Motivated by increasing levels of open ocean in the Arctic summer and the
lack of prior altitude-resolved studies, extensive aerosol measurements were
made during 11 flights of the NETCARE July 2014 airborne campaign from
Resolute Bay, Nunavut. Flights included vertical profiles (60 to 3000 m
above ground level) over open ocean, fast ice, and boundary layer clouds and
fogs. A general conclusion, from observations of particle numbers between 5
and 20 nm in diameter (
Surface temperatures within the Arctic are rising almost twice as fast as in any other region of the world. As a manifestation of this rapid change the summer sea ice extent has been retreating dramatically over the past decades with the possibility that the Arctic might be ice free by the end of this century (Boé et al., 2009) or even earlier (Wang and Overland, 2012). Arctic aerosol is well known to show a distinct seasonal variation with maximum mass concentrations and a strong long-range anthropogenic influence in winter and early spring. The phenomenon, known as Arctic Haze, was identified many years ago (e.g. Barrie, 1986; Heintzenberg, 1980; Rahn et al., 1977; Shaw, 1995), and has commanded renewed attention in recent years (e.g. Law et al., 2014; Quinn et al., 2007). During summer the Arctic is more isolated from remote anthropogenic sources and represents a comparatively pristine environment. The reason is that the Arctic front (e.g. Barrie, 1986), which provides a meteorological barrier for lower-level air mass exchange, moves north of many source regions during the summer months. Anthropogenic and biomass burning aerosols are transported to the Arctic during the summer, but increased aerosol scavenging helps maintain the pristine conditions near the surface (e.g. Browse et al., 2012; Croft et al., 2016a; Garrett et al., 2011).
Zhang et al. (2010) discussed the impacts of declining sea ice on the marine planktonic ecosystem, which includes increasing emissions of dimethyl sulfide (DMS) that may contribute to particle formation in the atmosphere (e.g. Charlson et al., 1987; Pirjola et al., 2000). Enhanced secondary organic aerosol from emissions of biogenic volatile organic compounds is also a possibility (Fu et al., 2009). Primary emissions of aerosol particles from the ocean, such as sea salt and marine primary organic aerosol, may also increase (Browse et al., 2014). Open water tends to increase cloudiness, which means that aerosol influences on clouds are likely to be more important. Over the Arctic the effects of aerosols on clouds are especially uncertain. Models have predicted that increasing numbers of particles may lead to overall warming (Garrett, 2004) when the atmosphere exists in a particularly low particle number state now referred to being “cloud condensation nuclei (CCN) limited” (Mauritsen et al., 2011), to an overall cooling effect when increasing numbers of particles are added to an atmosphere with more particles already present (Lohmann and Feichter, 2005; Twomey, 1974). It is important to characterise particle size distributions in this pristine environment to provide a baseline against which future measurements can be compared in a warming world. Indeed, Carslaw et al. (2013) highlighted the need to understand pre-industrial-like environments with only natural aerosols in order to reduce the uncertainty in estimations of the anthropogenic aerosol radiative forcing.
Primary sources, gas-to-particle formation processes, cloud processing, atmospheric ageing, mixing and deposition are all reflected in the size distribution. Therefore, measurements of aerosol size distributions are important for understanding the processes particles undergo in addition to their potential effects on clouds. The presence of ultrafine particles (UFPs) indicates recent production as their lifetime is of the order of hours. We focus this paper on ultrafine particles as these are an indication for in situ aerosol production processes in the Arctic. We also consider the growth of newly formed particles, as that determines how important they are for climate.
