Earthquake Forecasting Using Artificial Neural Networks

Abstract

Earthquake is one of the most devastating natural calamities that takes thousands of lives and leaves millions more homeless and deprives them of the basic necessities. Earthquake forecasting can minimize the death count and economic loss encountered by the affected region to a great extent. This study presents an earthquake forecasting system by using Artificial Neural Networks (ANN). Two different techniques are used with the first focusing on the accuracy evaluation of multilayer perceptron using different inputs and different set of hyper-parameters. The limitation of earthquake data in the first experiment led us to explore another technique, known as nowcasting of earthquakes. The nowcasting technique determines the current progression of earthquake cycle of higher magnitude earthquakes by taking into account the number of smaller earthquake events in the same region. To implement the nowcasting method, a Long Short Term Memory (LSTM) neural network architecture is considered because such networks are one of the most recent and promising developments in the time-series analysis. Results of different experiments are discussed along with their consequences. * Corresponding author network (Lakshmi and Tiwari, 2006). For this particular task, there are so many factors involved in the process that other model based approaches cannot accommodate as accurately as neural network does (Perol et al., 2017).This contribution has been peer-reviewed. https://doi.org/10.5194/isprs-archives-XLII-5-823-2018 | © Authors 2018. CC BY 4.0 License.

Assessment on maximum magnitude of natural and triggered earthquake when water is impounded in the mining pit: A case study in Nui Nho quarry, Vietnam based on gravity and magnetic data

“…The increase of pore pressure in rock increases the infiltration, thus reducing the friction of existing conformities or failure surfaces. As a result, it decreases the firmness and changes the elastic coefficient of rock, causes the fluctuation of preceding tension force, leading to the displacement of rock mass towards the tectonic failure surface [Gupta, 2002[Gupta, , 2011Gupta et al, 1972;Hagan et al, 1996;Hojjat and Panakkat, 2009;Kalpna and Chander, 2000;Kalpna and Gupta, 2008;Kalpna et al, 2016].…”

“…The feedforward neural network with backpropagation algorithm was applied to calculate the maximum magnitude of natural earthquake [Bhatia, 2018;Hagan et al, 1996;Hojjat and Panakkat, 2009;Pupkov et al, 2012;Trong et al, 2016]. Input data for calculation include: 1) Gradient value of elevation; 2) Gradient value of Bouguer gravity field (Vietnam Geophysical Divison, 2010); 3) Gradient of aeromagnetic anomaly (Vietnam Geophysical Divison, 1995); 4) Gradient value of isostatics [Trieu, 2010b]; 5) Gradient of sedimentary crust thickness [Hung, 2009]; 6) Gradient of crystalline basement depth [Hung, 2009]; and 7) Gradient of the Earth's crust thickness [Hung, 2009].…”

“…High-precision gravity method was chosen to determine earthquake source parameters in Nui Nho quarry, based on the opinions that rocks are strongly fractured when water penetrates the ground and changes pore pressure, which is the cause of earthquake [Gupta, 2002[Gupta, , 2011Gupta et al, 1972;Kalpna and Chander, 2000;Kalpna and Gupta, 2008;Kalpna et al, 2016]. Meanwhile, the strongly fractured zone is easily recognized through low value of normalized full gradient ( , ) compared to monolithic environment [Trong et al, 2016].…”

“…According to Bell and Nur [1978], the change of Coulomb stress(Δ ) caused by water impoundment in the reservoir is determined as follows: Δ = Δ − (Δ − Δ ), where Δ and Δ correspond to the changes of tangential stress and normal stress which are caused by the reservoir loading on the fault surface, Δ is the change of pore pressure, and is the coefficient of friction. The increase of Δ and the decrease of Δ mean that Δ has a positive value, which will stimulate the fault activity and vice versa.…”

“…Scientists have found that artificial lakes with considerable depths (e.g. over 90 m) can possibly cause earthquakes [Gupta, 2002[Gupta, , 2011Gupta et al, 1972;Kalpna and Chander, 2000;Kalpna and Gupta, 2008;Kalpna et al, 2016]. The weight of water mass buckles the basement rock of lake bottom, causing subsidence (the subsidence of lake bottom can be up to 1.5 ÷ 2 m), which is the rea-…”