Boundary layer concentrations and landscape scale emissions of volatile organic compounds in early spring

Abstract

Boundary layer concentrations of several volatile organic compounds (VOC) were measured during two campaigns in springs of 2003 and 2006. The measurements were conducted over boreal landscapes near SMEAR II measurement station in Hyytiälä, Southern Finland. In 2003 the measuremens were performed using a light aircraft and in 2006 using a hot air balloon. Isoprene concentrations were low, usually below detection limit. This can be explained by low biogenic production due to cold weather, phenological stage of the isoprene emitting plants, and snow cover. Monoterpenes were observed frequently. The average total monoterpene concentration in the boundary layer was 33 ppt _v . Many anthropogenic compounds such as benzene, xylene and toluene, were observed in high amounts. Ecosystem scale surface emissions were estimated using a simple mixed box budget methodology. Total monoterpene emissions varied up to 80 μg m⁻² h⁻¹, α-pinene contributing typically more than two thirds of that. These emissions were somewhat higher that those calculated using emission algorithm. The highest emissions of anthropogenic compounds were those of p/m xylene.

Full text

Page 1

Atmos. Chem. Phys., 7, 1869–1878, 2007
www.atmos-chem-phys.net/7/1869/2007/
© Author(s) 2007. This work is licensed
under a Creative Commons License.
Atmospheric
Chemistry
and Physics
Boundary layer concentrations and landscape scale emissions of
volatile organic compounds in early spring
S. Haapanala1, J. Rinne1, H. Hakola2, H. Hell´ en2, L. Laakso1, H. Lihavainen3, R. Janson4, C. O’Dowd5, and
M. Kulmala1
1University of Helsinki, Department of Physical Sciences, Helsinki, Finland
2Finnish Meteorological Institute, Air Chemistry Laboratory, Helsinki, Finland
3Finnish Meteorological Institute, Climate and Global Change Research, Helsinki, Finland
4Stockholm University, Department of Applied Environmental Science, Stockholm, Sweden
5National University of Ireland, Department of Physics, Galway, Ireland
Received: 4 October 2006 – Published in Atmos. Chem. Phys. Discuss.: 18 October 2006
Revised: 20 February 2007 – Accepted: 3 April 2007 – Published: 16 April 2007
Abstract. Boundary layer concentrations of several volatile
organic compounds (VOC) were measured during two cam-
paigns in springs of 2003 and 2006. The measurements were
conducted over boreal landscapes near SMEAR II measure-
ment station in Hyyti¨ al¨ a, Southern Finland. In 2003 the
measuremens were performed using a light aircraft and in
2006 using a hot air balloon. Isoprene concentrations were
low, usually below detection limit. This can be explained
by low biogenic production due to cold weather, phenolog-
ical stage of the isoprene emitting plants, and snow cover.
Monoterpenes were observed frequently. The average to-
tal monoterpene concentration in the boundary layer was
33pptv. Many anthropogenic compounds such as benzene,
xylene and toluene, were observed in high amounts. Ecosys-
tem scale surface emissions were estimated using a simple
mixed box budget methodology. Total monoterpene emis-
sions varied up to 80µgm−2h−1, α-pinene contributing typ-
ically more than two thirds of that. These emissions were
somewhat higher that those calculated using emission algo-
rithm. The highest emissions of anthropogenic compounds
were those of p/m xylene.
1Introduction
Atmospheric aerosol particles are important for the global
radiation budget (Seinfeld and Pandis, 1998; Twomey, 1991;
Ramanathan et al., 2001; Cess et al., 1995; Kurten et al.,
2003; Kulmala et al., 2004a). In addition to anthropogenic
sources, biogenic activities increase significantly the aerosol
load (e.g. Tunved et al., 2006). Substantial production of
new aerosol particles has been observed in forested boreal re-
Correspondence to: S. Haapanala
(sami.haapanala@helsinki.fi)
gion (M¨ akel¨ a et al., 1997; Kulmala et al., 2004b). The maxi-
mum of new aerosol particle formation in these areas occurs
in the spring (Dal Maso et al., 2005). Aerosol formation and
especially growth in rural areas are expected to be caused
mainly by terpenoid compounds and their oxidation products
(O’Dowd et al., 2002; Tunved et al., 2006). To understand
the details of these formation and growth processes, it is im-
portant to know the concentrations and sources of condens-
able vapors in the atmosphere. Atmospheric concentration of
any trace gas depends on transport, sources, sinks and chem-
istry of the compound in question. Surface- and boundary
layer concentrations of volatile organic compounds (VOC)
in European boreal region have been measured by various
investigators (e.g. Janson et al., 1992; Hakola et al., 2000,
2003, 2006a; Spirig et at., 2004; Rinne et al., 2005).
Classical nucleation theories suggest that high saturation
ratio of condensable vapors leads to nucleation. High sat-
uration ratio can be obtained by lowering the temperature,
or by increasing the concentration. The highest concentra-
tions of condensable compounds occurs usually just above
the canopy while lowest temperatures are observed at the top
of the boundary layer. In addition to saturation ratio, sev-
eral other factors have effects on nucleation. These include
relative humidity, pre-existing particle surface and turbulent
mixing. These partly contradicting requirements make it al-
most impossible to predict where in the boundary layer the
formation of new particles actually takes place, and which
are the most important factors affecting it.
VOC emissions can be estimated on different scales by
different techniques. Branch scale VOC emissions of typi-
cal tree species in the boreal areas have been measured us-
ing chambers by e.g. Janson (1993), Hakola et al. (1998,
2006b) and Tarvainen et al. (2005). On the ecosystem scale
VOC emissions from boreal forests have been measured
Published by Copernicus GmbH on behalf of the European Geosciences Union.

