Urban eddy covariance measurements reveal significant missing NOx emissions in Central Europe

Abstract

Nitrogen oxide (NOx) pollution is emerging as a primary environmental concern across Europe. While some large European metropolitan areas are already in breach of EU safety limits for NO2, this phenomenon does not seem to be only restricted to large industrialized areas anymore. Many smaller scale populated agglomerations including their surrounding rural areas are seeing frequent NO2 concentration violations. The question of a quantitative understanding of different NOx emission sources is therefore of immanent relevance for climate and air chemistry models as well as air pollution management and health. Here we report simultaneous eddy covariance flux measurements of NOx, CO2, CO and non methane volatile organic compound tracers in a city that might be considered representative for Central Europe and the greater Alpine region. Our data show that NOx fluxes are largely at variance with modelled emission projections, suggesting an appreciable underestimation of the traffic related atmospheric NOx input in Europe, comparable to the weekend-weekday effect, which locally changes ozone production rates by 40%.

Full text

1
Scientific RepoRts | 7: 2536 | DOI:10.1038/s41598-017-02699-9
www.nature.com/scientificreports
Urban eddy covariance
measurements reveal signicant
missing NOx emissions in Central
Europe
T. Karl
1, M. Graus1, M. Striednig1, C. Lamprecht1, A. Hammerle2, G. Wohlfahrt
2, A. Held3,
L. von der Heyden3, M. J. Deventer4, A. Krismer5, C. Haun6, R. Feichter7 & J. Lee8
Nitrogen oxide (NOx) pollution is emerging as a primary environmental concern across Europe.
While some large European metropolitan areas are already in breach of EU safety limits for NO2,
this phenomenon does not seem to be only restricted to large industrialized areas anymore. Many
smaller scale populated agglomerations including their surrounding rural areas are seeing frequent
NO2 concentration violations. The question of a quantitative understanding of dierent NOx emission
sources is therefore of immanent relevance for climate and air chemistry models as well as air pollution
management and health. Here we report simultaneous eddy covariance ux measurements of NOx,
CO2, CO and non methane volatile organic compound tracers in a city that might be considered
representative for Central Europe and the greater Alpine region. Our data show that NOx uxes are
largely at variance with modelled emission projections, suggesting an appreciable underestimation of
the trac related atmospheric NOx input in Europe, comparable to the weekend-weekday eect, which
locally changes ozone production rates by 40%.
e nitrogen cycle1 is essential for maintaining the oxidizing capacity of the atmosphere and regulating ozone
in the lower atmosphere2. Perturbations due to rapid industrialization and agricultural activities have led to a
signicant increase of atmospheric nitrogen oxides (NOx) during the 20th century3. A regionally intense buildup
of photochemical smog due to the presence of nitrogen oxides, CO and non-methane volatile organic com-
pounds (NMVOC) was rst identied in the US and attributed as the main cause of severe ozone pollution in
many areas4. Decades of subsequent research activities ranging from detailed laboratory5, 6 and smog chamber7–9
studies to large scale eld campaigns10–12 have led to a reasonably good mechanistic understanding of the for-
mation of tropospheric ozone, which is characterized by a complex nonlinear relationship between NOx and
reactive carbon species13. is interdependency gives regulators two key strategies to mitigate ozone pollution.
e eectiveness to control ozone thereby very much depends on the ratio between ambient OH reactivity and
NOx concentrations14, which can be described by relatively simple analytical relationships15. e development
of mechanistic regional16 and global air chemistry models17 has further given regulators and scientists powerful
tools to study tropospheric ozone formation16, where the mitigation of NOx emissions has emerged as one of the
key air pollution control strategies for ozone18–20 and more recently also for particulate matter with a diameter
of 1 μm or less (PM1)21. Due to the toxicity, nitrogen dioxide (NO2) is also regulated as a hazardous air pollutant
itself22. For example, in Europe regulatory action under the EU ematic Strategy on Air Pollution is in place to
limit urban street canyon NO2 concentrations to 40 μg/m3 per year (or 200 μg/m3/h on less than 18 days/year)23.
Current trends across European air quality networks show that regulatory thresholds of NO2 are violated at many
stations, which does not seem to be limited to large population centers anymore (ref. 24, SI). In fact many rural
1Institute of Atmospheric and Cryospheric Sciences, University of Innsbruck, Innsbruck, Austria. 2Institute of Ecology,
University of Innsbruck, Innsbruck, Austria. 3Atmospheric Chemistry, University of Bayreuth, Innsbruck, Germany.
4Department of Geography, University of California, Berkeley, USA. 5Abteilung Waldschutz, Amt der Tiroler
Landesregierung, Innsbruck, Austria. 6Abteilung Geoinformation, Amt der Tiroler Landesregierung, Innsbruck,
Austria. 7Amt für Verkehrsplanung, Umwelt, Magistrat III Stadt Innsbruck, Innsbruck, Austria. 8National Centre for
Atmospheric Science and Department of Chemistry, University of York, York, UK. Correspondence and requests for
materials should be addressed to T.K. (email: thomas.karl@uibk.ac.at)
Received: 3 November 2016
Accepted: 19 April 2017
Published: xx xx xxxx
OPEN

www.nature.com/scientificreports/
2
Scientific RepoRts | 7: 2536 | DOI:10.1038/s41598-017-02699-9
areas and smaller towns see NO2 concentration levels rivaling those of large metropolitan areas. Owing to the
spatiotemporal variability and uncertainty of dierent anthropogenic NOx sources, it is dicult to attribute emis-
sion uncertainties to specic sectors in complex bottom-up emission inventories25 or top-down remote sensing
assessments26. Recently evidence has accumulated that rapid shis in transportation fuels can have signicant
impacts on air quality27, 28. In Europe for example the question about the increasing penetration of Diesel cars
raises concerns as to what extent such a technological change has been counterproductive to mitigating atmos-
pheric NO2 pollution under new emission regulation standards19, 29. e United States environmental protection
agency’s (US EPA) notice of violation of the Clean Air Act to a German automaker regarding Diesel engines has
sparked a number of new real world driving (RDE) emission tests across Europe, which show signicant man-
ufacturer and vehicle specic variability30, 31. ese new data suggest that the impact on up-scaled average eet
emissions needed for accurate air quality predictions remains unclear32.
