The German Ultrafine Aerosol Network (GUAN) has been continuously measuring the particle number size distribution (PNSD) and equivalent black carbon (eBC) mass concentration since 2009 at 17 atmospheric observatories in Germany, covering all environments from roadside to high-Alpine environments. GUAN provides us an opportunity to reduce the knowledge gaps about the spatio-temporal variation of sub-micrometer particles in different size ranges and eBC mass. These data are not only highly valuable for air pollution and health studies but also can help to reduce the uncertainties in the climate model predictions. With these long-term multi-site-category measurements, it was investigated for the first time how pollutant parameters interfere with spatial characteristics and site categories. Based on this first investigation, the long-term changes in size-resolved particle number concentrations (PNC) and eBC mass concentration were investigate to evaluate the effectiveness of the emission mitigation policies in Germany. The emission and pollutants near ground can be frequently transported to the free troposphere (FT) in the mountain areas. To identify if the decreased emissions at lower-altitudes have affected the aerosol loading in the aged, well-mixed FT air over Central Europe, the long-term trends in PNC and eBC mass concentration were analyzed for the FT and planetary boundary layer (PBL) conditions separately, at two high-Alpine observation sites. In summary, this dissertation aims to answer the following related scientific questions:
Q1: How do the sub-micrometer PNSD, PNC, and eBC mass concentration interfere with spatial characteristics and site categories? (First publication)
In the first publication (Sun et al., 2019), the spatio-temporal variability of aerosol parameters including PNSD, PNCs, and eBC mass concentration from the GUAN network were investigated for the period 2009−2014. Significant differences in the pollutant concentration were observed among various site categories. The six-year median value of sub-micrometer PNC (diameter range 20–800 nm) varies between 900 and 9000 cm−3, while median eBC mass concentration varies between 0.1 and 2.3 μg m-3 in 17 observation sites. PNCs in different size ranges were found in different spatial variabilities.
A cross-correlation between PNSD and eBC mass concentration was analyzed to detect the influence of anthropogenic sources for different site categories. The size-dependent spatial variability analysis of PNCs extracted three size intervals: a higher spatial variability size range 10–30 nm, a transition size range 30–100 nm and a lower spatial variability size range 100–800 nm. Based on the evaluated spatial variability, the measured parameters at various sites were clustered by a hierarchical clustering approach, which revealed different spatial clusters for “source-driven” and “long-range transport” parameters. This result suggests that the traditional “site category” (i.e. urban, and regional background, etc.) concerning mainly the influence of local sources cannot always catch the variation of aerosol particle mass or number concentrations. The dominant factors for various parameter are different, leading to different variability and spatial distribution. The result of spatial clustering offers a sound scientific base to compare pollutant parameters measured in different locations and environments.
By assessing the relationship between the measured parameters and geographical distance between different sites, the spatial variability of the aerosol parameters follows the “First Law of Geography” that everything is related to everything else, but near things are more related than distant things (Tobler, 1970). However, different parameters show different sensitivities on geographical distance. The analysis provides an important reference for setting up an observation network with a specific research purpose and is also useful for the regional scale dispersion models or land-use regression models.
Q2: How do the sub-micrometer PNSD, PNC, and eBC mass concentration change at a decadal scale? Have the implementations of emission mitigation policies affected the observed decadal trend? (Second publication)
In the second publication (Sun et al., 2020), long-term trends in atmospheric PNCs and eBC mass concentration for a 10 years period (2009–2018) were determined for 16 sites of the GUAN, ranging from roadside to high-Alpine. To ensure the data consistency for the trend detection, a thorough and detailed data quality check and data cleaning for the large GUAN dataset was performed. Statistically significant decreasing trends were found for 85% of the parameters and observation sites indicating an overall decreasing trend in sub-micrometer PNC (except N[10−30]) and eBC mass concentration all over Germany.
Comparing the trends of measured parameters with the long-term change in total emission, we proofed that the observed trends of PNCs and eBC mass concentrations were mainly due to the emission reduction. The detailed diurnal and seasonal trends in eBC mass concentration and PNCs further confirmed that the observed decreasing trends were largely owing to the reduced emissions such as traffic emission, residential emission, and industry emission, etc. Moreover, the inter-annual changes of meteorological conditions and long-range transport pattern were found not to be the main reason for the decreases in pollutant parameters. This study suggests that a combination of emission mitigation policies can effectively improve the air quality over large spatial scales such as Germany. Given the relative novelty of the long-term measurements (PNSD, eBC mass concentration) in a network such as GUAN, the results proved to be quite robust and comprehensive.
