41 however if the uppermost active layer is

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: on effects, the model assumes that the cell contains a single composite well screen (or open interval) having a length equal to the sum of the lengths of all sections of screen, an effective well radius equal to the length-weighted average radius, and an effective cell-to-well conductance as described above. The composite well screen is generally assumed to be centered vertically halfway between the top of the highest individual section of well screen and the bottom of the deepest individual section of well screen (fig. 41). However, if the uppermost active layer is unconfined (or convertible) and contains a partially penetrating composite well screen, it is positioned in the cell so that the bottom of the composite screen is aligned vertically with the bottom of the lowermost screened interval within the cell (fig. 42). This assures that the well will continue to be connected to the cell if the calculated water-table elevation is above the bottom of the deepest part of the actual screen. Similarly, if the uppermost active layer is unconfined (or convertible) and contains a partially penetrating well screen defined by a partial penetration fraction for a node of a multi-node well, then the composite screen is positioned in the cell so that the bottom of the screen is aligned vertically with the bottom of the cell. Note also that in these cases, the user-defined partial penetration fraction and the initial saturated thickness of the cell are used to calculate the length of the screen. This length remains constant with Figure 41. Diagrammatic cross section showing wells with complex (nonuniform) well screens or open intervals within a model layer. Well A has uniform open intervals in model layers 1 and 3, but sections with two different well radii in model layer 2. Well B has two short well screens with different characteristics within a single model layer, which in MNW2 would automatically be replaced by an equivalent composite well screen that is partially penetrating (Well C). Figure 42. Schematic cross section of an unconfined aquifer showing a multi-node well and the relation of multiple screens in the uppermost model layer A, to the length and position of the reconstructed single composite screen in model layer 1. B, the position and length of the part of screen 2 in model layer 2 remain fixed. Model Features and Processes time and provides the basis for adjusting the partial penetration fraction if the water table changes position over time in a transient flow simulation. Given the effective cell-to-well conductance, the effective value of the combined well-loss terms in equation 15 can be computed as (39) An updated value for the cell-to-well conductance that includes the effects of partial penetration can subsequently be calculated using equation 25. MODFLOW allows the simulation to include a quasithree-dimensional representation of a ground-water system; typically, in this mode, a confining unit is not represented explicitly, but instead is represented with an adjusted vertical conductance value connecting the overlying layer with the underlying layer. If a multi-node well passes through such a nonsimulated confining unit, MNW2 assumes that blank casing exists within that interval. Flow Routing in Borehole Flow magnitude, flow velocity, and solute concentration can vary within a long borehole in which inflows and outflows vary along the well screen or open interval. Borehole flow meter surveys and sampling devices (for example, see Izbicki and others, 1999; Paillet, 2001) can collect such data in the field. If the model could calculate these variables, it would provide another type of dataset against which observations could be compared and the reliability of the model assessed or demonstrated. Therefore, the capability to route flow and solute through a multi-node well has been added. Because MNW2 does not compute head losses within a borehole, flow is routed using a relatively simple mass-balance approach similar to that used in the Streamflow Routing Package to route flow downstream from one stream reach to the next (Prudic, 1989; Prudic and others, 2004). MODFLOW calculates the flux between the aquifer and the well at every node. After the numerical solution is obtained for a given time step, the flow from one well node to the next is determined from the net of the exchange with the aquifer and inflow from the previous node. The MNW1 Package inherently assumes that if a pump is present in a long borehole, the pump intake is located above the first node of the multi-node well. The MNW2 Package adds the flexibility of specifying a pump intake (or discharge point for an injection well) location at any depth or associated with any single node of the multi-node well. This will not affect the calculated head in the well, but will, of course, affect how flow (and solute, if the GWT process is active) is routed through the borehole. Therefore, the net discharge from a well at the location of a pump intake represents another element that must be accounted for in the mass-balance approach...
View Full Document

This document was uploaded on 01/20/2014.

Ask a homework question - tutors are online