simulate high-frequency ground
motions using amplitude and random
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 3
phase spectra, providing a simple and
effective tool especially where strong-
motion data is limited, while Li Yousun
(1991) [4] proposed an efficient FFT-
based method to simulate multivariate
nonstationary Gaussian processes derived
from earthquake data, and Lui et al. (1971)
[3] developed a similar FFT-based
technique to compute earthquake-induced
structural responses considering soil-
structure interaction via two foundation
models. Trifunac et al. (1975) [2] defined
the duration of strong ground motion using
the mean-square integral and correlated it
with seismic parameters such as magnitude
and site geology, whereas Alsop et al.
(1966) [5] showed how the Cooley-Tukey
FFT algorithm significantly improves the
computational efficiency of seismic data
processing. Crockett (2018) [6]
highlighted the utility of FFT as a
statistical tool to distinguish cyclic from
anomalous geophysical time-series
phenomena, and Liu S.C. (1970) [7]
addressed the nonstationary nature of
earthquake behaviour using time-
dependent power spectral density analysis.
Bai and Zhu (2008) [8] supported the use
of wavelet transform over traditional
Fourier techniques in identifying the
nonstationary characteristics of earthquake
ground motions, and Power (1969) [9]
applied power spectral density analysis to
estimate earthquake magnitude and
explosive yield based on ground motion,
which Christian (1989) [10] also regarded
as a fundamental tool for understanding
the frequency content of motion in
complex structures. Oshima et al. (2022)
[11] developed a statistical method based
on Kullback–Leibler divergence to
improve seismic phase detection in single-
component records by analysing amplitude
distributions, while Jun Wei B. et al.
(2014) [12] used wavelet transforms to
achieve a more precise time-frequency
representation of seismic signals than
conventional methods. Iyama (1999) [13]
introduced a wavelet-based approach
linking wavelet coefficients to energy
input in structures, enabling time-
frequency analysis and simulation of
ground motions, and Teymur (1999) [14]
applied harmonic wavelet analysis to
identify localized energy variations in
nonstationary earthquake signals from the
1999 Kocaeli event. Kumar P. et al. (1997)
[15] reviewed the evolution and
geophysical applications of wavelet
transforms, predicting their increasing role
in complex seismic modelling, and Li X.
and Li Z. et al. (2016) [16] applied the
Hilbert–Huang Transform to capture
frequency shifts and energy build-up in
microseismic signals for rock burst
prediction. Finally, Safak (1999) [17]
proposed a discrete-time wave-propagation
method for seismic analysis of multistory
buildings on layered soils, offering
improved accuracy and modelling of
damping and soil-structure interaction
compared to traditional vibration-based
models. For instance, Samanta et al.
(2025) [18] examined probabilistic mass–
spin dynamics in astrophysical systems,
concepts that can similarly be applied to
earthquake wave data to explain stochastic
ground motion and rupture variability.
This study integrates advanced signal
processing methods to analyse the seismic
signals from the March 28, 2025,
Myanmar earthquakes, aiming to
characterize ground motion, compare
energy distribution between the two
events, and understand the underlying
frequency content and statistical behaviour
of the recorded seismic waves.
METHODOLOGY
Data Collection and Processing
Seismic waveform data were sourced from
the Seismological Facility for the
Advancement of Geoscience (SAGE), with
a focus on three primary components: BH1
(Broadband Horizontal 1, typically north-
south), BH2 (Broadband Horizontal 2,
typically east-west), and BHZ (Broadband
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 3
phase spectra, providing a simple and
effective tool especially where strong-
motion data is limited, while Li Yousun
(1991) [4] proposed an efficient FFT-
based method to simulate multivariate
nonstationary Gaussian processes derived
from earthquake data, and Lui et al. (1971)
[3] developed a similar FFT-based
technique to compute earthquake-induced
structural responses considering soil-
structure interaction via two foundation
models. Trifunac et al. (1975) [2] defined
the duration of strong ground motion using
the mean-square integral and correlated it
with seismic parameters such as magnitude
and site geology, whereas Alsop et al.
(1966) [5] showed how the Cooley-Tukey
FFT algorithm significantly improves the
computational efficiency of seismic data
processing. Crockett (2018) [6]
highlighted the utility of FFT as a
statistical tool to distinguish cyclic from
anomalous geophysical time-series
phenomena, and Liu S.C. (1970) [7]
addressed the nonstationary nature of
earthquake behaviour using time-
dependent power spectral density analysis.
Bai and Zhu (2008) [8] supported the use
of wavelet transform over traditional
Fourier techniques in identifying the
nonstationary characteristics of earthquake
ground motions, and Power (1969) [9]
applied power spectral density analysis to
estimate earthquake magnitude and
explosive yield based on ground motion,
which Christian (1989) [10] also regarded
as a fundamental tool for understanding
the frequency content of motion in
complex structures. Oshima et al. (2022)
[11] developed a statistical method based
on Kullback–Leibler divergence to
improve seismic phase detection in single-
component records by analysing amplitude
distributions, while Jun Wei B. et al.
(2014) [12] used wavelet transforms to
achieve a more precise time-frequency
representation of seismic signals than
conventional methods. Iyama (1999) [13]
introduced a wavelet-based approach
linking wavelet coefficients to energy
input in structures, enabling time-
frequency analysis and simulation of
ground motions, and Teymur (1999) [14]
applied harmonic wavelet analysis to
identify localized energy variations in
nonstationary earthquake signals from the
1999 Kocaeli event. Kumar P. et al. (1997)
[15] reviewed the evolution and
geophysical applications of wavelet
transforms, predicting their increasing role
in complex seismic modelling, and Li X.
and Li Z. et al. (2016) [16] applied the
Hilbert–Huang Transform to capture
frequency shifts and energy build-up in
microseismic signals for rock burst
prediction. Finally, Safak (1999) [17]
proposed a discrete-time wave-propagation
method for seismic analysis of multistory
buildings on layered soils, offering
improved accuracy and modelling of
damping and soil-structure interaction
compared to traditional vibration-based
models. For instance, Samanta et al.
(2025) [18] examined probabilistic mass–
spin dynamics in astrophysical systems,
concepts that can similarly be applied to
earthquake wave data to explain stochastic
ground motion and rupture variability.
This study integrates advanced signal
processing methods to analyse the seismic
signals from the March 28, 2025,
Myanmar earthquakes, aiming to
characterize ground motion, compare
energy distribution between the two
events, and understand the underlying
frequency content and statistical behaviour
of the recorded seismic waves.
METHODOLOGY
Data Collection and Processing
Seismic waveform data were sourced from
the Seismological Facility for the
Advancement of Geoscience (SAGE), with
a focus on three primary components: BH1
(Broadband Horizontal 1, typically north-
south), BH2 (Broadband Horizontal 2,
typically east-west), and BHZ (Broadband
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 4
Vertical). The dataset includes the P, PP,
S, and SS seismic phases generated by two
major earthquakes in Myanmar: a
magnitude Mw 6.7 earthquake (2025-03-
28 06:32:04 UTC, 21.698°N, 95.969°E,
depth 10.0 km) and a magnitude Mw 7.7
earthquake (2025-03-28 06:20:52 UTC,
22.001°N, 95.925°E, depth 10.0 km).
All data analysis was performed using
Python. The waveform data were
downloaded in SAC (Seismic Analysis
Code) format and read using the obspy
library. Preprocessing included handling
missing or NaN values, baseline
corrections, and normalization of the
signals. Signal processing techniques such
as Fast Fourier Transform (FFT), Power
Spectral Density (PSD), Wavelet
Transform (WT), and Probability Density
Function (PDF) analysis were employed to
examine the temporal and spectral features
of the seismic data.
Time-domain waveform plots were created
using matplotlib and obspy, while FFT and
PSD computations utilized scipy.fftpack
and scipy.signal.welch, respectively. Time-
frequency analysis was conducted using
the Continuous Wavelet Transform from
the pywt package. Statistical analysis of
waveform amplitudes was achieved using
Gaussian Kernel Density Estimation
(KDE) via scipy.stats.gaussian_kde to
generate the Probability Density Function
(PDF) plots.
