Comparison of PM<sub>10</sub> Sources Profiles at 15 French Sites Using a Harmonized Constrained Positive Matrix Factorization Approach
Receptor-oriented models, including positive matrix factorization (PMF) analyses, are now commonly used to elaborate and/or evaluate action plans to improve air quality. In this context, the SOURCES project has been set-up to gather and investigate in a harmonized way 15 datasets of chemical compoun...
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-06-01
|
| Series: | Atmosphere |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4433/10/6/310 |
| _version_ | 1828528318604902400 |
|---|---|
| author | Samuël Weber Dalia Salameh Alexandre Albinet Laurent Y. Alleman Antoine Waked Jean-Luc Besombes Véronique Jacob Géraldine Guillaud Boualem Meshbah Benoit Rocq Agnès Hulin Marta Dominik-Sègue Eve Chrétien Jean-Luc Jaffrezo Olivier Favez |
| author_facet | Samuël Weber Dalia Salameh Alexandre Albinet Laurent Y. Alleman Antoine Waked Jean-Luc Besombes Véronique Jacob Géraldine Guillaud Boualem Meshbah Benoit Rocq Agnès Hulin Marta Dominik-Sègue Eve Chrétien Jean-Luc Jaffrezo Olivier Favez |
| author_sort | Samuël Weber |
| collection | DOAJ |
| description | Receptor-oriented models, including positive matrix factorization (PMF) analyses, are now commonly used to elaborate and/or evaluate action plans to improve air quality. In this context, the SOURCES project has been set-up to gather and investigate in a harmonized way 15 datasets of chemical compounds from PM<sub>10</sub> collected for PMF studies during a five-year period (2012−2016) in France. The present paper aims at giving an overview of the results obtained within this project, notably illustrating the behavior of key primary sources as well as focusing on their statistical robustness and representativeness. Overall, wood burning for residential heating as well as road transport were confirmed to be the two main primary sources strongly influencing PM<sub>10</sub> loadings across the country. While wood burning profiles, as well as those dominated by secondary inorganic aerosols, present a rather good homogeneity among the sites investigated, some significant variabilities were observed for primary traffic factors, illustrating the need to better characterize the diversity of the various vehicle exhaust and non-exhaust emissions. Finally, natural sources, such as sea salts (widely observed in internal mixing with anthropogenic compounds), primary biogenic aerosols and/or terrigenous particles, were also found as non-negligible PM<sub>10</sub> components at every investigated site. |
| first_indexed | 2024-12-11T21:50:46Z |
| format | Article |
| id | doaj.art-04e3faf9ef7347aebf89132e5d50922a |
| institution | Directory Open Access Journal |
| issn | 2073-4433 |
| language | English |
| last_indexed | 2024-12-11T21:50:46Z |
| publishDate | 2019-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Atmosphere |
| spelling | doaj.art-04e3faf9ef7347aebf89132e5d50922a2022-12-22T00:49:28ZengMDPI AGAtmosphere2073-44332019-06-0110631010.3390/atmos10060310atmos10060310Comparison of PM<sub>10</sub> Sources Profiles at 15 French Sites Using a Harmonized Constrained Positive Matrix Factorization ApproachSamuël Weber0Dalia Salameh1Alexandre Albinet2Laurent Y. Alleman3Antoine Waked4Jean-Luc Besombes5Véronique Jacob6Géraldine Guillaud7Boualem Meshbah8Benoit Rocq9Agnès Hulin10Marta Dominik-Sègue11Eve Chrétien12Jean-Luc Jaffrezo13Olivier Favez14Univ. Grenoble Alpes, CNRS, IRD, INP-G, IGE (UMR 5001), 38000 Grenoble, FranceUniv. Grenoble Alpes, CNRS, IRD, INP-G, IGE (UMR 5001), 38000 Grenoble, FranceINERIS, Parc Technologique Alata, BP 2, 60550 Verneuil-en-Halatte, FranceIMT Lille Douai, Univ. Lille, UR SAGE, 59500 Douai, FranceUniv. Grenoble Alpes, CNRS, IRD, INP-G, IGE (UMR 5001), 38000 Grenoble, FranceUniv. Savoie Mont-Blanc, LCME, 73000 Chambéry, FranceUniv. Grenoble Alpes, CNRS, IRD, INP-G, IGE (UMR 5001), 38000 Grenoble, FranceAtmo AuRA, 69500 Bron, FranceAtmo Sud, 13294 Marseille, FranceAtmo Hauts de France, 59044 Lille, FranceAtmo Nouvelle Aquitaine, 33692 Merignac, FranceAtmo Normandie, 76000 Rouen, FranceAtmo Grand Est, 57070 Metz, FranceUniv. Grenoble Alpes, CNRS, IRD, INP-G, IGE (UMR 5001), 38000 Grenoble, FranceINERIS, Parc Technologique Alata, BP 