_{1}

^{*}

Earthquakes are more deadly than any other form of natural hazard. Because of this, scientists have been searching for accurate ways to predict earthquakes so that lives can be saved. Statistical analysis is useful methods of predicting earthquakes. They provide additional insights to the seismic hazard. In this work, the statistical analysis has been described through the study of “Region of Bangladesh”. A sample result from the statistical analysis which gives intermediate term prediction of earthquakes is given. At the end of the study, the prediction of near future earth-quakes is also investigated through the pictorial and tabular behaviors.

On March, 2011, an earthquake struck the town of Fokoshima, Japan. The earthquake lasted only few seconds, but many people were killed [

An earthquake is the result of a sudden release of energy in the Earth’s crust that creates seismic waves. The seismicity or seismic activity of an area refers to the frequency, type and size of earthquakes experienced over a period of time. Earthquakes are measured by Richter magnitude (ML) scale and moment magnitude (MW) scale. Earthquakes are classified by the depth at which they occur. There are two types of earthquakes: shallow and deep. Shallow earthquakes, which compose the majority of earthquakes, occur at depths down to 300 kilometers. Deep earthquakes occur at depths from 300 to 680 kilometers. Shallow earthquakes produce more damage. In addition, the mechanism that produces shallow earthquakes is known, while the mechanism that produces deep earthquakes is not fully understood. The information in this study refers only to shallow earthquakes, because they are more common, more destructive, and better understood.

Statistical analysis is an essential part of the scientific method, and it is especially important in earthquake prediction because public safety, public funds, and public belief are involved. A number of candidate statistical distributions have been proposed for computation of conditional probabilistic of future earthquakes, including the Laplace or double exponential (Utsu [

The purpose of this paper is to review earthquake zones in Bangladesh, and methods of predicting earthquakes. The statistical analysis of earthquakes has been discussed here of predicting earthquakes in Bangladesh. In order to demonstrate the statistical analysis, a case study will be reviewed: region in Bangladesh. An earthquake prediction states the probability of occurrence, time span, region, and magnitude range of the earthquake.

This work is divided into three main sections. In Section 2, we discussed the earthquake zones in Bangladesh according to seismic zones. An understanding of this section will help to understand how we could be used to predict an earthquake for particular regions in Bangladesh. The statistical analysis of earthquakes instructs in Section 3. Finally we specify discussion in Section 4.

According to geology and tectonics of Bangladesh and neighborhood five tectonics blocks can be identified which have been active in producing damaging earthquakes. These are 1) Bogra fault Zone 2) Tripura falt zone 3) Assam fault zone 4) Shillong plateau and 5) Sub Dauki fault zone [

The maximum magnitudes of earthquake that can be produced in different tectonic blocks are given in

Bangladesh and north-eastern India states are seismically very active and earthquake risks are increasingly seen as a major problem. During 1918-2007, twelve earthquakes with magnitude M ≥ 7 have affected parts of Bangladesh out of which two had their epicenters inside Bangladesh given in

Many methods for earthquake prediction have been suggested and some of these methods may be reliable. Statistical analysis is one of interesting method for predicting earthquake. Statistical analysis is when you look at the history of earthquakes in a given region and see if there is a recurrent or cyclical pattern of the earthquakes. If earthquakes in a given region have a recurrent pattern, then a long-term prediction can be made based on the recurrent pattern, see in

There were twenty nine earthquakes (ML 6 and above) strong enough to damage houses during 1918 to 2007.

. ML level in different tectonic block.