In this study, we analyse seismic
waveforms using a combination of spectral
and statistical techniques. The
methodologies employed include the Fast
Fourier Transform (FFT) for frequency
domain representation, Power Spectral
Density (PSD) estimation for power
distribution analysis, Continuous Wavelet
Transform (CWT) for time-frequency
localization, and Probability Density
Function (PDF) analysis using Kernel
Density Estimation (KDE) to characterize
the statistical distribution of waveform
amplitudes. These methods allow us to
capture both spectral and statistical
features of seismic signals recorded from
the Myanmar earthquake event.
Fast Fourier Transform (FFT) Spectra
The Fourier Transform is a fundamental
tool for analysing the frequency
components of a signal. The Discrete
Fourier Transform (DFT) is used to
transform a discrete-time signal from the
time domain into the frequency domain.
However, the direct computation of the
DFT has a computational complexity of
(
), making it inefficient for large
datasets. To overcome this, the Fast
Fourier Transform (FFT) algorithm is
used, which efficiently computes the DFT
with a reduced complexity of
( ).Given a discrete signal , - of
length , the DFT is computed as:
( ) ∑ ( )
.
where:
( ) is the DFT coefficient at
frequency index , representing the
spectral content of the signal at that
frequency.
( ) is the time-domain signal,
consisting of discrete samples.
is the total number of samples in the
signal.
represents the frequency bin index,
which corresponds to discrete
frequencies in the range , -
is a complex exponential
function that serves as a basis function
for frequency decomposition.
√ is the imaginary unit.
For each seismic waveform, we computed
the FFT and extracted the amplitude
spectrum by taking the absolute value of
the complex FFT coefficients:
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 4
Vertical). The dataset includes the P, PP,
S, and SS seismic phases generated by two
major earthquakes in Myanmar: a
magnitude Mw 6.7 earthquake (2025-03-
28 06:32:04 UTC, 21.698°N, 95.969°E,
depth 10.0 km) and a magnitude Mw 7.7
earthquake (2025-03-28 06:20:52 UTC,
22.001°N, 95.925°E, depth 10.0 km).
All data analysis was performed using
Python. The waveform data were
downloaded in SAC (Seismic Analysis
Code) format and read using the obspy
library. Preprocessing included handling
missing or NaN values, baseline
corrections, and normalization of the
signals. Signal processing techniques such
as Fast Fourier Transform (FFT), Power
Spectral Density (PSD), Wavelet
Transform (WT), and Probability Density
Function (PDF) analysis were employed to
examine the temporal and spectral features
of the seismic data.
Time-domain waveform plots were created
using matplotlib and obspy, while FFT and
PSD computations utilized scipy.fftpack
and scipy.signal.welch, respectively. Time-
frequency analysis was conducted using
the Continuous Wavelet Transform from
the pywt package. Statistical analysis of
waveform amplitudes was achieved using
Gaussian Kernel Density Estimation
(KDE) via scipy.stats.gaussian_kde to
generate the Probability Density Function
(PDF) plots.
In this study, we analyse seismic
waveforms using a combination of spectral
and statistical techniques. The
methodologies employed include the Fast
Fourier Transform (FFT) for frequency
domain representation, Power Spectral
Density (PSD) estimation for power
distribution analysis, Continuous Wavelet
Transform (CWT) for time-frequency
localization, and Probability Density
Function (PDF) analysis using Kernel
Density Estimation (KDE) to characterize
the statistical distribution of waveform
amplitudes. These methods allow us to
capture both spectral and statistical
features of seismic signals recorded from
the Myanmar earthquake event.
Fast Fourier Transform (FFT) Spectra
The Fourier Transform is a fundamental
tool for analysing the frequency
components of a signal. The Discrete
Fourier Transform (DFT) is used to
transform a discrete-time signal from the
time domain into the frequency domain.
However, the direct computation of the
DFT has a computational complexity of
(
), making it inefficient for large
datasets. To overcome this, the Fast
Fourier Transform (FFT) algorithm is
used, which efficiently computes the DFT
with a reduced complexity of
( ).Given a discrete signal , - of
length , the DFT is computed as:
( ) ∑ ( )
.
where:
( ) is the DFT coefficient at
frequency index , representing the
spectral content of the signal at that
frequency.
( ) is the time-domain signal,
consisting of discrete samples.
is the total number of samples in the
signal.
represents the frequency bin index,
which corresponds to discrete
frequencies in the range , -
is a complex exponential
function that serves as a basis function
for frequency decomposition.
√ is the imaginary unit.
For each seismic waveform, we computed
the FFT and extracted the amplitude
spectrum by taking the absolute value of
the complex FFT coefficients:
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 5
| ( )| √ ( ( ))
( ( ))
where, ( ( ))
and ( ( ))
are the
real and imaginary components of ( ),
respectively.
The frequency components were then
analyzed to identify dominant spectral
features, such as peak frequencies and
energy distribution, which are essential for
understanding seismic wave propagation
and characterizing subsurface structures.
Power Spectral Density (PSD)
Estimation
Power Spectral Density (PSD) provides a
measure of how the power of a signal is
distributed over different frequencies. We
estimated the PSD using Welch’s method,
which reduces noise and improves
statistical reliability. Welch’s method
divides the signal into overlapping
segments, computes the periodogram for
each segment, and averages them:
( )
∑|
( )|
Where,
( ) is the estimated power
spectral density,
( ) is the FFT of
segment, and is the number of segments.
A logarithmic representation of the PSD
helps in identifying energy distribution
patterns within the seismic signal,
highlighting dominant frequencies and
noise characteristics.
Probability Density Function (PDF)
To characterize the statistical distribution
of the seismic waveform amplitudes, we
utilized Kernel Density Estimation (KDE)
as a method to estimate the Probability
Density Function (PDF). Unlike
histograms, KDE provides a smooth,
continuous approximation of the
underlying distribution without relying on
binning, making it particularly suitable for
analysing continuous data such as seismic
signals.
Mathematically, the kernel density
estimate ̂( ) for a univariate dataset
*
+ is defined as: ̂( )
∑ .
/
where:
is the kernel function, typically a
Gaussian,
is the bandwidth, a smoothing
parameter that controls the width of
the kernel,
is the number of data points.
In our implementation, we used the
Gaussian kernel, which is defined as:
( )
√
For each seismic signal, the probability
density was estimated over 1,000 evenly
spaced amplitude values, producing a
high-resolution PDF that highlights key
statistical features such as symmetry,
modality, and the presence of outliers.
Wavelet Transform Analysis
Unlike the Fourier Transform, which
provides only frequency information, the
Wavelet Transform enables time-
frequency localization by analysing
variations in different scales. We
employed the Continuous Wavelet
Transform (CWT) using the Complex
Morlet wavelet function:
( )
√
where:
( ) is the Morlet wavelet function.
is the wavelet's standard deviation,
controlling time resolution.
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 5
| ( )| √ ( ( ))
( ( ))
where, ( ( ))
and ( ( ))
are the
real and imaginary components of ( ),
respectively.
The frequency components were then
analyzed to identify dominant spectral
features, such as peak frequencies and
energy distribution, which are essential for
understanding seismic wave propagation
and characterizing subsurface structures.
Power Spectral Density (PSD)
Estimation
Power Spectral Density (PSD) provides a
measure of how the power of a signal is
distributed over different frequencies. We
estimated the PSD using Welch’s method,
which reduces noise and improves
statistical reliability. Welch’s method
divides the signal into overlapping
segments, computes the periodogram for
each segment, and averages them:
( )
∑|
( )|
Where,
( ) is the estimated power
spectral density,
( ) is the FFT of
segment, and is the number of segments.
A logarithmic representation of the PSD
helps in identifying energy distribution
patterns within the seismic signal,
highlighting dominant frequencies and
noise characteristics.
Probability Density Function (PDF)
To characterize the statistical distribution
of the seismic waveform amplitudes, we
utilized Kernel Density Estimation (KDE)
as a method to estimate the Probability
Density Function (PDF). Unlike
histograms, KDE provides a smooth,
continuous approximation of the
underlying distribution without relying on
binning, making it particularly suitable for
analysing continuous data such as seismic
signals.
Mathematically, the kernel density
estimate ̂( ) for a univariate dataset
*
+ is defined as: ̂( )
∑ .
/
where:
is the kernel function, typically a
Gaussian,
is the bandwidth, a smoothing
parameter that controls the width of
the kernel,
is the number of data points.
