1 Introduction

Egyptian National Strong Motion Network (ENSMN) was installed in north Egypt. It was started in 2008 with one station at Helwan [1] and updated to 12 stations at the end of 2018 (Fig. 1), reaching 23 stations by 2024 to cover more area in highly population cities in north Egypt. The Egyptian National seismic network (ENSN, 80 stations) and the Egyptian National strong motion network (ENSMN, 23 stations) are used to monitor the seismic events in real-time data to the main center in Helwan (Fig. 1), to asses risk in Egypt. A moderate seismic activity characterizes Egypt. The plate tectonic activity between the African, Arabian, and Eurasian plates is the driving force controlling the seismicity of north Egypt and the East Mediterranean region (Fig. 1). The majority of earthquakes occurred at shallow depths. It coincided with the significant tectonic trends of the Gulf of Suez rift, Gulf of Aqaba–Dead Sea transform, Pelusium Shear System, and their connection with the Gulf of Suez–Cairo to Alexandria trend (Kebeasy [2]; Mahmoud [3]). Building up a strong motion database for the Egyptian sector came later with an increase in developments and constructions of tall buildings in new cities, which have its importance in serving as primary sources of information for engineering and seismologists. By analyzing the seismic motion records, engineers can better understand structure’s dynamic behaviour and estimate the seismic forces response. Building codes, land-use planning, and disaster preparedness can all be informed from a strong motion database.

Fig. 1
figure 1

Main tectonic features and Seismicity Ml > 3.5 in north Egypt and adjacent areas

Full size image

We examined the ground motion response spectra of the December 31, 2018 earthquake while taking into account the geologic conditions that prevail at the strong motion station sites. It was necessary to have clear information about the fault type, local or regional geology, site effect, and the nature of the dominant structure in the study area [4]. An earthquake with a magnitude of Mw = 3.8 and ML = 4.3 that struck the Cairo Suez area at the end of 2018 was recorded at the Egyptian strong motion and seismological networks, with the origin time 10:36:34 [5]. It is focal mechanism solution showed normal fault (Fig. 2) with a slight strike-slip component according to EL Gabry et al. [5] study. The large two historical events recorded in this area on September 29, 1984, and April 29, 1974 (CMT harved solution, [6]), with ML 4.6 in the Cairo-Suez, shear zone showed the same mechanism of faulting normal (Table 1) with strike-slip with main trend E-W to NE-SE. On August 24, 2002 a moderated earthquake with Mb = 4.3 was recorded east of the December 31, 2018 event, at 23:01 local time. The recorded PGA calculated by Abd EL-Aal et al. [7] at that TURA strong station about 50 km west of the KOT station was 2.25 gal (cm/s2). The located area of these events was affected by late Oligocene–early Miocene deformation related to the opening of the Suez rift as shown in the tectonic map (EGSMA updated in 2008, [8]; Fig. 2). This deformation was responsible for the E–W and NW–SE oriented normal faults [8, 9] associated with gentle folds affecting the upper Eocene and Miocene strata.

Fig. 2
figure 2

Geology map and focal mechanism of the December 31, 2018 earthquake and adjacent large significant events

Full size image
Table 1 The significant historical events focal mechanism in the studied area
Full size table

Sallam et al. [10] investigated the stratigraphic succession along the Cairo-Suez Road in the studied area. The central part of the Cairo–Suez district consists of nearly parallel, high mountain ridges trending in an E–W and NW– SE direction with lowlands in between. The Neogene (Miocene and Pliocene) rocks are characterized by the outcrops of Wadi Ghoweiba, and Wadi Hagul as well as north Jebel Iweibed, Jebel Homeyra and Jebel Gharra while the Quaternary deposits cover the Wadi floors. Low topographic regions (grabens) are filled with sediments from the Upper Eocene and Oligocene, whereas the mountains and high scarps that overlook both sides of the Qattamiya–Ain Sukhna Road are mostly formed by Middle Eocene rocks. Whereas the volcanic sheets, which are composed of basalts, were located in different locations in and around the Cairo–Suez region. They were noted at Gebel Abu Treifiya, encircling it on all sides except the northern outcrop, and they were also noted along the Qattamiya–Ain Sukhna Road, south of Gebel Umm Reheiat, where they formed noticeable hills. In the late Oligocene and early Miocene, the NW–SE and E-W normal faults rejuvenated, causing the basalts to be extruded [8]. Meneisy and Abdel Aal [11] dated these basalts and the result confirms the late Oligocene—early Miocene age.

