S U M M A R YWe relocate precisely micro-earthquakes induced by the Açu reservoir in Brazil and observe seismicity migration consistent with pore-pressure diffusion on a single fault zone. Fluids are believed to play a major role in triggering tectonic earthquakes; reservoir induced seismicity provides a natural laboratory in which to investigate the spatio-temporal evolution and triggering of earthquakes caused by fluid and pore-pressure diffusion.Between 1994 and 1997, 267 earthquakes (M L ≤ 2.1) were recorded and located beneath the Açu reservoir. The seismicity increased several months following annual water level peaks, implying that pore-pressure diffusion is the principal triggering mechanism. The small station spacing and very low-attenuation, Precambrian basement, rock enabled starting earthquake locations with uncertainties of only a few hundred metres. We relocate 155 earthquakes from the largest cluster at Açu using waveform cross-correlation to obtain groups of similar events. We use these groups to improve the pick accuracy (to subsample accuracy in 200 sample per second data), and then invert for more accurate hypocentral locations. Our uncertainties are on the order of 10 m, and our locations are more tightly clustered. We observe temporal migration of the earthquakes, both along strike, and to increasing depth. We observe a seismicity migration rate between 15 and 58 m d -1 . The rate is highest during the time of peak seismicity rate, and there is some suggestion that the rate decreases with increasing depth. Peak depth in seismicity is reached 175 d after the water peak, that is 192 d after the water low, and the maximum depth then decreases at a similar rate to the rate of increase. Our observations are consistent with triggering by pore-pressure diffusion within a heterogeneous fault zone with an average hydraulic diffusivity of ∼0.06 m 2 s -1 and fracture permeability of ∼6 × 10 −16 m 2 .

S U M M A R YWe calculate stress drop and rupture speed for M L ≤ 2.1 shallow reservoir induced earthquakes and find them to be similar to those of large, natural earthquakes. Previous studies have suggested that hydrofractures, mining and reservoir-induced earthquakes have lower average stress drop than natural tectonic earthquakes. This difference might result from the different tectonic setting or the shallower hypocentral depths of induced earthquakes. Alternatively, difficulties in correcting for attenuation and site effects in earlier studies may lead to underestimation of stress drop. In addition, most studies assume the rupture velocity of small reservoir induced earthquakes to be the same as for the large earthquakes. We analyse a set of 101 M L ≤ 2.1 earthquakes induced by changing water level in the Açu Reservoir, NE Brazil. The earthquakes are shallow, (depth <5 km) and the region has negligible natural seismicity. We use three different approaches to calculate the source parameters of the six largest (1.9 ≤ M L ≤ 2.1) earthquakes. We model the individual spectra to find corner frequency, frequency-independent Q, and long period amplitude. We use collocated small earthquakes as empirical Green's functions to calculate the spectral ratios, and determine the relative source time functions. Estimates of the source duration and corner frequency imply stress drops in the range of 26-179 MPa. These are similar to, or higher than tectonic earthquakes suggesting that the shallow hypocentral depth and the presence of water do not affect stress drop. We observe clear directivity for one of the earthquakes, and use the azimuthal variation in pulse width to estimate a rupture velocity of ≥0.6β.

S U M M A R YWe present the spatio-temporal evolution of seismicity recorded by eight three-component digital seismographs in operation continuously during a 3 yr period (1994 August to 1997 at Açu reservoir, NE Brazil. The Açu dam is a 34 m high earth-filled dam constructed in 1983 May on an area of Precambrian shield. Based on seismic monitoring between 1987 and 1989 using single-component analogue seismographs, previous workers concluded that the seismic activity was a case of reservoir-induced seismicity (RIS) associated with diffusion of pore fluid pressure beneath the reservoir. The digital data presented here reveal the seismic activity in remarkable detail with vertical and horizontal location errors ≈0.1 km. A total of 286 events were recorded by three or more stations and all occurred at a depth of <5 km. Using these data we demonstrate that the majority of the earthquake activity is clustered within several well-defined zones and that individual zones are active over discrete periods of time. Over the entire period of seismic monitoring between 1987 and 1997 there is no simple correlation between reservoir level and number of seismic events. Lateral migration of the locus of seismic activity in an unpredictable fashion is shown to be partly responsible for the poor correlation, as event detection is not uniform through time. We also show that the time delay between maximum water level and a subsequent increase in seismic activity varies systematically; longer time delays correspond to activation of an earthquake cluster with a greater average hypocentral depth. However, within any one cluster there is no correlation between time delay and depth. The 3-D distribution of seismic activity through time may only be explained in terms of triggering by the diffusion of pore fluid pressure if the rock properties (e.g. permeability, strength) are heterogeneous.

This version is available at https://strathprints.strath.ac.uk/43272/ Strathprints is designed to allow users to access the research output of the University of Strathclyde. Unless otherwise explicitly stated on the manuscript, Copyright © and Moral Rights for the papers on this site are retained by the individual authors and/or other copyright owners. Please check the manuscript for details of any other licences that may have been applied. You may not engage in further distribution of the material for any profitmaking activities or any commercial gain. You may freely distribute both the url (https://strathprints.strath.ac.uk/) and the content of this paper for research or private study, educational, or not-for-profit purposes without prior permission or charge.Any correspondence concerning this service should be sent to the Strathprints administrator: strathprints@strath.ac.ukThe Strathprints institutional repository (https://strathprints.strath.ac.uk) is a digital archive of University of Strathclyde research outputs. It has been developed to disseminate open access research outputs, expose data about those outputs, and enable the management and persistent access to Strathclyde's intellectual output.Journal of the Geological Society, London, Vol. 170, 2013, pp. 237 -247. doi: 10.1144 Rifting in intra-continental areas will commonly interact with a framework of pre-existing deformed zones. Although new structures form, structural inheritance may influence the location and geometry of tectonic deformation because strained rocks can act as weak zones localizing subsequent deformation (e.g. Watterson 1975;Sykes 1978;Handy 1989;Holdsworth et al. 2001). In particular, basins along rift margins are often inferred to be influenced by pre-existing structures (e.g. Daly et al. 1989;Lee & Hwang 1993;van Wees & Beekman 2000;Scheck-Wenderoth & Lamarche 2005). The connection between basins and pre-existing structures is frequently based on the similarity of the trends of seismically imaged rift faults to basement tectonic trends observed onshore (e.g. Roberts & Holdsworth 1999;Wilson et al. 2006). However, assessing the direct relationship between rift faults and pre-existing structures is often difficult because the basement is overlain by syn-to post-rift formations. The effect of pre-existing structures on fault architecture at scales of tens to hundreds of metres is less clear. Some studies suggest that pre-existing foliations resulting from crystal-plastic deformation exert a strong influence on fault zone architecture during later deformation (e.g. Beacom et al. 2001;Butler et al. 2008). Experimental work indicates that foliated rocks are mechanically anisotropic (Donath 1964;Shea & Kronenberg 1993), suggesting that foliations impart mechanical anisotropy or planes of weakness into the country rock that may be exploited by subsequent fracturing. However, analogues for the deeper sections of rift faults are rarely exposed, and the specific impacts of mechanical anisotropy for fault structures at exposure scale are con...