In our implementation, we used the
Gaussian kernel, which is defined as:
( )
√
For each seismic signal, the probability
density was estimated over 1,000 evenly
spaced amplitude values, producing a
high-resolution PDF that highlights key
statistical features such as symmetry,
modality, and the presence of outliers.
Wavelet Transform Analysis
Unlike the Fourier Transform, which
provides only frequency information, the
Wavelet Transform enables time-
frequency localization by analysing
variations in different scales. We
employed the Continuous Wavelet
Transform (CWT) using the Complex
Morlet wavelet function:
( )
√
where:
( ) is the Morlet wavelet function.
is the wavelet's standard deviation,
controlling time resolution.
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 6
is the central frequency of the
wavelet.
is the imaginary unit (
).
The first exponential term (
)
represents the oscillatory component.
The second exponential term (
) ensures time localization.
The CWT was applied to seismic signals
across multiple scales, enabling a
scalogram representation that maps signal
energy over time and frequency. This
method was instrumental in detecting
transient seismic events and frequency-
dependent variations over time.
RESULT AND DISCUSSION
Seismic waveforms from the two
significant Myanmar earthquakes on
March 28, 2025—Mw 7.7 at 06:20:52
UTC (22.001°N, 95.925°E) and Mw 6.7 at
06:32:04 UTC (21.698°N, 95.969°E), both
at a depth of 10.0 km—were analyzed
across three components: BH1 (first
horizontal), BH2 (second horizontal), and
BHZ (vertical). For the Mw 7.7 event, the
pre-event phase between 06:22:00 and
06:22:54 UTC is characterized by minimal
ground motion. At approximately 06:22:10
UTC, the arrival of the P-wave leads to a
sharp increase in amplitude, followed by
the PP-wave at 06:22:20 UTC, which
produces an additional rise associated with
internal reflections. A similar sequence is
observed for the Mw 6.7 event, where the
pre-event phase persists until about
06:33:57 UTC, after which the P-wave and
PP-wave arrive at 06:32:50 and 06:33:00
UTC, respectively. The amplitudes,
however, remain lower due to the smaller
magnitude. In both events, the BHZ
component displays stronger vertical
motion during the early wave arrivals,
whereas BH1 and BH2 capture more
pronounced lateral displacements as the
later phases develop. These observations
form the basis for the subsequent
frequency-domain and time–frequency
analyses.
Fig. 2: Seismic waveforms recorded for the Myanmar earthquakes on March 28, 2025:
Mw 7.7 event at 06:20:54 UTC and Mw 6.7 event at 06:32:04 UTC, shown across three
components for each event: (a) BH1 (first horizontal), (b) BH2 (second horizontal), and (c)
BHZ (vertical).
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 6
is the central frequency of the
wavelet.
is the imaginary unit (
).
The first exponential term (
)
represents the oscillatory component.
The second exponential term (
) ensures time localization.
The CWT was applied to seismic signals
across multiple scales, enabling a
scalogram representation that maps signal
energy over time and frequency. This
method was instrumental in detecting
transient seismic events and frequency-
dependent variations over time.
RESULT AND DISCUSSION
Seismic waveforms from the two
significant Myanmar earthquakes on
March 28, 2025—Mw 7.7 at 06:20:52
UTC (22.001°N, 95.925°E) and Mw 6.7 at
06:32:04 UTC (21.698°N, 95.969°E), both
at a depth of 10.0 km—were analyzed
across three components: BH1 (first
horizontal), BH2 (second horizontal), and
BHZ (vertical). For the Mw 7.7 event, the
pre-event phase between 06:22:00 and
06:22:54 UTC is characterized by minimal
ground motion. At approximately 06:22:10
UTC, the arrival of the P-wave leads to a
sharp increase in amplitude, followed by
the PP-wave at 06:22:20 UTC, which
produces an additional rise associated with
internal reflections. A similar sequence is
observed for the Mw 6.7 event, where the
pre-event phase persists until about
06:33:57 UTC, after which the P-wave and
PP-wave arrive at 06:32:50 and 06:33:00
UTC, respectively. The amplitudes,
however, remain lower due to the smaller
magnitude. In both events, the BHZ
component displays stronger vertical
motion during the early wave arrivals,
whereas BH1 and BH2 capture more
pronounced lateral displacements as the
later phases develop. These observations
form the basis for the subsequent
frequency-domain and time–frequency
analyses.
Fig. 2: Seismic waveforms recorded for the Myanmar earthquakes on March 28, 2025:
Mw 7.7 event at 06:20:54 UTC and Mw 6.7 event at 06:32:04 UTC, shown across three
components for each event: (a) BH1 (first horizontal), (b) BH2 (second horizontal), and (c)
BHZ (vertical).
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 7
The most critical phase of the Mw 7.7
earthquake occurred between 06:22:54 and
06:23:10 UTC, marked by the arrival of
the S-wave and its multiple reflection, the
SS-wave. The S-wave, arriving shortly
after the P-wave, generated the largest
ground displacement due to its transverse
motion, with a sharp amplitude spike at
approximately 06:23:11 UTC (indicated
by a blue vertical marker) that signaled the
onset of intense shaking. This phase
represented the peak amplitude and the
strongest ground motion of the event. The
subsequent SS-wave (highlighted by a
cyan marker) sustained shaking with
slightly reduced intensity, but the overlap
of S- and SS-waves produced complex
oscillatory behaviour and irregular
waveform fluctuations. From 06:24:10
UTC onward, the waveform entered a
dissipation phase, characterized by
gradually declining amplitudes as seismic
energy attenuated through scattering and
absorption in the crustal medium. By
06:27:00 UTC, amplitudes had decreased
significantly, marking the end of strong
shaking, though minor oscillations
persisted due to aftershocks and
reverberations. At around 06:30:00 UTC,
only low-level fluctuations remained,
indicating the conclusion of the primary
energy release. The most destructive
interval, between 06:23:10 and 06:23:50
UTC, corresponded to the peak seismic
energy release driven by fault rupture
dynamics and shallow earthquake depth.
During this period, the combined action of
strong P- and S-waves likely produced the
most severe ground motion, structural
damage, and secondary hazards such as
landslides and soil liquefaction, before
seismic energy gradually dissipated
outward from the source (see fig. 2).
The most critical phase of the Mw 6.7
earthquake occurred between 06:33:20 and
06:34:30 UTC, corresponding to the
arrival of the S-wave followed by the SS-
wave. The S-wave, succeeding the earlier
P- and PP-waves, produced a sharp
increase in ground motion, with peak
amplitude observed around 06:33:55 UTC.
This marked the onset of strong shaking,
though notably less intense than that of the
Mw 7.7 event. The SS-wave arrived
shortly afterward, sustaining the motion
with moderately reduced amplitude. The
interaction of these phases generated
oscillatory patterns that were evident but
less severe than those of the larger
earthquake. From approximately 06:35:10
UTC, the waveform entered the dissipation
phase, with amplitudes gradually
decreasing as seismic energy attenuated
through local geological structures. By
06:38:00 UTC, amplitudes had diminished
substantially, indicating the end of strong
motion, leaving only minor tremors and
reverberations. At around 06:40:00 UTC,
the signal stabilized, marking the
conclusion of significant energy release.
The peak energy interval, between
06:34:20 and 06:34:50 UTC, produced
localized strong shaking and possible
minor structural effects. Although the Mw
6.7 event exhibited lower energy output
and shaking intensity compared to the Mw
7.7 earthquake, it still displayed a well-
defined sequence of seismic phases and a
gradual decay of energy over time.
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 7
The most critical phase of the Mw 7.7
earthquake occurred between 06:22:54 and
06:23:10 UTC, marked by the arrival of
the S-wave and its multiple reflection, the
SS-wave. The S-wave, arriving shortly
after the P-wave, generated the largest
ground displacement due to its transverse
motion, with a sharp amplitude spike at
approximately 06:23:11 UTC (indicated
by a blue vertical marker) that signaled the
onset of intense shaking. This phase
represented the peak amplitude and the
strongest ground motion of the event. The
subsequent SS-wave (highlighted by a
cyan marker) sustained shaking with
slightly reduced intensity, but the overlap
of S- and SS-waves produced complex
oscillatory behaviour and irregular
waveform fluctuations. From 06:24:10
UTC onward, the waveform entered a
dissipation phase, characterized by
gradually declining amplitudes as seismic
energy attenuated through scattering and
absorption in the crustal medium. By
06:27:00 UTC, amplitudes had decreased
significantly, marking the end of strong
shaking, though minor oscillations
persisted due to aftershocks and
reverberations. At around 06:30:00 UTC,
only low-level fluctuations remained,
indicating the conclusion of the primary
energy release. The most destructive
interval, between 06:23:10 and 06:23:50
UTC, corresponded to the peak seismic
energy release driven by fault rupture
dynamics and shallow earthquake depth.
