1. Investigations of Methane Emissions from the Munich Oktoberfest 2018
Professorship of Environmental Sensing and Modeling
TUM Department of Electrical and Computer Engineering
Technical University of Munich
Investigations of Methane Emissions from the Munich Oktoberfest 2018
Motivation >
Approach >
Results >
Jia Chen (1), Florian Dietrich (1), Hossein Maazallahi (2,4), Dominik Winkler (1), Andreas Forstmaier (1),
Magdalena Hofmann (3), Hugo Denier van der Gon (4), and Thomas Röckmann (2)
(1) Environmental Sensing and Modeling, Technical University of Munich, Munich, Germany
(2) Institute for Marine and Atmospheric Research, Utrecht University, Utrecht, The Netherlands,
(3) Picarro B.V., ’s-Hertogenbosch, The Netherlands,
(4) Netherlands Organisation for Applied Scientific Research (TNO), Utrecht, The Netherlands
Jia Chen, Florian Dietrich et al. - Investigations of Methane Emissions from the Munich Oktoberfest 2018

2. 600 m 1000 m Motivation > Oktoberfest facts: > 6 Mio. Visitor
Professorship of Environmental Sensing and Modeling
TUM Department of Electrical and Computer Engineering
Technical University of Munich
Motivation >
Oktoberfest facts:
> 6 Mio. Visitor
Visitor density: ≈ 1 pers/m²
40 % of total energy provided by natural gas  CH4
4 km of temporal constructed gas pipelines
600 m
Approach >
Results >
1000 m
Jia Chen, Florian Dietrich et al. - Investigations of Methane Emissions from the Munich Oktoberfest 2018

3. Oktoberfest investigations 2018
Professorship of Environmental Sensing and Modeling
TUM Department of Electrical and Computer Engineering
Technical University of Munich
Oktoberfest investigations 2018
Motivation >
Wind
Approach >
Results >
Quantification of the emission number including a daily emission cycle of the Oktoberfest
Jia Chen, Florian Dietrich et al. - Investigations of Methane Emissions from the Munich Oktoberfest 2018

4. Investigations of Methane Emissions from the Munich Oktoberfest 2018
Professorship of Environmental Sensing and Modeling
TUM Department of Electrical and Computer Engineering
Technical University of Munich
Investigations of Methane Emissions from the Munich Oktoberfest 2018
Motivation >
Approach >
Results >
Jia Chen (1), Florian Dietrich (1), Hossein Maazallahi (2,4), Dominik Winkler (1), Andreas Forstmaier (1),
Magdalena Hofmann (3), Hugo Denier van der Gon (4), and Thomas Röckmann (2)
(1) Environmental Sensing and Modeling, Technical University of Munich, Munich, Germany
(2) Institute for Marine and Atmospheric Research, Utrecht University, Utrecht, The Netherlands,
(3) Picarro B.V., ’s-Hertogenbosch, The Netherlands,
(4) Netherlands Organisation for Applied Scientific Research (TNO), Utrecht, The Netherlands
Jia Chen, Florian Dietrich et al. - Investigations of Methane Emissions from the Munich Oktoberfest 2018

5. Motivation Motivation v Oktoberfest facts: > 6 Mio. Visitor
Professorship of Environmental Sensing and Modeling
TUM Department of Electrical and Computer Engineering
Technical University of Munich
Motivation
Motivation v
Oktoberfest facts:
> 6 Mio. Visitor
Visitor density: ≈ 1 pers/m²
40 % of total energy provided by natural gas  CH4
4 km of temporal constructed gas pipelines
Oktoberfest facts
Campaign 2017 setup
Campaign 2017 results
Approach >
Results >
Due to those numbers Oktoberfest could be a potential CH4 source
Jia Chen, Florian Dietrich et al. - Investigations of Methane Emissions from the Munich Oktoberfest 2018

6. Motivation – Measurement campaign 2017 setup
Professorship of Environmental Sensing and Modeling
TUM Department of Electrical and Computer Engineering
Technical University of Munich
Motivation – Measurement campaign 2017 setup
Motivation v
Ground-based remote sensing
FTIR spectrometer (EM27/SUN)
Differential column method
Covering Oktoberfest period
Oktoberfest facts
Campaign 2017 setup
Campaign 2017 results
Approach >
Results >
Jia Chen, Florian Dietrich et al. - Investigations of Methane Emissions from the Munich Oktoberfest 2018

7. Motivation – Measurement campaign 2017 results
Professorship of Environmental Sensing and Modeling
TUM Department of Electrical and Computer Engineering
Technical University of Munich
Motivation – Measurement campaign 2017 results
Motivation v
Before
During Oktoberfest
After
Oktoberfest facts
Campaign 2017 setup
Campaign 2017 results
Approach >
Results >
local time
local time
local time
Higher absolute GHG values and higher enhancement during the time of the Oktoberfest compared to the period before and after
Jia Chen, Florian Dietrich et al. - Investigations of Methane Emissions from the Munich Oktoberfest 2018

