Atmos. Chem. Phys., 17, 13611379, 2017 www.atmos-chem-phys.net/17/1361/2017/ doi:10.5194/acp-17-1361-2017 Author(s) 2017. CC Attribution 3.0 License.
Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model
Putian Zhou1, Laurens Ganzeveld2, llar Rannik1, Luxi Zhou1,a, Rosa Gierens1,b, Ditte Taipale3,4, Ivan Mammarella1, and Michael Boy1
1University of Helsinki, Department of Physics, P.O. Box 64, University of Helsinki, 00014 Helsinki, Finland
2Meteorology and Air Quality (MAQ), Department of Environmental Sciences, Wageningen University and Research Centre, Wageningen, the Netherlands
3University of Helsinki, Department of Forest Sciences, P.O. Box 27, University of Helsinki, 00014 Helsinki, Finland
4Estonian University of Life Sciences, Department of Plant Physiology, 51014 Kreutzwaldi 1, Estonia
anow at: US Environmental Protection Agency, Research Triangle Park, NC, USA
bnow at: Institute for Geophysics and Meteorology, University of Cologne, Cologne, Germany
Correspondence to: Putian Zhou (putian.zhou@helsinki.)
Received: 19 February 2016 Published in Atmos. Chem. Phys. Discuss.: 27 April 2016 Revised: 28 December 2016 Accepted: 3 January 2017 Published: 30 January 2017
Abstract. A multi-layer ozone (O3) dry deposition model has been implemented into SOSAA (a model to Simulate the concentrations of Organic vapours, Sulphuric Acid and Aerosols) to improve the representation of O3 concentration and ux within and above the forest canopy in the planetary boundary layer. We aim to predict the O3 uptake by a boreal forest canopy under varying environmental conditions and analyse the inuence of different factors on total O3 uptake by the canopy as well as the vertical distribution of deposition sinks inside the canopy. The newly implemented dry deposition model was validated by an extensive comparison of simulated and observed O3 turbulent uxes and concentration proles within and above the boreal forest canopy at SMEAR II (Station to Measure EcosystemAtmosphere Relations II) in Hyytil, Finland, in August 2010.
In this model, the fraction of wet surface on vegetation leaves was parametrised according to the ambient relative humidity (RH). Model results showed that when RH was larger than 70 % the O3 uptake onto wet skin contributed 51 % to
the total deposition during nighttime and 19 % during day
time. The overall contribution of soil uptake was estimated about 36 %. The contribution of sub-canopy deposition below 4.2 m was modelled to be 38 % of the total O3 depo
sition during daytime, which was similar to the contribution reported in previous studies. The chemical contribution to O3 removal was evaluated directly in the model simulations. Ac-
cording to the simulated averaged diurnal cycle the net chemical production of O3 compensated up to 4 % of dry depo
sition loss from about 06:00 to 15:00 LT. During nighttime, the net chemical loss of O3 further enhanced removal by dry deposition by a maximum 9 %. Thus the results indicated
an overall relatively small contribution of airborne chemical processes to O3 removal at this site.
1 Introduction
Tropospheric ozone (O3) is an important oxidant of many reactive species such as biogenic volatile organic compounds (BVOCs) emitted from the forest canopy (Bck et al., 2012; Smolander et al., 2014). It also plays a signicant role in the regulation of the atmospheric oxidation capacity by being one of the primary sources of the hydroxyl radical (OH), which is the most critical oxidant in the air (Mogensen et al., 2015). O3 also initiates the formation of Criegee intermediate (CI) radicals, which are crucial in tropospheric oxidation (Boy et al., 2013).
As an air pollutant, O3 can cause damage to human health (Kampa and Castanas, 2008) and affect ecosystem functioning via its various toxic impacts (Felzer et al., 2007). O3 can also alter the global radiative forcing as an important greenhouse gas (Stocker et al., 2013, chap. 2). Hence it is impor-
Published by Copernicus Publications on behalf of the European Geosciences Union.
1362 P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model
tant to understand the O3 budget, including its sources and sinks at local or site scale, in order to understand the global-scale implications.