During this period, the combined action of
strong P- and S-waves likely produced the
most severe ground motion, structural
damage, and secondary hazards such as
landslides and soil liquefaction, before
seismic energy gradually dissipated
outward from the source (see fig. 2).
The most critical phase of the Mw 6.7
earthquake occurred between 06:33:20 and
06:34:30 UTC, corresponding to the
arrival of the S-wave followed by the SS-
wave. The S-wave, succeeding the earlier
P- and PP-waves, produced a sharp
increase in ground motion, with peak
amplitude observed around 06:33:55 UTC.
This marked the onset of strong shaking,
though notably less intense than that of the
Mw 7.7 event. The SS-wave arrived
shortly afterward, sustaining the motion
with moderately reduced amplitude. The
interaction of these phases generated
oscillatory patterns that were evident but
less severe than those of the larger
earthquake. From approximately 06:35:10
UTC, the waveform entered the dissipation
phase, with amplitudes gradually
decreasing as seismic energy attenuated
through local geological structures. By
06:38:00 UTC, amplitudes had diminished
substantially, indicating the end of strong
motion, leaving only minor tremors and
reverberations. At around 06:40:00 UTC,
the signal stabilized, marking the
conclusion of significant energy release.
The peak energy interval, between
06:34:20 and 06:34:50 UTC, produced
localized strong shaking and possible
minor structural effects. Although the Mw
6.7 event exhibited lower energy output
and shaking intensity compared to the Mw
7.7 earthquake, it still displayed a well-
defined sequence of seismic phases and a
gradual decay of energy over time.
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 8
Fig. 3: Fast Fourier Transform (FFT) Spectra of the Magnitude 7.7 and 6.7 Myanmar
Earthquakes, for (a) BH1 Component, (b) BH2 Component, and (c) BHZ Component.
The Fast Fourier Transform (FFT) spectra
of the seismic signals from the Mw 7.7
Myanmar earthquake on March 28, 2025,
at 06:20:54 UTC were analyzed across
three channels, BH1, BH2, and BHZ,
representing the two horizontal and one
vertical components of ground motion. All
components exhibit a strong concentration
of energy in the low-frequency range,
particularly below 1 Hz, with the highest
spectral amplitudes occurring between 0
and 0.5 Hz. This indicates that the
earthquake’s dominant energy was
concentrated in the low-frequency band, a
characteristic feature of large-magnitude
events. These peaks reflect the influence of
primary seismic waves, P-waves, S-waves,
and especially surface waves, which
propagate efficiently at low frequencies,
leading to sustained shaking over large
distances (see fig. 3).
As frequency increases beyond 1 Hz,
spectral amplitudes decline sharply,
consistent with the general expectation that
large earthquakes radiate more energy at
low frequencies, while higher-frequency
components—often generated by
scattering, local site effects, or rupture
complexities—contribute relatively less
energy. The attenuation beyond ~2 Hz
suggests that most of the seismic energy
was concentrated early in the frequency
spectrum. The similarity of patterns across
all three components implies a relatively
uniform distribution of energy between
vertical and horizontal motions.
From an engineering perspective, the
dominance of low-frequency peaks poses
significant hazards for tall buildings, long-
span bridges, and flexible infrastructure
with natural frequencies below 1 Hz, as
resonance effects can amplify shaking and
damage potential. In contrast, smaller and
stiffer structures with higher natural
frequencies may be less affected by this
dominant low-frequency content, though
they could still respond to the weaker
high-frequency components present in the
spectrum.
Similarly, the FFT spectra of the Mw 6.7
Myanmar earthquake on March 28, 2025,
at 06:32:04 UTC, show a distinct
concentration of energy in the low-
frequency band, particularly between 0
and 0.5 Hz, across all three components,
BH1, BH2, and BHZ, representing
horizontal and vertical ground motions.
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 8
Fig. 3: Fast Fourier Transform (FFT) Spectra of the Magnitude 7.7 and 6.7 Myanmar
Earthquakes, for (a) BH1 Component, (b) BH2 Component, and (c) BHZ Component.
The Fast Fourier Transform (FFT) spectra
of the seismic signals from the Mw 7.7
Myanmar earthquake on March 28, 2025,
at 06:20:54 UTC were analyzed across
three channels, BH1, BH2, and BHZ,
representing the two horizontal and one
vertical components of ground motion. All
components exhibit a strong concentration
of energy in the low-frequency range,
particularly below 1 Hz, with the highest
spectral amplitudes occurring between 0
and 0.5 Hz. This indicates that the
earthquake’s dominant energy was
concentrated in the low-frequency band, a
characteristic feature of large-magnitude
events. These peaks reflect the influence of
primary seismic waves, P-waves, S-waves,
and especially surface waves, which
propagate efficiently at low frequencies,
leading to sustained shaking over large
distances (see fig. 3).
As frequency increases beyond 1 Hz,
spectral amplitudes decline sharply,
consistent with the general expectation that
large earthquakes radiate more energy at
low frequencies, while higher-frequency
components—often generated by
scattering, local site effects, or rupture
complexities—contribute relatively less
energy. The attenuation beyond ~2 Hz
suggests that most of the seismic energy
was concentrated early in the frequency
spectrum. The similarity of patterns across
all three components implies a relatively
uniform distribution of energy between
vertical and horizontal motions.
From an engineering perspective, the
dominance of low-frequency peaks poses
significant hazards for tall buildings, long-
span bridges, and flexible infrastructure
with natural frequencies below 1 Hz, as
resonance effects can amplify shaking and
damage potential. In contrast, smaller and
stiffer structures with higher natural
frequencies may be less affected by this
dominant low-frequency content, though
they could still respond to the weaker
high-frequency components present in the
spectrum.
Similarly, the FFT spectra of the Mw 6.7
Myanmar earthquake on March 28, 2025,
at 06:32:04 UTC, show a distinct
concentration of energy in the low-
frequency band, particularly between 0
and 0.5 Hz, across all three components,
BH1, BH2, and BHZ, representing
horizontal and vertical ground motions.
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 9
Like the Mw 7.7 event, this low-frequency
dominance reflects the influence of long-
period seismic waves, including P-waves,
S-waves, and surface waves. However, the
overall spectral amplitudes are
considerably lower, consistent with the
smaller energy release and shorter rupture
length of the Mw 6.7 earthquake.
Beyond 1 Hz, spectral amplitudes decline
sharply, with minimal energy above 2 Hz,
indicating limited high-frequency content,
a typical characteristic of moderate-
magnitude earthquakes. The similarity of
spectral behavior across all channels
suggests a relatively uniform energy
distribution and a symmetrical radiation
pattern from the seismic source.
The most prominent peaks, concentrated
within the 0.3–0.5 Hz range, raise potential
concerns for mid-rise structures whose
natural frequencies overlap with this band,
as resonance effects could amplify shaking
and structural stress. Nonetheless, the
relatively low amplitudes and absence of
strong high-frequency components imply a
reduced probability of widespread
structural damage, particularly for shorter
and stiffer buildings that are more
responsive to higher-frequency vibrations.
Fig. 4: Power Spectral Density (PSD) Analysis of Seismic Signals Recorded During the
Myanmar Earthquakes on March 28, 2025 — (a) BH1, (b) BH2, and (c) BHZ Components
for the Mw 7.7 Event at 06:20:54 UTC, and the Mw 6.7 Event at 06:32:04 UTC.
