WMOD00423

• Title: Currents…Developing Technology for Tomorrow's Challenges
• Date: 1/1/1993
• Weather Modification - Doc Type: Report

6

process of evaluating uncertainty associ-
ated with predictions of flow and
contaminant migration for use in risk
assessment (fig. 4). Results of such an
analysis aid decisions regarding the need

Integrated Interpolation

Geologic History
Depositional
Environment

Seismic Data

Fi~,ure 3. Each circle indicates the range of possible subsurface interpretations
fOI each data set alone. Interpretation of each set of data reveals a range of
possible subsurface configurations. When the data are integrated, only the
subsurface configurations which are possible interpretations of all the data re-
mElin as possible interpretations of the site. Each data set that is added to the
integrated interpolation further reduces the zone of overlap, thus reducing the
overall uncertainty.

to conduct further characterization. If fur-
ther characterization is deemed necessary,
then the analysis results can be used to
identify the type of data to collect and the
locations from which to collect these data.
Quantitative assessment of uncertainty pro-
vides a measure of the need for additional
work and avoids the shortcomings of quali-
tative assessments. Often, site engineers
argue that additional data make the prob-
lem more complex instead of simplifying
the situation. This conflict occurs because
the level of uncertainty was not recognized
at the start. More valid data can only tell
more about the site. Each new piece of
data would be expected to reduce subsur-
face uncertainty.

The development of this uncertainty pro-
gram will benefit studies of ground-water
contamination. Risk assessments must be
evaluated before remediation activities
begin. Migration of ground-water contami-
nants in a subsurface, heterogeneous
medium causes irregular contaminant dis-
tribution. Heterogeneity causes uncertainty
in identifying subsurface contaminant
plumes and predicting future pathways of
plumes. Accurate identification of problem
areas allows remedial action to be focused
on the problem areas. This approach al-
lows remediation to be better designed for
site conditions, thus reducing costs. A well-
defined distribution of contaminants and a
sufficient characterization of subsurface
properties will result in less expensive
remediation and characterization.
In summary, the program uses all site
data to make interpretations. Hard data,
such as borehole samples and physical test
results, are incorporated. Soft data, such
as geologic history of deposition, and
remotely collected geophysical data, are
also considered. Indicator theory, a method
of defining probability distributions in a
discrete marmer without assuming a priori
distribution, characterizes the data uncer-
tainty. Ground-water flow and contaminant
transport are simulated in the acceptable
configurations using other software.
Results of the flow and transport simula-
tions for all of the configurations are
compiled to produce a frequency distribu-
tion of head, velocity, or concentration at
selected locations and times. The technique
is currently being built into user-friendly
software which:

I

\;~
I

. Allows input of hard and soft fietd data.
. Facilitates statistical evaluation and
manipulation of the data. ,
. Facilitates geostatistical analysis of the
data.
. Produces alternative, equally likely sub-
surface configurations honoring the data.
. Accepts or interactively modifies the
evaluation.
. Inputs the spatial distribution of parame-
ters into a flow and transport model.
. Allows interactive editing of the flow
and transport models as necessary.