MICROEARTHQUAKES-RECORDING AND STUDY OF THEIR PARAMETERS FOR SELECTED REGIONS

Abstract

Microearthquake investigations involve collection and study of data for small magnitude events (MO). The rate of occurrence of small magnitude events being high as compared to that of moderate and big ones, their recording allows to extend the da,a base and lorn, ihe bas.s o. these slud.es. Permanent and portable local networks are employed to obtain data on the occurrence of small magnitude events. The ground motion due to microearthquakes .s feeble, rich in high frequencies and requires special instrumentation to monitor thorn. The important applications of those studies arc, the evaluation of local contemporary seismic activity, delineation of active faults, determination ol foca, mechamsm and stress drop parameters, evaluation ol mduced seismicty, mapping of geothermal sources, allow earthquake prediction research, seism.c hazard estimation of the critical Utilities (e.g., nuclear power plants, dam sites), mapping of aftershocks and estimat.on of velocity and Qstructure of the crust and upper mantle. This thesis is devoted to the study of local seismic activity of the regions surrounding three dam sites, employing small aperture portable local networks. Two of these regions fall in the Lesser Himalaya and the Sub Himalaya. The third region is located in the Peninsular India. The studies were earned ou, with the following objective* (i) To study the pattern of contemporary seismic activity of these sites and its comparison with historic seismicity. (ii) To identify similarities and differences in the pattern of seismic activity of the three sites-viz., two in the Himalaya and one in the Peninsular India. (UO To evaluate the potential of induced seismicity and suggest planning of seismological networks for future investigations. To achieve these objectives data on the occurrence of microearthauakes was collected from the period 1980 to 1983 employing smoked paper seismographs (MEQ-800) and L-*C or S-7000 seismometers. Atotal of fifteen months of data which consisted (ii) (iii) of six months at Chamera, six months at Navagam and three months at Jamrani dam sites were collected. These data sets form the basis of the study reported here. The locations of events have been obtained employing P-times. The (5-P) time has been used for obtaining origin time restricted locations. The earthquake location problem is nonlinear, however, the existing programs make use of linearisation employing Taylor series. In this study a technique to obtain locations by incorporating the second order partial derivatives as well, for a multilayer one dimensional earth model has been developed. The expressions for second order derivatives were incorporated in an existing linear least square program (LLS) which is modified version of the HYPOLAYR program (Eaton, 1969). This new program (NLLS) has been tested employing test earthquake data and synthetic data, and has been employed to locate events recorded in this study. The results have also been compared with the linear least square program (LLS). The events converge to same location employing NLLS and LLS programs. The convergence history employing LLS and NLLS methods remains by and large same for events within or close to the array. For events outside the array improvements in the convergence employing NLLS program have been obtained. For events located around Chamera dam site, NLLS method gave focal depths for about 14% more events as compa red to LLS method. Employing NLLS method more realistic assessment of uncertainty in the hypocenter solution is possible as there is a reduction in standard errors and length of the axes of ellipse governing the joint confidence region. In the region around Chamera dam site, lying in the Lesser Himalaya between the Main Boundary Thrust (MBT) and the Main Central Thrust (MCT), more than 1300 microearthquakes were recorded (Mn from - 2.8 to 3.4). Epicenters of 216 events scatter along the regional tectonic features. However, no events are located in the immediate vicinity of dam site. The activity is almost confined to the north of the MBT except for a cluster of epicenters at about 10 km northeast of dam site. This cluster falls to (iv) the northeast of the Ravi re-entrant structure. The distribution of epicenters is broadly in conformity to the historic seismicity. The estimated 'b' value of 0.46 from microearthquake data is in agreement with the 'b' value of 0.52 obtained from historic seismicity data. Generally, the focal depths of located events range between 10 to 15 km. The depth section across the strike direction of thrust sheets show the dipping zone with dip of about 22° towards northeast. Composite fault plane solutions (CFPS) for three sets of events indicate that the tectonic strain in the region is being released by both thrust as well as normal mode of faulting. The Jamrani dam site is located in the Sub Himalayan tectonic zone. The Krol Thrust which is homologous of the MBT, reported to be active from geological evidences, is exposed 0.5 km to the north of the dam site. The moderate magnitude earthquakes in this region are mostly confined to the Lesser Himalaya between the MBT and the MCT. No activity with (S-P) time ^7 sec is recorded. 