The Power Spectral Density (PSD)
analysis of seismic signals from the Mw
7.7 Myanmar earthquake on March 28,
2025, at 06:20:54 UTC, shows that across
all three components, BH1, BH2, and
BHZ, the earthquake’s energy was
predominantly concentrated in the low-
frequency range, with the PSD curves
exhibiting a steep decline in power as
frequency increased. Sharp peaks at very
low frequencies, with maximum values of
approximately for BH2 and
BHZ and around 10¹³ for BH1, correspond
to the dominant frequencies of primary
seismic waves such as P-waves and
surface waves, which typically control
long-period ground motion in large
earthquakes. Beyond 1 Hz, power
decreases steadily across all channels, with
PSD values dropping to about ⁸
in the 5–10 Hz range and further to
near 20 Hz, reflecting
attenuation of high-frequency components
through scattering, absorption, and crustal
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 9
Like the Mw 7.7 event, this low-frequency
dominance reflects the influence of long-
period seismic waves, including P-waves,
S-waves, and surface waves. However, the
overall spectral amplitudes are
considerably lower, consistent with the
smaller energy release and shorter rupture
length of the Mw 6.7 earthquake.
Beyond 1 Hz, spectral amplitudes decline
sharply, with minimal energy above 2 Hz,
indicating limited high-frequency content,
a typical characteristic of moderate-
magnitude earthquakes. The similarity of
spectral behavior across all channels
suggests a relatively uniform energy
distribution and a symmetrical radiation
pattern from the seismic source.
The most prominent peaks, concentrated
within the 0.3–0.5 Hz range, raise potential
concerns for mid-rise structures whose
natural frequencies overlap with this band,
as resonance effects could amplify shaking
and structural stress. Nonetheless, the
relatively low amplitudes and absence of
strong high-frequency components imply a
reduced probability of widespread
structural damage, particularly for shorter
and stiffer buildings that are more
responsive to higher-frequency vibrations.
Fig. 4: Power Spectral Density (PSD) Analysis of Seismic Signals Recorded During the
Myanmar Earthquakes on March 28, 2025 — (a) BH1, (b) BH2, and (c) BHZ Components
for the Mw 7.7 Event at 06:20:54 UTC, and the Mw 6.7 Event at 06:32:04 UTC.
The Power Spectral Density (PSD)
analysis of seismic signals from the Mw
7.7 Myanmar earthquake on March 28,
2025, at 06:20:54 UTC, shows that across
all three components, BH1, BH2, and
BHZ, the earthquake’s energy was
predominantly concentrated in the low-
frequency range, with the PSD curves
exhibiting a steep decline in power as
frequency increased. Sharp peaks at very
low frequencies, with maximum values of
approximately for BH2 and
BHZ and around 10¹³ for BH1, correspond
to the dominant frequencies of primary
seismic waves such as P-waves and
surface waves, which typically control
long-period ground motion in large
earthquakes. Beyond 1 Hz, power
decreases steadily across all channels, with
PSD values dropping to about ⁸
in the 5–10 Hz range and further to
near 20 Hz, reflecting
attenuation of high-frequency components
through scattering, absorption, and crustal
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 10
dissipation. The logarithmic scaling of
both PSD and frequency axes highlights
the broad spectral distribution,
emphasizing the dominance of low-
frequency energy. Although all three
components follow a similar overall trend,
BH2 generally exhibits the highest peak
power, likely influenced by local
geological effects, wave propagation
characteristics, or instrument sensitivity.
Importantly, the PSD results underline the
greater contribution of surface waves
relative to body waves in shaping the low-
frequency energy content, a factor of
critical importance for seismic hazard
assessment, engineering design, and early
warning systems, as such long-period
energy strongly affects tall buildings, long-
span bridges, and other flexible
infrastructure sensitive to resonance with
low-frequency oscillations (see fig 4).
Compared to the Mw 7.7 event, the PSD
values for the Mw 6.7 earthquake are
distinctly lower, generally peaking around
10¹² in the horizontal components and
slightly less in the vertical component,
consistent with the reduced energy release
and shorter rupture duration of a moderate-
magnitude event. The PSD plots display
clear peaks in the 0–0.5 Hz frequency
band, associated with long-period surface
waves and early-arriving P-waves, though
with amplitudes much smaller than those
of the larger earthquake. As frequency
increases, power declines rapidly, with
PSD values dropping by several orders of
magnitude beyond ~1 Hz and reaching
around 10⁶ near 5.5 Hz, highlighting the
absence of significant high-frequency
seismic energy. This lack of strong high-
frequency content is typical of
intermediate events and implies that while
long-period waves are present, their
destructive potential is considerably lower
than in major earthquakes. The
consistency across BH1, BH2, and BHZ
components suggests a uniform
distribution of energy in both horizontal
and vertical motions, reinforcing the
interpretation of a relatively symmetrical
radiation pattern. From an engineering
perspective, the dominant low-frequency
peaks indicate possible resonance risks for
mid-rise and low-rise structures with
similar natural frequencies, while the
limited high-frequency content reduces the
likelihood of widespread damage to
smaller, stiffer buildings.
Fig. 5: Wavelet Transform Analysis of Seismic Signals from the Magnitude 7.7 and 6.7
Myanmar Earthquakes Using Seismic Data: (a) BH1 Component, (b) BH2 Component, and
(c) BHZ Component Recorded at the IU.CHTO.00 Station.
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 10
dissipation. The logarithmic scaling of
both PSD and frequency axes highlights
the broad spectral distribution,
emphasizing the dominance of low-
frequency energy. Although all three
components follow a similar overall trend,
BH2 generally exhibits the highest peak
power, likely influenced by local
geological effects, wave propagation
characteristics, or instrument sensitivity.
Importantly, the PSD results underline the
greater contribution of surface waves
relative to body waves in shaping the low-
frequency energy content, a factor of
critical importance for seismic hazard
assessment, engineering design, and early
warning systems, as such long-period
energy strongly affects tall buildings, long-
span bridges, and other flexible
infrastructure sensitive to resonance with
low-frequency oscillations (see fig 4).
Compared to the Mw 7.7 event, the PSD
values for the Mw 6.7 earthquake are
distinctly lower, generally peaking around
10¹² in the horizontal components and
slightly less in the vertical component,
consistent with the reduced energy release
and shorter rupture duration of a moderate-
magnitude event. The PSD plots display
clear peaks in the 0–0.5 Hz frequency
band, associated with long-period surface
waves and early-arriving P-waves, though
with amplitudes much smaller than those
of the larger earthquake. As frequency
increases, power declines rapidly, with
PSD values dropping by several orders of
magnitude beyond ~1 Hz and reaching
around 10⁶ near 5.5 Hz, highlighting the
absence of significant high-frequency
seismic energy. This lack of strong high-
frequency content is typical of
intermediate events and implies that while
long-period waves are present, their
destructive potential is considerably lower
than in major earthquakes. The
consistency across BH1, BH2, and BHZ
components suggests a uniform
distribution of energy in both horizontal
and vertical motions, reinforcing the
interpretation of a relatively symmetrical
radiation pattern. From an engineering
perspective, the dominant low-frequency
peaks indicate possible resonance risks for
mid-rise and low-rise structures with
similar natural frequencies, while the
limited high-frequency content reduces the
likelihood of widespread damage to
smaller, stiffer buildings.
Fig. 5: Wavelet Transform Analysis of Seismic Signals from the Magnitude 7.7 and 6.7
Myanmar Earthquakes Using Seismic Data: (a) BH1 Component, (b) BH2 Component, and
(c) BHZ Component Recorded at the IU.CHTO.00 Station.
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 11
In Figure 5, the Wavelet Transform (WT)
analysis captures the detailed time,
frequency evolution of the seismic signal
generated by the Mw 7.7 Myanmar
earthquake, revealing a clear and sharp
concentration of high-energy wavelet
coefficients, represented by red and yellow
regions, across all three components (BH1,
BH2, and BHZ) beginning around
06:23:00 UTC. These high-energy zones
correspond to the arrival of P-waves,
followed by S-waves and surface waves,
which together mark the sudden burst of
seismic energy during the main rupture
phase. The vertical component (BHZ)
exhibits the most intense and sustained
energy, particularly at scales associated
with lower frequencies, indicating strong
vertical ground displacement that persists
throughout the early portion of the signal.