Aerosol size distributions including ultrafine particles (dp < 20 nm) have been measured before at different locations throughout the Arctic. Long-term studies at ground stations such as Alert, Nunavut (Leaitch et al., 2013), Ny Alesund and Zeppelin (Engvall et al., 2008a; Ström et al., 2003, 2009; Tunved et al., 2013), both on Svalbard and very recently in Tiksi, Russia (Asmi et al., 2016), and Station Nord, Greenland (Nguyen et al., 2016), indicate a strong seasonal dependence of the size distribution with the accumulation-mode aerosol dominating during the winter months and a shift to smaller particles during the summer months. New particle formation events are frequently observed from June to August. Ström et al. (2003) showed that the size distribution undergoes a rapid change from an accumulation mode dominated distribution during the winter months to an Aitken mode dominated distribution at the beginning of summer. Total number concentrations increase at the beginning of summer and roughly follow the incoming solar radiation on a seasonal scale suggesting that photochemistry is an important factor for new particle formation in the Arctic. At Ny Alesund maximum number concentrations occur in late summer and are explained by the Siberian tundra being a potential source of aerosol precursor gases (Ström et al., 2003) and marine biogenic sulfur (Heintzenberg and Leck, 1994). Analysis of air mass patterns for this region show that the shift in the size distributions is also accompanied by a change of source areas, with a dominance of Eurasian source areas in winter and North Atlantic air during summer (Tunved et al., 2013).
Particle measurements including aerosol size distributions were also
conducted from ice breaker cruises such as from the Swedish ice breaker
So far most studies that include size distribution measurements in the
summertime Arctic were conducted from ground stations or ship cruises. To
date there are only two studies that assess the altitude dependence of the
size distribution, i.e. one in the area of Svalbard
(Engvall et al., 2008b) and one from the
In this study we present data from aerosol size distribution measurements taken from an aircraft during a 3 week period in July 2014 in the high Arctic area of Resolute Bay, Nunavut, Canada. The flights focused on vertical profiles from as low as 60 m above the ground up to 3 km, as well as on low-level flights above different terrain such as fast ice, open ocean, polynyas and clouds. We focus especially on UFP (5–20 nm in diameter) and address the following questions: what are the concentrations of UFPs in the Arctic summertime, and what is their vertical distribution? What are the environmental conditions that favour occurrence of UFPs and is there evidence for growth of UFP to CCN sizes? Aside from the studies conducted near Svalbard, we believe this is the first aircraft study in the high Arctic to systematically address these specific questions. This work provides a comprehensive picture of UFPs observed during the campaign whereas a prior publication from Willis et al. (2016) detailed one UFP formation and growth event observed over Lancaster Sound.
The research aircraft
Aerosol was sampled through a stainless steel inlet mounted to the top of
the plane and ahead of the engines to exclude contamination. The tip of the
inlet consisted of a shrouded diffuser that provided nearly isokinetic flow.
Inside the cabin the intake tubing was connected to a stainless steel tube
(outer diameter of 2.5 cm, inner diameter of 2.3 cm) that carried the
aerosol to the back of the aircraft where it was allowed to freely exhaust
into the cabin so that the system was not over-pressured. The stainless
steel tube functioned as a manifold, off which angled inserts were used to
connect sample lines to the various instruments described below. In-flight
air was pushed through the line with a flow rate of approximately 55 L min
Trace gases (CO and H
Aircraft state parameters and meteorological measurements were performed with
an AIMMS-20 manufactured by Aventech Research Inc. at a very high sampling
frequency (> 40 Hz). The AIMMS-20 consists of three modules:
(1) the air data probe that measures the three-dimensional (3-D)
aircraft-relative flow vector (true air speed, angle-of-attack and sideslip)
and turbulence with a 3-D accelerometer; furthermore, temperature and
humidity sensors are contained within this unit and provide an accuracy and
resolution of 0.30 and 0.01
Particle number concentrations and particle size distributions were measured
with a TSI 3787 water-based ultrafine condensation particle counter (UCPC),
a Droplet Measurement Technology (DMT) Ultra-High Sensitivity Aerosol Spectrometer (UHSAS) and a Brechtel Manufacturing Incorporated (BMI)
scanning mobility system (SMS) coupled with a TSI 3010 condensation particle
counter (CPC). The UCPC detected particle concentrations of particles larger
than 5 nm in diameter with a time resolution of 1 Hz. The flow rate was set
to 0.6 L min
The BMI SMS was set to measure particle size distributions from 20 to
100 nm with a sample flow of 1 L min
CCN were measured with a DMT CCN counter (CCNC). The CCNC was operated behind a constant pressure inlet that was set to 650 hPa. The nominal supersaturation was held constant at 1 %. Calibrations prior to and during the campaign (for details see Leaitch et al., 2016) showed that a nominal supersaturation of 1 % at the reduced pressure translated into 0.6 % effective supersaturation.
