The inversion of pressure and flow rate data from a synthetic laboratory experiment
exhibiting two-phase flow effects demonstrates a variety of the inverse modeling capabilities
implemented in iTOUGH2. The example is discussed in the
iTOUGH2 Sample Problems and
iTOUGH2 Software Qualification.
In these reports, both the development of the iTOUGH2 input file and
the corresponding output file are described in great detail.
Pressure data from Gas-Pressure-Pulse-Decay (GPPD) experiments were
inverted to estimate absolute permeability, the Klinkenberg slip factor, and porosity
of a Graywacke core sample. The example shows the importance of the error analysis
for the identification of strong parameter correlations and systematic errors. The
example also demonstrates the potential advantage of using
robust estimators for minimizing the bias from systematic errors. The analysis is
described in Finsterle and Persoff [1997]
and Finsterle and Najita [1998].
Data commonly obtained during geothermal production
(pressures, temperatures, liquid and steam flow rates,,
enthalpies, etc.) can be used to calibrate a model of the geothermal reservoir.
The calibrated model can then be used to study and optimize future field management scenarios such as
reinjection of condensate. iTOUGH2 has been used to estimate hydrogeologic and thermal properties
as well as fracture spacing for a synthetic geothermal field. The inverse modeling results are discussed in detail
in Finsterle and Pruess [1995]. The use of
optimization techniques for geothermal management problems is outlined in
Finsterle and Pruess [1997].
An application to actual field data is described in
Finsterle et al. [1997].
iTOUGH2 has been used for the calibration of the three-dimensional site-scale
unsaturated zone model of Yucca Mountain. The data matched include saturations,
water potentials, and pneumatic pressures.
Simultaneous inversions of data from several boreholes were performed. The resulting parameter set was
then validated by comparing simulation results of a selected two-dimenional cross-section
of Yucca Mountain with independent data such as inferred gas flow patterns,
fast flow path, and temperature data.
The calibration is described in a YMP Milestone Report.
In situ measurements of water potentials at various distances from a drift wall
were performed by ETHZ and Nagra, Switzerland, to study the extent of the two-phase zone that develops
in the vicinity of a ventilated drift.
iTOUGH2 was used to calibrate the data in order to determine absolute permeability and two-phase flow
parameters of the granitic rock. The results are discussed in
Finsterle and Pruess [1995].
The error analysis provided by iTOUGH2 can be used to optimize the design of a laboratory experiment
or field test. Performing synthetic data inversions allows one to identify the test design that
potentially minimizes the uncertainty of the parameters to be determined. The approach is described in
Finsterle and Faybishenko [1999] for a laboratory experiment,
and in Finsterle and Pruess [1997] for a well test in a
tight formation containing gas.
iTOUGH2 can also used to optimize operational parameters of a remediation system.
For example, the injection and extraction rate of an air-sparging system can be automatically
adjusted such that a minimum of contaminant remains in the system. The forward model used here is T2VOC,
which calculates TCE migration in the NAPL, aqueous, and gas phase in the unsaturated zone of a highly
heterogeneous aquifer. Air injection beneath the water table is
simulated, and contaminated soil gas is extracted in the vadose zone. The objective function to be minimized contains
energy and treatment costs as well as a penalty for incomplete remediation. Other terms such as installation and
financial costs may be added. The optimized rates lead to a maximum contaminant removal in a given time.
Page updated: July 8, 1997