This preview shows page 1. Sign up to view the full content.
Unformatted text preview: e grid, then the flow between
the model domain and the well can vary greatly in magnitude
(and perhaps in direction) among the various nodes linked to
the single well. As in MNW1, the MNW2 Package assumes
that the hydraulic head within the well will equilibrate to a
single representative value. Because the heads in the aquifer
at various model nodes encompassing a multinode well will
vary depending on local and regional aquifer properties and
boundary conditions, a well can have nonuniform borehole
flow (or intraborehole flow), and the maximum borehole flow
rate can exceed the net withdrawal or injection rate specified
for the well. The MNW2 Package partitions the flux among
the various nodes connected to a multinode pumping well on
the basis of relative heads and hydraulic conductances (the
product of hydraulic conductivity and crosssectional area of
flow divided by the length of the flow path).
The conceptual model for flow through a long borehole
that is connected to multiple nodes of the model grid represents a substantial simplification of the actual hydrodynamics of such a system. A rigorous representation of the flow
dynamics within the borehole, such as analyzed by Cooley
and Cunningham (1979), is neither developed nor applied. In
fact, for simplicity and computational efficiency, it is assumed
that there are no head gradients within the borehole and that
a single value of hydraulic head (and water level) is effective
over the entire length of the borehole (Bennett and others,
1982; Fanchi and others, 1987), although Rutledge (1991)
discusses some hypothetical examples in which he calculates
head differences within a borehole of several feet [for example, a range of about 7 feet (ft) relative to a mean drawdown of
about 21 ft].
Under the simplifying assumptions of the MNW1 and
MNW2 Packages, at any level of the open or screened borehole, the flow between the well and the adjacent porous media
would be controlled by the head difference and the hydraulic
conductance between the well and the porous media for any
particular location (grid cell) where the well and porous media
are connected. When a well is open to two or more different
intervals in which the aquifer heads are different from each
other, as illustrated in figure 1, the well provides a pathway for
flow between the aquifers, and flow will occur in the borehole,
even in a nonpumped well, in response to head gradients in the
aquifer (and not to head gradients within the borehole, which
are not computed). Simulation of Pumping and Nonpumping Wells
In MODFLOW’s standard WEL Package, the discharge
of a well (Q) must be specified explicitly by the user (where
Q has dimensions of L3/T and is assumed to be negative in
sign for discharge). Also, a WEL Package well can only be 4 Revised MultiNode Well (MNW2) Package for MODFLOW GroundWater Flow Model connected to a single node of the grid. If a well is known to
discharge from a length of aquifer equivalent to more than one
model node or layer, then the user must determine an approach
for allocating the total discharge among the multiple model
nodes or layers such that the total flow (or net discharge) from
all model layers equals QNET:
(1)
where n is the total number of model nodes open to the well,
and m is the index of sequential node numbers of the multinode well.
In the MNW2 Package, however, the user specifies the
net discharge rate for the multinode well, and the code then
determines the layerbylayer (or nodebynode) flow rates
between the well and the aquifer using methods described in
the next section of this report. The total flow or discharge from
a well at the land surface is referred to as a “net” discharge
because, in complex hydrogeological settings, a single well
can have inflow in some parts of its open interval and outflow at other parts, with the difference being equal to the net
discharge. QNET can be negative (a pumping well), positive
(an injection well), or zero (an unpumped well or observation
well).
The total flow rate of a well (or net discharge) that is
specified in MNW2 is referred to in the input file as variable
Qdes, or the maximum desired flow rate for the well. This
designation is used because the actual flow rate at the wellhead
can be constrained by userspecified maximum and minimum
water levels at the well and by possible pumpcapacity constraints, as described in later sections of this report. If constraints are not specified, then the actual net discharge for the
well will equal the specified value of Qdes.
In constructing a well and installing a pump, the actual
location of the pump intake may be located at a depth or position determined for the conditions of that specific well. Therefore, the model allows the user to specify where within the
length of the borehole the pump intake is located. This is set
using the input variable PUMPLOC in input dataset 2b (appendix 1) and related variables in optional dataset 2e. If the user
does not specify a pump intake location...
View
Full
Document
This document was uploaded on 01/20/2014.
 Winter '14

Click to edit the document details