Seismi-tectonic studies in the whole SISZ (work carried out with Gudmundsson and Rögnvaldsson)

$\\$After the first study on Vördufell, the field studies have been carried out in September 1996 in the whole SISZ in order to collect fault slip data measurements and reconstruct paleostress tensor. The collection has been made in some selected sites in late Tertiary and Pleistocene lavas and hyaloclastites and in postglacial lavas. The areas investigated were the Skardsfjall, Hestfjall and Búrfell mountains, the Grímsnes area and the canyons of Stóra Laxá and Stóra Mástunga. This study has been completed by analysis of focal mechanisms of earthquakes in the same areas. We collected about 700 brittle tectonic data at 25 sites in Upper Pliocene to Holocene basalts and hyaloclastites. At each site the whole dataset reveals two opposite stress regimes, each including normal and strike-slip faulting. The analysis of the strike-slip faults in the major data subset indicates paleostress tensor with subhorizontal $\sigma _1$ and $\sigma _3$ axes, the trend of $\sigma _3$ being N E to N E (main trend: N E). The normal faults in the major subset characterize vertical $\sigma _1$ and horizontal $\sigma _3$, also trending N E to N E. In the minor data subsets, the direction of $\sigma _3$, for the strike-slip faults as well as for the normal ones, ranges from N E to N E (N E on average). The major stress regime includes 70% of the total population of faults. Of a total of 718 fault slip data, 55% indicate primarily normal-slip and 45% primarily strike-slip. The ratio normal/strike-slip faults is lower than in other areas of Iceland. Moreover, many normal faults were generated in the western rift zone segment, then drifted out of it. The above results indicate that the dominating stress field in the SISZ favours strike-slip faulting, with an horizontal $\sigma _3$ axis trending approximately WNW-ESE to NW-SE. We point out, however, that in addition, there is a contrasting minor stress field, characterized by approximately NNE-SSW to NE-SW extension. These results were compared to the stress regimes determined from earthquakes mechanisms. The determination of the stress regimes is done through calculation of stress tensors, involving the use of inverse methods. Preliminary analyses of focal mechanisms recorded by the SIL network revealed the presence of two contrasting stress regimes. The crucial problem of the choice between nodal planes of double couple focal mechanisms is solved in two ways. First, some methods do not require the choice between the nodal planes, such as for the right dihedra (P- and T- dihedra) method. Second, for those inversions requiring choices between nodal planes, and because the earthquakes studied are very shallow, the rich geological information on the fractured medium deserves attention. We use an integrated approach including four criteria; geophysical and geological: consideration of nodal plane attitudes as compared with geological discontinuities, comparison with recent fault mechanisms observed in the field, consideration of mechanical likelihood for each fault solution, and best fit criterion relative to the stress tensors calculated. The systematic use of these criteria for nodal plane selection within a weighted approach (reliability, magnitude, depth, etc.) allowed better determination of the stress regimes in the SISZ. As for faults collected in the field, major NW-SE extension is in agreement with the left-lateral behaviour of the E-W seismic zone. A minor NE-SW extension is attributed to rebound phenomena, local block accommodation and magmatic effects. For both these regimes, strike-slip mechanisms prevail, many normal and few reverse mechanisms being also present. Permutations between stress axes, $\sigma_2/\sigma_3$ and $\sigma_1/\sigma_2$, are common.