Cloud droplet sizes from 2 to 45
A DMT single-particle soot photometer (SP2) was deployed to measure refractory black carbon (rBC) number and mass concentrations. We refer to rBC mass concentrations as an indication of pollution influence. Calibrations with Aquadag soot were performed prior to and during the campaign. The lower size limit of detection of rBC particles by the SP2 was approximately 80 nm.
Sub-micron aerosol composition was measured with an Aerodyne high-resolution
time-of-flight aerosol mass spectrometer (HR-ToF-AMS;
e.g. DeCarlo et al., 2006). A
detailed description of the instrument is found in Willis et al. (2016). The
main purpose of the instrument was to measure non-refractory particulate
matter such as sulfate, nitrate, ammonium, methane sulfonic acid (MSA) and
the sum of organics. Detection limits were 0.009, 0.008, 0.004, 0.005 and
0.08
Carbon monoxide (CO) was measured with an Aerolaser ultra-fast carbon monoxide monitor model AL 5002 based on vacuum ultraviolet (VUV) fluorimetry, employing the excitation of CO at 150 nm. In situ calibrations were performed during flight at regular intervals (15–30 min) using a National Institute of Standards and Technology (NIST) traceable CO standard with zero water vapour concentration. CO mixing ratios were used as a relative indicator of aerosol influenced by pollution sources.
Water vapour (H
All particle data were averaged to 1 min intervals to match the time
resolution of the BMI SMS. Particle concentrations within different size
intervals were calculated. The notation
In order to obtain vertical profiles the data were averaged within altitude intervals. An average profile for a single flight was obtained by binning all data from the respective flight into altitude intervals of 100 m starting at the lowest flight altitude. In addition to data obtained during vertical profile flights, data acquired while flying at a constant level were also included. Average profiles containing data from more than one flight were calculated by averaging the respective single flight profiles.
Compilation of all flight tracks plotted on a satellite image from
4 July 2014. The image is taken from
Average size distributions were obtained by simply averaging each bin for
the desired time and altitude range. The size distributions measured by the
BMI SMS were used for particle sizes from 20 to 90 nm, and the distributions at
larger sizes are taken from the UHSAS. All particle concentrations are
expressed for ambient pressure conditions, i.e. they have not been adjusted
to standard temperature and pressure conditions. The
We used FLEXPART-WRF (Brioude et al.,
2013; website:
From 4 to 21 July 2014 11 flights were conducted out
of Resolute Bay (74.7
The ice/water coverage visible on the satellite picture is representative for the area during the first period. As can be seen, the ice edge was situated about 150 km east of Resolute Bay. It is clearly visible in the satellite image as a sharp line. The transition from a completely ice-covered region to open ocean was very abrupt during the first period. Only after a period of bad weather with high winds did the ice edge become less clear, and the region starting about 80 km east of Resolute Bay to about 200 km east was covered by fractured ice.
Median temperature, relative humidity (RH), wind speed, CO mixing
ratio and
Roughly 50 % of the flight time was within the inversion layer, and 50 % was in the free troposphere conducting altitude profile flights. A considerable amount of time was spent at 2800 m as this was the preferred altitude when travelling to a certain area. When clouds were present, the aircraft sampled them by slant profiling through the cloud in the case that clouds were above the boundary layer, or in the case clouds were within 200 m of the surface, by descending into the cloud as low as possible. Aerosol observations while inside cloud are excluded from the analysis here due to potential artefacts from droplets shattering on the outside inlet.
