43 if the pump intake is located in that cell the net

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: to routing flow in the borehole. 39 The flow routing algorithm is relatively simple. For any node of a multi-node well, routing starts at the node furthest from the wellhead, where by definition there is no inflow from a previous node, so the flux to or from the next node must equal the exchange with the aquifer. In general, the flow from any given node n to the next node closer to the wellhead (n-1) equals the inflow from the further node (n+1) plus or minus the flux between the aquifer and the well at that node (fig. 43). If the pump intake is located in that cell, the net discharge from the well must also be subtracted there. Expressed quantitatively, the flux between adjacent nodes is computed as (40) where Q is a volumetric flux (L3/T), Qn-1 is the flow from node n to the next node closer to the wellhead (positive in sign for flow towards the wellhead—which is upwards in a vertical well), Qn+1 is the flow from the further node to node n, QGW is the flux between the aquifer and the well at that node (negative sign for discharge from the aquifer and inflow to the well), and Qnet is the net external discharge directly from the well. Direct external discharge can occur only at the pump intake, which is located at either a user-specified node or (by default) above the first node closest to the wellhead (negative sign indicates discharge from the well). Figure 43. Schematic cross section illustrating components of flow into and out of well segments comprising a representative node (node n) of a multi-node well. If the GWT process is active, solute mass is also routed between well nodes, and the concentration in a node is calculated using a simple mixing formula, as described by Konikow and Hornberger (2006b, p. 6–8). This approach assumes fluid and solute mass storage within the borehole is negligible over the duration of the time step. Solute Transport Because intraborehole flow can have a substantial effect on nearby ground-water hydraulics, it is logical to infer that it 40 Revised Multi-Node Well (MNW2) Package for MODFLOW Ground-Water Flow Model can therefore affect solute (or contaminant) distribution in an aquifer system (Konikow and Hornberger, 2006a). This effect has been previously recognized, primarily with respect to the difficulty of obtaining representative water-quality samples from boreholes or monitoring wells with long open intervals or long well screens. Church and Granato (1996) note that borehole flow redistributes water and solutes in the aquifer adjacent to the well, increasing the risk of bias in water-quality samples. Reilly and others (1989) used flow simulations to demonstrate that contaminant monitoring wells with long screens may completely fail to fulfill their purpose in many ground-water environments because of intraborehole flow. They conclude that significant borehole flow can occur in wells with long screens even if they are in relatively homogeneous aquifers with very small vertical head differences in the aquifer. Lacombe and others (1995) noted that abandoned and improperly sealed boreholes may act as conduits for contaminant transport from contaminated zones to previously uncontaminated strata. Calculating such effects comprehensively requires that solute-transport processes in the ground-water system be simulated. To do this, Konikow and Hornberger (2006b) added the capability to represent multi-node wells to the GWT process of MODFLOW–2000. That capability has been extended to MNW2, and the code modified for the new flow-routing routines in MNW2, which allow a pump intake to be located in any node of the well. The solute-transport capability was tested and demonstrated by Konikow and Hornberger (2006a,b) using a variant of the Reilly problem in which an initial mass of contaminant was placed in the cells in layer 1 immediately upgradient from the long nonpumping borehole. Solute transport was then simulated in a transport subgrid that included 20 rows, 40 columns, and all 41 layers of the primary MODFLOW grid. The test assumed transient transport for one year in a steady-state flow field. These results were replicated with the new code to assure that MNW2 is compatible with the GWT process. In MNW2, if the GWT process is active and water is being injected into the well, the concentration in the injected fluid can be specified for the well for each stress period in dataset 4a. The Multi-Node Well Observation (MNWO) Package developed by Konikow and Hornberger (2006b) is now superseded by the MNWI Package documented in this report. This new output package for MNW2 controls the generation of additional separate output files listing detailed information about each multi-node well, including solute concentration information. These files can generate concentration profiles within a borehole and concentration changes over time at each node and in the net discharge of the well. Code Efficiency The features, processes, and options available for simulating multi-node wells offer the MODFLOW model user more flexibility and realistic conditions for simulating wells in ground-water s...
View Full Document

This document was uploaded on 01/20/2014.

Ask a homework question - tutors are online