TEMPORAL AND SPATIAL B-VALUE ANALYSIS USING PYTHON PROGRAMMING LANGUAGE IN "UC" FIELD UNDERGROUND MINING

• Main Author: Husein Shihab, Nabiel
• Format: Final Project
• Language: Indonesia
• Subjects: Geologi, hidrologi & meteorologi
• Online Access: https://digilib.itb.ac.id/gdl/view/40316
• Institution: Institut Teknologi Bandung

Mining activities can change the subsurface stress condition so that they can eventually induce microseismic which might cause a significant earthquake. Analysis of variation of b-value in space and time can be used as a guide for a seismic risk reduction. Before we calculate the b-value we must estimate the magnitude of completeness. Both magnitude of completeness and b-value vary spatially and time, so it is necessary to obtain both parameters automatically. In this study, author develop a program to estimate magnitude of completeness (Mc) and to calculate the variation of b-value in space and time automatically. The magnitude of completeness is estimated using maximum curvature, goodness of fit test, and Mc by b-value stability techniques. The b-value is calculated using Linear Model with Gaussian residual method and also Generalized Linear model with Poisson and Gaussian residual methods. The uncertainty of the calculated magnitude of completeness and b-value for each distribution are also presented. The program was created in a python programming language based on the algorithm given by Naylor et al. (2009) for calculating b-value and Mignan and Woessner (2012) for estimating magnitude of completeness. The sensitivity and robustness of the program were tested using earthquake catalogs from various fields. There are no significance differences between author’s program and author’s program references. Variation of b-value in time has not shown to be useful as a precursor of large events. Variation of b-value in space shows that the low b-value zone correlated with a significance magnitude events. There is a proportional relationship between b-value and number of event at number of event more than 2000 events.