GNGTS 2017 - 36° Convegno Nazionale

674 GNGTS 2017 S essione 3.2 of irrigation (September). At the beginning of each ERT monitoring time steps above, one ERT acquisition was conducted on the full 204 electrode setup (background). During July and September, when the irrigation took place, we also acquired a full acquisition on all 4 quarters at the end of the daily irrigation, on both T1 and T2 treatments. More frequent time-lapse acquisitions were performed on a hourly basis on the quarter surrounding the tree (q4 in both treatments). 3-D ERT data processing and inversion. All the ERT acquisitions took place with a ten- channel resistivity meter (Syscal Pro 72 Switch, IRIS Instruments) using the same acquisition scheme (skip-0 dipole-dipole). Direct and reciprocal resistance data were measured, to have an estimate of the data errors (Binley et al. , 1995) for each quarter (i.e. 72 electrodes). The ERT data processing consisted of data error identification and inversion of the resistance values using an Occam’s approach (R3t code, Binley, 2013), where the target mismatch between measured and computed resistance data is set according to the data error; more specifically, different inversion strategies were adopted: i) inversions aimed at producing 3-D ER “background” images in all quarters for both treatments; ii) inversions aimed at producing images of 3-D ER changes before and after irrigation (at a daily temporal scale), simultaneously for all quarters for both treatments; iii) time-lapse ER inversions of the individual quarters containing the trees (q4). Results and discussion. Ancillary data observed during the 3-D ERT monitoring. The results of the SWC monitoring for the PRD treatment show the expected alternating drying and wetting cycles on either side of the tree (i.e., east and west) after each switching event. In the T1 treatment, the SWC remained close to field capacity. Soil temperature variations were, on average, approximately 2°C during each ERT acquisition. Considering that the ER changes 2% per °C (Friedman, 2005), we conclude that in our case the temperature effect is negligible with respect to the effect of inferred SWC changes (Nijland et al. , 2010). The analysis of soil pore water indicates a moderate salinity, with EC 25 °C values in the range of 2-3 dS m -1 . Such values should not affect the sensitivity of our ERT measurements to SWC. During the ERT Fig. 1 - Small scale 3-D ERT monitoring scheme at (a) T1 and (b) T2 treatments.