A number of urban ux measurement sites for energy and CO2 have been established highlighting their
potential for surface-atmosphere exchange studies33, 34. In contrast, similar measurements for reactive gases are
oen still quite limited, owing to the complexity of the required measurement systems. A set of recently con-
ducted urban ground based and airborne NMVOC ux measurements revealed the usefulness to test bottom-up
emission inventories and revealed signicant discrepancies for some species35–39. Urban ux measurements for
NOx are even more scarce32, 40, 41 indicating that constraints on emission sources in urban areas can be quite
uncertain. Here we improve upon existing work, by simultaneously measuring NOx, selected tracer NMVOCs,
CO and CO2 leading to a well constrained ux dataset, that allows testing our understanding of prominent NOx
emission sources.
Results
Obtaining ensemble average statistics on eet emissions by eddy covariance ux measure-
ments. A comprehensive set of eddy covariance measurements for NOx, marker NMVOC, CO and CO2 at
an urban location allows a direct comparison of relative ux ratios with bottom-up emission sources. e study
site located in Innsbruck (N 47°15′51.50″, E 11°23′6.77″) at the center of the Inn valley, represents one of the
most strategically important Alpine crossing points for the transport of goods between Northern and Southern
Europe. Each year approximately 6 million vehicles42 pass through the east-west facing valley, which is about
10 km wide surrounded by mountain ridges about 2.5 km high. e valley topography leads to a very predictable
and pronounced wind system characterized by a topographic amplication factor (TAF) of about 343. Due to
the combination of signicant trac induced NOx emissions and increasingly stringent NO2 limit values, the
area is in non-attainment. Local authorities are facing legal proceedings by the European Commission for their
failure to control excessive levels of nitrogen oxides (Fig.S2), similar to many areas across Europe23. Tracer ux
relationships allow investigating to what extent urban emissions are caused by (a) trac, (b) urban residential
and (c) biomass burning/biofuel activities. Figure1 shows the diurnal evolution of Weekday (Tuesday-ursday)
and Sunday NOx uxes and concentrations along with mean trac count data at the site during the measurement
campaign (July – October, 2015). Median measured midday NOx mixing ratios in Innsbruck are comparable to
values reported for central London (10–14 ppbv), while corresponding observed uxes are about a factor of 3–4
lower (i.e. 3000–4000 ng/m2/s vs 700–1440 ng/m2/s)32. ese observations are consistent with the idea of an inten-
sication of air pollution proportional to TAF, and a corresponding eective lower air volume that pollutants are
being mixed into in steep valleys (Fig.S2, ref. 43). is comparison likely indicates that a much stronger reduction
of NOx emissions from the transport sector would be required in the Alps than for example in London in order
to achieve current air pollution standards along one of the busiest EU transport corridors across the Alps. Several
lines of evidence exclude signicant presence of biomass burning during the present study. e average ratio
between benzene and toluene uxes exhibited a typical value (3.3 ± 0.7; R2 = 0.85) characteristic for urban emis-
sion sources, dominated by fossil fuel combustion and evaporative/cold-start emissions. e correlation between
acetonitrile and benzene, toluene, NOx or CO2 uxes was low with an R2 of 0.07, 0.08, 0.02 and 0.06 respectively.
We also did not observe signicant excursions of other species recently suggested as additional biomass burning
markers44 such as furfural and furan showing a correlation coecient of R2 < 0.2 above their background uxes.
We observed an excellent correlation between CO2, benzene and NOx uxes (CO2/NOx: R2 = 0.86; benzene/NOx:
R2 = 0.75). e covariance between between NOx and CO2 (benzene) uxes yielded values of 0.91 (0.86). We
interpret these observations such that benzene, NOx and CO2 emissions are dominated by road trac with con-
tributions from residential combustion sources.
Benchmarking urban source emission ratios and inventories. To gain a more quantitative insight,
we investigated ux ratios between NOx and CO2 (FNOx/FCO2; Fig.2). e advantage of this approach is that it
allows determining the actual ensemble average of dierent emission sources based on measured ux ratios,
similar to an end-member un-mixing regression analysis. is allows us to compare our measurements to rela-
tive emission strengths reported in emission models and inventories. For large scale emission inventories (e.g.
grid cells > 1 km2) this approach also circumvents uncertainties related to assumptions of various downscaling
approaches. e observed FNOx/FCO2 ratios follow a diurnal cycle showing a ~40–50% variation throughout a day,
which reects the pronounced uctuation of trac activity across the city (e.g. ranging from about 97 vehicles/h
at night to 890 vehicles/h during daytime at a trac count station within the ux footprint). Since we can exclude
signicant industrial emissions within the ux footprint as well as biomass burning, the variation of FNOx/FCO2
should exhibit the characteristic behavior of city scale sources comprised of (1) a combination of vehicular emis-
sions and (2) residential/domestic combustion sources (e.g. oil and gas heating units). A minimization routine
(SI) allowed un-mixing these two end-members of the compositional data, reecting the actual emission ratios
for trac and urban residential combustion sources (Fig.2). e tted model (SI) can reproduce the diurnal cycle
and activity factors reasonably well, leading to a NOx/CO2 emission ratio for trac of 4.2(±0.3) × 10−3 ([mg/

www.nature.com/scientificreports/
3
Scientific RepoRts | 7: 2536 | DOI:10.1038/s41598-017-02699-9
m2/h] NOx/[mg/m2/h] CO2), and 0.20(±0.05) × 10−3 ([mg/m2/h] NOx/[mg/m2/h] CO2) for residential combus-
tion sources. ese calculated ratios are also depicted in Fig.2 by horizontal shaded blue and green lines. e
activity factors suggest that the ratio is dominated by trac, comprising about 85% of the activity averaged over
the entire day (and >95% during peak trac). We also investigated NOx/benzene ux source ratios revealing
Figure 1. Statistical plot of measured NOx uxes, mixing ratios and trac count data. e center dot shows
the ensemble median, where the box around it represents one standard deviation and whiskers the 25 and 75%
percentile. Individual extreme values are plotted as open circles. Panels A, B and C represent weekdays (i.e.