S.N. | Tectonic block | Maximum magnitude of earthquake |
---|---|---|

1 | Bogra fault Zone | 7.00 |

2 | Tripura fault zone | 7.00 |

3 | Assam fault zone | 8.50 |

4 | Shillong plateau | 7.00 |

5 | Sub Dauki fault zone | 7.00 |

Different magnitude levels of Earthquake zones in Bangladesh (Source: Banglapedia)

Three of them (ML 8.5 and above) were powerful enough to cause serious devastation. Even though it may be hard to gauge the extent of damage caused by the earthquakes, the following list is reliable 1918, 1923, 1927, 1930, 1932, 1933, 1934 (severe), 1935, 1936, 1938, 1940, 1941, 1943, 1954, 1955, 1956, 1957, 1958, 1959, 1964, 1970, 1975, 1980, 1993, 1994, 1995, 1997 (severe), 2004, 2007 (severe). Using statistical analysis of earthquake, the average interval among these twenty nine earthquakes is 3.179 years. However the time between 2007 and the last previous earthquake were 3 years. After doing statistical calculation, which assumes the random recurrences of earthquakes and we found that by 2007, there was a 62% probability of an earthquake that could damage structure. If only three severe earthquakes are considered, the average interval is 24.33 years. But by 2007, 10 years had elapsed since the last severe earthquake, the one in 1997.

Earthquake magnitude M ≥ 7 in Bangladesh

. Major earthquake in Bangladesh the last 100 years including date, magnitude, and epicenter.

Date | Name of earthquake | Magnitude | Epicenter |
---|---|---|---|

8 July, 1918 | Srimangal earthquake | 7.3 | Bangladesh-Tripura border |

9 September, 1923 | Meghalaya earthquake | 7.1 | Bangladesh-India border (Meghalaya) |

2 September, 1930 | Dubri earthquake | 7.1 | Dabigiri |

6 March, 1933 | India Bangladesh earthquake | 7.6 | India Bangladesh border |

15 January, 1934 | Bihar Nepal earthquake | 8.3 | Bihar-Nepal border |

11 February, 1936 | Bihar earthquake | 7.5 | North Bihar |

16 August, 1938 | Manipur Earthquake | 7.2 | Monipur near of Bangladesh |

23 October, 1943 | Assam earthquake | 7.2 | Hojai Assam |

21 March, 1954 | Monipur-Maynmar earthquake | 7.4 | Monipur-Maynmar border |

21 November, 1997 | Bandarban earthquake | 7.1 | Mizoram-Maynmar border |

26 December, 2004 | Cox’s Bazar earthquake | 7.0 | Bonda Aceh, Indonesia |

12 September, 2007 | Tsunami due earthquake (Cox’s Bazar) | 8.5 | Bengkula, Sumatra |

Again according to Vevsel Yilmaz et al. [

and the cumulative distribution and the reliability function are

and the mean of the two parameter Weibull distribution is

Equations (2), (3), α and β are fitting parameters of the distribution. Yilmuz [

Putting these values in Equation (1) we get

and earthquake data are well modeled by Weibull distribution in Equation (4). The mean occurrence period of earthquakes in Bangladesh which have magnitude equal to 6 or above is calculated approximately 2.95 years. So it is expected to have another earthquake having magnitude equal to 6 or above in Bangladesh is average 2.95 years later than the preceding one. Cumulative distribution function graph that shows occurrence risk of an earthquake having magnitude equal to 6 or more in t years after an earthquake occurred in region of Bangladesh with magnitude equal to or above 6 is given in

The graph of reliability function that gives the probability of having another earthquake with magnitude 6 or above in t years after the preceding one with the same magnitude in Bangladesh is given in

An unbiased statistical method of an earthquake prediction must involve adequate information that incorporates well known feature of earthquake occurrence. Statistical methods are useful for characterizing seismic hazard

Trend in number of earthquake each year (1918-2007)

The graph of the cumulative function F(t)

. Parameter estimation values and mean occurrence period.

. Parameter estimation values and mean occurrence period. | . Parameter estimation values and mean occurrence period. | . Parameter estimation values and mean occurrence period. |
---|---|---|

1.249 | 3.1742 | 2.95 |

The graph of reliability function R(t)

because earthquakes are, for all practical purposes, random phenomena. Seismic risk and earthquake occurrence probabilities can be estimated by using statistical distribution. In this article, I have examined the statistics of great future earthquake occurrence on the Bangladesh region by using statistical analyses.

The author thanks the editor and anonymous referee for thoughtful comments. The paper have been substantially improved due to their constructive feedbacks. Also grateful thanks are due to Tanjina for his help in data management.