Page 2

1870S. Haapanala et al.: VOC concentrations and emissions in early spring
Table 1. A summary of flights including start and end time of the measurement period, duration of that period, average canopy level air
temperature, and above canopy photosynthetic photon flux density (PPFD) during the measurements.
Date Start timeEnd timeDurationTemperaturePPFD
[dd/mm/yy][hh:mm ] [hh:mm][min][◦C][µmolm−2s−1]
QUEST II 2003, light aircraft
21 March 2003
25 March 03
26 March 03
27 March 03
28 March 03
28 March 03
28 March 03
2 April 2003
LABACET 2006, hot air balloon
10 March 2006
12 March 2006
13 March 2006
14 March 2006
17 March 2006
13:07
12:10
11:32
09:47
08:55
10:46
14:41
11:50
13:32
14:00
12:41
10:17
10:17
11:59
16:10
12:50
25
110
69
30
82
73
89
60
–3
4
6
3
2
4
5
–3
900
950
930
700
750
980
800
550
13:35
14:56
13:58
11:43
14:17
14:14
15:17
14:28
12:08
14:53
39
21
30
25
36
–9
–4
1
–2
5
550
550
630
810
500
13/03/06
14/03/06
17/03/06
Fig. 1. A map of the land use around SMEAR II station. The lo-
cation of the station is indicated with a white star in the middle of
the picture. The area shown is 40×40km2. Shades of green are
forests, yellow are agricultural lands, red are wetlands and blue are
water bodies. Black lines show the routes of three balloon flights.
Green stars indicate balloon locations when VOC sampling was
conducted. Map material: Copyright National Land Survey of Fin-
land 2002.
Table 2. The proportions of different land use categories in the area
of 1600km2around SMEAR II station.
Land use type Proportion [%]
built areas
wetlands
clear cut
deciduous forest
open land
agriculture
water bodies
mixed forest
pine dominated forest
spruce dominated forest
0.3
0.7
1.7
2.0
2.9
10.2
13.0
20.9
22.7
25.5
by e.g. Rinne et al. (1999, 2000a, 2000b) and Spanke et
al. (2001).These studies proved boreal vegetation to be
strong monoterpene emitters, with some sesquiterpene and
isoprene emissions as well.
Surface emissions on the landscape scale can be estimated
using boundary layer concentrations. Davis et al. (1994)
performed first measurements of landscape scale hydrocar-
bon emissions using a mixed-layer gradient technique. They
measured isoprene and monoterpene concentrations in the
lower part of the boundary layer above tropical rainforest
in Amazonas, Brazil and mixed pine-oak forest in Alabama,
USA. In the boreal region Spirig et al. (2004) have measured
boundary layer concentrations of isoprene and monoterpenes
using tethered balloon in the Southern Finland in summer-
Atmos. Chem. Phys., 7, 1869–1878, 2007www.atmos-chem-phys.net/7/1869/2007/

Page 3

S. Haapanala et al.: VOC concentrations and emissions in early spring1871
time. They used both mixed-layer gradient and mixed box
budget methods and estimated the surface fluxes of monoter-
penes to be between 180 and 300µgm−2h−1in August
2001. However, no previous measurements in these regions
in early spring, which is the maximum particle formation
season, have been reported. In addition, there is not much
information on the boundary layer concentrations of the an-
thropogenic compounds in the rural areas of northern Eu-
rope.
In the present study, we measured concentrations of sev-
eral non-methane hydrocarbons throughout the boundary
layer. In addition, we calculated estimates for the landscape
scale surface emissions of these compounds. Our measure-
ments were conducted in early spring when events of new
aerosol particle formation are often observed in boreal areas.
An extensive set of aerosol measurements were performed
simultaneously.
2Materials and methods
The measurements were performed during QUEST II (Quan-
tification of Aerosol Nucleation in the European Boundary
Layer) measurement campaign in 2003 (hereafter referred
to “QUEST II 2003”) and LABACET (LAgrangian Balloon-
borne Aerosol Characterization ExperimenT) experiment in
2006(hereafterreferredto“LABACET2006”). InQUESTII
2003 the measurements were carried out between 21 March
2003 and 2 April 2003 using a light aircraft. In LABACET
2006 the measurements were conducted between 10 March
2006 and 17 March 2006 using a hot air balloon as a mea-
surement platform. Table 1 summarizes the flights and some
of the basic atmospheric properties during them. Despite of
some longer flights, only samples taken within 45min were
accepted to represent one profile. Figure 1 shows three ex-
amples of hot air balloon routes and the location when VOC
sampling was conducted. In Fig. 2 altitude curves of the
same flights with VOC sampling points are shown. Due to
problems with GPS data, routes of all flights are not avail-
able.
The measurements took place near SMEAR II mea-
surement station, located in Southern Finland (61◦51?N,
24◦17?E, 180m a.s.l.).The area belongs to the south-
ern boreal zone. Vegetation consists mainly of coniferous
trees, dominated by Scots pine (Pinus sylvestris) and Nor-
way spruce (Picea abies). Figure 1 shows the different land
use categories around SMEAR II station. In Table 2 the pro-
portional abundances of different categories are given. The
land use data is derived from satellite photographs and for-
est inventories by National Land Survey of Finland and has
a resolution of 25×25m.
The annual mean temperature in the area is 3◦C. The
warmest month is July with mean temperature of 16◦C and
the coldest is February with mean temperature of –8◦C. The
annual mean precipitation is 700mm. These climatological
0 500100015002000
Time [s]
25003000 3500 4000
0
500
1000
1500
2000
2500
Altitude [m]
13−03−2006
0500100015002000
Time [s]
2500 300035004000
0
500
1000
1500
2000
2500
Altitude [m]
14−03−2006
0500 10001500 2000
Time [s]
2500 3000 3500 4000
0
500
1000
1500
2000
2500
Altitude [m]
17−03−2006
Fig. 2. Cruise altitude during the balloon flights shown in Fig. 1.
Green stars indicate balloon locations when VOC sampling was
started and red stars locations where sampling was stopped.
statisticsarefromJuupajoki-Hyyti¨ al¨ ameteorologicalstation,
located about 500m east from the SMEAR II measurement
station and the data represents period 1971–2000 (Drebs et
www.atmos-chem-phys.net/7/1869/2007/Atmos. Chem. Phys., 7, 1869–1878, 2007