Q3: Have the decreased PNC and eBC mass concentration due to emission mitigation policies at the lower-altitudes affected the background air in lower FT over Central Europe? (Third publication)
In the third publication (Sun et al., 2021), the long-term change of the eBC mass concentration and size-resolved PNCs were determined and analyzed at two high Alpine stations for the period 2009-2018: Schneefernerhaus at mountain Zugspitze in Germany (ZSF, 2671 m a.s.l.) and Jungfraujoch in Switzerland (JFJ, 3580 m a.s.l.). The trend analysis was performed for the FT and PBL-influenced conditions separately, aiming to assess whether the reduced emissions at lower-altitudes over Central Europe can affect the background air in the lower FT on a large spatial scale.
The FT and PBL conditions at the two stations were segregated using the adaptive diurnal minimum variation selection (ADVS) method. The result showed that the FT condition in cold months is more prevalent than in warm months. Overall, the FT conditions frequency was ~25% and 6% in the cold and warm seasons at ZSF, respectively. At JFJ, the frequency of FT was ~45% and 10% in these two seasons, respectively.
The PNC and eBC mass concentration showed a statistically significant decrease during PBL time. The observed decreasing trends in eBC mass concentration in the PBL-influenced condition are well consistent with the reported trends in total BC emission in Germany and Switzerland. For the FT conditions, decreases in PNC and eBC mass concentration over the years was detected at both sites, suggesting the background PNC and eBC mass in the lower FT over Central Europe has decreased as well. The implementation of emission mitigation policies is the most decisive factor but the weather pattern change over Central Europe also has contributed to the decreasing trends in FT condition.:List of Figures ……………………………………………………………………………………………..I
List of Tables ..……………………………………………………………………………………………..I
Abbreviations .……………………………………………………………………………………………II
1. Introduction …………………………………………………………………………………………….1
1.1 Role of atmospheric sub-micrometer aerosol particles…...………………………………………...1
1.2 Measurement of sub-micrometer particle number size distribution, particle number concentration,
and eBC mass concentration…..……………………………………………….………………………….2
1.3 Previous long-term observations of PNSD, PNC, and eBC mass concentration…………………...4
1.4 Objectives...………………………………………………………………………………………….6
2. Data and Method…..……………………………………………………………………………………9
2.1 The German Ultrafine Aerosol Network (GUAN) …………………………………………………9
2.1.1 Measurement sites in GUAN…………………………………………………………………..10
2.1.2 Instrumental set-up.……….………………………………………………………….…………14
2.1.3 Quality assurance.………………………………………………………………………….……16
2.1.4 Data coverage…..………………………………………………………………………………..17
2.2. High-Alpine observatory Jungfraujoch (JFJ)……………………………………………………..18
2.2.1 Measurement site……….……………………………………………………………………….18
2.2.2 Instrumentation ..………………………………………………………………………………..19
2.3 Data analysis methods……………………………………………………………………………….19
2.3.1 Agglomerative hierarchical clustering….……………………………………………………...19
2.3.2 Customized Sen’s slope estimator…………………..…………………………………………...21
2.3.3 Generalized least-square regression and autoregressive bootstrap confidence intervals (GLS-
ARB)…………………………………………………………………………………………………21
2.3.4 Seasonal Mann-Kendal test…..………………….………………………………………………22
2.3.5 Back-trajectory classification method….……………………………………………………...24
3. Results and Discussion…..………………………………………………………………………….27
3.1 First publication….…………………………………………………………………………………..27
3.1.1 Variability of black carbon mass concentrations, sub-micrometer particle number concentrations and size distributions: results of the German Ultrafine Aerosol Network ranging from city street to
High Alpine locations……………………………………...………………………………………...27
3.1.2 Supporting information..……………………………………………………………………….41
3.2 Second publication…………………………………………………………………………………..45
3.2.1 Decreasing trends of particle number and black carbon mass concentrations at 16 observational
sites in Germany from 2009 to 2018…..…………………………………………………………..45
3.2.2 Supporting information...……………………………………………………………………….66
3.3 Third publication……………………………...……………………………………………………..75
3.3.1 Long-term trends of black carbon and particle number concentration in the lower free
troposphere in Central Europe…………………………………………………………………………75
3.3.2 Supporting information..…….……………………………………………………………….92
4. Summary and Conclusions..………………………………………………………………………… 95
5. Outlook….…………………………………………………………………………………………...99
Appendix A….………………………………………………………………………………………...100
Bibliography…………………………………………………………………………………………...101
Acknowledgements….…………………………………………………………………………………115
| Identifer | oai:union.ndltd.org:DRESDEN/oai:qucosa:de:qucosa:77433 |
| Date | 18 January 2022 |
| Creators | Sun, Jia |
| Contributors | Universität Leipzig |
| Source Sets | Hochschulschriftenserver (HSSS) der SLUB Dresden |
| Language | English |
| Detected Language | English |
| Type | info:eu-repo/semantics/publishedVersion, doc-type:doctoralThesis, info:eu-repo/semantics/doctoralThesis, doc-type:Text |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0031 seconds