Meteorological conditions changed over the course of the campaign. Similar
conditions were encountered during the first part of the campaign (4–12 July, 6 flights), referred to as the “Arctic air mass
period” because air masses from within the Arctic dominated and the
atmosphere showed structures typical for the Arctic, such as a low boundary
layer height with thermally stable conditions, indicated by a near-surface
temperature inversion and frequent formation of low-level clouds. At this
time Resolute Bay was under the influence of high-pressure systems. Clear
sky with few or scattered clouds and low wind speeds dominated. Conditions
changed starting from 13 July when the region was influenced by
troughs of a low-pressure system located to the west above Beaufort Sea,
which eventually passed through Resolute Bay on 15 July bringing
along humidity, precipitation and fog. Intense fog and low visibility
impeded flying from 13 to 16 July. A short good weather
window in which the fog dissipated permitted flying again on 17 July
(referred to as “transition day”; one flight) just before Resolute Bay
came under influence of a pronounced low-pressure system located to the
south with its centre around King William Island (69.0
Vertical profiles of median temperature, relative humidity (RH), wind speed,
CO and
Within the BL, particle concentrations spanned over a wide range of
concentrations (max
Flight tracks colour coded by particle
concentrations.
Throughout the campaign we observed large variability in particle
concentrations (Fig. 3). We observed not only very clean air masses with
FLEXPART-WRF potential emissions sensitivities for each flight (using particle releases every 2 min along the flight track) that illustrate transport regimes during different periods of the campaign. The colour code indicates the residence time of air in seconds and the numbers represent the position of the plume centroid location in days prior to release (days 1–7).
UFP were very frequently present within the BL in high concentrations
(Fig. 3c). Here we refer to “bursts” of particles as a sudden and
relatively large increase in
The frequent presence of UFP agrees well with other studies made during the
Arctic summertime at several locations, such as at the ground stations in Ny
Alesund and Zeppelin
(Ström
et al., 2009; Tunved et al., 2013), at Alert
(Leaitch et al., 2013), and from
ship-based observations
(Chang
et al., 2011; Covert et al., 1996; Heintzenberg et al., 2006). However, such
a frequent presence of an UFP mode (65 % of the time > 200 cm
In the following sections, the vertical distribution of UFP and the size distributions are discussed in relation to meteorological conditions during the three distinct periods that characterised this campaign.
During this first period the study area was under the influence of a high-pressure
system. As illustrated by FLEXPART-WRF results (Fig. 4a and b),
air masses were either coming from the north extending to the east in the
Arctic Ocean or from the east passing over the open ocean in Lancaster Sound
and Baffin Bay. Both examples indicate that air masses resided within the
Arctic region at least 5 days prior to sampling. This is true for all
flights during this period. The very low CO mixing ratios (78 ppb
The Arctic air mass period was characterised by a very sharp contrast between the BL and the FT in terms of particle number concentrations and sizes (Fig. 5). The BL was characterised by a prominent layer of UFP from the surface to about 300 m with the highest concentrations closest to the surface (Fig. 5a). The height of the UFP layer coincides with the average height of the temperature inversion for this period (see temperature profile Fig. 2) and indicates that air masses were stably layered limiting exchange with the FT. This is supported by the observed lower turbulent mixing (i.e. turbulent kinetic energy) from boundary layer to the free troposphere during the campaign (Aliabadi et al., 2016a).
During this period we measured the highest concentrations of UFPs with the
1 min average up to 5300 cm
Average particle concentration data during the Arctic air mass
period.
The average
In Fig. 5b, the median size distribution shows that increases in UPF in the BL were frequent. The average size distribution shows that at times higher concentrations of particles extended up to about 80 nm, consistent with the suggestion above that some UFP particles experienced growth to larger sizes. A relevant case will be discussed in Sect. 4.3. Occasionally a mode of particles larger than 400 nm was present in the BL over open water (see Sect. 4.2), which was likely the product of primary oceanic emissions.
17 July marks the transition from dominance by Arctic air masses to a more
distant influence from southern air masses. The transition day consists of
only one flight in the area of Lancaster Sound, during which low
concentrations of particles larger than 20 nm were observed below 600 m;
e.g.
Average particle concentration data on the transition
day.
Average particle concentration data during the southern air mass
period.