Page 4

1872S. Haapanala et al.: VOC concentrations and emissions in early spring
70 8090
−20
−10
0
10
Temperature [°C]
2003
7080 90
−20
−10
0
10
2006
708090
20
40
60
80
100
RH [%]
70
LABACET
80 90
20
40
60
80
100
708090
0
500
1000
1500
Day of Year
PPFD [μmol m−2 s−1]
QUEST II
708090
0
500
1000
1500
Day of Year
Fig. 3. Air temperature and relative humidity at 8m height and pho-
tosynthetic photon flux density (PPFD) above forest canopy during
springs 2003 and 2006. Time periods of the measurement cam-
paigns are marked in the figures.
−5051015
0
200
400
600
800
1000
1200
1400
1600
1800
2000
Potential temperature [°C]
Altitude [m]
0
Water vapor mixing ratio [g kg−1]
1234
0
200
400
600
800
1000
1200
1400
1600
1800
2000
Fig. 4. Vertical profiles of potential temperature (left panel) and
water vapor mixing ratio (right panel) on 13 March 2006. The black
line shows subjectively approximated height of the boundary layer
top.
al., 2002).
At the SMEAR II measurement station many environmen-
tal parameters are measured routinely (Hari and Kulmala,
2005). These include various aerosol measurements, con-
centrations and surface fluxes of H2O, CO2and O3as well
as ordinary meteorological parameters. In Fig. 3 air temper-
ature and relative humidity measured at 8m height and pho-
tosynthetic photon flux density measured above forest during
springs 2003 and 2006 are shown. Spring 2003, with mean
temperature around zero, was clearly warmer than spring
2006.
Both aircraft and hot air balloon were equipped with tem-
perature and humidity sensors. These were used to determine
the top of the boundary layer. This was done subjectively
from the potential temperature and water vapour mixing ra-
tio data. The top of the boundary layer was assumed to be
at the height where water vapour mixing ratio has a strong
change or potential temperature gradient changes to positive.
In Fig. 4, one typical case (13 March 2006) is shown.
The measurement platform during QUEST II 2003 flights
was a DHC-6/300 Twin Otter STOL fixed-wing, twin engine
aircraft OH-KOG. The airflow for VOC samples was taken in
through a pitot tube located at the roof of the aircraft, in front
of the engines. Dynamic pressure of the pitot tube gener-
ated a flow through 3m of Teflon tubing which had an outlet
below the aircraft. The total airflow at the sample inlet was
much higher than required for VOC sampling devices. Sam-
pler devices were connected to this line. For more details
of the sampling system, see O’Dowd et al. (2007). The mea-
surement platform during LABACET 2006 flights was an Ul-
tramagic S-130 hot air balloon OH-SOL with an Ultramagic
C-6 gondola and a MK-21 burner. To avoid the contaminants
produced by the burner, the measurements were performed
only during descend of the balloon, when there is strong flow
of unaffected air from below the gondola and when the usage
of burner is minimal. In addition, the samples were collected
about two meters below the gondola base. From the data of
aerosol particle number concentration, temperature and hu-
midity, it was confirmed that these procedures were adequate
to guarantee contamination free measurements.
Light C2-C6hydrocarbons were sampled into 0.85l elec-
tro polished stainless steel canisters.
evacuated beforehand and pressurized during sampling us-
ing Teflon coated pump. The duration of one canister filling
was 60–180s. Chemical analysis of these samples was per-
formed using a gas chromatograph (HP-6890) with a flame
ionization detector (FID).
Heavier C5-C10hydrocarbons were trapped into cartridges
filled with Tenax-TA and Carbopack-B adsorbents. The sam-
ples were taken using 10-min sampling time and constant
flow of about 0.26l per minute in the QUEST II campaign.
During the LABACET experiment only 1- to 4-min sampling
times were possible. Therefore we had to use a sampling
flow of about 0.45l per minute. At this high flow rate a
breakthrough can be a significant problem and hence we used
two cartridges installed in series. The concentrations anal-
ysed from these two cartridges were summed to yield the
total concentration. On average 90% of the measured total
monoterpene concentration was trapped in the first cartridge.
The adsorbent samples were analyzed using automatic ther-
modesorption device (Perkin-Elmer ATD-400) connected to
a gas chromatograph (HP-5890) and a mass-selective detec-
tor (HP-5972). For more details on the analysis systems used
for airborne samples, see Hakola et al. (2000).
During the QUEST II 2003 campaign, additional surface
layer concentration measurements of monoterpenes were
conducted at the top of the SMEAR II tower, above the
The canisters were
Atmos. Chem. Phys., 7, 1869–1878, 2007www.atmos-chem-phys.net/7/1869/2007/

