Abstract:
Knowledge of the spatial distribution of geotechnical parameters in the near surface is essential in engineering geology. Latest developments in engineering geophysics and direct push-based (DP) data acquisition provide a large set of techniques for non-invasive and in situ data recording for high-resolution parameterisation. Different traditional but sparse geotechnical surveying and analysis tools are typically used at selected locations such as drillings and sieve analysis, respectively. State-of-the-art laboratory techniques and supporting field measurements are used to gather the relevant soil properties for soil type classification helping to assign this information to homogeneous sections of the ground (layers). Site-specific data interpretation however becomes challenging because actual field conditions (in situ) differ from those appropriated in the laboratory, owing to the high spatial heterogeneity of near-surface unconsolidated sediments. We performed intensive fieldwork at two test sites (Löbnitz and Taucha) representing typical construction grounds in Central Germany. We tested electrical resistivity tomography, ground penetrating radar, refraction seismic, multichannel analysis of surface waves, and mobile DP-based seismic traveltime tomography. A combination can overcome the deficiencies restrictions of the particular individual methods, compensating the deficiencies of each method, helping therefore to minimise any drawbacks or limitations that depend on the contrast of and between the physical properties, which each technique is sensitive to. In developing mobile seismic tomography, we overcame prior restrictions imposed by existing on-site boreholes by using DP-devices as carrier systems. The derived geotechnical parameters allows us to carry out uncertainties evaluated by additional applied DP-methods gathering high-resolution data for ground truthing. Furthermore, we show that DP-based in situ-obtained soil colour yields information about the vertical stratigraphic pattern. So far, no methods exist that stipulate how best to handle such high-resolution data from colorimeter probes. We present improvements of direct data acquisition, numerical transformation, filtering, and interpretation. We found that filtered colour surrogates provide more detailed information about the soil which corresponds to its geological set-up. The results help us gain a new understanding of soil colours as a technically reliable proxy that is applicable in geotechnical site characterisation. The findings encourage an enable the reliable characterisation of a highly heterogeneous ground, especially for appraising information uncertainty at different scales. Compared with traditional sparse geotechnical measurements, we obtain more information for definition of clearly homogeneous sections (layers). The combined data interpretation compensates for any disadvantages of a single method. Thus, we expect a significant positive impact for near-surface characterisation in the frame of engineering geological investigations.