TUE-THU; composite of 609 individual data points) and panels D, E and F depict Sundays (193 individual data
points).
Figure 2. Diurnal cycle of median trac count data (black line - le axis), measured (blue circles) and model
tted (dashed blue line) mass ux ratio of NOx/CO2 (right axis). e corresponding calculated end members for
trac and residential combustion ratios are indicated by blue and green horizontal lines. e shading reects
one standard deviation. In addition colored dashed lines show predictions using xed emission ratios from
COPERT (magenta) and ACCMIP (red).

www.nature.com/scientificreports/
4
Scientific RepoRts | 7: 2536 | DOI:10.1038/s41598-017-02699-9
comparable dierences as observed for FNOx/FCO2. e CO2 ux weekend-weekday eect and CO/CO2 ratios close
to a recent road tunnel study45, all imply that a biogenic inuence on CO2 uxes due to photosynthetic uptake
or respiration can be considered negligible at this site. e obtained NOx/CO2 ux ratio for trac is signicantly
larger than predicted by a number of state of the art emission inventories and emission standards (Table1).
Generally, we observe 50–70% higher NOx emissions relative to CO2 from road trac than what is calculated
with the most recent trac emission models. In these detailed bottom – up models, mobile source emissions
are treated for dierent engine sizes and fuels, that, in the past, relied on standardized protocols obtained in test
facilities, but were recently updated based on a number of RDE tests30, 31. Exhaust from modern gasoline pow-
ered engines, despite higher ignition temperatures than those powered by Diesel, can be eectively treated for
NO along with NMVOC (and CO) using a three way catalytic (TWC) converter. TWC treatment can lead to a
10 fold reduction of NO. is has been hard to achieve for Diesel powered cars, which nowadays mostly rely on
selective catalytic reduction due to the high air to fuel ratio during combustion. Generally, dierent combustion
and exhaust treatment characteristics result in signicantly higher NOx/CO2 emission ratios for Diesel powered
cars than for gasoline. e modelled eet average contribution suggests that at least 90% of urban NOx emissions
should originate from Diesel driven vehicles at the present location. 85% is modelled to be emitted by the passen-
ger car eet based on the COPERT model and TRACCS database (ref. 46, SI). e current Austrian passenger car
eet comprises about 50% Diesel cars and the percentage across Europe grew at a substantially faster rate com-
pared to the US47. Based on our measurements the current average Austrian car eet emits about 36 times more
NOx per CO2 molecule compared to the US TIER II emission standard and a factor of 8–10 more than Euro 6
emission standards31. Factoring in dierences in fuel economy between the European and US car eet, this would
equate to about an order of magnitude more NOx emissions per travelled distance compared to newly introduced
emission standards. How comparable are these results to other European countries? Diesel engines dominate
the European passenger car market: 55% of all newly registered vehicles in the EU were powered by Diesel-fuel
in 201242; the penetration of Diesel cars of the two largest European economies bordering the Alps ranges from
30% (Germany) to 70% (France)42, more than an order of magnitude higher than in the US47. When comparing
measured NOx/CO2 ux ratios with current inventories used for IPCC atmospheric chemistry/climate48 and air
quality models49 we obtain a discrepancy up to about a factor of 3–4 (Table1). Incidentally, a recent comprehen-
sive model evaluation has suggested signicant discrepancies between regionally modelled and observed surface
NOx concentrations that seemed worse (e.g. by a factor of 2 during summers) over Europe than over the US25.
While a number of uncertainties (agricultural emissions, vertical mixing, biomass burning emissions, deposition)
can potentially result in modelled concentration biases in these models, our measurements suggest that road
transport related biases likely contribute signicantly to these discrepancies as they account for about half of the
total NOx emissions across Europe50.
What is the impact on atmospheric chemistry? e weekend – weekday eect (Fig.1) allows to gain
insight into changing NOx and NMVOC uxes on ozone production in more detail. In Austria, heavy duty vehi-
cle trac (trucks heavier than 7.5t and all road trains) is banned between Saturday 15:00 and Sunday 22:00 and
on public holidays between midnight and 22:00. Trac count data generally show a pronounced dierence in
Measurement/inventory ratio or
Measurement/emission standard ratio NOx/CO2 NOx/CO NOx/benzene
INNAQS/COPERT1,#
..
.
1 716
18
..
.
39
31
47
INNAQS/HBFA3.22
..
.
1 514
16
N/A
INNAQS/ACCMIP (trac)3
..
.
4 028
53
*
..
.
2 117
25
INNAQS/EMEP (trac)#
..
.
2 926
33
*N/A
INNAQS/US Tier II4
..
.
36 033 4
38 6
N/A
NOx/CO2 PC and LCV < 1305 kg NOx/CO2 LCV 1305 kg–3500 kg
INNAQS/Euro65Diesel
..
.
4 743
50
..
.
4 541
48
Petrol .
.
.
126
11 7
13 5
.
.
.
195
18 1
20 9
INNAQS/Euro55Diesel
..
.
2 119
23
..
