Automatic inspection and analysis of digital waveform images by means of convolutional neural networks

Analyzing seismic data to get information about earthquakes has always been a major task for seismologists and, more in general, for geophysicists. Recently, thanks to the technological development of observation systems, more and more data are available to perform such tasks. However, this data “grow up” makes “human possibility” of data processing more complex in terms of required efforts and time demanding. That is why new technological approaches such as artificial intelligence are becoming very popular and more and more exploited. In this paper, we explore the possibility of interpreting seismic waveform segments by means of pre-trained deep learning. More specifically, we apply convolutional networks to seismological waveforms recorded at local or regional distances without any pre-elaboration or filtering. We show that such an approach can be very successful in determining if an earthquake is “included” in the seismic wave image and in estimating the distance between the earthquake epicenter and the recording station.

Average error determination in the calculation of earthquake epicenter coordinates

Objective. The study is aimed at determining the dependence of the average error in calculating the epicenter coordinates of an earthquake on errors in measuring the velocities of seismic waves for various methods of seismic event localization. Error distribution investigation for the method for determining the earthquake hypocenter coordinates using the Cassinian oval. Methods. The problem was solved using statistical methods: methods of frequency and regression analyzes, means comparison method, and uniform search method. Results. A relationship between the accuracy of measuring the velocities of seismic waves when determining the coordinates of an earthquake epicenter were established for four different earthquake hypocenter coordinates calculation methods. A method for determining the earthquake hypocenter coordinates using the fourth-order figure of the Cassinian oval was proposed. The error distribution density of the Cassinian oval method was compared with the ones of other methods. Conclusion. The results obtained make it possible to choose one or another method for calculating the hypocenter coordinates depending on the specific area in which a seismic event occurred and the locations of seismic sensors.

EARTHQUAKE PREDICTION EVALUATION BASED ON VLF DATA USING A NOVEL INTERSECTION-UNION METHOD

Electromagnetic phenomena, especially those in the Very Low Frequency/Low Frequency (VLF/LF) bands are promising for short-term earthquake prediction. Seismo-ionospheric perturbations cause a variety of changes in different receiver-transmitter VLF/LF signal paths. Therefore, independent and simultaneous observations at different points thus in different VLF/LF signal propagation paths are necessary to better predict the earthquake. Most of the previous research on VLF data have been based on one path or limited number of paths which examined perturbations in the time domain and less attention has been paid to estimate the location of the earthquake. In the present research, using wavelet analysis, the temporal variations of seismo-ionospheric perturbations and the approximate time of earthquake are predicted. Clear disturbances are observed two weeks before the Kumamoto earthquake happened in Japan in 2016. The novelty of this study is to present an approach called Intersection-Union method to predict earthquake location. Based on the geometry of a VLF/LF network, the Intersection-Union method was introduced to estimate the earthquake epicenter. This method is based on the overlay of earthquake occurrence probable areas. With simultaneous use of different propagation paths by the Intersection-Union method, an area with a radius of about 300 km was determined as the probable location of the earthquake epicenter. The accuracy of the proposed method is 300 km compared with 1000 km accuracy of other earthquake location prediction scenarios.

Changes in subsidence and uplift and the nighttime land surface temperature anomaly related the distance to the earthquake epicenter and the faults using Sentinel and MODIS imageries

When an earthquake occurs, the faults of the region usually heat the rocks and soil of the region due to their movements. The purpose of this study was to analyze the uplift, subsidence, cloud cover and changes in nighttime land surface temperature (nLST) anomalies around faults and the earthquake epicenter in Kermanshah, Iran (date and time of earthquake 12 November, 2017 at 18:18 Coordinated Universal Time (UTC) and at 21:48 Iranian time(. Heat changes were investigated by considering the effect of other cooling factors such as vegetation (EVI), land altitude and soil moisture, rainfall and water areas. Using the MODIS sensor product, the amount of cloud cover and cooling factors were obtained. Using sentinel 1A the amount of earth uplift and subsidence were calculated. The results showed that using statistical analysis, a significant difference was observed in the nighttime land surface temperature around the faults and around the uplift and subsidence on the night of the accident, before and after night of earthquake. However, there was no significant difference between nighttime temperature and changes in the rate of spatial variation of cooling factors. It was found that the earthquake caused an increase in temperature at the fault and earthquake epicenter location. It also causes changes in height such as uplift and subsidence. Cloud cover situation showed before the earthquake, the cloud density was high and after the earthquake, the cloud density decreased. Crises managers can consider these results for monitoring metropolices for more readiness before earthquake accordance.

Determination of the Earthquake Epicenter from Line-of-Sight Displacement Images Obtained by Sentinel-1A/B Radar Data Using the GMTSAR Software Package

This article described the technology of determining earthquake epicenter with radar remote sensing on the example of Sentinel-1A/B. To determine the epicenter of the earthquake, the Earth's crust displacements were analyzed using radar remote sensing data obtained for the ascending and descending flight orbits. Coordinates of Earthquake epicenters were found according to line-of-sight displacement images via its maximum value. Displacement of the Earth's crust was obtained by processing in the GMTSAR package in the VirtualBox virtual machine of the Linux Ubuntu 16.04 operation system. Two earthquakes that occurred in 2020 were studied to determine the accuracy of finding epicenters using the ascending and descending orbits Sentinel-1A/B. These earthquakes occurred in Western Xizang, China, and Doganyol, Turkey. The maximum deviation from the officially registered epicenter coordinates was 15.38 km for Doganyol and 3.2 km for the Western Xizang earthquake. The negative displacement was 90 mm for Doganyol and 50 mm for Western Xizang.

Spatial modeling of ground subsidence susceptibility along Al-Shamal train pathway in Saudi Arabia

Al-Shamal train pathway, which is extended between Saudi Arabia and Jordan, is prone to geo-hazards due to the geological features, proximity to faults, earthquake epicenter, and the human activities along the pathway. The objectives of this study are to shed light on the ground subsidence susceptibility along Al-Shamal train pathway in Qarrayat city in Saudi Arabia and develop a ground subsidence susceptibility model to determine the prone areas to the impacts of ground subsidence to mitigate and avoid the loss of life and property. This study integrated the various data types to map the subsidence susceptibility along Al-Shamal train pathway. Nine ground subsidence causative parameters were selected as subsidence controlling factors in the study area including lithology, land cover/land use, elevation, slope, aspect, annual average rainfall, distance to faults, distance to earthquake epicenter, and distance to streams. The analytical hierarchy process is applied to obtain accurate weight to each criterion through the distribution of online Google form questionnaire to experts in different expertise and get their judgments on the weights of ground subsidence causative parameters in the study area. A subsidence susceptibility index was derived by classifying susceptible maps into five classes, namely, very low, low, moderate, high, and very high using the statistical distribution analysis. The results revealed that the study area is subjected to moderate susceptibility with about 32.56. A total of 29.8 and 11.52% of the study area had very low and low susceptibilities, respectively, and 8.44 and 17.68% had very high and high susceptibilities, respectively. The results were validated using the receiver operating characteristic using previous ground subsidence locations. The area under the curve showed 0.971, which is equivalent to 97.1%. Consequently, the findings of the study are thought to be beneficial to managers and decision makers for future planning, mitigating, and preventing subsidence in the study area.