 |
| Titel |
Overview of receptor-based source apportionment studies for speciated atmospheric mercury |
| VerfasserIn |
I. Cheng, X. Xu, L. Zhang |
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| ISSN |
1680-7316
|
| Digitales Dokument |
URL |
| Erschienen |
In: Atmospheric Chemistry and Physics ; 15, no. 14 ; Nr. 15, no. 14 (2015-07-17), S.7877-7895 |
| Datensatznummer |
250119907
|
| Publikation (Nr.) |
 copernicus.org/acp-15-7877-2015.pdf |
|
|
|
|
|
| Zusammenfassung |
Receptor-based source apportionment studies of speciated atmospheric mercury
are not only concerned with source contributions but also with the influence
of transport, transformation, and deposition processes on speciated
atmospheric mercury concentrations at receptor locations. Previous studies
applied multivariate receptor models including principal components analysis
and positive matrix factorization, and back trajectory receptor models
including potential source contribution function, gridded frequency
distributions, and concentration–back trajectory models. Combustion sources
(e.g., coal combustion, biomass burning, and vehicular, industrial and waste
incineration emissions), crustal/soil dust, and chemical and physical
processes, such as gaseous elemental mercury (GEM) oxidation reactions,
boundary layer mixing, and GEM flux from surfaces were inferred from the
multivariate studies, which were predominantly conducted at receptor sites in
Canada and the US. Back trajectory receptor models revealed potential impacts
of large industrial areas such as the Ohio River valley in the US and
throughout China, metal smelters, mercury evasion from the ocean and the Great
Lakes, and free troposphere transport on receptor measurements.
Input data and model parameters specific to atmospheric mercury receptor
models are summarized and model strengths and weaknesses are also discussed.
Multivariate models are suitable for receptor locations with intensive air
monitoring because they require long-term collocated and simultaneous
measurements of speciated atmospheric Hg and ancillary pollutants. The
multivariate models provide more insight about the types of Hg emission
sources and Hg processes that could affect speciated atmospheric Hg at a
receptor location, whereas back trajectory receptor models are mainly ideal
for identifying potential regional Hg source locations impacting elevated Hg
concentrations. Interpretation of the multivariate model output to sources
can be subjective and challenging when speciated atmospheric Hg is not
correlated with ancillary pollutants and when source emissions profiles and
knowledge of Hg chemistry are incomplete. The majority of back trajectory
receptor models have not accounted for Hg transformation and deposition
processes and could not distinguish between upwind and downwind sources
effectively. Ensemble trajectories should be generated to take into account
the trajectory uncertainties where possible. One area of improvement that
applies to all the receptor models reviewed in this study is the greater
focus on evaluating the accuracy of the models at identifying potential
speciated atmospheric mercury sources, source locations, and chemical and
physical processes in the atmosphere. In addition to receptor model
improvements, the data quality of speciated atmospheric Hg plays an equally
important part in producing accurate receptor model results. |
| |
|
| Teil von |
|
|
|
|
|
|
|
|