.
2 018
21
Petrol .
.
.
126
11 7
13 5
.
.
.
195
18 1
20 9
INNAQS/Euro45Diesel
..
.
1 514
16
..
.
1 413
15
Petrol .
.
.
4 744
50
.
.
.
145
13 5
15 5
INNAQS/Euro35Diesel
..
.
07507
08
..
.
0 9085
10
Petrol
..
.
2 523
27
..
.
180
16 7
19 3
Table 1. Measurement – inventory comparison (i.e. measured/modelled ux (emission) ratio). e uncertainty
range is given by sub- and superscripts. PC: passenger cars; LCV: light commercial vehicle. 1COPERT
emission model (SI). 2HBFA 3.2 (SI). Comparison with HBFA 3.3, that was published during the copy editing
phase,yielded an averagebias of 1.2. 3ACCMIP – (SI). 4US EPA22. 5EEA24. *For inventories that do not explicitly
report CO2, we converted data using the measured midday range of CO to CO2 ux ratios (3.6 to 4.6 ppbv/
ppmv), which fall close to a recent evaluation based on a road tunnel study45. #Data are trend adjusted for 2015
according to GAINS (http://gains.iiasa.ac.at/models/)23.

www.nature.com/scientificreports/
5
Scientific RepoRts | 7: 2536 | DOI:10.1038/s41598-017-02699-9
driving habits resulting in a factor of 1.9 ± 0.2 lower vehicle counts on Sunday than on weekdays. We calculated
typical ozone production rates for midday-aernoon conditions (11–16 h LT), when photochemistry peaks. e
corresponding NOx uxes are a factor of 2.1 ± 0.2 lower on Sundays, closely matching observations of vehi-
cle activity. Benzene and toluene uxes, representing the variation of anthropogenic NMVOC emissions, were
lower by a factor of 1.8 ± 0.3 and 2.0 ± 0.3 respectively. CO2 uxes changed by a similar factor of 2.3 ± 0.5. e
weekend-weekday comparison provides an independent conrmation that these pollutant emissions are domi-
nated by trac activity during the day. Average NOx concentrations are a factor of 2.5 ± 0.2 lower during Sundays.
Incidentally, the observed weekend – weekday reduction is comparable to the observed measurement-inventory
discrepancies or the eect if an entire car eet was converted from a Euro 5 (0.18 g/km) to a Euro 6 (0.08 g/km)
NOx emission standard. Our measurements therefore allows us to benchmark such a hypothetical regulatory
action in the real atmosphere. Changes in local ozone production are calculated following procedures outlined
before15, 51, where the sensitivity of local ozone production can be approximated by the ratio of radical termina-
tion (LN) processes (e.g. NO2 + OH) and photochemical radical production (Q):
δδ δ
δδ δ
≈
−−
+−
L
Q
ONOJ
ONMVOC NO
22
3(1)
Nxx
xx
Here the δ symbol indicates the relative change between weekday and Sunday, J represents the photolysis rates,
and Ox = O3 + NO2. All terms on the right side can be inferred from measurements of the weekend eect, where
the anthropogenic change of NMVOCs is assumed to follow benzene. ere is evidence of a non-neglegible bio-
genic NMVOC (BVOC) presence at the site (e.g. 20–50% of the NMVOC reactivity) and these BVOCs do not
exhibit any anthropogenically related variation between weekdays and weekend. We apportioned the change of
NMVOC reactivity therefore into a biogenic and anthropogenic part using data from the PTR-QiTOF-MS instru-
ment. To achieve this, we estimated the total anthropogenic NMVOC reactivity from known urban concentration
ratios52, 53 scaled to benzene and compared this to the measured reactivity of BVOC, in particular the reactive
biogenic marker species isoprene and monoterpenes. is allowed to obtain upper (δNMVOC = δbenzene) and
lower (δNMVOC = δbenzene × [reactivity anthropogenic NMVOC]/[reactivity BVOC]) bounds for LN/Q. It can
further be shown15 that the sensitivity of local ozone production can be related to LN/Q according to:
=−
−
dPO
dNO
ln ()
ln()
1
1
(2a)
x
L
Q
L
Q
3
3
2
1
2
N
N
=−
dPO
dNMVOC
ln ()
ln()
1
(2b)
L
QL
Q
3
1
2
1
2
N
N
e calculated range for LN/Q delimits a ratio between 0.82 and 0.92 (theoretical maximum = 1) and places the
site in a NOx inhibited chemical regime for ozone production during weekdays15, 51. Incidentally these aernoon
values are systematically higher than reported for a rural site in Northern Italy with comparable NOx concentra-
tions, and more closely follow conditions reported for Milan in 199854. Our ux data clearly demonstrate that the
weekend eect of ozone production found in Innsbruck is largely a direct result of lower weekend NOx emissions.
As a consequence of the dominating nitrogen chemistry for radical loss calculated via eqs1 and 2, local ozone
production will therefore increase by 39% to 70% proportional to a reduction of NOx mixing ratios in Innsbruck.