The horizontal components (BH1 and
BH2) also display significant energy with
relatively symmetric distributions,
suggesting well-developed and balanced
ground motion along both axes. Across all
three components, the high-energy content
decays rapidly after 06:23:39 UTC,
transitioning into lower-magnitude
coefficients, depicted in cooler blues, that
dominate the remainder of the plots and
represent the persistence of low-energy
surface and coda waves. The temporal
alignment of energy spikes in all channels
points to a coordinated release of energy
during the initial rupture, followed by a
rapid attenuation that is clearly traceable
along the time axis. Notably, the BHZ
panel maintains a broader vertical spread
of energy across multiple scales,
underscoring the dominant role of vertical
motion during this event. Meanwhile, the
consistent structure and intensity of the
wavelet patterns in BH1 and BH2
highlight the substantial and balanced
nature of horizontal shaking. Taken
together, the combination of strong
vertical energy in BHZ and comparably
robust horizontal energy in BH1 and BH2
indicates a multifaceted rupture process in
which both uplift and lateral ground
motions were prominently expressed
within a narrow, concentrated time
window.
The Wavelet Transform (WT) analysis of
the Mw 6.7 Myanmar earthquake reveals a
distinct yet less intense time–frequency
structure compared to the Mw 7.7 event,
with high-energy wavelet coefficients (red
and yellow) appearing across all three
components, BH1, BH2, and BHZ,
primarily during the 06:33:43 to 06:34:35
UTC interval. This concentrated band
corresponds to the arrival of the main
seismic phases, particularly the P-waves
and more prominently the S-waves, which
drive the initial burst of shaking and
account for the sharp rise in energy during
this period. The BHZ component exhibits
a compact and shorter-duration burst of
energy relative to the Mw 7.7 event,
reflecting vertical ground displacement but
at a comparatively lower magnitude. The
horizontal components, BH1 and BH2,
display well-timed and balanced energy
spikes, indicating significant horizontal
ground motion along both perpendicular
axes. The strongest energy regions are
tightly clustered around the onset of strong
shaking, after which a rapid transition into
lower-energy coefficients (shown in cooler
blues) occurs beyond 06:34:36 UTC,
marking the dissipation of seismic energy.
This transition is consistently observed
across all three panels, confirming the
short-lived nature of strong ground motion
in this event. While BH1 and BH2 show
nearly symmetrical wavelet energy
distributions, reinforcing the balanced
horizontal shaking, the BHZ panel retains
a vertically stretched concentration of
energy at select scales, underscoring the
contribution of vertical displacement,
albeit less pronounced than in the Mw 7.7
earthquake.
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 11
In Figure 5, the Wavelet Transform (WT)
analysis captures the detailed time,
frequency evolution of the seismic signal
generated by the Mw 7.7 Myanmar
earthquake, revealing a clear and sharp
concentration of high-energy wavelet
coefficients, represented by red and yellow
regions, across all three components (BH1,
BH2, and BHZ) beginning around
06:23:00 UTC. These high-energy zones
correspond to the arrival of P-waves,
followed by S-waves and surface waves,
which together mark the sudden burst of
seismic energy during the main rupture
phase. The vertical component (BHZ)
exhibits the most intense and sustained
energy, particularly at scales associated
with lower frequencies, indicating strong
vertical ground displacement that persists
throughout the early portion of the signal.
The horizontal components (BH1 and
BH2) also display significant energy with
relatively symmetric distributions,
suggesting well-developed and balanced
ground motion along both axes. Across all
three components, the high-energy content
decays rapidly after 06:23:39 UTC,
transitioning into lower-magnitude
coefficients, depicted in cooler blues, that
dominate the remainder of the plots and
represent the persistence of low-energy
surface and coda waves. The temporal
alignment of energy spikes in all channels
points to a coordinated release of energy
during the initial rupture, followed by a
rapid attenuation that is clearly traceable
along the time axis. Notably, the BHZ
panel maintains a broader vertical spread
of energy across multiple scales,
underscoring the dominant role of vertical
motion during this event. Meanwhile, the
consistent structure and intensity of the
wavelet patterns in BH1 and BH2
highlight the substantial and balanced
nature of horizontal shaking. Taken
together, the combination of strong
vertical energy in BHZ and comparably
robust horizontal energy in BH1 and BH2
indicates a multifaceted rupture process in
which both uplift and lateral ground
motions were prominently expressed
within a narrow, concentrated time
window.
The Wavelet Transform (WT) analysis of
the Mw 6.7 Myanmar earthquake reveals a
distinct yet less intense time–frequency
structure compared to the Mw 7.7 event,
with high-energy wavelet coefficients (red
and yellow) appearing across all three
components, BH1, BH2, and BHZ,
primarily during the 06:33:43 to 06:34:35
UTC interval. This concentrated band
corresponds to the arrival of the main
seismic phases, particularly the P-waves
and more prominently the S-waves, which
drive the initial burst of shaking and
account for the sharp rise in energy during
this period. The BHZ component exhibits
a compact and shorter-duration burst of
energy relative to the Mw 7.7 event,
reflecting vertical ground displacement but
at a comparatively lower magnitude. The
horizontal components, BH1 and BH2,
display well-timed and balanced energy
spikes, indicating significant horizontal
ground motion along both perpendicular
axes. The strongest energy regions are
tightly clustered around the onset of strong
shaking, after which a rapid transition into
lower-energy coefficients (shown in cooler
blues) occurs beyond 06:34:36 UTC,
marking the dissipation of seismic energy.
This transition is consistently observed
across all three panels, confirming the
short-lived nature of strong ground motion
in this event. While BH1 and BH2 show
nearly symmetrical wavelet energy
distributions, reinforcing the balanced
horizontal shaking, the BHZ panel retains
a vertically stretched concentration of
energy at select scales, underscoring the
contribution of vertical displacement,
albeit less pronounced than in the Mw 7.7
earthquake.
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 12
Fig. 6: Probability Density Function (PDF) comparison of seismic waveform amplitudes for
the Mw 7.7 (left column) and Mw 6.7 (right column) Myanmar earthquakes across three
components (BH1, BH2, BHZ) recorded at the IU.CHTO.00 seismic station.
In the left column of Figure 6, which
corresponds to the Mw 7.7 earthquake, the
probability density function (PDF) plots
for the three components, BH1, BH2, and
BHZ, exhibit sharp, highly concentrated
peaks centred near zero amplitude, with
narrow and symmetric distribution tails.
This distribution indicates that the seismic
amplitudes during this event were strongly
focused on the mean, reflecting intense but
relatively coherent ground motion. Among
the three, the BH2 component displays the
tallest peak, slightly exceeding those of
BH1 and BHZ, suggesting a marginally
stronger horizontal amplitude response in
that direction, though the overall
distribution patterns remain consistent
across all components. The sharpness and
symmetry of the curves highlight a
dominant high-magnitude, short-duration
burst of seismic energy, consistent with the
concentrated wavelet signatures observed
earlier. The BHZ component, with its
narrow and elevated distribution, further
reinforces the role of vertical ground
motion, consistent with thrust faulting or
subduction-related rupture mechanisms
that typically involve significant vertical
displacement.
By contrast, the right column of Figure 6,
representing the Mw 6.7 earthquake,
reveals broader and less regular PDF
curves across BH1, BH2, and BHZ, with
visibly lower and wider peaks than those
seen in the Mw 7.7 case. The amplitude
range is reduced, on the order of 10⁶
compared to 10⁷ for the larger event—
indicating substantially lower energy
release and weaker ground motion. The
curves exhibit asymmetry and multiple
fluctuations, suggesting scattered
amplitude values and a more irregular
waveform distribution, potentially linked
to the smaller rupture extent or the
influence of secondary seismic phases.
Among these, the BH2 component shows
the most pronounced yet irregular peak,
pointing to slightly greater energy in that
horizontal direction. The less concentrated
and wider PDFs characterize the moderate
magnitude of this earthquake, reflecting a
longer or more dispersed shaking sequence
rather than the sharp, pulse-like energy
release observed in the Mw 7.7 event.
Collectively, these PDF patterns confirm
the lower intensity yet dynamically
complex nature of the Mw 6.7 earthquake,
with both horizontal and vertical
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 12
Fig. 6: Probability Density Function (PDF) comparison of seismic waveform amplitudes for
the Mw 7.7 (left column) and Mw 6.7 (right column) Myanmar earthquakes across three
components (BH1, BH2, BHZ) recorded at the IU.CHTO.00 seismic station.