2, 60550 Verneuil-en-Halatte, FranceReceptor-oriented models, including positive matrix factorization (PMF) analyses, are now commonly used to elaborate and/or evaluate action plans to improve air quality. In this context, the SOURCES project has been set-up to gather and investigate in a harmonized way 15 datasets of chemical compounds from PM<sub>10</sub> collected for PMF studies during a five-year period (2012−2016) in France. The present paper aims at giving an overview of the results obtained within this project, notably illustrating the behavior of key primary sources as well as focusing on their statistical robustness and representativeness. Overall, wood burning for residential heating as well as road transport were confirmed to be the two main primary sources strongly influencing PM<sub>10</sub> loadings across the country. While wood burning profiles, as well as those dominated by secondary inorganic aerosols, present a rather good homogeneity among the sites investigated, some significant variabilities were observed for primary traffic factors, illustrating the need to better characterize the diversity of the various vehicle exhaust and non-exhaust emissions. Finally, natural sources, such as sea salts (widely observed in internal mixing with anthropogenic compounds), primary biogenic aerosols and/or terrigenous particles, were also found as non-negligible PM<sub>10</sub> components at every investigated site.https://www.mdpi.com/2073-4433/10/6/310PMsource apportionmentaerosolssimilarity assessmentuncertainties |
| spellingShingle | Samuël Weber Dalia Salameh Alexandre Albinet Laurent Y. Alleman Antoine Waked Jean-Luc Besombes Véronique Jacob Géraldine Guillaud Boualem Meshbah Benoit Rocq Agnès Hulin Marta Dominik-Sègue Eve Chrétien Jean-Luc Jaffrezo Olivier Favez Comparison of PM<sub>10</sub> Sources Profiles at 15 French Sites Using a Harmonized Constrained Positive Matrix Factorization Approach Atmosphere PM source apportionment aerosols similarity assessment uncertainties |
| title | Comparison of PM<sub>10</sub> Sources Profiles at 15 French Sites Using a Harmonized Constrained Positive Matrix Factorization Approach |
| title_full | Comparison of PM<sub>10</sub> Sources Profiles at 15 French Sites Using a Harmonized Constrained Positive Matrix Factorization Approach |
| title_fullStr | Comparison of PM<sub>10</sub> Sources Profiles at 15 French Sites Using a Harmonized Constrained Positive Matrix Factorization Approach |
| title_full_unstemmed | Comparison of PM<sub>10</sub> Sources Profiles at 15 French Sites Using a Harmonized Constrained Positive Matrix Factorization Approach |
| title_short | Comparison of PM<sub>10</sub> Sources Profiles at 15 French Sites Using a Harmonized Constrained Positive Matrix Factorization Approach |
| title_sort | comparison of pm sub 10 sub sources profiles at 15 french sites using a harmonized constrained positive matrix factorization approach |
| topic | PM source apportionment aerosols similarity assessment uncertainties |
| url | https://www.mdpi.com/2073-4433/10/6/310 |
| work_keys_str_mv | AT samuelweber comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach AT daliasalameh comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach AT alexandrealbinet comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach AT laurentyalleman comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach AT antoinewaked comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach AT jeanlucbesombes comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach AT veroniquejacob comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach AT geraldineguillaud comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach AT boualemmeshbah comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach AT benoitrocq comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach AT agneshulin comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach AT martadominiksegue comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach AT evechretien comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach AT jeanlucjaffrezo comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach AT olivierfavez comparisonofpmsub10subsourcesprofilesat15frenchsitesusingaharmonizedconstrainedpositivematrixfactorizationapproach |