Page 292 - Materials Chemistry, Second Edition

L1644_C07.fm Page 264 Monday, October 20, 2003 12:10 PM

Moriguchi and Terazono (2000) present an approach for Japan in which mete-
orological conditions are set to be equal for all examples. Nigge (2000) offers a
method for statistically determined population exposures per mass of pollutant that
considers short-range and long-range exposure separately and allows taking into
account dispersion conditions and population distributions using sufficiently detailed
data in a systematic way. Impact indicators are derived that depend on the settlement
structure class and the stack height. However, a general framework that is also valid
for other receptors is not proposed; the dispersion conditions are considered to be
equal for all classes in the case study and a quite simple dispersion model is used.
In this study a general applicable framework for site-dependent impact assess-
ment by statistically determined receptor exposures per mass of pollutant is proposed
that addresses some of the mentioned shortages. Receptor density and dispersion
conditions are related to form a limited number of representative generic spatial
classes, as suggested by Potting and Hauschild (1997) and recommended by Udo
de Haes et al. (1999). The basis for the classification is statistical reasoning, assuming
no threshold (Potting 2000). For each class and receptor, incremental receptor expo-
sures per mass of pollutant can be calculated. Finally, the incremental exposures are
transformed into damage estimations.
The general framework was applied to the case of population exposures due to
airborne emissions in the Mediterranean region of Catalonia, Spain. A differentiation
was made with regard to dispersion conditions, stack height and atmospheric resi-
dence time; a sophisticated dispersion model was applied and a geographic infor-
mation system (GIS) was used.

7.2 GENERAL FRAMEWORK FOR SITE-DEPENDENT
IMPACT ANALYSIS
Udo de Haes (1996) distinguished the following dimensions of impact information
that are relevant for life-cycle impact assessment (LCIA): information concerning
(1) effect, (2) fate and exposure, (3) background level, and (4) space. The first two
dimensions directly refer to the cause–effect chain. The last two can be interpreted
as additional conditions related to the processes in the considered chain.
Based on Udo de Haes’ (1996) proposal, Wenzel et al. (1997) suggested the
relation of dimensions of impact information and levels of sophistication presented
in Figure 7.1. The first dimension in this figure is in accordance with the proposal
of Udo de Haes (1996); only the exposure information has been removed. In this
way, the second dimension covers all information connected to the source (emission,
distribution/dispersion, and concentration increase). The third dimension comprises
the third dimension from Udo de Haes (1996) and, additionally, all other types of
information about the receiving environment and/or target system (background con-
centration, exposure increase, sensitivity of the target system, etc.).
With each of the three dimensions, the characterization modeling addresses
different levels of sophistication. In the effect analysis, the sophistication increases
from the use of legal threshold standards via the application of NOEC (no observable
effect concentration) to the integration of slope in the impact factors. These factors

© 2004 CRC Press LLC