While the role of highly oxygenated molecules (HOMs) in new particle formation (NPF) and secondary organic aerosol (SOA) formation is not in dispute, the interplay between HOM chemistry and atmospheric conditions continues to draw significant research attention. During the Influence of Biosphere-Atmosphere Interactions on the Reactive Nitrogen budget (IBAIRN) campaign in September 2016, profile measurements of neutral HOMs below and above the forest canopy were performed for the first time at the boreal forest SMEAR II station. The HOM concentrations and composition distributions below and above the canopy were similar during daytime, supporting a well-mixed boundary layer approximation. However, much lower nighttime HOM concentrations were frequently observed at ground level, which was likely due to the formation of a shallow decoupled layer below the canopy. Near the ground HOMs were influenced by the changes in the precursors and oxidants and enhancement of the loss on surfaces in this layer, while the HOMs above the canopy top were not significantly affected. Our findings clearly illustrate that near-ground HOM measurements conducted under stably stratified conditions at this site might only be representative of a small fraction of the entire nocturnal boundary layer. This could, in turn, influence the growth of newly formed particles and SOA formation below the canopy where the large majority of measurements are typically conducted.
Highly oxygenated molecules (HOMs), a subgroup of the oxidation products of volatile organic compounds (VOCs) identified by their high oxidation states, have been recognized as important precursors for organic aerosol in the atmosphere (Ehn et al., 2014). They have also been found to enhance new particle formation (NPF) and growth (Kulmala et al., 2013; Zhao et al., 2013; Ehn et al., 2014; Bianchi et al., 2016; Kirkby et al., 2016; Tröstl et al., 2016). The importance of HOMs has been confirmed in ambient environments, especially in monoterpene-dominated regions such as the boreal forest (Kulmala et al., 2013; Ehn et al., 2014), but also in high-altitude mountain regions (Bianchi et al., 2016) and in rural areas (Jokinen et al., 2014; Kürten et al., 2016). In laboratory studies, HOM formation has been observed from various precursor molecules (Ehn et al., 2017), including both biogenic and anthropogenic emissions (Molteni et al., 2018).
The direct observation of HOMs has only recently become possible, following
the developments of the atmospheric-pressure-interface time-of-flight
(APi-TOF, measures the naturally charged HOMs) (Junninen et al., 2010) and chemical-ionization
atmospheric-pressure-interface time-of-flight (CI-APi-TOF, measures the
neutral HOMs) (Jokinen et al., 2012) mass
spectrometers. Ehn et al. (2010) and Bianchi et al. (2017)
found that the naturally charged HOM clusters could be observed every night
in the boreal forest during spring. Out of the observed ambient mass
spectra, a significant part could be reproduced in a chamber by introducing
the monoterpene
Further investigations of HOM formation chemistry have been carried out in
both laboratory and field studies. Based on current understanding from
laboratory experiments, the formation of HOMs involves three main steps:
(1) initial formation of peroxy radicals (
Beyond those chemical pathways, varied meteorological conditions are also
factors influencing the MT and oxidants at different heights above the
forest floor. Unsurprisingly, the oxidants producing HOMs (e.g.,
The first measurements of the HOM concentrations at two different heights (36 m and 1.5 m a.g.l.) are presented and discussed. The influence of boundary layer dynamics on the HOMs at these different heights at SMEAR II station is analyzed and characterized in conjunction with auxiliary turbulence and micrometeorological measurements.
The measurements were performed at the SMEAR II station (Station for
Measuring Ecosystem–Atmosphere Relations) in the boreal forest in
Hyytiälä, southern Finland (61
Concentration of HOMs was measured with two nitrate-ion-based CI-APi-TOF mass
spectrometers. The CI-APi-TOF measuring at higher altitude was deployed at
the top of a 35 m tower located
In comparison to the direct determination of TE (Heinritzi et al., 2016), this method increases the uncertainty in the quantification of HOM concentrations. However, as mentioned, a more accurate knowledge of the exact HOM concentrations would not influence the main findings of this study.
The relative transmission curve between the two CI-APi-TOF mass
spectrometers, determined during a period of strong turbulent mixing.
Intercomparison results using a permeation tube containing
trinitrotriazinane (
The MT, trace gases, and meteorological parameters were continuously
monitored at the different heights (4.2, 8.4, 16.8, 33.6, 50.4,
67.2, 101, and 125 m) on a 126 m mast
The Influence of Biosphere-Atmosphere Interactions on the Reactive Nitrogen
budget (IBAIRN) campaign was conducted from 1 to 25 September 2016. After
data quality checks, only the measurements collected after 5 September were
used. Figure 2 shows the overall time series of the meteorological parameters
measured at ground and tower levels, including the temperature, RH, global
radiation, concentrations of trace gases, MT, and total HOMs (Zha, 2018). The
weather was generally sunny and clear during the campaign except for a few
cloudy (10, 15, and 22–23 September) and drizzling (24 and 25 September)
days. The mean air temperature and RH observed at ground level were
The overall time series of the measured trace gases, meteorological parameters, and estimated total HOM concentrations at the ground (blue) and tower (red) levels.