It will decrease between 70% to 85% proportional to a reduction of NMVOC or CO. e impact on gross ozone
production (P(O3)) was also investigated independently using the Leeds Master Chemical Mechanism (MCM)
as a photochemical box model (refs 13 and 55, SI, Fig.3). e model simulates an increase of P(O3) by 40% (from
2.5 to 3.5 ppbv/h) on Sundays due to a 2 fold reduction of NOx, giving an explanation for the observed weekend
eect, which results in an increase of local aernoon ozone concentrations up to 24% (Fig.4). e model simula-
tion suggests a maximum increase of P(O3) up to 3 times (2.5 ppbv/h to 7.8 ppbv/h), if solely NOx concentrations
were decreased from current levels to about 2 ppbv (Fig.3). Below 2–4 ppbv of NOx, ozone production would
enter the NOx sensitive regime and gradually lead to decreasing P(O3). Next, we setup the MCM as a diurnally
constrained 0-dimensional diluting box model (SI) to study the sensitivity with respect to changes in modelled
ozone concentrations between weekdays and weekends. e model is thereby fully constrained by methane, CO,
measured NMVOC, NO, photolysis rates, and PBL height along with ancilliary meteorological variables (e.g. tem-
perature, humidity). NO2 and ozone are initialized by their measured initial concentrations, but are then allowed
to freely adjust during the model run. We chose a spin-up time of 3 days and used model results from the last day
of simulation to compare with measured ozone concentrations. Figure4 shows the diurnal patterns of modelled
and observed ozone concentrations. Observed (modelled) peak ozone concentrations are 49 ± 2 (52) ppbv and
43 ± 2 (44) ppbv on weekends and weekdays respectively. Modelled and observed weekend-weekday dierences
averaged over daytime hours correspond to 7.2 ± 1 and 7.6 ± 1.2. While the model reproduces the general diur-
nal ozone cycle and midday peak reasonably well, it underestimates absolute nighttime concentrations exhibiting
a stronger diurnal cycle. We attribute this shortcoming mainly to poor knowledge and constraints on entrainment
and advection processes that dominate observed nighttime distributions of ozone. While this approach has its
limitations, we believe it does a sucient job for investigating relative daytime changes of ozone concentrations
caused by the weekend eect. On average, daytime weekday concentrations are about 7–8 ppbv lower than on
weekends. ese results can be compared to two quite contrasting areas. In Mexico City51, ozone concentrations
change very little despite similar relative variations of NOx mixing ratios between weekday and weekend, which

www.nature.com/scientificreports/
6
Scientific RepoRts | 7: 2536 | DOI:10.1038/s41598-017-02699-9
can be attributed to a completely NOx saturated environment. On the other hand, a quite pronounced variation is
found in the southern California Air basin56, where average ozone concentrations are observed to change almost
twice as much compared to the present study. We interpret these observations such that they reect the change
in ozone production eciency (Fig.4), when going from environments with NOx saturated conditions to more
VOC limited conditions. In this context a number of explanations of the weekend eect on ozone mixing ratios
are oen discussed51: (1) dierent timing of NOx emissions on weekends and associated chemical repartitioning,
(2) carryover of previous day pollutants at the surface and alo, (3) higher weekend VOC emissions, (4) higher
weekend photolysis frequencies due to lower aerosol loadings, (5) changes in biogenic emissions due to a dierent
radiation regime caused by lower aerosol loadings and (6) lower overall weekend NOx emissions. We show for the
rst time, that, for Innsbruck, at least 85% of the NOx concentration change, leading to a 7–8 ppbv ozone increase
on weekends, can be directly associated with a change of the overall local atmospheric NOx ux.
Discussion
Continued urbanization in conjunction with rapid technological changes in the mobility sector poses a chal-
lenge for accurate up to date predictions of pollutant emissions. By constraining the actual uxes of nitrogen
oxides, NMVOC, CO2 and CO into the atmosphere our measurements provide an observationally based expla-
nation why NOx concentrations have hardly declined since the introduction of EURO 3 emission standards in
Central Europe. While the technological shi towards Diesel passenger cars might have helped curb CO2 emis-
sions through better fuel economy compared to gasoline powered cars in the past, it created a widespread prob-
lem of NO2 pollution across Europe23 that does not seem to be exclusively limited to the largest metropolitan
and industrialized areas. e presented ux measurements indicate that trac related NOx emissions in current
operational air quality models can be signicantly underestimated by up to a factor of 4 across countries exhib-
iting a sizeable fraction of Diesel powered cars in their eet. As Diesel fuels (including bio-diesel) could account
for 70% of the growth in transportation fuels by 2040, with signicant demand in Asian markets according to
industry projections57, a better understanding of the uncertainty in associated changes of NOx uxes and ozone
chemistry will therefore be important for future environmental impact studies. Our measurements show that pro-
jected signicant decreases in European NOx emissions from the mobile transport sector will lead to conditions
improving NO2 exposure limits, but could locally increase ozone levels on the short term. Concomitant reduction
measures for NOx and NMVOC (CO) might therefore still prove most eective to avoid parallel increases of
local ozone levels due to new NOx emission standards. is might be particularly important in areas where top-
ographic amplication can lead to a stronger accumulation of air pollutants than over at land and be a relevant
consideration for mountainous mega-cities58. Using the observed weekend eect as proxy for underestimated
NOx emissions (i.e. a factor 2–4 dierence), models would overestimate P(O3) by 30–40% under the observed
NOx inhibited – VOC limited regime and underestimate P(O3) downwind, once NOx concentrations fall below
2–4 ppbv. Here we demonstrate that parallel ux measurements of a wide range of chemical species can be used to
benchmark urban emission sources, complement traditional approaches and signicantly improve uncertainties
inherent to bottom-up scaling in atmospheric chemistry models.
Figure 3. Calculated ozone production rates (solid lines) and derivatives (dashed lines) as a function of NOx.
Blue and cyan represent base cases for the study site, taking into account changes in O3, NOx and NMVOCs
concentrations. e solid black line follows the observed weekend-weekday trajectory.