In our paper (1) we made a questionnaire for this earthquake to determine the intensity values across the felt zones and (2) then, we calculated the PGA, PGV, maximum displacement and pseudo-spectral acceleration using seven stations KOT, ZAG, BNH, MANS, ISM, NUB, and ANS. (2) to point out some basic properties of the tectonic environments governing earthquakes within the new capital city. (3) identify the response design spectrum at recorded earthquake stations, (3) finally we calculated the H\V ratio obtained from earthquake data at the seven stations to calculate the predominant frequency and compared them with average shear velocities measured (Vs30) at 4 different types of soils.

2 Methodology and data analysis

This event was recorded by six Titan Nanometric stations and only one Reftek station, installed in and around the Nile Delta, Egypt (Fig. 1, Table 2). The sensitivity factor of the sensor was 0.0625 × 10–6 g and 1 × 10–6 respectively. The sample rate was adjusted for these stations at 200 samples per second.

Table 2 Recording stations of the December 31, 2018 earthquake
Full size table

Our analysis includes three mains: The first step is collecting data for constructing the intensity map for the December 2018 earthquake. We used advanced social media technology like Google Questionnaire, Facebook, EMSC website (European Mediterranean Seismological Centre, [12]). We also collected intensity information from people present at the time of the event in NRIAG Qattamiya observatory which is located close to the source. The collected information was evaluated to determine the intensity with the radius of the circle ~ 52 km from the epicenter of the December 31, 2018, event (Table 3). An intensity map (Fig. 3) gives us the shaking pattern reflected by the December 31, 2018 earthquake damage and the effects on humans from the earthquake event.

Table 3 Intensities of the December 31 2018 earthquake collected from Questionaries, EMSC, based on the MMI scale
Full size table
Fig. 3
figure 3

Intensity values Map recorded by populations in circular radius ~ 50 km

Full size image

In the second step of analysis, we calculated the main spectra parameters, which are acceleration with corresponding time histories (PGA) & Fourier amplitude spectrum (Figs. 4 and 5), peak ground velocity (PGV), and relative displacement (RD), pseudo acceleration (PSAA), Aria's intensity (Table 3). We have calculated these acceleration parameters for the seven stations using the MATLAB code OpenSeismo code developed by Papazafeiropoulos and Plevris [13], which is summarized in Table 4.

Fig. 4
figure 4

Maximum acceleration with time series recorded at stations a KOT (E-W); b ANS (Z); c ISM (N–S); d ZAG (N–S), e BNH (Z); f MANS (E–W); g NUB (N–S)

Full size image
Fig. 5
figure 5

Fourier amplitude spectrum with time series recorded at stations includes a KOT (E-W); b ANS (Z); c ISM (N–S); d ZAG (N–S), e BNH (Z); f MANS (E–W); g NUB (N–S)

Full size image
Table 4 Shows spectra parameters calculated by the MATLAB code of the December 31 2018 earthquake
Full size table

A primary resource for defining these accelerogram parameters is the spectrum of recorded seismic events [14], and the subsequent extrapolation of those data points to higher energies as shown in Table 3 for the December 31, 2018 earthquake. The response spectrum methodology can be used to calculate the maximum amplitudes of a group of simple damped harmonic oscillator’s response to a certain ground-motion time history [15, 16]. The pseudo acceleration (PSAA) is a measure of the maximum elastic spring force per unit of mass. Relative displacement (RD) represents the maximum value of the relative displacement of the simple system during vibratory motion. Peak ground velocity (PGV) gives an approximate index of the greatest velocity, relative to its base, of the center of mass of the resonant simple structure. PGV can also be related to the maximum energy absorbed in the spring. For low dam**, the pseudo relative velocity (PSRV) spectrum provides an upper bound to the Fourier amplitude spectrum [17]. While the Arias instrumental intensity [18] and Housner and Jennings [19] were defined as the sum of total energy per unit weight stored on the population of the undamped oscillator’s uniform. It is distributed concerning frequency at the end of the duration of earthquake motion [18] as:

$$I a =\frac{\pi }{2g}\underset{0}{\overset{td}{\int }}{(a\left(t\right))}^{2}dt$$
(1)