Page 5

S. Haapanala et al.: VOC concentrations and emissions in early spring1873
Table 3. A summary of sampling methods during the projects.
MeasurementSampling
method
Sampling
rate [ml min−1]
flowSampling time
[min]
Analytical
method
Measured compounds
QUEST II 2003
GroundTenax-TA 50120thermodesorption-
GC-MS
GC-FID
thermodesorption-
GC-MS
monoterpenes
Aircraft
Aircraft
canister
Tenax-TA
Carbopack-B
1
10
benzene, isoprene
monoterpenes,
matic hydrocarbons
& 260aro-
LABACET 2006
Balloon
Balloon
canister
Tenax-TA
Carbopack-B
1–3
1–4
GC-FID
thermodesorption-
GC-MS
benzene, isoprene
monoterpenes,
matic hydrocarbons
& 450aro-
forest canopy.
TA at 50mlmin−1for 2h per sample. A sampler system
with timers and solenoid valves was used to enable sampling
around the clock for the duration of the campaign. The sam-
ples were analysed in the laboratory by ATD-GC-MS. For
more details on the sampling and analysis system of surface
measurements, see Janson et al. (2001). Table 3 summarizes
all sampling procedures during these projects.
In order to estimate the magnitude of the surface emission,
we used a simple mixed box budget method (e.g. Guenther
et al., 1996). In this method, the mixed boundary layer is
treated as a closed and well mixed box where sources (sur-
face emission) and sinks (chemical loss and entrainment) are
in balance. Here we assume that these processes are constant
in time and space, causing the mean concentrations of the
compounds to be constant as well. The entrainment flux at
the top of the boundary layer can be estimated using jump
model where flux equals to product of concentration differ-
ence across the boundary layer top and the growth rate of
the boundary layer. Guenther et al. (1996) and Spirig et
al. (2004) showed that in their studies neglecting entrainment
would lead to less than 20% underestimation in the surface
flux. Due to lack of good estimates of boundary layer growth
rate, the entrainment flux at the top of the boundary layer is
neglected in this study. This leads, on average, to similar un-
derestimation in the surface flux as in the studies cited above,
as will be shown in Results and discussion. After these as-
sumptions we can write
These samples were collected on Tenax-
F = zS,
where F is the surface flux, z is the height of the boundary
layer and S is the chemical loss rate.
The chemical loss rates for different hydrocarbons were
estimated from their reactions with ozone (O3) and hydroxyl
radical (OH). Ozone concentration was measured at the top
of the SMEAR II mast, 67m above the ground level. Al-
though measured in the lowest part of the boundary layer,
thisdatawasassumedtorepresentthewholemixedboundary
(1)
layer with reasonable accuracy. This data was used directly
to estimate the loss rate of hydrocarbons caused by ozone.
The typical monthly mean daytime concentrations of OH
were adopted directly from Hakola et al. (2003). They cal-
culated daytime concentration of OH using a photochemical
model. The model was initialized using data from year 2000,
measured in the vicinity of SMEAR II station. The mean
OH concentrations were 0.013 ppt and 0.030 ppt for March
and April, respectively. Real instantaneous values may dif-
fer significantly from these typical values. From the differ-
ences between mean concentrations of successive months in
the spring, it was estimated that the actual concentration may
differ from mean values even with a factor of about 3. This
conclusion is supported by the model data presented by Boy
et al. (2005). Reactions of hydrocarbons with nitrate radical
(NO3) were ignored because they are important only during
night time. Loss rates were calculated from
?cO3kO3+ cOHkOH
where cVOCis the concentration of a particular VOC com-
pound, cO3and cOHare the concentrations of O3and OH,
respectively, and kO3and kOHare the corresponding second
order rate coefficients. Average boundary layer concentra-
tions were obtained by trapezoid integrals. Integration was
conductedfromthelowestmeasuredpointuptothetopofthe
boundary layer. The uppermost concentration measurement
was extrapolated to represent the concentration at the top of
the boundary layer. The concentrations that were below the
detection limit of the chemical analysis were converted to the
value of detection limit divided by two.
S = cVOC
?,
(2)
www.atmos-chem-phys.net/7/1869/2007/ Atmos. Chem. Phys., 7, 1869–1878, 2007

Page 6

1874S. Haapanala et al.: VOC concentrations and emissions in early spring
59%
5%
34%
3%
SURFACE LAYER
46%
12%
21%
21%
BOUNDARY LAYER
α−pinene
β−pinene
Δ3−carene
camphene
Fig. 5.
QUEST II 2003 in the surface layer (left panel) and upper in the
boundary layer (right panel).
Relative abundances of different monoterpenes during
3 Results and discussion
3.1Boundary layer concentrations of volatile organic com-
pounds
Statistics of the VOC concentrations measured during the
whole campaigns are presented in Table 4. Isoprene concen-
tration exceeded the detection limit (≈10pptv) only in two
samples. Isoprene is known to be emitted by many boreal
plant species (e.g. Hakola et al., 1998; Hell´ en et al., 2006;
Haapanala et al., 2006) directly from synthesis which is light
and temperature dependent (Guenther et al., 1993). The ab-
sence of isoprene can be explained by three factors: early
time of year, cold weather at the time of the measurements,
and relatively low coverage of isoprene forming plants. Due
to early time of year isoprene emitting deciduous trees did
not have leaves yet and ground vegetation, such as sphag-
nummossesinwetlands, wascoveredbysnow. Coldweather
alone could have explained low concentration of isoprene on
most of the days.
Monoterpenes were observed frequently throughout the
boundary layer. Figure 5 shows relative abundances of dif-
ferent monoterpenes in the surface layer and in the boundary
layer. In this comparison only data from QUEST II 2003 was
used. The most abundant monoterpene was α-pinene with
average daytime surface concentration of 37pptvand bound-
ary layer concentration of 18pptv. The second most abun-
dant monoterpenewas?3-carene. Contribution ofcamphene
was strongly increased upwards. This is, at least partly, ex-
plainedbythedifferencesinreactivitiesofdifferentmonoter-
penes. Camphene has lowest reactivity against OH and O3
of the monoterpenes analyzed (Atkinson, 1994). The av-
erage total monoterpene concentration in the surface layer
was 63pptv. Hakola et al. (2003) measured the average to-
tal surface layer concentrations of monoterpenes to be about
80pptvin March 2001 at SMEAR II. The average tempera-
ture at this time was –5◦C. This value as well as the monoter-
pene distribution is quite similar compared to our results.
Also Rinne et al. (2000a) measured surface layer monoter-
pene distribution at SMEAR II in August 1998. They found
almost similar distribution with somewhat higher proportion
of α-pinene and lower proportion of ?3-carene. The total
concentration in their measurement was about 500pptv.
Higher in the boundary layer the average total monoter-
pene concentrations were 34pptvand 32pptvduring QUEST
II 2003 and LABACET 2006, respectively.
al. (2004) measured average monoterpene concentration
to be 37pptv within mixed layer in August 2001 above
Hyyti¨ al¨ a. This concentration is close to those values mea-
sured in the present study. Also the monoterpene distribu-
tion was quite close to our results except for the significant
limonene concentrations measured by Spirig et al. (2004).
Hakola et al. (2003) found out that limonene concentrations
in the surface layer have stronger seasonal variation than
other monoterpenes, with higher concentrations during sum-
mer and fall.
In Fig. 6 examples of vertical gradients of monoterpene
concentrations are shown. The gradients are seldom well be-
having although clear decreasing trend upwards can be seen.
Figure 6b shows how monoterpene concentrations suddenly
dropped at nearly constant altitude. Some of these variations
in the gradients may be explained by changes of vegetation
inside measurement footprint during movement of the mea-
surement platform. A large part of the variation is explained
by analytical uncertainties of 17% up to 61% for different
monoterpenes. The magnitudes of analytical uncertainties
are calculated as mean relative standard deviation of paral-
lel samples taken regularly at the SMEAR II station. Due to
short sampling times in the present study, real uncertainties
are likely to be even higher.
In addition to biogenic compounds, the concentration data
of four VOCs of mainly anthropogenic origin are given in Ta-
ble 4. The highest concentrations are those of benzene, being
179ppt and 148ppt during QUEST II 2003 and LABACET
2006, respectively. Figure 7 showsselected verticalgradients
of benzene. For comparison, Hakola et al. (2006a) observed
average benzene concentrations in the surface layer to be
211pptvand 28pptvin the winter and summer, respectively.
Those surface air measurements were done at a rural site in
the Northern Finland. Xylene concentrations varied a lot,
average concentration being close to that of benzene. Emis-
sions of these anthropogenic compounds can be assumed to
be roughly constant throughout the year, but slower chemi-
cal degradation in the winter causes about ten-fold concen-
trations.
Spirig et
3.2Estimates of surface emissions
The total landscape scale emission of monoterpenes var-
ied between 5±4 and 39±11µgm−2h−1during QUEST
II 2003 and between 0±10 and 79±17µgm−2h−1during
Atmos. Chem. Phys., 7, 1869–1878, 2007www.atmos-chem-phys.net/7/1869/2007/

