Analysis of the induced seismicity of the Lacq gas field (Southwestern France) and model of deformation

Abstract

S U M M A R YThe goal of this paper is to propose a model of deformation pattern for the Lacq gas field (southwest of France), considering the temporal and spatial evolution of the observed induced seismicity. This model of deformation has been determined from an updating of the earthquake locations and considering theoretical and analogue models usually accepted for hydrocarbon field deformation. The Lacq seismicity is clearly not linked to the natural seismicity of the Pyrenean range recorded 30 km farther to the south since the first event was felt in 1969, after the beginning of the hydrocarbon recovery. From 1974 to 1997, more than 2000 local events (M L < 4.2) have been recorded by two permanent local seismic networks. Unlike previously published results focusing on limited time lapse studies, our analysis relies on the data from 1974 to 1997. Greater accuracy of the absolute locations have been obtained using a well adapted algorithm of 3-D location, after improvement of the 3-D P-wave velocity model and determination of specific station corrections for different clusters of events. This updated catalogue of seismicity has been interpreted taking into account the structural context of the gas field. The Lacq gas field is an anticlinal reservoir where 3-D seismic and borehole data reveal a pattern of high density of fracturing, mainly oriented WNW-ESE. Seismicity map and vertical cross-sections show that majority of the seismic events (70 per cent) occurred above the gas reservoir. Correlation is also observed between the orientation of the pre-existent faults and the location of the seismic activity. Strong and organized seismicity occurred where fault orientation is consistent with the poroelastic stress perturbation due to the gas recovery. On the contrary, the seismicity is quiescient where isobaths of the reservoir roof are closed to be perpendicular to the faults. These quiescient areas as well as the central seismic part are characterized by a surface subsidence determined by repeated levelling profiles. Moreover, the temporal evolution of the distribution of the seismicity clearly exhibits a spatial migration from the centre to the boundaries of the reservoir. We conclude that the entire field is strained but this deformation is seismically expressed only where faults are parallel to the isobaths of the reservoir roof and where these faults plunge towards outside the field according to one of the two theoretical deformation models considered in our study. Then we propose a temporal scenario of deformation along the principal axis of seismic deformation.

Active Tectonics of the Pyrenees: A review

“…In the Aquitaine basin, more than 2000 clustered local low-moderated events (M L < 4.2) were induced by the extraction of gas in the Lacq field since 1969 (Grasso and Feignier, 1990;Maury et al, 1992;Segall et al, 1994;Souriau and Pauchet, 1998;Bardainne et al, 2008). To the east of Iruña, a 4.6 mbLg magnitude earthquake was widely felt on September 18 of 2004 in an area lacking significant seismic activity in recent times.…”

Resolving source mechanisms of microseismic swarms induced by solution mining

“…For instance, Godano et al (2012) found evidence for thrust faulting induced from solution mining in the Arkema-Vauvert salt field (France), which occurred mostly along pre-existing fault structures. In addition, Bardainne et al (2008) documented dominant thrust faulting above a gas filled anticline structure in the Lacq gas field (France).…”

Absolute earthquake locations using 3-D versus 1-D velocity models below a local seismic network: example from the Pyrenees

“…This seismicity for the period 1974-1997 has been relocated with a local 3-D velocity model (Guyoton et al 1992;Bardainne et al 2008).…”

“…The reference is based on a NLLoc 3-D absolute earthquake location process (Bardainne et al 2008) using a flat earth approximation with simple geographic conversion to kilometres. A high resolution 3-D P velocity model (Guyoton et al 1992) and a constant V P /V S ratio of 1.73 has been used.…”

“…Two local permanent local seismic networks were recording ground motions during this period. The 3-D absolute earthquake location using NLLoc in a high resolution 3-D P velocity model from Bardainne et al (2008) has been used as reference for comparison. 1030 earthquakes are compared.…”