www.nature.com/scientificreports/
7
Scientific RepoRts | 7: 2536 | DOI:10.1038/s41598-017-02699-9
Methods
Instrumentation. NMVOC. A PTR-QiTOF instrument (Ionicon, Austria) was operated in hydronium
mode at standard conditions in the dri tube of 112 Townsend. e instrument was set up to sample ambient air
from a turbulently purged “3/8” Teon line. Every seven hours, zero calibrations were performed for 30 minutes
providing VOC free air from a continuously purged catalytical converter though a setup of soware controlled
solenoid valves. In addition, known quantities of a suite of VOC from a 1 ppm calibration gas standard (Apel
& Riemer, USA) were periodically added to the VOC free air and dynamically diluted into low ppbv mixing
ratios. NOx: A dual channel chemiluminescence instrument (CLD 899 Y; Ecophysics) was used for NO and NOx
measurements. e instrument was equipped with a metal oxide converter operated at 375C. e instrument
was operated in ux mode acquiring data at about 5 Hz, similar to measurements performed over a pasture59
and forest60. A NO standard was periodically introduced for calibration. Zeroing was performed once a day
close to midnight. CO2, H2O: A closed path eddy covariance system (CPEC 200; short inlet, enclosed IRGA
design; Campbell Scientic) measured three dimensional winds along with CO2 and H2O. An additional 3D sonic
anemometer (CSAT3; Campbell Scientic) was available for turbulence measurements at an alternative height
level (Fig.S3). Calibration for CO2 was performed once a day. O3: Ozone concentrations were obtained from a
closed-path UV photometric analyser (APOA-370, Horiba, Japan); CO: CO measurements were available for a
limited amount of time in August 2015. Ambient mole fractions of carbon monoxide (CO) were measured with
a quantum cascade laser spectrometer (CWQC-TILDAS-76-D, Aerodyne, USA) with a 76 m path length optical
cell at a wavenumber of ca. 2190 cm−1. e QCL was operated at a pressure of ca. 4 kPa using a built-in pressure
controller and temperature of the optical bench and housing controlled to 35 °C. Fitting of absorption spectra
at 2 Hz, storing of calculated dry mole fractions, switching of zero/calibration valves, control of pressure lock,
correction for band broadening and other system controls were realized by the TDLWintel soware (Aerodyne,
USA) run on a PC synchronized in time with the system collecting the anemometer data using the NTP soware
(Meinberg, Germany).
Eddy covariance data analysis. e eddy covariance method is derived from the scalar budget equation aer
Reynolds decomposition, and in its simplest form for horizontally homogeneous ows normal to the surface,
where the mean vertical motion of wind (
w
) can be considered 0, relates the measured surface-atmosphere
exchange ux (F) to the covariance between vertical wind and concentration uctuation according to:
=′′wcF,
where w′ represents the vertical uctuation of wind speed, and c′ the concentration uctuation. Brackets denote
the averaging interval. e ensemble average used here is 30 minutes. Fluxes were selected according to standard
quality control criteria, such as raw data despiking, correcting for high and low pass ltering biases, applying a
stationarity test and test on developed turbulent conditions (e.g. u* ltering)61. In addition we parsed the data to
make sure that the ux footprint would reect a representative urban area (SI).
Figure 4. Observed and modelled changes in ozone concentrations on weekdays (Tuesday – ursday) and
weekends (Sunday). Panel A depicts a statistical boxplot showing weekend-weekday dierences during daytime,
which are represented by the colored sections in panels B and C.

www.nature.com/scientificreports/
8
Scientific RepoRts | 7: 2536 | DOI:10.1038/s41598-017-02699-9
References
1. Fowler, D. et al. Eects of global change during the 21st century on the nitrogen cycle. Atmos. Chem. Phys. 15, 13849–13893 (2015).
2. Mons, P. S. et al. Tropospheric ozone and its precursors from the urban to the global scale from air quality to short-lived climate
forcer. Atmos. Chem. Phys. 15, 8889–8973 (2015).
3. Intergovernmental Panel on Climate Change (IPCC) “Climate Change 2014: Synthesis eport. Contribution of Woring Groups I,
II and III to the Fih Assessment eport of the Intergovernmental Panel on Climate Change”, (IPCC, 2014, Geneva, https://www.
ipcc.ch/report/ar5/syr/).
4. Haagen-Smit, A. J., Bradley, C. E. & Fox, M. M. Ozone formation in photochemical oxidation of organic substances. Indust. and
Engineering Che mistr y 45, 2086–2089 (1953).
5. Calvert, J. G. Gaseous reactions. Annual Review of Physical Chemistry 11, 41–64 (1960).
6. Atinson, . inetics and mechanism of the gasphase reactions of the hydroxylradical with organic compounds under atmospheric
conditions. Chem. Rev. 86, 69–201 (1985).
7. Leone, J. A., Flagan, . C., Grosjean, D. & Seinfeld, J. H. An outdoor smog chamber and modeling study of toluene-NOx
photooxidation. Int. J. of Chem. Kinetics 17, 177–216 (1985).
8. Martin, M., Pons, M., Wirtz, ., Etzorn, T. & Senzig, J. An outdoor smog chamber study of the gas-phase chemistry of toluene and
xylenes in NOx/air systems, in e oxidizing capacity of the troposphere (Proceedings of the seventh European symposium on
physicochemical behavior of atmospheric pollutants, European Commission, Luxembourg, 1996) pp. 350–354.
9. arl, M. et al. inetic Study of the OH-isoprene and O3-isoprene reaction in the atmosphere simulation chamber SAPHI. GRL. 31,
doi:10.1029/2003GL019189 (2004).
10. Meagher, J. F., Cowling, E. B., Fehsenfeld, F. C. & Parhurst, W. J. Ozone formation and transport in the southeastern United States:
Overview of the SOS Nashville Middle Tennessee Ozone Study. J. Geophys. Res. 103, 22213–22223 (1998).
11. idley, B. A. & obinson, E. e Mauna-Loa-Observatory Photochmsitry Experiment. J. Geophys. Res. 97, 10285–10290 (1992).
12. Jacob, D. J. et al. Transport and Chemical Evolution over the Pacic (TACE-P) aircra mission: Design, execution, and rst results.