Page 7

S. Haapanala et al.: VOC concentrations and emissions in early spring1875
010 203040 5060
0
200
400
600
800
1000
25/03/2003 12:40
Concentration [ppt]
Altitude [m]
a)
α−pinene
β−pinene
Δ3−carene
camphene
0 1020304050 60
0
200
400
600
800
1000
28/03/2003 09:50
Concentration [ppt]
Altitude [m]
b)
α−pinene
β−pinene
Δ3−carene
camphene
0 10203040 50 6070
0
200
400
600
800
1000
14/03/2006 12:00
Concentration [ppt]
Altitude [m]
c)
α−pinene
β−pinene
Δ3−carene
camphene
Fig. 6. Examples of vertical gradients of different monoterpene
species. Dashed line indicates the height of boundary layer. Panels
(a) and (b) originates from QUEST II 2003 aircraft flights and panel
(c) from LABACET 2006 balloon flight.
125 150175200 225
0
200
400
600
800
1000
Concentration [ppt]
Altitude [m]
25/03/2003 12:40
28/03/2003 09:50
14/03/2006 11:45
Fig. 7. Concentrations of benzene in the boundary layer on three
days and corresponding boundary layer heights.
LABACET 2006.
sions was α-pinene (see Fig. 8 and Table 5). Emission did
not show clear dependence on the surface temperature (see
Fig. 9) which can be due to very low temperatures and re-
sulting low concentrations as compared to analytical uncer-
tainties. The uncertainty estimates were obtained by stan-
dard error propagation. Uncertainties used in this analysis
were: Boundary layer height 10%, concentrations 17%–61%
depending on the compound, O3concentration 5%, OH con-
centration 300%, reaction rate constants 10%.
We did not have observations on the growth rate of
the boundary layer but from the boundary layer heights it
was assumed to be less than 0.01ms−1. Using the aver-
age boundary layer concentration of monoterpenes during
LABACET 2006 (0.17µgm−3) entrainment flux of less than
6µgm−2h−1is obtained. Compared to average surface flux
of monoterpenes (38.4µgm−2h−1) this means underestima-
tion of about 16%.
We calculated the landscape scale monoterpene emission
potential using temperature dependent emission algorithm
(Guenther et al., 1993) with commonly used temperature de-
pendency factor β=0.09◦C−1. The standard emission poten-
tial obtained was 145µgm−2h−1. For comparison, we cal-
culated the average landscape scale emissions using the same
algorithm (Guenther et al., 1993), the land use data presented
in Table 2, and emission potentials and foliar biomass den-
sities for different forest types used by Lindfors and Lau-
rila (2000). The resulting landscape scale emission potential
of monoterpenes was 575µgm−2h−1, which leads to some-
what higher emissions than those derived in this paper. These
curves are shown in the Fig. 9. Spirig et al. (2004) reported
landscape scale emissions to be lower than ecosystem scale
emissions measured before at the same place, which supports
our results.
The dominant compound in the emis-
www.atmos-chem-phys.net/7/1869/2007/Atmos. Chem. Phys., 7, 1869–1878, 2007

