Chemistry Reference
In-Depth Information
which is of high importance for the determination of the influence of wood burning
emissions on ambient air quality.
2.2 Emission Factors: Examples for Austria, Germany
and Switzerland
As indicated above, emissions of PM from wood burning (i.e. emission factors) are
highly dependent on various factors. In order to provide some insight into the
variability of PM emission factors from wood burning typical for the Alpine region,
results from studies in Austria, Germany and Switzerland are summarised.
For Austria latest emission factors are only available from the late 1990s [
28
,
62
].
At that time, automatic boilers were hardly considered. The publications give
estimates for the total suspended particulates from wood burning which are
148 kg/TJ for small stoves and 90 kg/TJ for central heating systems. Emission
data from 2005 is available from studies in Germany [
29
]. In this report, device-
specific wood burning emissions factors are given, the reported emission factors
range from 57 kg/TJ for pellet stoves to 142 kg/TJ for and manual loaded boilers.
For Switzerland particulate emissions from the most common wood burning
devices are the following: open stoves 100 kg/TJ, tiled stoves 100 kg/TJ and wood
chips (
1,000 kW) 90 kg/TJ [
63
,
30
]). Further detailed and device-specific wood
emissions are given in the corresponding publications.
<
2.3 PM Source Apportionment Methods
Several studies have recently focused on determination of the contributions of the
main sources of PM and the total carbonaceous fraction (carbonaceous matter, CM)
in ambient air at receptor sites. Although the contribution of WB to PM and CM
cannot be determined directly, various methods for source apportionment based on
statistical approaches and on single or multiple specific tracers are available and
have been applied in the past. All of these methods have advantages and limitations,
a problem that is common to all of them is - that validation of the results is not
possible due to the absence of a “correct” determination method. Therefore,
plausibility checks and comparisons of results obtained with different methods
are important for quality control.
A widely used approach for estimation of source contributions at receptor sites is
receptor modelling [
31
,
32
]. In receptor models, source contributions are estimated
based on the measurements of various chemical constituents in a sufficiently large
number of ambient PM samples, often filter samples that are collected during 24 h.
Depending on the available knowledge about the main sources, CMB or multivari-
ate statistical models can be applied: CMB requires a priori knowledge of the
chemical profile of all relevant sources,
i.e.
the percentage of the chemical
