Forecasting seismicity rate in the north-west Himalaya using rate and state dependent friction law

Abstract

In this study, rate and state Coulomb stress transfer model is adopted to forecast the seismicity rate of earthquakes (M W 5) in the north-west Himalaya region within the testing period 2011-2013. Coulomb stress changes (DCFF), considered to be the most critical parameter in the model, exhibit stress increase in the whole study region, excluding the Chaman fault of the Kirthar range where significant stress shadow has been observed. The estimated background seismicity rate varies in the range 0.0-0.7 in the region, which is preoccupied by low aftershock duration of <50 years. Furthermore, a low b-value that varies between 0.54 and 0.83 is observed in Kirthar ranges, Karakoram fault and Pamir-Hindukush region. However, areas like Hazara syntaxis of the northern Pakistan and northern Pamir of the Eurasian plate exhibit higher b-values in the range 1.23-1.74. Considering constant constitutive properties of the faults (i.e. As = 0.05 MPa), our forecast model for variable DCFF and heterogeneous b-value successfully captures the observed seismicity rate of earthquakes. Results have been verified using statistical S-test. However, the model fails to capture the observed seismicity rate during the period when reconstructed for average b-value to be 0.86 and no change in DCFF (DCFF = 0).

Forecast of Earthquake Magnitude for North-West (NW) Indian Region Using Machine Learning Techniques

“…α 𝑖 and α 𝑖 * are the Lagrange multipliers obtained through solving the dual form of the optimization problem. The Earthquake dataset is partitioned into five distinct datasets, corresponding to the proposed zones outlined by Chingtham et al (2017). For each zone, a Support Vector Regression (the Support Vector Regressor) model is constructed.…”