The estimated total HOM concentration is representative for the overall
concentration level of HOMs and is here defined as the sum of the detected
signals among ions from
The estimated total HOM concentrations at the two heights were not different
during the day (mean
Comparison between ground (
Mean mass spectra with the averaging periods of daytime (09:00–15:00) and nighttime (21:00–03:00) at the ground and tower levels.
Figure 4 shows the mean mass spectra (in unit mass resolution, UMR, for
Summary of the “Non-inversion night” and “Inversion night” types.
The nighttime HOMs at ground level are likely influenced by transport
processes below the canopy since the estimated total HOM concentrations were
found much lower on the nights when temperature inversions were observed. To
further investigate the potential impact of such micrometeorological
phenomena on ground-level HOMs, the nights during the campaign without
precipitation or instrument failure were selected (14 nights in total) and
categorized into two types based on the occurrence of temperature inversions:
(1) the “non-inversion night” type included seven nights when no
temperature inversion was recorded; (2) the “inversion night” type category
consisted of seven nights that had encountered temperature inversions, and
the ground temperatures were generally
Table 1 shows the overall statistics including the mean and median values of
the temperatures,
Two individual nights representing the non-inversion night and inversion
night types were selected and further compared. Figure 5a shows the time
series of the meteorological parameters, trace gases, and HOMs measured at
ground and tower levels of one selected night for the non-inversion night
type (11–12 September, from 21:00 to 03:00). A
number of measures can be used to assess the local atmospheric stability
conditions at a given layer. These measures are commonly based on either the
Obukhov length and its associated atmospheric stability parameter or a
Richardson number (flux based, gradient based, or bulk). Because of its
simplicity and the availability of high-resolution mean air temperature
profiles, the bulk Richardson number (
Selected HOMs representing the major HOM types (and formation pathways) were summed up and categorized into four groups, as shown in Table 2. Each pathway might be influenced differently by boundary layer dynamics and micrometeorological processes. In this study, OH-initiated HOMs were assumed negligible due to the very low OH level in the nocturnal boundary layer.
Compositions of selected HOMs and their main oxidants (Yan et al., 2016).
All the HOM groups in Fig. 5a show stable patterns, and good agreement is
observed between the ground and tower measurements in the first half of the
night. Variations were observed when air mass change occurred at around
01:00, as indicated by the drop of
Figure 5b shows the time series of the trace gases, MT, and HOM groups of
both ground and tower measurements during an inversion night case
(8–9 September, from 21:00 to 03:00).
The ground
Schematic figure showing how vertical mixing, vertical advection,
and horizontal advection influence ground-level
A previous study by Alekseychik
et al. (2013) at the SMEAR II station showed that nocturnal decoupled air layers
were frequently (with a fraction of 18.6 % based on a long-term dataset)
observed under high-
During the non-inversion night, the ground
Mass defect (MD) plots of the selected
Therefore, the differences between the ground and tower measurements were
due to the joint effects of (i) decoupling between the stably stratified
near-ground layer and the canopy top and the consequent formation of a
shallow layer, (ii) weakening of advective and turbulent flux transport
terms thereby inhibiting mass exchange between the ground decoupled layer
and the remaining nocturnal boundary layer, and (iii) an increased surface area-to-volume ratio within the decoupled layer thereby enhancing
Hourly changes of
the ratios between estimated tower and ground HOM concentrations from 12:00
on 11 September to 11:00 on 12 September (non-inversion night). Markers are
sized by ground HOM concentrations and colored by the
Examination of the selected HOMs was useful and efficient to assess
the changes in HOMs; however, such an analysis might only indicate the major
formation pathways. Hence, it was also worthwhile to have a holistic view of
the entire mass spectra and all the detected HOMs. The mass defect (MD) plot
(Fig. 7) separates all identified compounds according to their exact masses
on the
Hourly changes of the ratios between estimated tower and ground HOM
concentrations from 12:00 on 8 September to 11:00 on 9 September (inversion
night). Markers are sized by ground HOM concentrations and colored by the O3
difference between tower and ground
Several limitations still exist in this study. From the measurement side, one
major concern was the comparability between our two CI-APi-TOF mass
spectrometers. In the worst case, our conclusion might be biased if
instrument responses changed due to some parameter that correlated with the
observed inversions. The main parameters in this case would be ambient
temperature and RH. As both instruments were located in
temperature-controlled containers and the sample flow was mixed 1 : 2 with
dry sheath air in the CI-APi-TOF drift tube, neither of these were expected
to yield such large changes. However, for confirmation, we compared the
detailed spectral evolution during days and nights of the study. Figure 8
shows an example of hourly changes of the ratios between tower and ground
