The Use of Exponential Distribution Model to Estimate Recurrence Periods of Earthquakes in Zimbabwe

Abstract

This study estimated recurrence periods of earthquakes in Zimbabwe using exponential distribution model. The data for this study were extracted from a catalogue, the Advanced National Seismic System (ANSS) owned by Northern California Earthquake Data Centre UC Berkeley, USA. The selected data consisted of earthquake events with magnitude Mb  4.0 for the study area from 1st January, 1901 to 31st December, 2001 (100 years) with focal depth from 0 – 700 km within latitudes 15°S – 22°S and longitudes 25°E – 34°E. A total number of 81 events were employed in the study. The formulated hypothesis was tested using Chi square test and Independent t-test. The findings of this study revealed that experimental distribution of earthquakes have no significant difference with theoretical distribution of earthquakes in Zimbabwe. This implies that Zimbabwe earthquake data follow exponential distribution. The return periods for magnitude 4.2 and 6.2 were estimated to be 4.00 and 47.48 years respectively. It has been observed that as the magnitude increases towards higher magnitude, the return period increases except at magnitude 4.7 where it decreased. Therefore, the occurrence of minor to light earthquakes is more frequent than stronger ones: therefore, the probability of occurrence of earthquakes of low magnitude (up to 4.0) is higher than for earthquakes with magnitude of 6.0 and above. As a result, Zimbabwe may not likely experience any serious earthquake (magnitude 6.0 or greater) until the year 2048, considering that the last 6.0 magnitude event - with an estimated return period of 47.48 years - occurred in 2001. Nevertheless, earthquake occurrence cannot be predicted with certainty yet: earthquakes are in fact naturally unpredictable, due to sensitivity of catalogues to small events, saturation of magnitudes and differences in data collection by seismic stations and networks.