384 events (Mn from 0 to 3.4) with (S-P) times mostly between 10 to 25 sec falling outside the array, have been recor ded. Epicenters for 188 events and focal depths for only 3 events could be obtained. Most of these events cluster about 120 km northeast of the dam site in the known region of seismicity in the vicinity of MCT. No events were located in the vicinity of the MBT. Other studies also indicate infrequent seismic activity along this section of the MBT (Ni and Barazangi, 1984). Nine events are located southeast of the array in the Ganga foredeep. This intraplate activity may have significance because these locations fall in the close vicinity of moderate earthquake with normal faulting and nodal plane parallel to the Himalayan arc. The focal depths obtained for three events agree with the well determined earlier reported focal depths of moderate earthquakes for this section of the Himalaya. Thus, the pattern of microearthquake activity is found to be in agreement with that of historic seismicity. The region of Navagam dam site is located in the Shield region of Peninsular India and has experienced infrequent and sporadic small to moderate earthquakes. 258 micro( v) earthquakes with magnitude (Mn from - 2.4 to 3) mostly below one were recorded. Epicen ters of 106 well located events fall around the region of dam site. The activity is bound to the south by the Piplod fault. Small clusters of activity about 8 and 10 km northeast and north northwest of the dam site are also mapped. The overall spread of activity follows the Narmada rift zone. The focal depths range between 2 to 22 km with majority of events confined between 10 to 15 km which could be interpreted as the depth where maximum stresses are prevailing in the rift zone. Depth sections indicate, in general, that hypocenters located close to the Narmada, Piplod and Surpan faults are having nearly vertical trends which conform to the rift valley environment. CFPS for two sets of events indicate normal faulting with pressure axes nearly vertical or plunging 55° due E64°S. This pattern of faulting is also attesting to the rift structure but does not agree with the reported fault plane solutions of moderate earthquakes for the region of Peninsu lar India. From microearthquake data estimated value of 'b' is 1.19. The microearthquake studies around three dam sites have enhanced the available knowledge and allowed improved understanding of current seismic activity. Both Chamera and Jamrani dam sites which are located in similar regional tectonic setting are exhibiting very different contemporary seismic activity. The ambient strain generating process in environ of Himalaya is the convergence of Indian and the Eurasian plates. Therefore, differences of local seismicity in region along the Himalaya such as Chamera and Jamrani will most likely be due to local variations of this strain field. The high level of observed seismic activity around Chamera dam site may be due to the presence of the Ravi re-ent rant structure to the southwest of the dam site where higher strain accumulation could take place locally. Hie Chamera site is located 50 km to the west of the epicenter of the great Kangra earthquake of 1905, where the plate has already ruptured in this section of Himalaya. Since the dam site falls in the rupture zone of this earthquake, the observed activity in this area roulcl be influenced by the occurrence of this earthquake which is expected to have caused concentration of strains at the edge of this rupture zone. (vi) The lack of seismic activity in the region around 3amrani may be due to various factors such as, the region may be undergoing creep, the low level of strains in the region and the dilatancy hardening. To identify specific factor for this observation neither adequate data base nor additional evidence is available. However, it is to be noted that 3amrani darn site lies between the rupture zones of two great earthquakes. Navagam site is located in the Narmada rift zone and microearthquake survey has brought out that this part of the rift zone is infested with small magnitude events. This region appears to be characterised by low to moderate strain built up as compared to the two other sites as interpreted from the comparison of 'b' values. Small magnitude induced seismicity is reported from dams (e.g., Tarbella dam in northern Pakistan) in the Himalaya, whereas in Peninsular India, Koyna and Bhatsa dams have shown induced seismicity. Apart from other considerations on the basis of CFPS and direction of pressure axes the potential of induced seismicity appears to be more at Navagam dam site amongst the three dam sites. The recording has provided very useful short time preimpounding sample of seismicity and with suitable future recording the phenomena of induced seismi city may be better understood around these regions.