HOMs, over a 24 h period without nighttime temperature inversion
(11 September). During this period, ambient temperatures changed from
19.1
In contrast, during a 24h period with nighttime temperature inversion
(8 September, shown in Fig. 9), the ratios agreed well only during daytime
(from 12:00 to 17:00, and 09:00 to 11:00 on the next day). Between these two
periods, temperature and RH were most of the time in the same range as on
11 September (when no strong deviations were observed), but now the HOM
behavior changed dramatically between the two heights. The ratios increased
from
Figures 8 and 9 clearly imply that the large differences between ground and
tower HOM concentrations were driven by temperature inversions and consequent
changes in the composition of the air in the two detached layers. Large
changes in HOMs were observed only when the ground temperature was lower than
the tower temperature and when the ozone concentration at ground level was
several parts per billion lower than at the tower (shown as a color scale in
Figs. 8 and 9). Absolute temperatures or RHs at the two heights were not able
to explain the changes. As a concrete example, good agreement was observed at
07:00 on 12 September, while ambient temperatures were low (ground and tower
temperatures were 9.3 and 8.6
From the micrometeorology side, the contribution from the potential
micrometeorological processes in the layer between 1.5 and 4.2 m (between
the sampling heights of the ground HOMs and other parameters) could not be
estimated with the current experiment design (i.e., only two measurement
heights). Similarly, the influence from horizontal advection could not be
entirely ruled out as a reason for the reduced ground-level HOM
concentrations (and other significantly changed species) because of the
possible horizontal inhomogeneity of HOM precursors and oxidants below the
canopy. However, our conclusion was confirmed by the incompatibility between
the increasing ground-level MT and
Highly oxygenated molecules (HOMs) were measured above the canopy and at
ground level (below the canopy) in a boreal forest environment during the
IBAIRN campaign that took place in September 2016. Boundary layer dynamics
and micrometeorology were found to be important factors that influence the
abundance of and trends in HOMs at ground level, by perturbing both their
sources and sinks. In the well-mixed boundary layer (e.g., during daytime or
nights without strong inversion), HOM concentrations and other measured
species were overall similar between the ground and tower measurements. In
contrast, much lower ground-level HOM concentrations were observed when
nighttime temperature inversion and formation of a decoupled layer occurred
below the canopy. On the one hand, the production of the ground-level HOMs could
be affected by the decreasing
We have presented the first detailed measurements of HOMs below and above the canopy across a wide range of atmospheric stability conditions. The results highlight the significance of near-ground boundary layer dynamics and micrometeorological processes on the ambient HOMs, showing that ground-based HOM measurement at this site might not be representative for the entire nocturnal boundary layer. Conventionally, field measurements of HOMs and other parameters are mostly performed close to the ground, and the possible effect of boundary layer dynamics and micrometeorological processes to the HOM concentrations has rarely been considered. Aerosol particle growth and SOA formation rates at ground level are likely to be influenced by the reduced HOM concentrations on the inversion nights. However, there are still limitations due to current experiment design, such as horizontal separation in instrument setup, or the uncertainties from using point measurements at two heights to infer larger-scale exchange. Clearly, more vertical and planar measurements of HOMs are needed to confirm the emerging picture presented here. Influence of boundary layer dynamics should be better characterized and evaluated in future field campaigns.
The time series of the measured trace gases, meteorological
parameters, and estimated total HOM concentrations at the ground and tower
levels will be available at the end of December 2018
(
The supplement related to this article is available online at:
QZ, MR, and ME designed the study. QZ, CY, HJ, MR, NS, JA, and SS collected the data. QZ and CY analyzed the data. QZ wrote the paper. All coauthors contributed to data interpretation and commented on the paper.
The authors declare that they have no conflict of interest. Douglas R. Worsnop is affiliated with Aerodyne Research, who produce the CI-APi-TOF instruments used in this study.
This work was supported by the IBAIRN project, the Academy of Finland Center of Excellence in Atmospheric Science, European commission Actris2 and Actris PPP, the European Research Council (grant 638703-COALA), transnational access from ENVRI plus, and the SMEAR II technical team. Qiaozhi Zha thanks ATM-DP (Doctoral Program in Atmospheric Sciences) graduate programs, John Crowley and the Max Plank Institute in association with the IBAIRN proposal, and the tofTools team for providing tools for mass spectrometry analysis. Gabriel Katul acknowledges the support from the US National Science Foundation (NSF-EAR-1344703, NSF-AGS-1644382), the U.S. Department of Energy (DE-SC0011461), and the University of Helsinki for supporting a 3-month sabbatical leave at the Division of Atmospheric Sciences. Otso Peräkylä thanks the Vilho, Yrjö and Kalle Väisälä Foundation. Edited by: Kyung-Eun Min Reviewed by: two anonymous referees