Page 8

1876S. Haapanala et al.: VOC concentrations and emissions in early spring
3.3 Estimates of production rate of condensable vapors
We estimated the production rate of condensable vapors from
the total oxidation rate of VOCs (Eq. 2). We assumed aver-
age formation yield of 8% from monoterpenes (e.g. Yu et al.,
1999; Hoppel et al., 2001) and 4% from aromatic compounds
(e.g. Odum et al., 1996).The average total production
rate of condensable vapors was 0.6×104moleculescm−3s−1
which is about half of the values reported by Spanke et
al. (2001) and Spirig et al. (2004) at the same site. How-
ever, they both conducted measurements in August when
temperature and hence emissions of biogenic compounds are
substantially higher. In addition, oxidant levels are signifi-
cantly higher at that time of the year. To study the annual
cycle of production of condensable vapors from monoter-
penes (α-pinene, β-pinene, ?3-carene, camphene, sabinene
and limonene) we used the concentration data measured by
Hakola el al. (2003). From these data we calculated the
monthly production rate in the similar manner for surface
layer at about midday. In March the surface layer production
turns out to be about threefold compared to our results rep-
resenting the whole boundary layer. Further we can see the
strong annual cycle peaking in the June–July, when tempera-
ture induced emissions of monoterpenes are highest together
with fast oxidation due to strong solar radiation.
4 Conclusions
Boundary layer concentrations of volatile organic com-
pounds were measured over boreal forests during early
springs of 2003 and 2006. Due to short sampling times used
in LABACET 2006 experiment, those results are somewhat
more uncertain. Because of the ability to maintain stable
flight altitude, aircraft seems to be better measurement plat-
form than hot air balloon for this kind of work.
Despite of cold weather, the boundary layer concentra-
tions of monoterpenes were at the same level than those mea-
sured in August at the same site during earlier studies. Iso-
prene, however, was almost absent. Although monoterpene
concentrations were at the same level than in late summer,
the production of condensable vapors was significantly re-
duced due to slower chemistry. Therefore, the aerosol for-
mation events can not be explained solely by condensable
vapors produced from those VOC compounds measured in
this study. The concentrations of anthropogenic compounds
were at the same level than those observed in the wintertime
during earlier studies. Likewise for the biogenic compounds,
the high concentrations are explained by the slow OH chem-
istry due to early time of year.
Landscape scale surface emissions were estimated using a
simple mixed box method. The vertical profiles were not
always well-behaving and therefore we did not use gradi-
ent methods. In the summer, the boundary layer might be
somewhat better mixed and thus allows the usage of these
a−pineneb−pineneD3−carenecamphene
0
5
10
15
20
25
30
Flux [μg m−2 h−1]
QUEST II 2003
LABACET 2006
Fig. 8. Average fluxes of different monoterpene species during the
two experiments.
−10 −505 10
0
10
20
30
40
50
60
70
80
90
100
Temperature [°C]
Monoterpene flux [μg m−2 h−1]
QUEST II 2003
LABACET 2006
Fit
Algorithm prediction
Fig. 9.
and fit to that dataset using temperature dependency coefficient
β=0.09◦C−1. Solid black line shows algorithm prediction.
Sum flux of monoterpenes versus surface temperature
methods. Emissions of monoterpenes were lower than pre-
dicted by the ecosystem scale measurement data. Fitting
of the measurement data to the Guenther emission algo-
rithm yield to the landscape scale emission potential of only
145µgm−2h−1, which is considerably lower than one de-
rived from previous emission measurements.
to the systematic underestimation caused by the mixed box
method, this may indicate lower emission potential due to
early season. The highest monoterpene emissions were those
of α-pinene and ?3-carene, which is in line with previous
measurements conducted over boreal forests.
In addition
Acknowledgements. J. Paatero is acknowledged for the mea-
surements during aircraft flights.
L. Kulmala, T. Ruuskanen and T. Gr¨ onholm are acknowledged for
hot air balloon flights. S. Haapanala is grateful to the Academy
M. Sipinen, R. Lampinen,
Atmos. Chem. Phys., 7, 1869–1878, 2007www.atmos-chem-phys.net/7/1869/2007/