On this day, occasional bursts of UFP up to 1400–1900 cm
During this period the region was under the influence of a low-pressure
system centred south of Resolute Bay. FLEXPART-WRF air mass trajectories
(Fig. 4d and e) indicate a prevalence of air masses from the south
potentially affected by wild fires (see Fig. S2). At the
beginning of this period on 19 July (Fig. 4d), air mass
trajectories suggest the strongest influence from the south while towards
the end of the period on 21 July (Fig. 4e) FLEXPART-WRF indicates
that southern air masses mixed with air masses coming off Greenland. Near-surface
temperatures were higher than during the previous periods (Fig. 2),
and temperature inversions were less pronounced (2–4
Average profiles of particle concentrations above ice, open water and cloud. The number of data points for each specific profile is 130 above water, 216 above cloud and 123 above water.
UFP were observed less frequently than during the Arctic air mass period and
in lower concentrations (Fig. 7). Bursts of UFP above 1000 cm
The southern air mass period clearly shows different aerosol characteristics
within the near-surface layer than compared to the Arctic air mass period
and the transition day. Average concentrations of particles larger than 40 nm
were the highest within the boundary layer and decreased with altitude
(Fig. 7a). This is in sharp contrast to the cleaner boundary layers
observed before. Whereas concentrations of particles larger than 40 nm were
Case study from 8 July flight. Time series of flight altitude and
illustration of the surface including cloud coverage
During the southern air mass period, three important factors had changed
compared to both prior periods. (1) Air mass back trajectories had clearly
shifted to the south and potentially transported emissions from wild fires
located in the Northwest Territories (Fig. S2) into the
region, which might mix into the boundary layer. (2) The
Within the FT the size distributions shows a bimodal character with a minima at 60–80 nm, which may indicate the air masses experienced cloud processing. This is likely, given the presence of the low-pressure system bringing moister and warmer air masses. The bimodal size distribution is different from the average size distribution during the Arctic air mass period when drier air masses from within the Arctic dominated.
We investigated the potential influence of different underlying water surfaces on the occurrence of UFP by examining in detail the time periods when we were flying at altitudes at or below 500 m during the Arctic air mass period. We distinguish between three water surfaces: ice-covered areas (including ice edge and ice covered with melt ponds), open ocean (including polynyas) and low-level clouds (including both cloud above water and cloud above ice). Here we point out that the case “cloud” does not include in-cloud flight times but only flight periods when above cloud top without actually entering the cloud (confirmed by a zero signal in a liquid cloud probe, FSSP100). An altitude of 500 m was chosen to include time periods when we were flying above low-level clouds and to capture mostly flights within the boundary layer where a local influence of the terrain below was likely. During the Arctic air mass period, there was a clear separation between ice and open water over Lancaster Sound with east of the ice edge completely ice free, where west of the ice edge the ocean was seamlessly covered by fast ice (see satellite picture in Fig. 1).
Each average profile above the different water surfaces exhibits unique
features (Fig. 8). Above ice the highest concentrations of UFP (average:
400 cm
The flight on 8 July provides a case study illustrating that the occurrence of UFP is confined to the BL suggesting a surface source of UFP and that the appearance of UFP is promoted by cloud. We consider the altitude dependence of the UFP within the BL in relation to air mass history and cloud.
On this flight we first flew out into Lancaster Sound west of Resolute Bay,
turned around and descended into the BL above the ice. Here, we focus on the
time period from 15:50 UTC (descent into the BL) to 17:20 UTC where we
travelled from west to east and remained within the BL but stayed out of
cloud as shown in Fig. 9; see also Fig. S2. The later part
of the flight focused on in situ cloud properties and is discussed elsewhere
(Leaitch et al., 2016). The weather was
sunny with low-level clouds starting around 150 km over ice and west of the
ice edge in Lancaster Sound. The clouds had formed over the water and were
blown over the ice where they were dissipating
(Leaitch et al., 2016). In the entire
area the atmosphere was characterised by a surface temperature inversion
extending vertically up to about 300 m with
UFP were present throughout the BL with the highest concentrations at the
lowest altitudes and decreasing concentrations towards the top of the BL
(Fig. 9b). In contrast, larger particles (e.g.