In the left column of Figure 6, which
corresponds to the Mw 7.7 earthquake, the
probability density function (PDF) plots
for the three components, BH1, BH2, and
BHZ, exhibit sharp, highly concentrated
peaks centred near zero amplitude, with
narrow and symmetric distribution tails.
This distribution indicates that the seismic
amplitudes during this event were strongly
focused on the mean, reflecting intense but
relatively coherent ground motion. Among
the three, the BH2 component displays the
tallest peak, slightly exceeding those of
BH1 and BHZ, suggesting a marginally
stronger horizontal amplitude response in
that direction, though the overall
distribution patterns remain consistent
across all components. The sharpness and
symmetry of the curves highlight a
dominant high-magnitude, short-duration
burst of seismic energy, consistent with the
concentrated wavelet signatures observed
earlier. The BHZ component, with its
narrow and elevated distribution, further
reinforces the role of vertical ground
motion, consistent with thrust faulting or
subduction-related rupture mechanisms
that typically involve significant vertical
displacement.
By contrast, the right column of Figure 6,
representing the Mw 6.7 earthquake,
reveals broader and less regular PDF
curves across BH1, BH2, and BHZ, with
visibly lower and wider peaks than those
seen in the Mw 7.7 case. The amplitude
range is reduced, on the order of 10⁶
compared to 10⁷ for the larger event—
indicating substantially lower energy
release and weaker ground motion. The
curves exhibit asymmetry and multiple
fluctuations, suggesting scattered
amplitude values and a more irregular
waveform distribution, potentially linked
to the smaller rupture extent or the
influence of secondary seismic phases.
Among these, the BH2 component shows
the most pronounced yet irregular peak,
pointing to slightly greater energy in that
horizontal direction. The less concentrated
and wider PDFs characterize the moderate
magnitude of this earthquake, reflecting a
longer or more dispersed shaking sequence
rather than the sharp, pulse-like energy
release observed in the Mw 7.7 event.
Collectively, these PDF patterns confirm
the lower intensity yet dynamically
complex nature of the Mw 6.7 earthquake,
with both horizontal and vertical
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 13
components contributing to its ground
motion profile.
CONCLUSIONS
The comprehensive analysis conducted in
this study provides important insights into
the seismic behaviour of the two major
earthquakes that struck Myanmar on
March 28, 2025—Mw 7.7 and Mw 6.7—
using advanced signal processing
techniques applied to waveform data from
the Seismological Facility for the
Advancement of Geoscience (SAGE). By
integrating Fast Fourier Transform (FFT),
Power Spectral Density (PSD), Wavelet
Transform (WT), and Probability Density
Function (PDF) analyses on three-
component seismic recordings (BH1, BH2,
BHZ), we successfully characterized the
temporal, spectral, and statistical features
of ground motion with high precision.
The findings reveal fundamental
differences in the response of the two
events, governed by their magnitudes,
rupture mechanics, and propagation
characteristics. The Mw 7.7 earthquake
released substantially higher seismic
energy, marked by prolonged shaking and
large-amplitude waveforms, especially in
the vertical (BHZ) and horizontal (BH2)
components. Its signals were dominated by
low-frequency content, primarily in the 0–
0.5 Hz range, which is particularly
hazardous for tall and flexible structures
due to resonance effects. In contrast, the
Mw 6.7 earthquake, despite occurring
under similar tectonic conditions,
generated more localized and transient
bursts of energy with lower amplitudes
and relatively weaker frequency content
above 1 Hz, indicative of a shorter rupture
duration and reduced structural risk.
The FFT and PSD results quantitatively
confirmed these differences, showing
sharp spectral attenuation with frequency
in both events, but more pronounced in the
smaller earthquake. The Mw 7.7 event
exhibited peak spectral amplitudes on the
order of , further underscoring
its dominance in low-frequency energy
and potential for widespread impact. The
WT analysis highlighted the contrasting
time-frequency evolution of the two
events: the Mw 7.7 rupture produced
dense, sustained clusters of high-energy
wavelet coefficients, whereas the Mw 6.7
rupture was characterized by a compact,
short-lived energy burst. This behaviour
suggests differing fault rupture dynamics
and supports the interpretation of
significant subduction-related vertical
displacement in the larger event. Finally,
the PDF analysis reinforced these
distinctions, with the Mw 7.7 earthquake
showing sharp, symmetric peaks near zero
amplitude, reflecting intense and coherent
ground motion, while the Mw 6.7
earthquake displayed broader, more
irregular distributions, indicative of
weaker and less uniform shaking.
Conflicts of Interest
The authors declare that there are no
conflicts of interest related to this work.
Authors’ Contributions
All authors contributed equally to the
conception, analysis, and writing of this
study.
REFERENCE
1. Boore, D. M. (2003). Simulation of
ground motion using the stochastic
method. Pure and Applied
Geophysics, 160, 635–676.
https://doi.org/10.1007/PL00012553
2. Trifunac, M. D., & Brady, A. G.
(1975). A study on the duration of
strong earthquake ground motion.
Bulletin of the Seismological Society
of America, 65(3), 581–626.
3. Liu, S. C., & Fagel, L. W. (1971).
Earthquake interaction by fast
Fourier transform. Journal of the
Engineering Mechanics Division,
97(4), 1223–1237.
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 13
components contributing to its ground
motion profile.
CONCLUSIONS
The comprehensive analysis conducted in
this study provides important insights into
the seismic behaviour of the two major
earthquakes that struck Myanmar on
March 28, 2025—Mw 7.7 and Mw 6.7—
using advanced signal processing
techniques applied to waveform data from
the Seismological Facility for the
Advancement of Geoscience (SAGE). By
integrating Fast Fourier Transform (FFT),
Power Spectral Density (PSD), Wavelet
Transform (WT), and Probability Density
Function (PDF) analyses on three-
component seismic recordings (BH1, BH2,
BHZ), we successfully characterized the
temporal, spectral, and statistical features
of ground motion with high precision.
The findings reveal fundamental
differences in the response of the two
events, governed by their magnitudes,
rupture mechanics, and propagation
characteristics. The Mw 7.7 earthquake
released substantially higher seismic
energy, marked by prolonged shaking and
large-amplitude waveforms, especially in
the vertical (BHZ) and horizontal (BH2)
components. Its signals were dominated by
low-frequency content, primarily in the 0–
0.5 Hz range, which is particularly
hazardous for tall and flexible structures
due to resonance effects. In contrast, the
Mw 6.7 earthquake, despite occurring
under similar tectonic conditions,
generated more localized and transient
bursts of energy with lower amplitudes
and relatively weaker frequency content
above 1 Hz, indicative of a shorter rupture
duration and reduced structural risk.
The FFT and PSD results quantitatively
confirmed these differences, showing
sharp spectral attenuation with frequency
in both events, but more pronounced in the
smaller earthquake. The Mw 7.7 event
exhibited peak spectral amplitudes on the
order of , further underscoring
its dominance in low-frequency energy
and potential for widespread impact. The
WT analysis highlighted the contrasting
time-frequency evolution of the two
events: the Mw 7.7 rupture produced
dense, sustained clusters of high-energy
wavelet coefficients, whereas the Mw 6.7
rupture was characterized by a compact,
short-lived energy burst. This behaviour
suggests differing fault rupture dynamics
and supports the interpretation of
significant subduction-related vertical
displacement in the larger event. Finally,
the PDF analysis reinforced these
distinctions, with the Mw 7.7 earthquake
showing sharp, symmetric peaks near zero
amplitude, reflecting intense and coherent
ground motion, while the Mw 6.7
earthquake displayed broader, more
irregular distributions, indicative of
weaker and less uniform shaking.
Conflicts of Interest
The authors declare that there are no
conflicts of interest related to this work.
Authors’ Contributions
All authors contributed equally to the
conception, analysis, and writing of this
study.
REFERENCE
1. Boore, D. M. (2003). Simulation of
ground motion using the stochastic
method. Pure and Applied
Geophysics, 160, 635–676.
https://doi.org/10.1007/PL00012553
2. Trifunac, M. D., & Brady, A. G.
(1975). A study on the duration of
strong earthquake ground motion.
Bulletin of the Seismological Society
of America, 65(3), 581–626.
3. Liu, S. C., & Fagel, L. W. (1971).
Earthquake interaction by fast
Fourier transform. Journal of the
Engineering Mechanics Division,
97(4), 1223–1237.
