A B S T R A C TA traveltime inversion technique is applied to model the upper ∼40 m of the subsurface of a glaciated shield rock area in order to calculate static corrections for a multi-azimuth multi-depth walk-away vertical seismic profile and a surface seismic reflection profile. First break information from a seismic refraction survey is used in conjunction with a ray-tracing program and an iterative damped least-squares inversion algorithm to create a two-dimensional model of the subsurface. The layout of the seismic survey required crooked seismic lines and substantial gaps in the source and receiver coverage to be accounted for. Additionally, there is substantial topographical variation and a complex geology consisting of glaciofluvial sediment and glacial till overlying a crystalline bedrock. The resolution and reliability of the models is measured through a parameter perturbation technique, normalized χ 2 values, root means square traveltime residuals and comparison to known geology.
I N T R O D U C T I O NThe town of Outokumpu, in northern Karelia, Finland, is the site of a historical but now depleted base metal mine. Renewed exploration of the area has been spurred in part by the recent acquisition of a crustal-scale seismic reflection profile through Finland acquired as part of the FIRE (Finnish Reflection Experiment) project (Kukkonen et al. 2006; SorjonenWard 2006), which prompted a reassessment of the regional geology. The profile near Outokumpu revealed an unexpected series of horizontal reflectors that appeared to be related to the Outokumpu assemblage exposed at the surface to the northwest and this further motivated the drilling of the International Continental Scientific Drilling Program (ICDP) wellbore to a depth of 2.5 km (Kukkonen and the Outokumpu Deep Drilling Working Group 2007). High-resolution seismic reflection/refraction profiles and multi-azimuth multi-depth walk-away vertical seismic profiles (VSP) were obtained as * E-mail: schijns@ualberta.ca part of this study. However, the analyses of these data were hampered by severe (10-70 ms) static time shifts caused by a combination of the large velocity contrast between the underlying metamorphic rock and Quaternary glacially deposited sands and gravels and the rapidly fluctuating surface topography.Conventional static correction techniques failed to satisfactorily correct either the surface reflection/refraction or the VSP data. Here, an alternative strategy that determines the source and receiver static corrections via seismic refraction traveltime inversion is presented. The utility of these static corrections is illustrated by the improvement to the moveout curves observed in a multi-azimuth, multi-depth VSP survey and the improved imaging of a seismic reflection profile.