TITLE:
Characterization of Groundwater Pollution Sources with Unknown Release Time History
AUTHORS:
Om Prakash, Bithin Datta
KEYWORDS:
Groundwater Pollution; Pollution Source Identification; Unknown Pollution Source Starting Time; Linked Simulation Optimization; Simulated Annealing
JOURNAL NAME:
Journal of Water Resource and Protection,
Vol.6 No.4,
March
28,
2014
ABSTRACT:
Characterizations of unknown groundwater pollution sources in terms of
source location, source flux release history and sources activity initiation
times, from sparse observation concentration measurements are a challenging
task. Optimization-based methods are often applied to solve groundwater
pollution source characterization problem. These methods are effective only
when the starting times of activity of the sources are precisely known, or the
possible time window within which the sources activity actually start is known with
a fair degree of certainty. However, in real life scenarios, the starting time
of the activity of the sources is either unknown or can lie anywhere within a
time window of years or decades. Absence of any prior information about the
span of time window, within which the sources become active, makes existing
source identification methodologies inefficient. As an alternative, an
optimization-based source identification model is proposed, to simultaneously
estimate source flux release history and sources activity initiation times. The
method considers source flux release history and sources activity initiation
times as explicit decision variables, optimally estimated by the decision
model. Performance of the developed methodology is evaluated for an
illustrative study area having multiple sources with different source activity
initiation times, missing observation data and transient flow conditions. These
evaluation results demonstrate the potential applicability of the proposed
methodology and its capability to correctly estimate the unknown source flux
releasing history and sources activity initiation times.