Air mass histories at these locations determined from FLEXPART-WRF (Fig. 10) indicate the following:
To the west of Resolute Bay (point B), Lancaster Sound air masses had
been mixed with air masses from the north. This is also confirmed by the
local wind directions indicating winds coming from the northwest sector
(Fig. 10a), and it is consistent with the associated change in cloud. Near the top of the BL, air masses had descended recently (< 3 h)
into the BL (Fig. 10c point C and point F). In contrast, deeper within
the BL at points B and D, air masses had descended into the BL earlier
(
Aerosol composition shows a clear difference between the aerosol in the FT
and the BL. The aerosol sulfate rapidly decreases as we enter the BL around
16:00 UTC, while aerosol organic mass concentrations show an initial relative
increase followed by an absolute increase towards the east (Fig. 9c).
Within the BL, aerosol organics and sulfate mass loadings show a pattern
similar to
To explain these observations, we hypothesise that the smaller particle mode
is formed by nucleation and growth occurring within the BL and especially in
cloud vicinity. UFP concentrations near cloud top have been reported before
(e.g. Radke and Hobbs, 1991, Wiedensohler et al., 1997, Clarke et al., 1999;
Garrett et al., 2002; Hegg et al., 1990; Mauldin et al., 1997) and it is
suggested that nucleation in near-cloud regions is favoured by the low
surface areas, possibly due to cloud scavenged aerosol, moist air and a high
actinic flux. Indeed, near cloud top, where we observed an increase of UFP
extending up to almost 50 nm, the conditions for nucleation and growth are
ideal. We speculate that the availability of precursor gases is provided by
the long residence time (
The event at point E occurs where the aircraft was between cloud top and the top of the BL, where no increases in UFP were observed before or after. It may be that the aircraft descended slightly but sufficiently into the cloud-influenced area, which looks to be 25–40 m above cloud top (Fig. 9g), but also at that point we were in vicinity of Prince Leopold Island, which is a bird sanctuary and many bird colonies nest at the 260 m high cliff. FLEXPART-WRF and the in situ wind measurement show that air masses to a large extent were directly coming off the island (Fig. 10, point E) suggesting a connection between the appearance of UFP and possible emissions from the fauna of the island. The increase of particle phase ammonium (Fig. 9d) at the same time supports this connection and nucleation of particles from biogenic precursors emitted by bird colonies are documented (Weber et al., 1998; Wentworth et al., 2016, Croft et al., 2016b).
Alternatively, it should be considered that evaporating fog and cloud droplets may also act as a primary source of UFP (e.g. Heintzenberg et al., 2006; Karl et al., 2013; Leck and Bigg, 1999). Karl et al. (2013) suggested a combined pathway that involves the emission of UFP by fog and cloud droplets, together with secondary processes enabling growth of these particles. For our observations we have no reason to assume that nucleation does not occur since conditions are ideal but we cannot rule out that nanoparticles are emitted by the possibly evaporating cloud droplets onto which gases then condense.
In conclusion the aerosol mass within the near-surface layer is dominated by
organics relative to sulfate, while at just a slightly higher altitude
sulfate is clearly increased and increases further above the inversion
layer. A high organic content coincides with increases in UFP particles,
especially at times when also growth into the size range up to 50 nm is
indicated. Similarly the MSA-to-sulfate ratio shows a peak at the lowest
altitudes with maximum values in the vicinity of clouds that coincide with a
long residence time (
Correlations between CCN and particle concentrations for the full study period.
CCN concentrations were measured at a supersaturation of 0.6 %. The
vertical profiles of CCN concentrations (Fig. 11a) show patterns similar
to those of larger particles. In the very clean boundary layer of the Arctic
air mass period and the transition day CCN concentrations are equally low
(
A central question is whether and to what degree the CCN are influenced by
the UFP. Two factors help with addressing this question; (1) particles as
small as 20 nm and in general much smaller than the average 80 nm size
associated with the CCN at 0.6 % will nucleate cloud droplets in the clean
environment of the summer Arctic (Leaitch et al., 2016), and (2) there is
evidence here that increases in particles larger than 20 nm are associated
with increases in the UFP, particularly for UFP influenced by clouds (e.g.
Fig. 8). Figure 12 shows regressions of CCN with UFP,
This study presents airborne observations of ultrafine particles (UFP)
during the Arctic summertime. In total, 11 flights were conducted in July 2014 in
the area of Resolute Bay situated in the Canadian Archipelago. The location
allowed access to open water, ice-covered regions and low clouds. Flights
focused around the ice edge in Lancaster Sound including open waters to the
east, the ice-covered region to the west and polynyas north of Resolute Bay.