Where g is acceleration due to gravity.  Thus, duration and amplitude considered in the definition of Ia. The significant duration D5_95 represents the time interval between the 5% and 95% of the Arias intensity. It is a measure used in seismology to quantify the duration of strong ground shaking during an earthquake. i.e. if you have a time series of Arias intensity values, you can calculate the duration between the time when the intensity reaches 5% of its maximum value and the time when it reaches 95% of its maximum value.

Controlling the frequency content of a strong ground motion requires an understanding of its Fourier spectrum. An approximate periodic function that has an earthquake history [with a tail of zeros] may be expressed as

$$x(t) = c_{0} + \sum\limits_{n = 1}^{\infty } {c_{n} } \sin (\omega_{n} t + \phi_{n} )$$
(2)

where \(c_{n}\) and \(\phi_{n}\) are the amplitude and phase angle, respectively, of the nth harmonic in the Fourier series. The Fourier amplitude spectrum (FAS) is a plot of \(c_{n}\) versus \(\omega {}_{n}\). It shows how the amplitude of the motion varies with frequency. The frequency at which a structure vibrates the strongest is known as the resonant frequency, which is related to building engineering. This may be determined using the Fourier spectrum (Fig. 5, with the maximum Fourier spectrum defined as shown in Table 4).

In the third analysis step, the horizontal-to-vertical spectral ratio (HVSR) was calculated from the data from the waveform earthquake of December 31, 2018, using the Geopsy software (Wathelet et al. [20]). The H/V curves are computed in this study by choosing the S-wave and the coda information for the signals received after the S-wave arrival. We considered analysis H\V of ground motion earthquake for the S-wave part window because S-wave site responses are considered significantly dependent on the local geological structure and have higher amplification than work with P-wave and noise (Pilz et al. [21]). The H/V curves are computed by combining the Fourier spectra of horizontal and vertical components of an earthquake recording starting at the S-wave arrival. The computing analysis of the Fourier spectrum is applied for each signal, after which the different signals are processed to obtain an equivalent spectrum, with a peak corresponding to the fundamental frequency (F0) of the soil according to a given amplification (A0). This method was first proposed by Nakamura in 1989 to determine the site's amplification function and resonance frequencies of S-wave produced by unconsolidated sediments overlying high-velocity bedrock. The H/V spectrum was calculated according to the following equation (Delgado et al. [22]).

$$\frac{H}{V}=\frac{\sqrt{{FNS}^{2 }+{FEW}^{2 }}}{2FZ}$$

FNS and FEW are Fourier amplitude spectrum of the horizontal components, FZ is the vertical components.

The HVSR steps included in data processing using the HVSR technique in this study are as follows: (a) splitting the data into time blocks with 2048 data points each at a sampling rate of 200 samples per second; (b) applying Fast Fourier transformation to the three component waveform data to transform them into the frequency domain using Geopsy software; (c) Spectral amplitudes were smoothed using the Konno and Ohmachi [23, 24], ensuring smoothing of numerical instabilities while preserving the major features of the earthquake spectra that were considered valid only when the S/N ratio is greater than 3. (d) figuring out the power spectrum for three-component records of the same sensor; (e) estimating the HVSR, which is calculated as the square root of the ratio of the two merged horizontal components and the vertical component power spectra [25, 26]; (f) defining the amplification factor and predominant frequency by monitoring changes in the H/V ratio with frequency and determining the results (Fig. 6).

