Preliminary groundwater modelling by considering the interaction with superficial water: Aosta plain case (northern Italy)

Abstract

The study is developed through scientific cooperation between the University of Milano-Bicocca and the Regional Agency for Environmental Protection (ARPA) of the Valle d’Aosta Region. Its aim is to produce a decision-support tool to help the Public Administration’manage groundwater and public water supply. The study area is the plain of Aosta, between the cities of Aymavilles and Brissogne; in this area groundwater represents the main source of public water supply. The valley is oriented east-west, along the Baltea for a length of 13.1 km and a width of 4.6 km. The textural and hydrogeological properties of the deposits are strictly connected to glacial deposition and to the subsequent sedimentary processes which took place in glacial, lacustrine and fluvial systems. The study is based on available well information in the Aosta plain - including water wells (133) and piezometers (121) - which have been coded and stored in the well database TANGRAM,. The database facilitates interpretation of the well data, and it allows three-dimensional mapping of subsurface hydrogeological characteristics through database codification and ordinary kriging interpolation. The study is designed to achieve two objectives. The first is to provide the Aosta Public Authorities with a well database in order to simplify groundwater management. The second is to provide Public Authorities with a groundwater flow model of the local aquifer. The model integrates surface and subsurface flows in order to fully account for all important stresses, both natural and anthropogenic, on the groundwater system. It provides a tool for testing hypotheses (such as the impact of new wells) and thereby allows science-based management of the aquifer resource.

Numerical Modeling of Remediation Scenarios of a Groundwater Cr(VI) Plume in an Alpine Valley Aquifer

“…Within the study area, this river collects water from 14 tributaries, among which the most important is the Buthier Creek, which crosses the town of Aosta from north to south (Figure 1a). The geological and hydrogeological features of this area were exhaustively described by many previous studies [20][21][22][23][24][25][26][27][28][29][30], so only a brief summary is given in the following. The aquifer system is formed by fluvioglacial, alluvial, lacustrine, and fan deposits, which overlay a deep crystalline basement.…”

“…The modeling of these scenarios is based on previous groundwater flow and Cr(VI) transport models [21,22]. The latter is hereafter called Scenario 0.…”

“…The resulted 3D distribution of K values was then discretized into 10 zones; each zone was defined through the analysis of the frequency distribution of K. Finally, K values of each zone were adjusted during flow model calibration, leading to maximum and minimum values of 475 and 0.0864 m/day for K x and K y , and 47.5 and 0.00864 m/day for K z , respectively [21]. The following boundary conditions were used [22]: (a) no flow (Neumann boundary condition) for reconstructing the "V" shape of the crystalline basement; (b) general head boundary (GHB; Cauchy boundary condition) for simulating the groundwater inflow and outflow through model margins at the western and eastern model edges, respectively; (c) recharge (Neumann boundary condition) through four zones with maximum and minimum values of 1.69 × 10 −3 and 6.02 × 10 −4 m/day [22]; these values were calculated as the water average surplus obtained by the Thornthwaite-Mather's method, starting from the average monthly precipitation and temperature data; (d) Neumann boundary conditions through the multinode well (MNW) package for simulating the active wells in the area (i.e., 27 wells with~63,000 m 3 /day of total discharge); (e) Cauchy boundary conditions through the streamflow-routing (SFR2) package [41] for simulating surface water bodies, in particular, the Dora Baltea River. Concerning the latter, since the SFR2 package is not supported by MT3DMS, before running the transport model, the SFR2 was substituted with the DRAIN package along the gaining segments of the Dora Baltea River.…”

“…Concerning the latter, since the SFR2 package is not supported by MT3DMS, before running the transport model, the SFR2 was substituted with the DRAIN package along the gaining segments of the Dora Baltea River. The flow model was run in steady-state conditions [22]. The flow model calibration was done by trial and error using 25 head and 1 flow targets, referred to January 2009, that can be considered as representative of average hydraulic conditions of the studied aquifer, since maximum and minimum hydraulic heads are registered during summer and spring, respectively [22].…”

“…The flow model was run in steady-state conditions [22]. The flow model calibration was done by trial and error using 25 head and 1 flow targets, referred to January 2009, that can be considered as representative of average hydraulic conditions of the studied aquifer, since maximum and minimum hydraulic heads are registered during summer and spring, respectively [22]. The values of K and riverbed conductivity were the most sensitive, and they were adjusted in the calibration process.…”