A new probabilistic seismic hazard assessment has been performed for the Democratic Republic of Congo (DRC) and surrounding areas. The DRC encompasses both intra-plate and active tectonic areas associated with the Congo Craton and the western branch of the East African Rift System, respectively. The seismic hazard assessment is based on the new Sub-Saharan-Global Earthquake Model Sub-Saharan–Global Earthquake Model earthquake catalogue with homogeneous magnitude representation (Mw) created by augmenting available global catalogues (e.g. International Seismological Center (ISC)-Reviewed, ISC–GEM, GCMT) with information from local agencies and regional projects, particularly from the AfricaArray network. This catalogue spans from 1900 to 2015. The initial declustered catalogue has 782 events. The historical earthquake record is sparse with significant variation in completeness over time across different regions. After taking the completeness of the catalogue into account, the final declustered catalogue used to calibrate the magnitude-frequency distribution of events used for the seismic hazard assessment spans 55 years (from 1960 to 2015) with 398 events and a magnitude of completeness of about 4.5. The maximum credible magnitude of earthquakes was determined using the entire catalogue from 1900 to 2015. The seismotectonic zonation into 15 seismic source areas was done on the basis of the regional geological structure, neotectonic fault systems, basin architecture and distribution of thermal springs and earthquake epicentres. Also, consideration was given to a regional strain rate model developed for the East African Rift by Saria et al. (Journal of Geophysical Research, 119, 3584–3600, 2014) in the frame of the GEM Strain Rate Project. Tectonic information was derived mostly from the scientific literature and by integration of available datasets. The current area source model consists of a total of 15 seismic zones distributed over 6 main tectonic groups that we assume to have comparable rheological and mechanical behavior with respect to the underlying crustal geology. The definition of these groups is essential for the regional calibration of b values. The b value is expected to be regionally stable with variations less than the uncertainty limits, while the activity rate λ is liable to vary substantially from one seismic source zone to another. The Gutenberg–Richter seismic hazard parameters were determined using Aki’s maximum likelihood method Aki (Bulletin of the Earthquake Research Institute, Tokyo University, 43, 237–239, 1965) and Weichert’s method Weichert (Bulletin of the Seismological Society of America, 70, 1337–1346, 1980), and compared with extension of the Aki-Utsu b value estimator for incomplete catalogues Kijko and Smit (Bulletin of the Seismological Society of America, 102, 1183–1287, 2012). Hazard computations were performed using the OpenQuake engine (version 2.7.0-1). The peak ground acceleration (PGA) and spectral acceleration at periods of 0.05, 0.1, 0.2, 0.5, 1 and 2 s was calculated using four ground motion prediction equations (GMPEs): two for active shallow crust Chiou et al. (Earthquake Spectra, 30, 1117–1153, 2014); Akkar et al. (Bulletin of Earthquake Engineering, 12, 359–387, 2014) and two for stable continental conditions Atkinson et al. (Bulletin of the Seismological Society of America, 96, 2181–2205, 2006; Pezeshk et al. (Bulletin of the Seismological Society of America, 101, 1859–1870, 2011) for soil sites corresponding to Vs 30 = 600, 760 and 1500 m/s at 11 cities of the DRC and surrounding areas. The results are consistent with those obtained using the regional frequency-independent attenuation law of Mavonga (Physics of the Earth and Planetary Interiors, 62, 13–21, 2007b) developed in the Kivu Rift segment and others Twesigomwe (Journal of African Earth Sciences, 24, 183–195, 1997), and Jonathan (Some aspects of seismicity in Zimbabwe and Eastern and Southern Africa. M.Sc. thesis. Institute of solid earth physics, Univ. Bergen, Bergen, Norway, pp. 100 (1996)), using Crisis 2012 software.
Read full abstract