Flow and transport in porous media can be modelled at very different scales: from the molecular scale (length scale or model resolution, l, approximately equal to 10-10 m), to the microscopic scale (pore-grain scale, l 10-5 ÷10-3 m), the macroscopic scale (representative elementary volume, l 10-3 ÷ 10-1 m), the megascopic scale used for local field problems (l 1 ÷ 10 m) and the gigascopic scale at which regional aquifer systems are studied (l 102 ÷ 103 m). The link between the megascopic and gigascopic scales is important for practical applications and is discussed in this communication with reference to the flow models developed by our research group in the last years for the aquifer system of the metropolitan area of Milano. In fact we have developed several nested models, which consider areas varying between 20 and 700 km2 and with grid spacing varying between 100 and 1500 m. Models developed and applied at different scales can reproduce different features and require different data for the calibration and validation. This communication will give practical examples about the following problems: the adequacy of data (data accuracy; spatial and temporal density of data point; data effectiveness in representing phenomena at the appropriate scale); model calibration with multiple sets of data; model validation through the "forecasting" of natural undisturbed conditions of the aquifer system.
Some Lessons from Modeling Ground Water Flow in the Metropolitan Area of Milano at Different Scales
Romano E;
2007
Abstract
Flow and transport in porous media can be modelled at very different scales: from the molecular scale (length scale or model resolution, l, approximately equal to 10-10 m), to the microscopic scale (pore-grain scale, l 10-5 ÷10-3 m), the macroscopic scale (representative elementary volume, l 10-3 ÷ 10-1 m), the megascopic scale used for local field problems (l 1 ÷ 10 m) and the gigascopic scale at which regional aquifer systems are studied (l 102 ÷ 103 m). The link between the megascopic and gigascopic scales is important for practical applications and is discussed in this communication with reference to the flow models developed by our research group in the last years for the aquifer system of the metropolitan area of Milano. In fact we have developed several nested models, which consider areas varying between 20 and 700 km2 and with grid spacing varying between 100 and 1500 m. Models developed and applied at different scales can reproduce different features and require different data for the calibration and validation. This communication will give practical examples about the following problems: the adequacy of data (data accuracy; spatial and temporal density of data point; data effectiveness in representing phenomena at the appropriate scale); model calibration with multiple sets of data; model validation through the "forecasting" of natural undisturbed conditions of the aquifer system.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.