American Journal of Climate Change, 2013, 2, 138-146
http://dx.doi.org/10.4236/ajcc.2013.22014 Published Online June 2013 (http://www.scirp.org/journal/ajcc)
Global River Basin Modeling and Contaminant Transport
Rakesh Bahadur, Christopher Ziemniak, David E. Amstutz, William B. Samuels
Center for Water Science and Engineering, Science Applications International Corporation, McLean, USA
Email: samuelsw@saic.com
Received January 4, 2013; revised February 6, 2013; accepted February 15, 2013
Copyright © 2013 Rakesh Bahadur et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ABSTRACT
Using geographic information system techniques, elevation derived datasets such as flow accumulation, flow direction,
hillsope and flow length were used to delineate river basin boundaries and networks. These datasets included both
HYDRO1K (based on 1 km resolution DEM) and HydroSHEDs (based on 100 meter Shuttle Radar Topography Mis-
sion). Additional spatial data processing of global landuse and soil type data were used to derive grids representing soil
depth, texture, hydraulic conductivity, water holding capacity, and curve number. These grids were input to the Geospa-
tial Stream Flow model to calculate overland flow (both travel time and velocity). The model was applied to river ba-
sins across several continents to calculate river discharge and velocity based on the use of satellite derived rainfall esti-
mates, numerical weather forecast fields, and geographic data sets describing the land surface. Model output was com-
pared to historical stream gauge observations as a validation step. The stream networks with associated discharge and
velocity are used as input to a riverine water contamination model.
Keywords: Hydrology; Geographic Information Systems; Surface Water
1. Introduction
Modeling the fate and transport of waterborne contami-
nants in rivers and watersheds requires fundamental
knowledge of the hydrologic cycle. The processes are
well known and hydrologic models have been developed.
The limiting factor in applying these models is the un-
derlying data. Data sources for rivers and watersheds in
the United States have been integrated with models [1-5]
to simulate both deliberate and accidental releases.
However, for applications outside the US, little or no
waterborne modeling has been done for chemical, bio-
logical or radiological constituents.
The physical processes involved in watershed analysis
start with the deposition of water on the earth’s surface
as rain or snow. The liquid water (including snow melt)
then moves over the surface forced by gravity to seek the
lowest point in the terrain. As the liquid flows over the
surface, some of it percolates into the soil. The fraction
going into the soil depends on the land cover, soil texture
and saturation, which in turn depends on the rate at
which the soil dries out due to evapotranspiration.
The application of transport models is dependent on
the availability of data to implement the modeling and to
verify model fidelity. To apply complex models to wa-
tersheds, simplification and adaptation are necessary to
address the complexity of each individual modeling do-
main. For a given setting, some terms in the governing
equations are less important than others, allowing simpli-
fication and a more efficient implementation. However,
over-simplification can result in simulation models that
are far removed from the physical, chemical, or biologi-
cal characteristics of water bodies.
Global river flows are an important input (boundary
condition) to estuarine, coastal and oceanic models.
Real-time river flow [6] is also a critical input to river
models used to portray transport and dispersion of toxic
contaminants released deliberately or accidentally on-
shore. In the absence of a network of real-time river
gages, as is available in the US, alternative means are
required for calculating the flow of drainage streams and
rivers.
Two models (GeoSFM and ICWater) were used, re-
spectively, to create drainage networks (with associated
flows and velocities) and to perform contaminant trans-
port based on these networks. The GeoSFM processes
and datasets are described in section 2 below. The appli-
cation of ICWater for downstream contaminant transport
and dispersion is discussed in Section 4.
2. Hydrologic Data Processing
In this study, the first step in the process was to assemble
hydrologic and terrain data sets from remotely sensed
Copyright © 2013 SciRes. AJCC