O3 is produced via photochemical reactions in the presence of precursor gases, e.g. volatile organic compounds (VOCs), CO (carbon oxide), OH and NOx (nitric oxide and nitrogen dioxide), or transported downward from stratosphere and is removed mainly near the Earths surface.For vegetated surfaces a large part of the removal process occurs via stomatal uptake on leaf surface and non-stomatal uptake on plant canopies and soil surface (Wesely, 1989;Ganzeveld and Lelieveld, 1995; Altimir et al., 2006; Rannik et al., 2012; Launiainen et al., 2013), as well as depletion by chemical reactions (Kurpius and Goldstein, 2003;Wolfe et al., 2011). In this study we focus on the O3 removal and production processes within and immediately above the canopy, more particularly on the O3 uptake by boreal forest which covers 33 % of global forest land (Ruckstuhl et al., 2008).
For vegetation, the uptake of O3 depends on the turbulence intensity above and within the canopy, the diffusive transfer in the quasi-laminar boundary layer over the leaf surface, the biological properties of the plants, surface wetness condition and soil type (Ganzeveld and Lelieveld, 1995). Among them the effect of canopy wetness on O3 deposition has attracted a lot of attention in previous studies (e.g. Massman, 2004;Altimir et al., 2006). For different vegetation types and under different environmental conditions the surface wetness can enhance or reduce O3 deposition (Massman, 2004). For a boreal forest, a number of studies have revealed an enhancement of the O3 uptake under dew or high humidity conditions. For example, Lamaud et al. (2002) reported that dew on canopy surface signicantly increased the O3 uptake at night and in the morning over a pine stand. Altimir et al. (2006) also found that the condensed moisture on the surfaces enhanced the non-stomatal O3 uptake in a Scots pine forest when ambient relative humidity (RH) was over 60 70 %. Similarly to Altimir et al. (2006), Rannik et al. (2012) revealed a strong sensitivity of the nighttime O3 uptake to
RH. The enhancement of O3 uptake on wet leaf surface has been explained by previous studies, as both the microstructure of the leaf surface and the hydrophilic compounds existing on the leaf surface are able to facilitate the formation of the water lms or clusters, although the foliage surface itself is hydrophobic (Altimir et al., 2006). As a result, the different dissolved compounds like organics in the solution formed on leaf surface could react with O3 and thus enhance the O3 uptake (Altimir et al., 2006).
In addition, the boreal forest emits a large portion of BVOCs (Rinne et al., 2009), which are considered to play a signicant role in non-stomatal removal of O3 by oxidation (Kurpius and Goldstein, 2003; Goldstein et al., 2004;Wolfe et al., 2011). For example, Fares et al. (2010) found the correlation between the oxidation products of monoterpenes and O3 non-stomatal ux at a ponderosa pine stand
in California, USA, indicating that the gas-phase reactions of O3 with BVOCs were mostly responsible for O3 nonstomatal loss. In a model study, Wolfe et al. (2011) suggested that the non-stomatal O3 uptake at the same Californian site could be explained by considering the role of O3 destruction with the presence of very reactive BVOCs. Consequently, further analysis of the role of non-stomatal removal of O3 also strongly depends on the improvement of BVOCs measurement. However, the inuence of this gas-phase chemical removal process may vary among different sites. A study by Rannik et al. (2012), who conducted a detailed analysis of a long-term O3 deposition ux measurement at the same site as in this study (Station to Measure EcosystemAtmosphere Relations II (SMEAR II), a boreal forest station in Hyytil, Finland), indicated that, at the currently known strength of BVOC emissions, the air chemistry of BVOCs was not likely an important O3 sink term at this site.
