Determination of point and extended source parameters of 5 April 2017 Sefid-Sang earthquake (Ml 6.0) in time and frequency domains using KIWI tools

KIWI (KInematic Waveform Inversion) is a recently developed multi-step inversion tools at the Institute of Geophysics of University of Hamburg. The main aim of developing this method is to perform moment tensor inversion retrieving the point and extended source parameters in regional distances. In KIWI tools, point and kinematic source parameters are retrieved in a sequential process in three inversion steps in time and frequency domains using different inversion methods, parts of waveforms and so on. After the point source inversion done, the method retrieves the radiation pattern, including fault plain parameters, Scalar moment and centroid depth. Also, for large enough earthquakes (Mw>5.5), extended source inversion retrieves finite source parameters such as rupture directivity, rupture area and velocity, rise and rupture time, average slip and nucleation point regarding to the point source centroid location. KIWI tools uses pre-calculated Greens functions, hence, the inversion process is quite fast. Due to the same reason, this method is rendered for automatic real-time retrieval of point and extended source parameters. In general, we can highlight the most important characteristics and applications of KIWI tools as follows: ability of easy implementation for real-time retrieval of source parameters, stability of inversion, rapid directivity detection, no requirements of aftershocks and foreshocks, no limitation in depth and magnitude and ability of retrieving reliable results even in absence of accurate velocity model used to build the Green’s functions and large stations azimuthal gap. In this research, we introduce the KIWI tools and use its applications to study of the April 5, 2017 (Ml 6.0) Sefidsang-Fariman earthquake. The data used in this research were recorded by permanent broadband stations of International Institute of Earthquake Engineering and Seismology (IIEES) and some global broadband stations from IRIS network at a minimum epicenteral distance of 200 kilometers. To have a better evaluation of KIWI tools functionality, we made inversion of source parameters using six different set of information (including IASP91 and IRSC velocity models and the mentioned set of data). Then, the information set including IRSC velocity model and all available data considered as the optimum one. Comparing the obtained results using the optimum set of information and the remaining sets, Maximum difference in centroid depth, Latitude and Longitude is 1.9 kilometer, 0.23 and 0.5 degree related to information sets including only IRIS network data, while there is a good consistency in retrieved focal mechanisms. After all, it is tried to run a sensitivity test using the optimum information set to have a better assessment on KIWI tools stability in source parameters analysis. Based on the achieved results, the erroneous input parameters (e.g. Latitude, Longitude and Depth) had a low influence on our optimum results. The final results in this research represents the centroid of earthquake in a shallow depth (7.1 km) with a magnitude slightly larger than those published by other institutions like USGS (Mw 6.2). Retrieved focal mechanism shows mainly reverse faulting with small dextral strike-slip component dipping north-east which is in a good accordance with the Kashafrood fault characteristics as the closest active fault to the epicenter. Also, extended source inversion revealed mostly unilateral source directivity toward SE with a rupture area, rupture time and approximate average sleep of 300 km², 9.3 seconds and 16 cm.