J. Geophys. Res. 108, 1–19 (2003).
13. Jenin, M. E., Saunders, S. M. & Pilling, M. J. The tropospheric degradation of volatile organic compounds: A protocol for
mechanism development. Atmos. Environ. 31, 81–104 (1997).
14. Brune, W. H. et al. Ozone production chemistry in the presence of urban plumes, Faraday Discussions, doi:10.1039/c5fd00204d
(2015).
15. leinman, L. I. et al. Dependence of ozone production on NO and hydrocarbons in the troposphere. GRL 24, 2299–2302 (1997).
16. Im, U. et al. Evaluation of operational on-line-coupled regional air quality models over Europe and North America in the context of
AQMEII phase 2. Part I: Ozone. Atmos. Environm. 115, 404–420 (2015).
17. Ehhalt, D. & Prather, M. “Atmospheric Chemistry and Greenhouse Gases” (Intergovernmetnal Panel on Climate Change, 3rd
assessment report, Chapter 4, 2001; http://www.grida.no/publications/other/ipcc_tar/).
18. de Gouw, J. A., Parrish, D. D., Frost, G. J. & Trainer, M. educed emissions of CO2, NOx, and SO2 from US power plants owing to
switch from coal to natural gas with combined cycle technology. Earths Future 2, 75–82 (2014).
19. Ehlers, C. et al. Twenty years of ambient observations of nitrogen oxides and specied hydrocarbons in air masses dominated by
trac emissions in Germany, Faraday Discussions, doi:10.1039/c5fd00180c (2015).
20. Carslaw, D. C. & Beevers, S. D. New Directions: Should road vehicle emissions legislation consider primary NO2? Atmos. Environ.
38, 1233–1234 (2004).
21. iendler-Scharr, A. et al. Ubiquity of organic nitrates from nighttime chemistry in the European submicron aerosol. GRL 43,
7735–7744 (2016).
22. U. S. Environmental Protection Agency, (EPA), “NAAQS Scientific and Technical Information”, https://www.epa.gov/naaqs/
nitrogen-dioxide-no2-primary-air-quality-standards retrieved on 07-01-2016, (2016).
23. iese wetter, G. et al. Modelling NO2 concentrations at the street level in the GAINS integrated assessment model: projections under
current legislation. Atmos. Chem. Phys. 14, 813–829 (2014).
24. European Environment Agency, “Air quality in Europe – 2015” (Tech. ep. 5/2015, ISBN 978-92-9213-702-1, doi:10.2800/62459,
2015).
25. Giordano, L. et al. Assessment of the MACC reanalysis and its inuence as chemical boundary conditions for regional air quality
modeling in AQMEII-2. Atmos. Environ. 115, 371–388 (2015).
26. Hecel, A. et al. Inuence of low spatial resolution a priori data on tropospheric NO2 satellite retrievals. Atmos. Meas. Technol. 4,
1805–1820 (2011).
27. Salvo, A. & Geiger, F. M. eduction in local ozone levels in urban Sao Paulo due to a shi from ethanol to gasoline use. Nature.
Geoscience. 7, 450–458 (2014).
28. Madronich, S. Atmospheric chemistry: Ethanol and ozone. Nature Geoscience 7, 395–397 (2014).
29. Carslaw, D. C. Evidence of an increasing NO2/NOx emissions ratio from road trac emissions. Atmos. Environm 39, 4793–4802
(2005).
30. Bundesministerium für Verehr und digitale Infrastrutur (BMVI), “Bericht der Untersuchungsommission “Volswagen” 2016”
(https://www.bmvi.de/SharedDocs/DE/Publiationen/LA/bericht-untersuchungsommission-volswagen.html, retrieved 2016)
[the easiest access to this source is via the UL].
31. Franco, V., Posada Sanches, F., German, J. & Moc, P. “eal-world exhaust emissions from modern diesel cars, Part 2” (Tech. eport:
e International Council on Clean Transportation, www.theicct.org, Washington, DC, 2014).
32. Lee, J. et al. Measurement of NOx uxes from a tall tower in central London, U and comparison with emission inventories. Environ.
Sci. & Tech 49, 1025–1034 (2015).
33. Christen, A. Atmospheric measurement techniques to quantify greenhouse gas emissions from cities. Urban Climate 10, 241–260
(2014).
34. Nemitz, E., Hargreaves, . J., McDonald, A. G., Dorsey, J. . & Fowler, D. Meteorological measurements of the urban heat budget
and CO2 emissions on a city scale. Environ. Sci. Tech. 36, 3139–3146 (2002).
35. Velasco, H. E. et al. Eddy covariance ux measurements of pollutant gases in urban Mexico City. Atmos. Chem. Phys. 9, 7325–7342
(2009).
36. Valach, A. C., Langford, B., Nemitz, E., Macenzie, A. . & Hewitt, C. N. Seasonal and diurnal trends in concentrations and uxes
of volatile organic compounds in central London. Atmos. Chem. Phys. 15, 7777–7796 (2015).
37. arl, T. et al. Emissions of volatile organic compounds inferred from airborne ux measurements over a megacity. Atmos. Chem.
Phys. 9, 271–285 (2009).
38. Langford, B., Davison, B., Nemitz, E. & Hewitt, C. N. Mixing ratios and eddy covariance ux measurements of volatile organic
compounds from an urban canopy (Manchester, U). Atm. Chem. Phys 9, 1971–1987 (2009).
39. antala, P. et al. Anthropogenic and biogenic inuence on VOC uex at an urban bachground site in Helsini, Finland. Atm os.
Chem. Phys. 16, 7981–8007 (2016).
40. Marr, L. C., Moore, T. O., lapmeyer, M. E. & illar, M. B. Comparison on NOx uxes measured by eddy covariance to emission
inventories and land use. Environ. Sci. & Techn 47, 1800–1808 (2013).