During the last 2 decades, several numerical models have been developed to study and simulate O3 dry deposition processes under different climatic and environmental conditions.Many of them have implemented the big-leaf framework following the Wesely (1989) approach, which can be coupled to regional or global models to estimate the O3 deposition ux on large scales (e.g. Hardacre et al., 2015). However, the big-leaf approach does not consider explicitly the role of in-canopy interactions between biogenic emissions, chemistry, turbulence and deposition. Therefore, more detailed multi-layer models including the role of these in-canopy interactions have been developed and applied to analyse in-canopy deposition-related mechanisms (e.g. Ganzeveld et al., 2002b; Rannik et al., 2012; Launiainen et al., 2013). These multi-layer canopy exchange models have also been coupled to large-scale models, e.g. a global chemistryclimate model system (Ganzeveld et al., 2002a), or have been implemented in column models with detailed vertically separated layers (e.g. Wolfe and Thornton, 2011).
In this study a multi-layer O3 dry deposition model was implemented into the one-dimensional (1-D) chemical transport model SOSAA (a model to Simulate the concentrations of Organic vapours, Sulphuric Acid and Aerosols). This deposition model was based on the dry deposition representation originally described in Ganzeveld and Lelieveld (1995) and Ganzeveld et al. (1998) and implemented in the Multi-Layer Canopy CHemistry Exchange Model (MLCCHEM; Ganzeveld et al., 2002b). This canopy exchange system in MLC-CHEM was already applied in a single column model on the analysis of site-scale exchange processes (Ganzeveld et al., 2002b; Seok et al., 2013), as well as in a global chemistryclimate model system on the analysis of atmospherebiosphere exchange processes (Ganzeveld et al., 2002a, 2010).
Furthermore, the long-term continuous measurements and extensive campaigns at SMEAR II have provided a vast amount of data with complementary information on micrometeorology as well as O3 uxes and concentrations, which
Atmos. Chem. Phys., 17, 13611379, 2017 www.atmos-chem-phys.net/17/1361/2017/
P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model 1363
Figure 1. Vertical proles of all-sided LAI (leaf area index) and LAD (leaf area density), as well as the diagram of resistance analogy method used in the O3 dry deposition model. The overstorey layers and the bottom layer are considered separately. The bottom layer includes the broad-leaved understorey vegetation and soil surface. rac is the resistance representing the turbulent transport from the reference height of the understorey vegetation to the soil surface. rbs is the soil boundary layer resistance. rsoil is the soil resistance. rb is the quasi-laminar boundary layer resistance above the leaf surface. rveg represents the resistance to vegetation leaves, which is plotted on the right-hand side in detail. For broad leaves, the resistance to the side with (rveg1) or without (rveg2) stomata is computed separately. rstm is the stomatal resistance and rmes is the mesophyllic resistance. rcut is the cuticle resistance and rws is the resistance to wet skin. fwet is the wet skin fraction. All the variables are dened for each layer. Note that here LAI is the all-sided leaf area index for each layer. The symbols are also explained in the text and Table A1.
are highly appropriate for validating the new model and investigating more detailed processes. We selected a featured month August 2010 for such an extensive evaluation of the model because this month was characterised by exceptional hot and dry conditions in the rst 2 weeks, which possibly represented a future climate at this site (Williams et al., 2011), then followed by 2 cooler weeks. This study is a starting point of investigating gas dry deposition processes by using SOSAA. We aim to evaluate not only quantitatively O3 uxes and concentration proles but also the role of individual deposition processes at this site. This is a prerequisite for a further analysis of BVOCs deposition and chemistry in the follow-up research.
In the following section, a detailed description of the measurement and model will be shown. The comparisons between simulated and observed meteorological quantities, O3 uxes above the canopy and O3 concentration proles are described in Sect. 3, as well as the discussion about O3 ux proles and the impact of air chemistry. Finally, a summary is given in Sect. 4.
