The Datacrossing DSS
Groundwater contamination is a problem of growing concern
to water resources managers worldwide. Tools and methodologies for optimal
water resources management and for environmental impact analysis are needed
in order to evaluate natural hazards, presence of contaminant sources,
etc..
In this project, an integrated software system has been developed to
predict the impact of point/non point source pollution and to
quantify the uncertainty
due to the model application.
The system is a suit of partly in-house developed modules composed of GIS
technologies,
mesh generators, hydrological and pre/post processing codes within a Data-GRID
environment.
The GRID infrastructure has been
designed,
using SRB technologies in order to assess data and computing resources within the
distributed
environment (physical resources found at great geographical distance have been
connected).
The Datacrossing DSS can be applied to any aquifer, being the numerical model
and the applications adaptable at every soil media, hydro-geological or
climate condition.
The DSS permits to:
- query the data, and visualize maps;
- simulate the hydrodinamics of the aquifer;
- identify the location of a point/non point source pollution given in situ
measurements;
- predict the time of the spill;
- help hydrologists plan the best location of monitoring wells;
plan and control remidiation strategies, and alert users of possible
contamination.
The Modules
Data and resources are stored in the nodes of a SRB federated system,
dynamically assessed by each environmental application and displayed via a mapserver application.
The data fluxes are controlled and optimised.
- Data Manager
An interface written in PHP submits and controls the running jobs. It also
permits to import datasets into a user space within the SRB filesystem along
with a set of metadata. Metadata are automatically applied to distinguish owners,
context and content. Users can assess the SRB filesystem (the complexity is hidden
to the users), retrieve and display maps om the web-GIS.
- Simulation Manager
This tool works in two steps: users, through the Mapserver application, can
identify the source of pollution within the modeling domain (the source of
contamination is highlighted on a map). The corresponding simulation is
retrieved from the SRB filesystem, and results are displayed on the web
viewer. The time evolution of the contamination plume can be dynamically
mapped and overlaid with any geographical layers (e.g. all pumping wells).
Data transfer is a higly demanding task, therefore data flows need to be
controlled and optimised. The application access and process the data sets
where they are found (in the UNICA and CRS4 Zones) and only the selected
simulations are effectively retrieved and displayed on the Web GIS.
- Network
Analyst
This tool, managed by the Data manager, is aimed at designing a water-quality
monitoring network based on the physical characteristics of the aquifer. The
questions we seek to address with this module are where and when to probe the
groundwater system. The hydro-geological model, including solver, subsurface
geometry and physical description, is used to predict and improve the
acquisition coverage. Such coverage is time dependent. The aquifer domain is
divided into elements tagged by a Boolean value at different times. The true
value makes out the elements controlled by the monitoring network at each time
step. The controlled area is expected to increase with time until a steady
state is reached. Results are saved in a vector map format, stored in the SRB
filesystem, and can be displayed on the GIS web viewer.
- Source
Analyst.
Through the Data Manager, users upload the file that contains the position of
the monitoring wells and the measured concentration values of the pollutant.
The Source Pollution Analyst interpolates the simulated nodal mesh
concentrations generated by the code in the wells. In general, the 3D mesh
does not overlay with the monitoring network. The quadratic norm between the
measured concentrations and the simulated ones is computed, at each time step,
for each well, and this is repeated for all simulations. Each element of the
computational domain is modeled as a possible source of pollution. Ensemble
statistical indicators are then used to evaluate the performance of each
simulation in reproducing the real contamination. The DSS detects which mesh
elements are the most likely sources of the observed contamination and
estimate the time of the leakage event. A vector map is created, stored in the
SRB filesystem and displayed on the web viewer.
- Saltwater
contamination
Saltwater contamination scenarious (e.g. alternative exploitation schemes can
be compared) can be run using this module. Managers can design management
startegies and evaluate their effect on the salt water intrusion dynamics.
General considerations
Weak points: results will largely depend on the number and the position of the wells within the
computational domain and the reliability of measurements. The conceptual model is the key point to
correctly reproduce the hydrodynamics of the system.
Strong points: the DSS relies on a objective procedure. The advection, diffusion and dispersion
process is described with a mathematical physically based model.
The DSS can be used interactively only in the
Log in section.
|
|