UFP were observed within all regions and above all terrains with the highest
concentrations encountered in the boundary layer immediately above cloud and
open water. It is shown that UFP occur most frequently (> 65 %
of the time) and with the highest concentrations (up to 5300 cm
The frequent presence of UFP in the boundary layer over open water and low
clouds and the enhanced number concentrations at the lowest altitudes sampled
indicate a surface source, such as the ocean, for the UFP gaseous
precursors. This is especially true during the Arctic air mass period when
the sampling region was pristine and not influenced by pollution.
FLEXPART-WRF simulations indicate that air masses had resided within the
Arctic region at least 5–7 days prior to sampling. During this time UFP were
restricted to the boundary layer and no UFP events were observed aloft,
thereby excluding that these UFP form in the free troposphere and subside
into the near-surface layer (e.g. Clarke et al.,
1998; Quinn and Bates, 2011). At the same time we observed an extremely
clean boundary layer (surface area of
Chlorophyll
Relating observations of UFP to the surface below during the Arctic air mass
period revealed that the highest UFP concentrations occurred above low-level
cloud and open water with averages of 1040 and 560 cm
Overall, the summertime Arctic is an active region in terms of new particle formation, occasionally accompanied by growth. The value of these altitude profiles across a wide spatial extent, performed for the first time in this campaign, is that they demonstrate that this activity is largely confined to the boundary layer, and that the dominant source of small particles to the boundary layer does not arise by mixing from aloft but most likely from marine sources. For future studies, the relative impact of such natural sources of UFP needs to be evaluated with respect to potential new sources, such as those that may arise with an increase in shipping.
NETCARE (Network on Climate and Aerosols, 2015,
The authors declare that they have no conflict of interest.
The authors would like to thank a large number of people for their contributions to this work. We thank Kenn Borek Air, in particular the pilots Kevin Elke and John Bayes and the aircraft engineer Kevin Riehl. We are grateful to John Ford, David Heath and the University of Toronto machine shop for safely mounting our instruments on racks for aircraft deployment. We thank Jim Hodgson and Lake Central Air Services in Muskoka, Jim Watson (Scale Modelbuilders, Inc.), Julia Binder and Martin Gerhmann (Alfred Wegener Institute, Helmholtz Center for Polar Marine Research, AWI), and Mike Harwood and Andrew Elford (Environment and Climate Change Canada, ECCC) for their support of the integration of the instrumentation and aircraft. We gratefully acknowledge Carrie Taylor (ECCC), Bob Christensen (U of T), Lukas Kandora, Manuel Sellmann and Jens Herrmann (AWI), Desiree Toom, Sangeeta Sharma, Dan Veber, Andrew Platt, Anne Marie Macdonald, Ralf Staebler and Maurice Watt (ECCC) for their support of the study. We thank the Biogeochemistry department of MPIC for providing the CO instrument and Dieter Scharffe for his support during the preparation phase of the campaign. The authors J. L. Thomas and K. S. Law acknowledge funding support from the European Union under Grant Agreement no. 5265863 – ACCESS (Arctic Climate Change, Economy and Society) project (2012–2015) and TOTAL SA. Computer simulations were performed on the IPSL mesoscale computer centre (Mésocentre IPSL), which includes support for calculations and data storage facilities. We thank the Nunavut Research Institute and the Nunavut Impact Review Board for licensing the study. Logistical support in Resolute Bay was provided by the Polar Continental Shelf Project (PCSP) of Natural Resources Canada under PCSP Field Project no. 218614, and we are particularly grateful to Tim McCagherty and Jodi MacGregor of the PCSP. Funding for this work was provided by the Natural Sciences and Engineering Research Council of Canada through the NETCARE project of the Climate Change and Atmospheric Research Program, the Alfred Wegener Institute, Helmholtz Center for Polar and Marine Research and Environment and Climate Change Canada. Edited by: L. M. Russell Reviewed by: J. Heintzenberg and two anonymous referees