2 Methods
2.1 Site
All the measurement data used in this study were from
SMEAR II located in Hyytil, Finland (61 51[prime] N, 24 17[prime] E;
181 m a.s.l.) (Hari and Kulmala, 2005). The boreal coniferous forest is relatively homogeneous around the station in all the directions within 200 m, 75 % covered by Scots pine
(Pinus sylvestris) and the rest covered by Norway Spruces (Picea abies) and deciduous trees (Bck et al., 2012). The understorey vegetation mainly consists of lingonberry (Vaccinium vitis-idaea) and bilberry (Vaccinium myrtillus) with a mean height of 0.20.3 m. The forest oor is covered by dense mosses, mostly Dicranum polysetum, Hylocomium splendens and Pleurozium schreberi. Underneath is a 5 cm layer of humus in soil (Kolari et al., 2006; Kulmala et al., 2008). In 2010, the tree height reaches around 18 m. The all-sided leaf area index (LAI) is about 7.5 m2 m2, including 6.0 m2 m2 overstorey vegetation, 0.5 m2 m2 un
derstorey vegetation and 1 m2 m2 moss layer (Launiainen
et al., 2013). The vertical proles of LAI and leaf area density (LAD) are shown in Fig. 1.
2.2 Measurements
The measurement data at SMEAR II are currently publicly available in the data server maintained by AVAA open data publishing platform (http://avaa.tdata.fi/web/smart/smear
Web End =http://avaa.tdata./web/smart/smear ), which was originally introduced in Junninen et al. (2009). Several observed quantities used in this study are available at 4.2, 8.4, 16.8, 33.6, 50.4 and 67.2 m a.g.l., including air temperature (measured by Pt100 sensor), air water content (LI-COR LI-840 infrared light absorption analyser) and O3 concentration (TEI 49C ultraviolet light absorption analyser). Other observed quantities include the photosynthetically active radiation (PAR, 400700 nm) (LI-COR Li-190SZ quantum sensor) measured at 18 m, PAR (array of four LI-COR Li-190SZ sensors) measured at 0.6 m, net radiation (Reeman MB-1 net radiometer) at 67 m, O3 ux
www.atmos-chem-phys.net/17/1361/2017/ Atmos. Chem. Phys., 17, 13611379, 2017
1364 P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model
(Gill Solent HS 1199 sonic anemometer and Unisearch Associates LOZ-3 gas analyzer) at 23 m, friction velocity (Gill So-lent 1012R anemometer/thermometer) at 23 m, sensible and latent heat uxes (H and LE) (Gill Solent 1012R and LICOR LI-6262 gas analyzer) at 23 m and soil heat ux (Hukseux HFP01 heat ux sensors).
In this study the measured O3 uxes were calculated over 30 min averaging period using the EddyUH software (Mammarella et al., 2016) and according to standard methodology (for more details see Rannik et al., 2012). Other variables were also half-hour averaged to t the model time step for both input and output. The air temperature (T ), RH and O3 concentration were linearly interpolated using the observations collected at a height of 16.8 and 33.6 m to arrive at the estimated parameter values at 23 m to allow a direct comparison of the model results with the measurements or being used as input for the model. The missing observed data points of T , RH and O3 were gap-lled with the method described in
Gierens et al. (2014).
The measured O3 uxes were ltered based on the fact that previous studies showed that the measured uxes had large errors under very low turbulence (Rannik et al., 2006).The threshold of such low turbulence conditions was usually set according to the measured friction velocity on top of the canopy in the range of 0.1 to 0.25 m s1 (Altimir et al., 2006;
Rannik et al., 2012; Launiainen et al., 2013). Here the ob-served O3 uxes were excluded when u [lessorequalslant] 0.2 m s1, which
was proposed by Rannik et al. (2012). In addition, the O3 ux measurements were ltered out when precipitation occurred within preceding 1 h. Previous studies used a more strict criteria for such a lter that the preceding 12 h should keep dry to ensure dry canopy conditions (Altimir et al., 2006; Launiainen et al., 2013). However, in this study the fraction of wet canopy skin was taken into account and consequently we applied the ltering criteria of 1 h. Overall, 60 % of O3 ux data were available compared to 87 % prior to ltering.
Here we should notice that the uxes determined by the eddy-covariance (EC) technique were affected by the stochastic nature of turbulence, revealing as the random uncertainty of 30 min average uxes. For the EC measurement the random uncertainty was typically on the order of 10 to a few tens of percent. For the O3 turbulent ux measurement at the same site Keronen et al. (2003) presented the random error statistics, dened as 1 standard deviation of the random uncertainty of turbulent ux, ranging from about 10 to 40 %.
