GNGTS 2013 - Atti del 32° Convegno Nazionale

Comparison between ultrashallow reflection and refraction tomography in a geotechnical case study R. Balia Dipartimento di Ingegneria Civile, Ambientale e Architettura-DICAAR, Università degli Studi di Cagliari, Italy Introduction. From the last decades of the past century, an intense research to investigate the possibility and convenience of using the ultra-shallow reflection technique in engineering and environment - both with P- and SH-wave - has been developed, as witnessed by the abundant literature including theoretical papers and technical papers dealing with real case studies (e.g. Steeples et al. , 1997; Ghose et al. , 1998; Miller et al. , 1998; Miller and Xia, 1998; Steeples and Miller, 1998; Baker et al. , 2000; Deidda and Balia, 2001; Balia et al. , 2001, 2003; Balia and Gavaudò, 2003; Schmelzbach et al. , 2005; Balia and Littarru, 2010). Almost at the same time there has been a significant evolution of refraction seismology, with the final transition from classical techniques, based on the analysis and interpretation of the traveltime curves, to refraction tomography (e.g. White, 1988; Moser, 1991; Hole, 1992; Mandal, 1992; Boschetti et al. , 1996; Sheehan et al. , 2005). Presently, there is the feeling that, in spite of the continuous development of the shallow and ultra-shallow reflection, the use of this technique for operational, non-scientific purposes is still relatively uncommon. Very likely, this is also due to the greater ease of use and the effectiveness attained with refraction tomography so that in most of the cases, especially in engineering and environmental problems in which more often than not the targets are just a few meters deep, the latter technique is preferred. As known, ultra-shallow reflection requires specific instruments and field procedures: high natural frequency geophones, when available, placed at very short spacing, CMP cables, very short shot-point spacing; the problems become even more challenging in the processing: strong interference with refractions, ground-roll and air-wave, data sets generally characterized by poor signal-to-noise ratio, great difficulty for compensation of static effects and for velocity analysis, and so on. Of course, a good processing package is strictly necessary. Even refraction tomography requires a good processing package, but: geophone interval and shot interval can be less, the spread management is less demanding and nowadays processing simply requires the coordinates of detectors and shot-points, and obviously the first-arrival times; moreover the most complete processing packages allow to process jointly surface data and down-hole/up-hole/cross-hole data. The aim of this paper is to show, analyze and discuss the results obtained along one same profile by means of shallow reflection and refraction tomography respectively, using the same base materials and, obviously, the respective processing packages. Experimental site and instruments. The experimental site is located in the old town of the city of Cagliari (Sardinia, Italy). In this case, the accurate knowledge of the subsoil to a depth of 10-15 m from the ground level was requested for designing an underground car parking. Apart from the road paving (0.3 m thick) the near-surface geological scheme of the site is constituted, top to bottom, by: 1) a more or less compacted backfill made up of sand, gravel and abundant clay, with thickness from zero to several meters, 2) a bedrock constituted by Miocene argillaceous limestone, more or less fractured and weathered, more than twenty meters thick, belonging to the intermediate member of the so-called “Miocene Cagliaritano”, and named “Tramezzario” (Barrocu et al. , 1981). The excavation for the parking should have a depth of about 12 m, and therefore the aim of the survey was to define the depth to Miocene limestone bedrock and to identify both characteristics and possible changes especially within the bedrock, at least in the first fifteen meters from ground surface. The ground surface of the site is regular and the line along which seismic data acquisition has been performed, about North-South oriented, has a dip of 5%. The basic equipment employed for the two surveys is listed below: 12 GNGTS 2013 S essione 3.1