41. Vaughan, A. . et al. Spatially resolved ux measurements of NOx from London suggest signicantly hagher emissions than
predicted by inventories. Faraday Discu ssions 189, 455–472 (2016).

www.nature.com/scientificreports/
9
Scientific RepoRts | 7: 2536 | DOI:10.1038/s41598-017-02699-9
42. EUOSTAT, “Energy, transport and environment indicators” (Tech. eport 201, European Commission, ec.europa.eu, ISBN 978-92-
79-49471-0, doi:10.2785/547816, 2015).
43. Vergreiner, I. & Dreiseitl, E. Valley winds and slope winds – observations and elementary thoughts. Meteorol. Atmos. Phys. 36,
24–286 (1987).
44. Coggon, M. M. et al. Emissions of nitrogen-containing organic compounds from the burning of herbaceous and arboraceous
biomass: Fuel composition dependence and the variability of commonly used nitrile tracers. GRL, doi:10.1002/2016GL070562
(2016).
45. Popa, M. E. et al. Vehicle emissions of greenhouse gases and related tracers from a tunnel study: CO: CO2, N2O: CO2, CH4: CO2,
O2:CO2 ratios, and the stable isotopes 13C, 18O in CO2 and CO. Atmos. Chem. Phys. 14, 210–2123 (2014).
46. Papadimitriou, G. et al. “TACCS - Transport data collection supporting the quantitative analysis of measures relating to transport
and climate change” (Tech. ep. EMISIA No. 13.E.025.V1, 2013).
47. Harley, . A., Marr, L. C., Lehner, J. . & Giddings, S. N. Changes in motor vehicle emissions on diurnal to decadal time scales and
eects on atmospheric composition. Environ. Sci. Tech. 39, 5356–5362 (2005).
48. Lamarque, J. F. et al. Historical (1850–2000) gridded anthropogenic and biomass burning emissions of reactive gases and aerosols:
methodology and application. Atmos. Chem. Phys. 10, 7017–7039 (2010).
49. Vestreng, V. et al. Evolution of NOx emissions in Europe with focus on road transport control measures. Atmos. Chem. Phys 9,
1503–1520 (2009).
50. eis, S. et al. A. oad trac emissions –predictions of future contributions to regional ozone levels in Europe. Atmos. Environ. 34,
4701–4710 (2000).
51. Stephens, S. et al. Weely patterns of Mexico City’s surface concentrations of CO, NOx, PM10 and O3 during 1986–2007. Atmos.
Chem. Phys. 8, 5313–5325 (2008).
52. Hellen, H., Haola, H. & Laurila, T. Determination of source contributions of NMHCs in Helsini (60 degrees N, 25 degrees E)
using chemical mass balance and the Unmix multivariate receptor models. Atmos. Environm 37, 1413–1424 (2003).
53. B orbon, A. et al. Emission ratios of anthropogenic volatile organic compounds in northern mid-latitude megacities: Observations
versus emission inventories in Los Angeles and Paris. J. Geophys. Res. 118, 2041–2057 (2013).
54. Spirig, C., Neel, A., leinman, L. I. & Hjorth, J. NOx versus VOC limitation of O3 production in the Po valley : Local and integrated
vew based on observations, J. Geophys. Res. 107, doi:10.1029/2001JD000561 (2002)
55. Wolfe, G. & Thornton, J. The Chemistry of atmosphere-forest exchange (CAFÉ) model – Part 1: Model description and
characterization. Atmos. Chem. Phys. 11, 77–101 (2011).
56. im, S. W. et al. Modeling the weely cycle of NOx and CO emissions and their impacts on O3 in the Los Angeles – South Coast air
Basin during the CalNex 2010 eld campaign. J. Geophys. Res. Atmos. 121, 1340–1360 (2016).
57. ExxonMobil, “e Outloo for Energy: A view to 2040” (Techn. eport, http://corporate.exxonmobil.com/en/energy/energy-
outloo/introduction/a-view-to-2040 retrieved on 1st of August 2016).
58. Molina, L. T. et al. An overview of the MILAGO 2006 Campaign: Mexico City emissions and their transport and transformation.
Atmos. Chem. Phys. 10, 8697–8760 (2010).
59. Marx, O., Brümmer, C., Ammann, C., Wol, V. & Freibauer, A. TANC – a novel fast-response converter to measure total reactive
atmospheric nitrogen. Atmos. Meas. Tech. 5, 1045–1057 (2012).
60. ummel, U., Ammann, C., Gut, A., Meixner, F. X. & Andreae, M. O. Eddy covariance measruements of nitric oxide ux within an
Amazonian rainforest, J. Geophys. Res. 107, doi:10.1029/2001JD000520 (2002).
61. Foen, T., Leuning, ., Oncley, S. P., Mauder, M. & Aubinet, M. Corrections and data quality, Eddy Covariance: A Practical Guide to
Measurement and Data Analysis (luwer Academic Publishers, Dortrecht, ed. 1, 2010), pp. 85–131.
Acknowledgements
T.K. received support from the EC Seventh Framework Program (Marie Curie Reintegration Program, “ALP-
AIR”, grant no. 334084).
Author Contributions
Karl, T. and Graus, M. conceived the experiment. Karl, T., Graus, M., Striednig, M., Lamprecht C., Hammerle,
A., Wohlfahrt, G., Held, A., von der Heyden, L., and Deventer J. designed and performed the INNAQS eld
experiment, interpreted the data and wrote the manuscript with inputs from all co-authors; Lee, J. provided input
on the interpretation of NOx ux data; Krismer A. provided air quality data, Feichter R. performed and analyzed
trac ow data; Haun C. provided the bottom up stationary emission source data; Karl T. performed the model
simulations.
Additional Information
Supplementary information accompanies this paper at doi:10.1038/s41598-017-02699-9
Competing Interests: e authors declare that they have no competing interests.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
© e Author(s) 2017