2.3 Classication of time period
Previous studies showed that in pine forest RH could enhance non-stomatal O3 uptake (Lamaud et al., 2002; Altimir et al., 2006; Rannik et al., 2012), especially during nighttime (Rannik et al., 2012). Hence in order to further analyse the impact of RH, the data were separated into different groups according to daytime (D) and nighttime (N) as well as RH measured inside the canopy, representing the daytime with high
humidity condition (DH), daytime with low humidity condition (DL), nighttime with high humidity condition (NH) and nighttime with low humidity condition (NL). The data points were considered as daytime when the sun elevation angle was larger than 10 and as nighttime when the sun elevation angle was smaller than 0 . The RH threshold value was set to 70 % as in previous studies (Altimir et al., 2006;Rannik et al., 2012), so a period is in high humidity condition when all the measured RH values inside the canopy are higher than 70 % and a period is in low humidity condition when all the measured RH values inside the canopy are lower than 70 %. For O3 ux, all was used to represent the time period with all available data after ltering described in section 2.2.
2.4 Model description
2.4.1 SOSAA
SOSAA is a 1-D chemical transport model which couples different modules to simulate the emissions of BVOCs, chemical reactions of organic and inorganic compounds in the air, transportation of trace gases and aerosol particles, as well as the aerosol processes within and above the canopy in the planetary boundary layer. It was rst introduced as SOSA by Boy et al. (2011) based on the 1-D version of SCADIS (SCAlar DIStribution; Sogachev et al., 2002). After that an aerosol module based on UHMA (University of Helsinki Multicomponent Aerosol model; Korhonen et al., 2004) was implemented by Zhou et al. (2014), resulting in its name being changed to SOSAA. The current version of SOSAA includes ve modules. The meteorology module is based on SCADIS. Emissions of BVOCs from the canopy are calculated by the Model of Emissions of Gases and Aerosols from Nature (MEGAN; Guenther et al., 2006). The Master Chemical Mechanism version 3.2 (MCMv3.2) (http://mcm.leeds.ac.uk/MCM
Web End =http: http://mcm.leeds.ac.uk/MCM
Web End =//mcm.leeds.ac.uk/MCM ) has been implemented to provide chemistry information. The nucleation, condensation, coagulation and deposition of aerosol particles are described by UHMA. In this study a gaseous compound dry deposition module has been implemented into SOSAA. SOSAA has already been applied and veried in several studies (e.g. Kurtn et al., 2011; Mogensen et al., 2011, 2015; Boy et al., 2013;Bck et al., 2012; Smolander et al., 2014; Zhou et al., 2015).
In SOSAA, the horizontal wind velocity (u and v), T , specic humidity (qv), turbulent kinetic energy (TKE) and the specic dissipation of TKE (!) are computed every time step (10 s) by prognostic equations. In order to represent the local to synoptic-scale effects, u, v, T and qv near and within the canopy are nudged to local measurement data at SMEAR II station with a nudging factor of 0.01. A TKE! parametrisation scheme is used to calculate the turbulent diffusion coefcient (Kt) (Sogachev, 2009):
Atmos. Chem. Phys., 17, 13611379, 2017 www.atmos-chem-phys.net/17/1361/2017/
P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model 1365
Kt = C[notdef]
TKE
! , (1)
"
TKE, (2)
where " is the dissipation rate of TKE and C[notdef] (0.0436) is a closure constant. Hence the turbulent ux of a quantity X (Ft,X) can be computed as
Ft,X = Kt
@X
! =
@z , (3)
where upward uxes are positive and vice versa. Specically, the H and LE at each model layer are computed as
H = Cp,airairKh
@T
@z + d
, (4)
@z , (5)
where Cp,air (1009.0 J kg1 K1) is the specic heat capacity at constant pressure. air (1.205 kg m3) is the air density, which is a constant in the model. d (0.0098 K m1) is the lapse rate of dry air. Lv (2.256 [notdef] 106 J kg1) is the latent heat
of vaporisation for water. Kh is the turbulent eddy diffusivity for heat uxes, which is derived from Kt according to the atmospheric stability.
