GNGTS 2017 - 36° Convegno Nazionale

676 GNGTS 2017 S essione 3.2 monitoring surveys, the daily average tree transpiration reached 1.9 mm d -1 in T1, and 0.9 mm d -1 in the T2. Seasonal changes shown by ERT data. Fig. 2 shows the 3-D-ER (Ω m) images derived from the background acquisitions of June, July and September 2015 in T1 (Fig. 2a) and T2 (Fig. 2b), along with the ER profiles averaged within selected soil layers of the investigated soil volume (Fig. 2c for T1 and Fig. 2d for T2). At the end of the irrigation season (September), the mean reduction of ER, in the investigated soil profile, was of 69% in T1 and 38% in T2. This is consequence of the adopted irrigation regimes (i.e. full irrigation versus PRD). Overall, the most notable features emerging from the background inversions in Fig. 2 are the high ER (above 100 Ω m) areas located especially at depths between 0.4 and 1.0 m at the beginning of the irrigation phase. One of the most interesting aspects concerning the patterns of high ER in Fig. 2 is that they seem all to evolve substantially over time. This is a strong evidence against the widely spread belief that most of the electrical signal from roots comes from their large lignified structures (Amato et al., 2008; Rossi et al., 2011). In fact, the effect of large roots can be mistaken for the associated effects of strong soil drying (due to RWU) that roots may exert on the nearby soil. Results from our work seem to point towards the latter explanation. Fig. 3 - a) Time-lapse ER ratio volume at a selected time step (after the end of the irrigation, time 03) with respect to the background condition (before irrigation, time 00); b) tree transpiration rate (mm h-1), irrigation and ERT surveys timing are displayed in the graph in function of time. Data refers to the full-irrigated treatment (T1) on 15 Jul 2015.