GNGTS 2017 - 36° Convegno Nazionale

622 GNGTS 2017 S essione 3.2 Laboratory and field GPR water content measures of typical Central Apennines sandy soils M. Ercoli, L. Di Matteo, C. Pauselli, P. Mancinelli, A. Cannata Dipartimento di Fisica e Geologia, Università di Perugia, Italy Evaluation of water content (WC) of fine grained unsaturated soils plays an important role in several fields (civil engineering, recharge of aquifers, triggering of landslides and flooding, agricultural irrigation and soil erosion, remediation of contaminated lands, etc). The estimation of the volumetric water content (θ v ) is an important parameter related to gravimetric water content and dry density of soils (Di Matteo et al. , 2009). The integration of different techniques may represent a powerful solution for WC definition (Revil et al. , 2012) and also GPR data are commonly used for WC estimation (Tosti and Slob, 2015). In this work, we focus on the calibration of the moisture content through GPR, by the comparison with point-based measurements. Materials and methods. We analyzed two sandy soils widely outcropping in central Italy (Fig.1). The first (S A ) was sampled in a fluvial quarry located on the north side of the Nera river, arising from limestone formations of the Umbria-Marche succession. The second (S B ) was sampled on the left bank of the Tiber River alluvial plane, originated by erosion of the flyschoid units (Marnoso Arenacea Fm.). Geotechnical properties of these soils (specific gravity, grain size distribution, compaction properties, etc.) have been recently investigated by Di Matteo et al. (in press). For WC estimation, laboratory analyses were carried out on both S A and S B, by mixing each soil with tap water and left to moisten for 24 hours, then repeatedly adding more water to obtain different gravimetric water contents. We then created a physical model to work in controlled conditions and measure the dielectric permittivity ε r . We filled up a cylindrical PVC soil column, manually compacting the sand to obtain a thickness of 0.30 m using a cylindrical standardized hammer, repeating the procedure with different water content values. After positioning a metal plate under the soil column (De Chiara et al. , 2014), we collected n°154 Common Offset (CO) GPR radargrams (1 GHz antenna), in order to measure ε r values and evaluate WC increase. We determined the ε r values using the picked TWT first arrivals and then θ v using Topp’s equation (Topp et al. , 1980). GPR laboratory results. Results from the tests show an increase of ε r values due to the decrease of the EM velocity, generated by progressively major water contents. The θ v values, calculated through the Topp’s law, show that the GPR technique is sensitive to WC increase. The values obtained in dry conditions are in agreement with literature data (ε r = 3-6) (Jol, 2009). The higher WC produced higher ε r estimations (ε r < 15), in the expected range of 10< ε r <30 Fig. 1 - Location map of sites used for soils sampling (SA – Conca Ternana alluvial plane – Nera River, SB – Tiber River alluvial plane). Meaning of legend of lithological map (Umbria Region, Central Italy): 1) recent and ancient fluvial- lacustrine deposits; 2) volcanic deposits; 3) flyschoid rocks; 4) calcareous and marly-silici-calcareous rocks.