The upper boundary values of u, v, T and qv are constrained by the ERA-Interim reanalysis dataset provided by the European Centre for Medium-Range Weather Forecasts (ECMWF; Dee et al., 2011). Above the canopy, the incoming direct and diffuse global radiations measured at SMEAR II station, and the long-wave radiation obtained from the ERA-Interim dataset are read in to improve the energy balance closure. Then the reection, absorption, penetration and emission of three bands of radiation (long-wave, near-infrared and PAR) at each layer inside the canopy are explicitly computed according to the radiation scheme proposed by Sogachev et al. (2002). At the lower boundary, the measured soil heat ux at SMEAR II is used to further improve the representation of surface energy balance. All the input data are interpolated to match the model time for each time step. With the input data, the mass and energy exchange between atmosphere and plant cover (including the soil underneath) and the radiation attenuation inside the canopy are optimal to simulate the micrometeorological drivers of O3 deposition at this site.
In current SOSAA, a modied version of MEGAN has been used to simulate the emissions of BVOCs from the trees. The emissions of some important BVOCs are included, e.g. monoterpenes ( -pinene, -pinene, [Delta1]3-carene, limonene, cineol and other minor monoterpenes OMT), sesquiterpenes (farnesene, -caryophyllene and other minor sesquiterpenes OSQ) and 2-methyl-3-buten-2-ol (MBO). The chemistry mechanism is from MCMv3.2, including necessary inorganic reactions and the full MCM oxidation paths
for methane (CH4), isoprene, MBO, -pinene, -pinene, limonene and -caryophyllene. We have also included the rst-order oxidation reactions with OH, O3, NO3 for cineole, [Delta1]3-carene, OMT, farnesene and OSQ. The related chemical reactions of stabilised Criegee intermediates (sCIs) with updated reaction rates from Boy et al. (2013) are also taken into account in current simulations. For more details about emissions and chemistry we refer to Mogensen et al. (2015).
2.4.2 Multi-layer O3 dry deposition model
A gas dry deposition model has been implemented into SOSAA to investigate the inuence of the dry deposition processes on the atmospherebiosphere gas exchange and in-canopy gas concentrations. In this study we focus on the O3 dry deposition since it is the basis of calculating the uptake of other trace gases, including BVOCs (Wesely, 1989). In this multi-layer dry deposition model the O3 deposition ux is calculated at each layer as
Fi = [O3]i [notdef] Vd,i (i = 1, ..., N), (6) where F is the O3 deposition ux (g m2 s1), [O3] is the O3 concentration (g m3) and Vd is the layer-specic conductance (m s1). The subscript i represents layer index. Layer 1 is the bottom layer including the soil surface and the understorey vegetation where the moss layer is considered as part of the soil surface for simplicity. The overstorey layers 2 to N include only vegetation surface, where N is the layer index at the canopy top.
Vd is calculated for bottom layer (layer 1) and overstorey layers (layers 2 to N) differently. In addition, the deposition onto dry and wet parts of the leaf surface is considered separately. In overstorey layers, only the deposition onto leaves is taken into account, while in the bottom layer the additional pathway of deposition onto the soil surface exists. Thus,
Vd,i = LAIiVdveg,i + i1Vdsoil, (7)
Vdveg,i =
1rveg,i , (8)
LE = LvKh
@qv
Vdsoil =
1rac + rbs + rsoil
. (9)
Here LAIi is the all-sided leaf area index for each layer (m2 m2). The Kronecker delta i1 ( i1 = 1 when i = 1;
i1 = 0 when i [negationslash]=1) is introduced here to simplify the for
mula. Vdveg,i is the layer-specic leaf surface conductance and Vdsoil is the soil conductance.
rveg is the leaf surface resistance which represents how O3 nally deposits onto different parts of leaf surface (Fig. 1). It can be calculated at each layer for needle leaves as
rveg = rb +
11/(rstm + rmes) + (1 fwet)/rcut + fwet/rws
. (10)
For broad leaves, O3 can deposit on a side without stomata or a side with stomata, and hence rveg is computed in a different way as
www.atmos-chem-phys.net/17/1361/2017/ Atmos. Chem. Phys., 17, 13611379, 2017
1366 P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model
rveg = 2
sensitivity analysis for rsoil will be shown in Sect. 2.6. The diagram of the resistance analogy parametrisation method described above is shown in Fig. 1. All the symbols are also explained and listed in Table A1.