Page 9

S. Haapanala et al.: VOC concentrations and emissions in early spring1877
Table 4. Statistics of the average VOC concentrations in pptv. Data of QUEST II 2003 represents measurements during 8 different flights
and data of LABACET 2006 represents measurements during 5 different flights. For further details of the flights see Table 1.
DL UC QUEST II 2003 surface layer
meanstdev
QUEST II 2003 boundary layer
mean stdev
LABACET 2006 boundary layer
meanstdev minmaxminmaxminmax
isoprene
α-pinene
β-pinene
?3-carene
camphene
?monot.
benzene
toluene
p/m xylene
o xylene
10
14
2
4
4
4%
23%
17%
27%
61%
–
37
3
21
2
63
–
–
–
–
–
21
5
19
5
–
b.d.l.
b.d.l.
b.d.l.
b.d.l.
–
62
12
50
15
b.d.l.
15
4
8
7
34
179
108
123
48
b.d.l.
9
2
4
4
b.d.l.
b.d.l.
b.d.l.
b.d.l.
b.d.l.
b.d.l.
37
9
18
21
6
20
1
10
1
32
148
61
81
102
2
10
2
9
2
b.d.l.
14
b.d.l.
4
b.d.l.
10
37
4
25
5
4
46
68
85
16%
34%
49%
30%
–
–
–
–
–
–
–
–
–
–
–
–
39
72
94
36
101
38
26
11
257
285
329
123
46
27
27
36
81
46
68
85
199
108
130
166
DL = typical detection limit, UC = analytical uncertainty in %, - = not measured, b.d.l. = below detection limit
Table 5. Statistics of the VOC emissions in µg m−2h−1. Data of QUEST II 2003 represents measurements during 8 different flights and
data of LABACET 2006 represents measurements during 5 different flights. For further details of the flights see Table 1.
QUEST II 2003
mean
LABACET 2006
meanstdevminmax stdevminmax
isoprene
α-pinene
β-pinene
?3-carene
camphene
?monot.
benzene
toluene
p/m xylene
o xylene
b.d.l.
13.1
1.7
3.9
1.5
20.2
1,1
1,0
3,3
0,9
b.d.l.
8.2
1.4
3.2
1.7
b.d.l.
3.3
0.3
b.d.l.
b.d.l.
b.d.l.
24.7
4.3
8.6
4.9
1.7
27.4
0.4
8.6
0.3
38.4
0.4
0.9
4.6
4.2
0.6
16.2
1.0
8.0
0.5
b.d.l.
16.1
b.d.l.
2.4
b.d.l.
2.8
55.5
2.2
22.2
1.1
0,6
0,4
2,4
0,6
0,3
0,3
1,5
0,1
2,0
1,7
8,7
2,3
0.2
0.5
1.7
1.6
0.2
0.6
3.3
3.0
0.6
1.7
7.7
7.0
b.d.l. = below detection limit
of Finland for financial support (project 206162).
was partly supported by the European Comission under contract
EVK2-CT2001-00127 (QUEST).
This work
Edited by: K. H¨ ameri
References
Atkinson R.: Gas-phase tropospheric chemistry of organic com-
pounds, J. Phys. Chem. Ref. Data Monog., 2, 216 pp, 1994.
Boy, M., Kulmala, M., Ruuskanen, T.M., Pihlatie, M., Reissell,
A., Aalto, P. P., Keronen, P., Dal Maso, M., Hell´ en, H., Hakola,
H., Jansson, R., Hanke, M., and Arnold, F.: Sulphuric acid clo-
sure and contribution to nucleation mode particle growth, At-
mos. Chem. Phys., 5, 863–878, 2005, http://www.atmos-chem-
phys.net/5/863/2005/.
Cess, R. D., Zhang, M. H., Minnis, P., Corsetti, L., Dutton, E., For-
gan, B. W., Garber, D. P., Gates, W. L., Hack, J. J., Harrison, E.
F., Jing, X., Kiehl, J. T., Long, C. N., Morcette, J.-J., and Pot-
ter, G. L.: Absorption of solar radiation by clouds: Observations
versus models, Science, 267, 496–503, 1995.
Dal Maso, M., Kulmala, M., Riipinen, I., Wagner, R., Hussein,T.,
Aalto, P. P., and Lehtinen, K. E. J.: Formation and growth of
fresh atmospheric aerosols: eight years of aerosol size distribu-
tion data from SMEAR II, Hyyti¨ al¨ a, Finland, Boreal Environ.
Res., 5, 323–336, 2005.
Davis, K. J., Lenschow, D. H., and Zimmerman, P. R.: Biogenic
nonmethane hydrocarbon emissions estimated from tethered bal-
loon observations, J. Geophys. Res., 99, 25587–25598, 1994.
Drebs, A., Nordlund, A., Karlsson, P., Helminen, J., and Rissa-
nen, P.: Climatological statistics of Finland 1971–2000. Finnish
Meteorological Institute, Helsinki, 99 pp, ISBN 951-697-568-2,
2002.
Guenther, A. B., Zimmerman, P. R., Harley, P. C., Monson, R. K.,
and Fall, R.: Isoprene and Monoterpene Emission Rate Variabil-
ity: Model Evaluations and Sensitivity Analyses, J. Geophys.
Res., 98, 12609–12617, 1993.
Guenther, A., Zimmerman, P., Klinger, L., Greenberg, J., Ennis, C.,
Davis, K., Pollock, W., Westberg, H., Allwine, G., andGeron, C.:
Estimates of regional natural volatile organic compound fluxes
www.atmos-chem-phys.net/7/1869/2007/Atmos. Chem. Phys., 7, 1869–1878, 2007