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 14
4. Li, Y., & Kareem, A. (1991).
Simulation of multivariate
nonstationary random processes by
FFT. Journal of Engineering
Mechanics, 117(5), 1037–1058.
5. Alsop, L. E., & Nowroozi, A. A.
(1966). Faster Fourier analysis.
Journal of Geophysical Research,
71(22), 5482–5483.
6. Crockett, R. G. M. (2018, April). The
Fast Fourier Transform (FFT) as a
statistical tool for time-series
analysis. In European Geosciences
Union (EGU) General Assembly
2018.
7. Liu, S. C. (1970). Evolutionary
power spectral density of strong-
motion earthquakes. Bulletin of the
Seismological Society of America,
60(3), 891–900.
8. Bai, Q., Zhu, F., Kang, Y., & Bian, J.
(2008, July). Power spectral density
estimation of seismic wave based on
wavelet transform. In 2008 Chinese
Control and Decision Conference
(pp. 4600 –4603). IEEE.
https://doi.org/10.1109/CCDC.2008.
4597945
9. Power, D. V. (1969). Analysis of
earth motions and seismic sources by
power spectral density. Bulletin of
the Seismological Society of
America, 59(3), 1071–1091.
10. Christian, J. T. (1989). Generating
seismic design power spectral
density functions. Earthquake
Spectra, 5(2), 351–368.
https://doi.org/10.1193/1.1585724
11. Oshima, M., & Takenaka, H. (2022).
Enhancement of direct waves based
on the probability density function of
seismic wave amplitudes.
Geophysical Journal International,
231(1), 327 –354.
https://doi.org/10.1093/gji/ggab499
12. Jun-Wei, B., Zuo-Ju, W., Zhi-Jia,
W., Fang, O., & Yuan, C. (2014,
June). Wavelet transform and its
application in earthquak e
engineering. In 2014 Fifth
International Conference on
Intelligent Systems Design and
Engineering Applications (pp. 1126–
1128). IEEE.
https://doi.org/10.1109/ISDEA.2014.
270
13. Iyama, J., & Kuwamura, H. (1999).
Application of wavelets to analysis
and simulation of earthquake
motions. Earthquake Engineering &
Structural Dynamics, 28(3), 255–
272.
https://doi.org/10.1002/(SICI)1096-
9845(199903)28:3<255::AID-
EQE811>3.0.CO;2-K
14. Teymur, B., Madabhushi, S. P., &
Newland, D. E. (2001). Application
of wavelet theory in the analysis of
earthquake motions recorded during
the Kocaeli earthquake. In
International Conferences on Recent
Advances in Geotechnical
Earthquake Engineering and Soil
Dynamics, Turkey, 1999.
15. Kumar, P., & Foufoula‐Georgiou,
E. (1997). Wavelet analysis for
geophysical applications. Reviews of
Geophysics, 35(4), 385–412.
https://doi.org/10.1029/97RG00427
16. Li, X., Li, Z., Wang, E., Feng, J.,
Chen, L., Li, N., & Kong, X. (2016).
Extraction of microseismic
waveforms characteristics prior to
rock burst using Hilbert–Huang
transform. Measurement, 91, 101–
113.
https://doi.org/10.1016/j.measuremen
t.2016.05.047
17. Şafak, E. (1999). Wave-propagation
formulation of seismic response of
multistory buildings. Journal of
Structural Engineering, 125(4), 426–
437.
https://doi.org/10.1061/(ASCE)0733-
9445(1999)125:4(426)
18. Dolai, R., & Samanta, D. (2025).
Stochastic modeling of compact
binary coalescences for gravitational
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 14
4. Li, Y., & Kareem, A. (1991).
Simulation of multivariate
nonstationary random processes by
FFT. Journal of Engineering
Mechanics, 117(5), 1037–1058.
5. Alsop, L. E., & Nowroozi, A. A.
(1966). Faster Fourier analysis.
Journal of Geophysical Research,
71(22), 5482–5483.
6. Crockett, R. G. M. (2018, April). The
Fast Fourier Transform (FFT) as a
statistical tool for time-series
analysis. In European Geosciences
Union (EGU) General Assembly
2018.
7. Liu, S. C. (1970). Evolutionary
power spectral density of strong-
motion earthquakes. Bulletin of the
Seismological Society of America,
60(3), 891–900.
8. Bai, Q., Zhu, F., Kang, Y., & Bian, J.
(2008, July). Power spectral density
estimation of seismic wave based on
wavelet transform. In 2008 Chinese
Control and Decision Conference
(pp. 4600 –4603). IEEE.
https://doi.org/10.1109/CCDC.2008.
4597945
9. Power, D. V. (1969). Analysis of
earth motions and seismic sources by
power spectral density. Bulletin of
the Seismological Society of
America, 59(3), 1071–1091.
10. Christian, J. T. (1989). Generating
seismic design power spectral
density functions. Earthquake
Spectra, 5(2), 351–368.
https://doi.org/10.1193/1.1585724
11. Oshima, M., & Takenaka, H. (2022).
Enhancement of direct waves based
on the probability density function of
seismic wave amplitudes.
Geophysical Journal International,
231(1), 327 –354.
https://doi.org/10.1093/gji/ggab499
12. Jun-Wei, B., Zuo-Ju, W., Zhi-Jia,
W., Fang, O., & Yuan, C. (2014,
June). Wavelet transform and its
application in earthquak e
engineering. In 2014 Fifth
International Conference on
Intelligent Systems Design and
Engineering Applications (pp. 1126–
1128). IEEE.
https://doi.org/10.1109/ISDEA.2014.
270
13. Iyama, J., & Kuwamura, H. (1999).
Application of wavelets to analysis
and simulation of earthquake
motions. Earthquake Engineering &
Structural Dynamics, 28(3), 255–
272.
https://doi.org/10.1002/(SICI)1096-
9845(199903)28:3<255::AID-
EQE811>3.0.CO;2-K
14. Teymur, B., Madabhushi, S. P., &
Newland, D. E. (2001). Application
of wavelet theory in the analysis of
earthquake motions recorded during
the Kocaeli earthquake. In
International Conferences on Recent
Advances in Geotechnical
Earthquake Engineering and Soil
Dynamics, Turkey, 1999.
15. Kumar, P., & Foufoula‐Georgiou,
E. (1997). Wavelet analysis for
geophysical applications. Reviews of
Geophysics, 35(4), 385–412.
https://doi.org/10.1029/97RG00427
16. Li, X., Li, Z., Wang, E., Feng, J.,
Chen, L., Li, N., & Kong, X. (2016).
Extraction of microseismic
waveforms characteristics prior to
rock burst using Hilbert–Huang
transform. Measurement, 91, 101–
113.
https://doi.org/10.1016/j.measuremen
t.2016.05.047
17. Şafak, E. (1999). Wave-propagation
formulation of seismic response of
multistory buildings. Journal of
Structural Engineering, 125(4), 426–
437.
https://doi.org/10.1061/(ASCE)0733-
9445(1999)125:4(426)
18. Dolai, R., & Samanta, D. (2025).
Stochastic modeling of compact
binary coalescences for gravitational
Journal of Earthquake Science and Soil Dynamics Engineering
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 15
wave analysi s. SSRN.
https://doi.org/10.2139/ssrn.5289415
Volume 8 issue 3, Sep-Dec 2025
e-ISSN:3048-6017
HBRP Publication Page 1-15 2025. All Rights Reserved Page 15
wave analysi s. SSRN.
https://doi.org/10.2139/ssrn.5289415
Categories
GeneralDownload
Download Slideshow Get the original presentation fileQuick Actions
Statistics
- Views 0
- Slides 15
- Favorites 1
- Age 70 days
Related Slideshows
22
Pray For The Peace Of Jerusalem and You Will Prosper
RodolfoMoralesMarcuc
35 views
26
Don_t_Waste_Your_Life_God.....powerpoint
chalobrido8
38 views
31
VILLASUR_FACTORS_TO_CONSIDER_IN_PLATING_SALAD_10-13.pdf
JaiJai148317
34 views
14
Fertility awareness methods for women in the society
Isaiah47
31 views
35
Chapter 5 Arithmetic Functions Computer Organisation and Architecture
RitikSharma297999
30 views
5
syakira bhasa inggris (1) (1).pptx.......
ourcommunity56
31 views