Beauheim 1987 - Waste Isolation Pilot Plant - U.S. Department of ...
Beauheim 1987 - Waste Isolation Pilot Plant - U.S. Department of ...
Beauheim 1987 - Waste Isolation Pilot Plant - U.S. Department of ...
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
with subscripts:<br />
i = individual flow period<br />
j = individual flow period<br />
n = number <strong>of</strong> flow periods considered.<br />
In general, the fewer the number <strong>of</strong> points used in calculating the derivative, the more accurate it will be.<br />
Three-point derivatives, calculated using only the nearest neighbor on either side <strong>of</strong> a point, usually provide<br />
enough resolution to distinguish most important features. However, excessive noise in the data sometimes<br />
makes it necessary to use five- or seven-point derivatives, or various "windowing" procedures, to obtain a<br />
smooth curve. Unfortunately, these procedures may also smooth out some <strong>of</strong> the features <strong>of</strong> the curve needed<br />
for interpretation.<br />
The type curves published by both Gringarten et al. (1979) and Bourdet et al. (1984) were derived for<br />
flow-period (drawdown) analysis. In general, the curves can also be used for buildup-period analysis, so long<br />
as it is recognized that, at late time, buildup data will fall below the drawdown type curves because <strong>of</strong><br />
superposition effects.<br />
If the test analysis is to be performed manually, the drawdown or buildup data are plotted as pressure change<br />
since drawdown or buildup began (Ap) versus elapsed time since drawdown or buildup began (t) on log-log<br />
paper <strong>of</strong> the same scale as the type curves. The derivative <strong>of</strong> the pressure change is also plotted using the same<br />
vertical axis as the Ap data. The data plot is then laid over the type curves and moved both laterally and<br />
vertically, so long as the axes remain parallel, until a match is achieved between the data and pressure and<br />
pressure-derivative curves with the same CDe2s value. When the data fit the curves, an arbitrary match point is<br />
selected, and the coordinates <strong>of</strong> that point on both the data plot, t and Ap, and on the type-curve plot, pD and<br />
t&D, are noted. The permeability-thickness product is then calculated from a rearrangement <strong>of</strong> Eq (A-1):<br />
PD<br />
kh = 141.2qBp -<br />
AP<br />
(A-8)<br />
The groundwater-hydrology parameter transmissivity, T, is related to the permeability-thickness product by<br />
the following relationship, modified from Freeze and Cherry (1979):<br />
where:<br />
T = k hpg/p<br />
p = fluid density, M/L3<br />
g = gravitational acceleration, L/T~<br />
p = fluid viscosity, M/LT<br />
(A-9)<br />
When T is given in ft2/day, kh is given in millidarcy-feet,p is given in g/cm3, g is set equal to980.665crn/s2, andp<br />
is given in centipoises, Eq (A-9) becomes:<br />
T = 2.7435 x khplp (A-10)<br />
149