Page 10

1878S. Haapanala et al.: VOC concentrations and emissions in early spring
from enclosure and ambient measurements, J. Geophys. Res.,
101, 1345–1359, 1996.
Haapanala, S., Rinne, J., Pystynen, K.-H., Hell´ en, H., Hakola, H.,
and Riutta, T.: Measurements of hydrocarbon emissions from
a boreal fen using the REA technique, Biogeosci., 3, 103–112,
2006.
Hakola, H., Rinne, J., and Laurila, T.: The hydrocarbon emission
rates of tea-leafed willow (Salix phylicifolia), silver birch (Be-
tula pendula) and european aspen (Populus tremula), Atmos. En-
viron., 32, 1825–1833, 1998.
Hakola, H., Laurila, T., Rinne, J., and Puhto, K.: The ambient con-
centrations of biogenic hydrocarbons at a northern European, bo-
real site, Atmos. Environ., 34, 4971–4982, 2000.
Hakola, H., Tarvainen, V., Laurila, T., Hiltunen, V., Hell´ en, H., and
Keronen, P.: Seasonal variation of VOC concentrations above a
boreal coniferous forest, Atmos. Environ., 37, 1623–1634, 2003.
Hakola, H., Hell´ en, H., and Laurila, T.: Ten years of light hydrocar-
bons (C2-C6) concentration measurements in background air in
Finland, Atmos. Environ., 40, 3621–3630, 2006a.
Hakola, H., Tarvainen, V., B¨ ack, J., Ranta, H., Bonn, B., Rinne, J.,
and Kulmala, M.: Seasonal variation of mono- and sesquiterpene
emission rates of Scots pine, Biogeosci., 3, 93–101, 2006b.
Hari, P. and Kulmala, M.:Station for measuring ecosystem-
atmosphere relations (SMEAR II), Boreal Environ. Res., 5, 315–
322, 2005.
Hell´ en, H., Hakola, H., Pystynen, K.-H., Rinne, J., and Haapanala,
S.: C2-C10 hydrocarbon emissions from a boreal wetland and
forest floor, Biogeosci., 3, 167–174, 2006.
Hoppel, W., Fitzgerald, J., Frick, G., Caffrey, P., Pasternack, L.,
Hegg, D., Gao, S., Leaitch, R., Shantz, N., Cantrell, C., Albrec-
cinski, T., Ambrusko, J., and Sullivan, W.: Particle formation
and growth from ozonolysis of α-pinene, J. Geophys. Res., 106,
27603–27618, 2001.
Janson, R.: Monoterpene concentrations in and above a forest of
Scots pine. J. Atmos. Chem., 14, 385-394, 1992.
Janson, R.: Monoterpene emissions from Scots pine and Norwegian
spruce. J. Geophys. Res., 98, 2839–2850, 1993.
Janson, R., Rosman, K., Karlsson, A., and Hansson, H.-C.: Bio-
genic emissions and gaseous precursors to forest aerosols, Tellus
B, 53, 423–440, 2001.
Kulmala, M., Suni, T., Lehtinen K. E. J., Dal Maso, M., Boy, M.,
Reissell, A., Rannik, ¨U., Aalto, P., Keronen, P., Hakola, H.,
B¨ ack, J., Hoffmann, T., Vesala, T., and Hari, P.: A new feed-
back mechanism linking forests, aerosols, and climate, Atmos.
Chem. Phys., 4, 557–562, 2004a.
Kulmala, M., Vehkam¨ aki, H., Pet¨ aj¨ a, T., Dal Maso, M., Lauri, A.,
Kerminen, V.-M., Birmili, W., and McMurry, P. H., Formation
and growth rates of ultrafine atmospheric particles: a review of
observations, J. Aerosol Sci., 35, 143–176, 2004b.
Kurten, T., Kulmala, M., Dal Maso, M., Suni, T., Reissell, A.,
Vehkam¨ aki, H., Hari, P., Laaksonen, A., Viisanen, Y., andVesala,
T.: Estimation of different forest-related contributions to the ra-
diative balance using observations in Southern Finland, Boreal
Environ. Res., 8, 275–285, 2003.
Lindfors, V. and Laurila, T.: Biogenic volatile organic compound
(VOC) emissions from forests in Finland, Boreal Environ. Res.,
5, 95–113, 2000.
M¨ akel¨ a, J. M., Aalto, P., Jokinen, V., Pohja, T., Nissinen, A., Palm-
roth, S., Markkanen, T., Seitsonen, K., Lihavainen, H., and Kul-
mala, M.: Observations of ultrafine aerosol particle formation
and growth in boreal forest, Geophys. Res. Lett., 24, 1219–1222,
1997.
O’Dowd, C. D., Aalto, P, H¨ ameri, K., Kulmala, M., and Hoffmann,
T.: Atmospheric particles from organic vapours, Nature, 416,
497–498, 2002.
O’Dowd, C., Yoon, Y., Junkerman, W., Aalto, P., Kulmala, M., Li-
havainen, H., and Viisanen Y.: Airborne measurements of nu-
cleation mode particles I: coastal nucleation and growth rates,
Atmos. Chem. Phys., 7, 1491–1501, 2007, http://www.atmos-
chem-phys.net/7/1491/2007/.
Odum, J., Hoffmann, T., Bowman, F., Collins, D., Flagan, R.,
and Seinfeld, J.: Gas/particle partitioning and secondary organic
aerosol yields, Environ. Sci. Technol., 30, 2580–2585, 1996.
Ramanathan, V., Crutzen, P. J., Kiehl, J. T., and Rosenfeld, D.:
Aerosol, climate and the hydrological cycle, Science, 294, 2119–
2124, 2001.
Rinne, J., Hakola, H., and Laurila, T.: Vertical fluxes of monoter-
penes above a Scots pine stand in the boreal vegetation zone,
Phys. Chem. Earth (B), 24, 711–715, 1999.
Rinne, J., Hakola, H., Laurila, T., and Rannik,¨U.: Canopy scale
monoterpene emissions of Pinus sylvestris dominated forests,
Atmos. Environ., 34, 1099–1107, 2000a.
Rinne, J., Tuovinen, J.-P., Laurila, T., Hakola, H., Aurela, M., and
Hyp´ en, H.: Measurements of hydrocarbon fluxes by a gradient
method above a northern boreal forest, Agric. Forest. Meteorol.,
102, 25–37, 2000b.
Rinne, J., Ruuskanen, T. M., Reissell, A., Taipale, R., Hakola,
H., and Kulmala, M.: On-line PTR-MS measurements of atmo-
spheric concentrations of volatile organic compounds in a Eu-
ropean boreal forest ecosystem, Boreal. Environ. Res., 10, 425–
436, 2005.
Seinfeld, J. H. and Pandis, S. N.: Atmospheric Chemistry and
Physics: From Air Pollution to Climate Change, Wiley, New
York, 1998.
Spanke, J., Rannik,¨U., Forkel, R., Nigge, W., and Hoffmann, T.:
Emission fluxes and atmospheric degradation of monoterpenes
above a boreal forest: field measurements and modelling, Tellus
B, 53, 406–422, 2001.
Spirig, C., Guenther, A., Greenberg, J. P., Calanca, P., and
Tarvainen, V.:Tethered balloon measurements of biogenic
volatile organic compounds at a Boreal forest site, Atmos.
Chem. Phys., 4, 215–229, 2004, http://www.atmos-chem-
phys.net/4/215/2004/.
Tarvainen, V., Hakola, H., Hell´ en, H., B¨ ack, J., Hari, H., and Kul-
mala, M.: Temperature and light dependence of the VOC emis-
sions of Scots pine, Atmos. Chem. Phys., 5, 989–998, 2005,
http://www.atmos-chem-phys.net/5/989/2005/.
Tunved, P., Hansson, H.-C., Kerminen, V.-M., Str¨ om, J., Dal Maso,
M., Lihavainen, H., Viisanen, Y., Aalto, P. P., Komppula, M., and
Kulmala, M.: HighNaturalAerosolLoadingoverBorealForests,
Science, 312, 261–263, 2006.
Twomey, S.: Aerosols, clouds and radiation, Atmos. Environ., 25A,
2435–2442, 1991.
Yu, J., Cocker, D., Griffin, R., Flagan, R., and Seinfeld, J.: Gas-
phase oxidation of monoterpenes: Gaseous and particular prod-
ucts, J. Atmos. Chem., 34, 207–258, 1999.
Atmos. Chem. Phys., 7, 1869–1878, 2007 www.atmos-chem-phys.net/7/1869/2007/