Contaminant Fate and Reactive Transport in Groundwater - Dealing with Contaminated Sites: From Theory Towards Practical Application - page 868

boundary and initial conditions, errors associated with field measurements, et
cetera, and these are increased by the typical availability of only sparse spatial
and temporal field datasets. Different methods have been proposed for uncer-
tainty analysis in reactive transport models including Monte Carlo simulations
and stochastic models (e.g. Hill and Tiedeman 2007 ; Zheng and Bennet 2002 ).
It is important to note that the short list of guidelines for model application given
above should not be regarded as a merely sequential procedure, but rather as an iter-
ative process where the outcomes of a given action are made available for preceding
or subsequent steps. This allows a continuous improvement of the understanding of
the contaminant fate and transport at a specific site. Thereby, one of the most impor-
tant characteristics of a contaminant transport model - its capability of providing a
quantitative framework to integrate site-specific information and to test and refine
the site conceptual model - will be fully exploited.
In the following sections, practical applications of contaminant reactive trans-
port modeling are illustrated for a typical petroleum hydrocarbon contamination
scenario and for a field study at a landfill site, where ammonium was the principal
contaminant of concern.
19.4 Reactive Transport Scenarios
Several two-dimensional reactive transport scenarios were simulated with the aim of
evaluating the influence of different parameters on contaminant fate and transport.
To this purpose, the fate of a toluene plume in a shallow unconfined aquifer was
investigated. The influence of different groundwater flow, contaminant transport and
kinetic parameters on plume migration was assessed through a sensitivity analysis.
In Fig. 19.7 the geometry and the conceptual model of toluene transport in
groundwater is shown. As depicted in this figure, the modeled aquifer is consid-
ered a two-dimensional unconfined system with groundwater entering the domain
from the left boundary and flowing, with a dominant horizontal component, towards
the right. Toluene dissolving from a LNAPL source in the upper groundwater layer
was considered as model contaminant. As shown above (Table 19.1 ), the release
of an oxidizable organic contaminant, such as toluene, in a pristine aquifer can acti-
vate multiple biogeochemical processes (Chapelle 2001 ; Hunter et al. 1998 ). For the
sake of simplicity the attention is focused exclusively on the aerobic degradation of
toluene, in this modeling example. As the toluene plume migrates down-gradient, a
reactive fringe forms in particular at the lower edge of the plume. At the fringe the
two reactants, i.e. the oxidizable contaminant (electron donor) and oxygen (electron
acceptor) mix, allowing aerobic degradation of toluene to proceed. As highlighted
by previous modeling studies (e.g. Cirpka et al. 1999 ; Ham et al. 2004 ; Liedl et al.
2005 ; Maier and Grathwohl 2006 ; Rolle et al. 2005 ), under steady state conditions,
the principal process controlling the mixing of reactants is transverse dispersion.
The presence of mixing controlled enhanced biodegradation activity at the plume