8. Measurement approach Motivation > Approach v Results >
Professorship of Environmental Sensing and Modeling
TUM Department of Electrical and Computer Engineering
Technical University of Munich
Measurement approach
Motivation >
Approach v
Measurement
Multiple Gaussian plume model
Plume selection algorithm
Forward and inverse model
Results >
Waking and biking around the area of the Oktoberfest
Sensor: Picarro GasScouter G4302 (cavity-ring-down laser spectrometer)
Jia Chen, Florian Dietrich et al. - Investigations of Methane Emissions from the Munich Oktoberfest 2018

9. Modeling approach – Multiple Gaussian plume model
Professorship of Environmental Sensing and Modeling
TUM Department of Electrical and Computer Engineering
Technical University of Munich
Modeling approach – Multiple Gaussian plume model
Motivation >
Tents as sources
Every tent as Gaussian plume source
Overlapping plumes of the 16 biggest tents with each other 0°
Approach v
Wind
Measurement
Multiple Gaussian plume model
-90°
90°
Plume selection algorithm
Forward and inverse model
Results >
180°
Framework to determine the emission number of an area source
Jia Chen, Florian Dietrich et al. - Investigations of Methane Emissions from the Munich Oktoberfest 2018

10. Modeling – Plume selection
Professorship of Environmental Sensing and Modeling
TUM Department of Electrical and Computer Engineering
Technical University of Munich
Modeling – Plume selection
Motivation >
Raw data: Possibility that several plumes exist within a single round
Plume detection by low-pass filtering (Kaiser window)
Separating into plumes
Comparing plume angle range with forward model: the plume with the higher overlapping ratio is chosen
Approach v
Measurement
Multiple Gaussian plume model
Plume selection algorithm
Forward and inverse model
Results >
Determining best suited plume for each round
Less sensitive to wind errors
Jia Chen, Florian Dietrich et al. - Investigations of Methane Emissions from the Munich Oktoberfest 2018

11. Modeling approach – Forward and inverse model
Professorship of Environmental Sensing and Modeling
TUM Department of Electrical and Computer Engineering
Technical University of Munich
Modeling approach – Forward and inverse model
Motivation >
CH4 concentrations with respect to the angle
Forward model consisting of 16 overlapping Gaussian plumes
Scale the model such that the area underneath model and measurements curve is equal
𝐸 𝑂𝑘𝑡𝑜𝑏𝑒𝑟𝑓𝑒𝑠𝑡 = 𝐸 𝑝𝑟𝑖𝑜𝑟 ⋅ 𝑘 𝑠𝑐𝑎𝑙𝑖𝑛𝑔
Approach v
Measurement
Multiple Gaussian plume model
Plume selection algorithm
Forward and inverse model
Results >
Forward model is agreeing well with the measurements
Jia Chen, Florian Dietrich et al. - Investigations of Methane Emissions from the Munich Oktoberfest 2018

12. Results – Emission distribution
Professorship of Environmental Sensing and Modeling
TUM Department of Electrical and Computer Engineering
Technical University of Munich
Results – Emission distribution
Motivation >
Total distribution
Comparison weekend vs. weekday
Approach >
Results v
Emission distribution
Source attribution
Study comparison
Gaussian shaped emission distribution
Overall emissions: 7.3 ± 0.6 µg(m²s)-1
Weekend emissions significantly higher than weekday emissions
Jia Chen, Florian Dietrich et al. - Investigations of Methane Emissions from the Munich Oktoberfest 2018

13. Results – Source attribution
Professorship of Environmental Sensing and Modeling
TUM Department of Electrical and Computer Engineering
Technical University of Munich
Results – Source attribution
Motivation >
Emission trend is not directly correlated with amount of visitors
Tents seem to be the emission sources (good correlation between measurement and model)
Biogenic human emissions are too weak to explain the total emission numbers
Approach >
Results v
Emission distribution
Source attribution
Study comparison
Emissions likely related to gas leakages, incomplete combustion, etc.
Jia Chen, Florian Dietrich et al. - Investigations of Methane Emissions from the Munich Oktoberfest 2018

14. Results – Comparison to other studies
Professorship of Environmental Sensing and Modeling
TUM Department of Electrical and Computer Engineering
Technical University of Munich
Results – Comparison to other studies
Motivation >
Boston: 0.59 µg/(m²s) (McKain et al. 2015)
Chino: 16 µg/(m²s) (Chen et al. 2016)
Human biogenic: 0.04 µg/(m²s) (Keppler et al. 2016)
Approach >
Results v
Emission distribution
Source attribution
Study comparison
Boston
city
emission
Chino
Dairy
farm
Oktober-
fest
Human
biogenic
emission
Oktoberfest is a significant methane source
Jia Chen, Florian Dietrich et al. - Investigations of Methane Emissions from the Munich Oktoberfest 2018