Fig. 6
figure 6

HVSR analysis for the December 31, 2018, earthquake for several stations, including a KOT, b ISM, c ZAG, d BNH, e MANS, and f NUB. The dotted lines represent the standard deviation, while the solid represents the average H\V

Full size image

3 Results and discussions

The fact that site effects have the potential to magnify seismic motions within the ground’s uppermost layer greatly makes them a significant factor in characterizing seismic ground motions [7]. Direct information can be inferred from calculations of the peak ground acceleration with a time history (Table 1, Fig. 3). Ground motion duration is determined by the amplitude of the ground motion, the distance to the epicenter, and the local site characteristics [27]. The high level of released energy on December 31 2018, M w = 3.8 was at partially high acceleration values at the near station. In the near station, the peak ground acceleration located at the Qattmayia (KOT station, 10 km from epicenter) observed in the (E–W), (N-S) and (Z) directions were 16, 14, and 10 gals, respectively. At the intermediate distance, ZAG station (68.5 km) has values of the peak of acceleration slightly higher than Anshas as shown in Table 4. The peak ground accelerations were the highest at E–W channels at all stations except BNH, ZAG and NUB stations. The ZAG and NUB stations’ peak ground acceleration showed the maximum in N–S, while Banha’s maximum acceleration peaked at the vertical component. This was a result of the directivity and the epicentral distance of the earthquake (Khellafi et al. [27], Bechtoula and Ousalem, [28], Pacor et al. [29]).

The H/V spectral ratio is a technique used to identify the fundamental resonant frequency of a sedimentary layer with qualitative amplification [30]. Chaaraoui et al. [31] and Khedr et al. [32] have demonstrated the use of this ratio in determining the fundamental frequency of soft soils by noting the nearly systematic disappearance of the vertical component of Rayleigh wave motion at the fundamental S-wave resonant frequency [33]. Nakamura [34] assessed both the fundamental resonance frequency (fo) and the ground motion amplification affected by a surface sedimentary layer at a certain site. The peak amplitude (Ao) is attributed to the contrast in shear-wave velocity between an upper soft sediment layer and a harder layer underlying it [35]. The physical parameters of the entire earthquake-site-structure system are correlated with the damage when analyzing the patterns of damage in an earthquake (Panzera et al. [36]). The frequency-dependent response of both the soil-rock column and the structure is engaged in soil-structure interaction, which causes damage in many earthquakes. Many earthquakes generate damage due to the interaction of soil and structure, including the frequency-dependent response of both the soil-rock column and the structure (Panou et al. [37]). The response of the soil-rock column, known as site amplification, is controversial because the soil has strain-dependent properties that affect the way the soil column filters the input body and surface seismic waves, modifying the amplitude and phase spectra and the duration of the surface ground motion [38, 39]. HVSR analysis was performed for the December 31 2018 earthquake at stations including KOT, ANS, ISM, ZAG, BNH, MANS, and NUB (Table 5 and Fig. 5). We performed the HVSR analysis because no complete geotechnical information is available to characterize the studied strong motion station sites. The inferred stratigraphy has also been compared to the geology map. The average shear wave velocity for the top 30 m (Vs30) is collected from previous studies, as shown in Table 4, to identify the soil type and compare the results obtained from H/V.

Table 5 Amplification factor and Fundamental frequency obtained from earthquake record and average shear wave velocity for the top 30 m (V30)
Full size table

The investigation of the HVSR analysis obtained from the earthquake data shows that fundamental frequency (F0) values vary between 0.7 and 4.2 HZ while the amplification factor (A0) ranges from 1.6 to 5.8. The highest amplification A0 found in BNH, MANS, and ZAG stations ranges from 5.8, 4, 4 and F0 values from 3, 4.2, 1. MANS, ZAG, and BNH stations are characterized by silt soil type with Vs30 = 170 to 195 m/s, class E, D. The ANS site is characterized by soil type C with A0 = 4 and F0 = 1.1. Two peaks were observed at the KOT station, the first one having A0 = 2.4 and F0 = 0.5 HZ, which is less than the second one, which had A0 = 3.8 and F0 = 3 Hz. The First peak might be related to relative superficial impedance contrasts, which is also supported by Bravo [42]. The second peak could be related to local geological conditions, as inferred from the geological map. Hard to moderate limestone soil with Vs30 measured at this site equals 307 m/s of soil type C. While the ISM station has A0 = 2.8 and F0 = 0.5 HZ. ISM station is located in sand sediments, and the measured Vs30 varies from 290 to 500 m/s in class type C. The NUB station has little amplification, and it was constructed in a rock site with Vs30 = 760 to 1500 m/s and the A0 = 1.6 with fundamental frequency = 2 Hz. These gentle amplification in basement rocks may be related to many factors, such as the degree of weathering, the thickness of the rock is low or the presence of fractures or tectonics situations (Chaaraoui et al. [43], Pilez et al. [21]).

