Computer-based models and databases
can predict and evaluate the environmental impact of an industrial
development from the planning stage, through operation, to decommissioning
and reclamation. Although the partners developed new environmental
management tools for mining, these have applications in many other
The project brought together two environmentally-conscious mining
companies and two universities skilled in environmental modelling
and computer mapping. The result is a computer-based environmental
management system built around a geographical information system.
Both mining companies use the new software and are pleased with its
Mines and quarries are an essential part
of Europe's industry, but they are often seen as bad neighbours.
All over the world the mining industry is under pressure to reduce
pollution from visual impact, noise, dust and fumes, road traffic
and the contamination of water supplies.
Good planning and environmental management can go a long way towards
eliminating these negative factors, but historically this has not
been easy. The tendency has been to treat each aspect of the environmental
impact as a separate phenomenon, whereas in reality many of the
factors are linked in complex ways.
What the mining companies needed was a tool that would allow them
to predict the potential impacts of their activities, from the early
days of planning right through to the closure of the mine. This
would help them choose the least harmful options at every stage.
Also important was the ability to view all this data, whether predicted
or measured, as an integrated whole instead of a series of isolated,
and sometimes contradictory, fragments.
The environmental management system developed during this project
combines the predictive capabilities of numerical models with new
tools for manipulating and analysing spatial data. The system has
simplified environmental impact prediction, monitoring and data
analysis by coordinating them behind a single computer 'front end'.
Getting a bird's eye view
A map is often the simplest and clearest way to display predicted
or measured concentrations of pollutants in the environment. Potential
water pollution due to acid rock drainage from waste dumps, for
instance, can be shown as a series of concentration contours spreading
out through an aquifer. The map also helps engineers assess environmental
impact by relating the levels of pollution to the geographic and
demographic characteristics of the area.
Transparent overlays on printed maps have been used for many years,
but their capabilities are rather limited. In 1990 researchers at
London's Imperial College realised that a geographical information
system (GIS) combined with suitable numerical models would provide
the ideal tool for effective environmental management in mining.
By 1992 the Imperial College scientists were ready to start developing
a full-scale system and validate it using real data. As partners
they chose two mining companies: Sociedade Mineira de Neves-Corvo
(Somincor), a copper and tin mine in Portugal, and Outokumpu Zinc
Tara Mines, which produces zinc and lead in Ireland. The Instituto
Superior Tecnico in Lisbon also joined the team to supply expertise
in geostatistical modelling.
The importance of modelling
One of the main objectives of BE-5174 was to develop environmental
prediction models and to integrate them with the GIS.
Imperial College developed a model of the effects of blasting in
a mine. This allows the system to predict accurately the vibration
levels at any point on the map, and so to design blasting regimes
to remain below the prescribed vibration levels and reduce their
impact on nearby people and buildings.
The Imperial College researchers also produced an air pollution
model for point source emissions, while their colleagues at the
Instituto Superior Tecnico did the same for area source emissions.
The models are similar to the Gaussian plume dispersion models used
by the US Environmental Protection Agency, but with some significant
improvements. By modelling the complex behaviour of dispersion in
the atmospheric boundary layer, the Imperial College researchers
improved the accuracy of the predictions.
They also simplified the meteorological data requirements of the
model. This is important in the minerals extraction industry as
operations are often located in remote areas where full-scale meteorological
monitoring may be impractical.
The Instituto Superior Tecnico concentrated on water, developing
a river quality model and methods of analysing groundwater quality
from monitored water quality indicators. Using geostatistics, a
discipline originally developed in the mining industry for estimating
ore reserves, the researchers are able to estimate pollutant levels
accurately while reducing the amount of data that needs to be collected.
By identifying the critical points on the map and taking measurements
here, instead of at random, the system helps make the best use of
environmental technologists' time in the field. The predictive capability
of the environmental management system was further increased by
the development of groundwater and solute transport model at Imperial
Putting it on [e1]the map
Just as important as the modelling is the ability to interpret
and analyse the results. The researchers at Imperial College used
ARC/INFO, a widely used GIS, as a basis for the system's graphical
user interface. They then integrated the various models with the
GIS in a simple, flexible architecture that links the predictive
power of the models with the spatial analysis and mapping capabilities
of the GIS. The system runs on a standard workstation.
Environmental managers wanting to assess the effects of air pollution
on the surrounding plant life, for instance, might combine a map
of different vegetation types with predicted pollution concentration
contours. The system can also show how environmental impacts vary
with time, using either predictions from the models or real historical
data from monitoring programmes.
Wider distribution within the mining world
Both Tara and Somincor are using the new system to manage their
day-to-day activities. Developing a comprehensive environmental
database for each mine has needed considerable effort, but both
companies feel that the work has been worthwhile and that they have
benefited from the project.
The developers are now keen to see the software used by the mining
community worldwide. Imperial College is currently exploring various
alternatives so that the prototype system can be turned into an
industry-standard package with the necessary technical support.
For the future, say the developers, the system could benefit from
more work on river water quality modelling, perhaps using neural
networks. It would be possible, they say, to predict environmental
quality indicators such as acidity and conductivity in the river
from information about the mine's effluent and past data on river
quality and flowrates. Imperial College is currently developing
a far-field noise prediction model to extend the system's capability.
An environmentally acceptable mine is possible, believes the industry,
but convincing the public of this has historically not been easy.
Good environmental management will help improve both the industry's
performance and its reputation.