282
Air Pollution and Turbulence: Modeling and Applications
10.4 FINAL COMMENTS
One of the main aims of European environmental policy is to improve air quality in Europe. The Framework Directive on air quality assessment and management addresses air quality near major sources and in cities. Other instruments, such
as the National Ceiling Directive, the Large Combustion Plant Directive, and the
Gothenburg Protocol, address long-range trans-boundary issues as well as local
ones. The effective implementation of European legislation requires a strategy for
policy decision support based on modeling applications that are powerful tools for
air quality management.
The key aspect is the use of models to attribute measured concentrations to the
sources from which the pollution may have been emitted, and to assess the environmental impact of pollutants or pollution reduction strategies. The presented case
study is an example of the application of modeling tools to the assessment and evaluation of air quality management strategies. The analysis of the modeling results reveals
a reasonable simulation of ozone levels. The statistical evaluation allowed the conclusion that despite the relative deviation from the observed concentrations, the modeling system presents good correlation coeffi cients in most of the air quality stations.
The simulation of the Porto and Lisbon domains indicates the importance of refi ning
the grid to achieve better detail and defi nition of results. Concerning the 2010 scenario simulation of the National Emission Ceiling Program, the signifi cant decrease
in ozone concentration is noticeable when the limits are compared with the baseline
year of 2001. That decrease is more signifi cant in the urban areas of Lisbon and Porto.
However, the need is well recognized to deepen the study in order to improve results
which means, in practice, updating the emission inventory for instance and its spatiotemporal disaggregation, and refi ning the initial and boundary conditions.
Modeling results should be carefully analyzed and interpreted. The uncertainties
associated with the input data, and with the model itself, need to be properly evaluated before their predictions can be used with confi dence. Dispersion is primarily
controlled by turbulence, which is random by nature, and thus cannot be precisely
described or predicted by means of basic statistical properties. As a result, there is
spatial and temporal variability that occurs naturally in the observed concentration
fi eld. On the other hand, uncertainty in the model results could also be due to factors such as errors in the input data and model formulation. Because of the effects of
uncertainty and its inherent randomness, it is not possible for an air quality model
ever to be “perfect,” and there is always a base amount of scatter that cannot be
removed (Chang and Hanna, 2004).
Notwithstanding these limitations, air quality models should be considered as
powerful tools to predict the fate of pollutant gases or aerosols upon their release
into the atmosphere, and to infl uence decisions with signifi cant public health and
economic consequences.
ACKNOWLEDGMENTS
The authors are grateful for the collaboration of L. Salmim and would like to thank
her for her participation in the case study modeling work. Also the collaboration of
© 2010 by Taylor and Francis Group, LLC
Air Pollution and Turbulence: Modeling and Applications
10.4 FINAL COMMENTS
One of the main aims of European environmental policy is to improve air quality in Europe. The Framework Directive on air quality assessment and management addresses air quality near major sources and in cities. Other instruments, such
as the National Ceiling Directive, the Large Combustion Plant Directive, and the
Gothenburg Protocol, address long-range trans-boundary issues as well as local
ones. The effective implementation of European legislation requires a strategy for
policy decision support based on modeling applications that are powerful tools for
air quality management.
The key aspect is the use of models to attribute measured concentrations to the
sources from which the pollution may have been emitted, and to assess the environmental impact of pollutants or pollution reduction strategies. The presented case
study is an example of the application of modeling tools to the assessment and evaluation of air quality management strategies. The analysis of the modeling results reveals
a reasonable simulation of ozone levels. The statistical evaluation allowed the conclusion that despite the relative deviation from the observed concentrations, the modeling system presents good correlation coeffi cients in most of the air quality stations.
The simulation of the Porto and Lisbon domains indicates the importance of refi ning
the grid to achieve better detail and defi nition of results. Concerning the 2010 scenario simulation of the National Emission Ceiling Program, the signifi cant decrease
in ozone concentration is noticeable when the limits are compared with the baseline
year of 2001. That decrease is more signifi cant in the urban areas of Lisbon and Porto.
However, the need is well recognized to deepen the study in order to improve results
which means, in practice, updating the emission inventory for instance and its spatiotemporal disaggregation, and refi ning the initial and boundary conditions.
Modeling results should be carefully analyzed and interpreted. The uncertainties
associated with the input data, and with the model itself, need to be properly evaluated before their predictions can be used with confi dence. Dispersion is primarily
controlled by turbulence, which is random by nature, and thus cannot be precisely
described or predicted by means of basic statistical properties. As a result, there is
spatial and temporal variability that occurs naturally in the observed concentration
fi eld. On the other hand, uncertainty in the model results could also be due to factors such as errors in the input data and model formulation. Because of the effects of
uncertainty and its inherent randomness, it is not possible for an air quality model
ever to be “perfect,” and there is always a base amount of scatter that cannot be
removed (Chang and Hanna, 2004).
Notwithstanding these limitations, air quality models should be considered as
powerful tools to predict the fate of pollutant gases or aerosols upon their release
into the atmosphere, and to infl uence decisions with signifi cant public health and
economic consequences.
ACKNOWLEDGMENTS
The authors are grateful for the collaboration of L. Salmim and would like to thank
her for her participation in the case study modeling work. Also the collaboration of
© 2010 by Taylor and Francis Group, LLC