Seismosignal software is used to calculate the response-spectrum analysis for the December 31, 2018 event (Fig. 7). The response spectrum is defined by measuring pseudo-spectral acceleration as a function of the structural period for a specific time history and dam** level. Response-spectrum analysis sheds light on dynamic behaviour [44]. It’s well known that parameters such as soil type [44], epicentral distance, earthquake magnitude, duration and characteristics of seismic source significantly affect the shape and amplitude of the response spectrum [25]. Structures with shorter periods experience more acceleration as shown in Fig. 7a in the KOT station, while longer periods experience more displacement. Structural performance objectives should be considered during initial design and response-spectrum analysis. It is essential to compare the Response design spectrum with the design code building [25]. The periods play a crucial role in structural design and seismic analysis. The acceleration value at 0.2 s represents the maximum acceleration response of structure during an earthquake at short period, while the acceleration value at 1 s gives insights into structural response at this longer time period and it accounts for the dynamic properties of the building, including its fundamental period of vibration [42]. The acceleration values at time 0.2 s can be obtained from Fig. 7a, b, c, d, e, f, g as 23.3, 2.8, 0.3,3.1,3,0.8,0.8 Gals while acceleration values at time 1 s equal to 15, 0.3, 0.5, 8, 0.2, 1, 0.1 Gals for KOT, ANS, ISM, ZAG, BNH, MANS, NUB stations respectively.

Fig. 7
figure 7

Response spectrum with time in seconds recorded at stations includes a KOT (E–W); b ANS (Z); c ISM (N–S); d ZAG (N–S), e BNH (Z); f MANS (E–W); g NUB (N–S)

Full size image

4 Conclusions and recommendations

These kinds of studies help in update the building code after recording a new significant seismic event by a strong motion accelerograph [25] and combining it with the near field normalized pseudo acceleration response. Our findings demonstrate that local site conditions play an important role in modifying the shape of the response spectrum. The Egyptian strong motion network was constructed to understand the peak ground motion resulting from recording earthquakes in highly populated areas in Egypt. The ENSMN recorded an earthquake with a magnitude Mw = 3.8 at the end of the year 2018 near the New Administrative Egyptian city. We calculated Peak ground acceleration and velocity, displacement and Pseudo acceleration spectra, significant duration Aria’s intensity, Fourier spectrum and mean period using the openSeismo MATLAB code [11]. Our main results reflected that the values had been affected by the soil type (site effects), type of fault, and directivity in each strong motion station. The maximum recorded acceleration was found at the KOT station, 10 km from the epicenter, with a value of 16.38 Gal, which characterized Vs30 = 307 m/s and amplification factor A0 = 3.8 and F0 = 3 Hz. The high amplification factor at KOT station is expected to be due to the station site being characterized by low frequency-low velocity, as proposed by Stambouli et al. [46]. The maximum observed intensity is (IV). The HVSR analysis for the Egyptian national strong motion stations reflected a high amplification factor at BNH, MANS, ZAG, ANS Stations which are characterized by Nile silty sedimentary soil with the lowest value at ISM and NUB Stations as it was characterized by sand deposits and weathered or fractured rocks sites respectively.

The October 12 1992 earthquake which occurred ~ 60 km west of December 31, 2018 which have a moment magnitude of 5.8 with great damage to Cairo buildings. At that time, there is no strong motion stations to record this earthquake. Earthquake accelerograms are needed for engineering calculations; therefore, the simulation of such accelerograms is becoming increasingly important to estimate the ground motion value of past large earthquakes [14]. Omar et al. [45] simulated the ground motion at the same site of the KOT station with a 30-Gal value. It is crucial to observe the maximum ground motion with time history and successive records of seismic activities to update the building code and achieve sustainable development in newly constructed cities.

The new capital administrative city has been constructed recently in Egypt, with many tall buildings (Skyscrapers), and the increase of strong motion accelerographs is highly required in the near future around the new city and along the tallest building floor and up levels. It will act as disaster mitigation by building an early warning system and structure health monitoring. The buildup of structural health monitoring helps us monitor changes in the material and geometric properties of engineering structures in new cities, including bridges and high buildings.