In the model the evolution of O3 concentration is calculated for each layer by the prognostic equation
@ [O3]
@t =
@ @z
[parenleftbigg]
1 rveg1 +
1 rveg2
[parenrightbigg]
, (11)
rveg1 = rb +
1(1 fwet)/rcut + fwet/rws
, (12)
rveg2 = rb +
11/(rstm + rmes) + (1 fwet)/rcut + fwet/rws
. (13)
Here rb is the quasi-laminar boundary layer resistance over the leaf surface, which depends on molecular diffusivity and horizontal wind speed (Meyers, 1987), and rstm is the stomatal resistance which is derived from the stomatal resistance for water vapour (rstm,H2O) by using a factor of the molecular diffusivity ratio:
rstm =
DH2O
DO3
Kt @ [O3] @z
VdvegLAD + VdsoilAs
[parenrightbig]
[O3]
+Qchem, (17)
where the rst term on the right-hand side represents the vertical mixing of O3. The second term is the sink by dry deposition, which is non-zero only inside the canopy. The last one is chemistry production and loss of O3 for each model layer.
As (m2 m3) is the soil area index, which is the ratio between soil area and the model grid volume; hence it is non-zero only at the bottom layer, which includes the soil surface. All the other chemical compounds are also computed following this prognostic equation. According to Eq. (3) the O3 turbulent ux Ft in the model can be obtained as
Ft = Kt
@ [O3]
rstm,H2O. (14)
Here DH2O and DO3 are the molecular diffusivities of water vapour and O3, respectively. rstm,H2O is computed by
SCADIS module in SOSAA and also used to calculate LE and thus the energy balance (Sogachev et al., 2002). rmes is the mesophyllic resistance, which can be ignored for O3 (0 s m1). rcut (105 s m1) is the cuticle resistance and rws (2000 s m1) represents the uptake on leaf wet skin. Their values are taken from Ganzeveld and Lelieveld (1995). Canopy wetness is represented by the fraction of wet skin fwet, which is determined by RH (Lammel, 1999; Wu et al., 2003):
fwet =
8 >
<
>
:
@z . (18)
2.5 Model set-up
In this study the newly implemented O3 dry deposition module was applied to simulate the time period from 1 to 31 August 2010 (Julian day 213 to 243). The model column domain was set from 0 m at ground surface up to 3000 m with 51 layers logarithmically congured, including the whole planetary boundary layer and part of the free atmosphere on top of it.We also constrained the model with the site-specic vegetation cover properties as presented before in Sect. 2.1. The overstorey layers only included needle-leaved part of Scots pine trees above 0.3 m. Below that there was the under-
storey vegetation and ground surface. Since the understorey consisted of vegetation with leaves instead of needles, the parametrisation method for the understorey vegetation was considered the same as that for broad-leaved species. In order to secure a more accurate representation of canopy wetness which was also relevant to the calculation of the layer-specic conductance for O3, RH values inside the canopy were calculated from the measured absolute humidity and simulated air temperature.
In addition, to secure a realistic simulation of O3 in a column model like SOSAA we also forced the models O3 concentration at 23 m to resemble the observed value every time step; the O3 concentrations at other levels were then calculated by Eq. (17). In this way, we implicitly added the role of advection in determining the O3 concentration above the canopy. The gap-lled observed values which were used for the forcing are shown in Fig. 2b.
Several sensitivity cases have been conducted in this study (Table 1). In the case BASE all the parameters and
Atmos. Chem. Phys., 17, 13611379, 2017 www.atmos-chem-phys.net/17/1361/2017/
1 0.9[lessorequalslant]RH RH 0.7
0.2 0.7[lessorequalslant]RH < 0.9 0 RH < 0.7
. (15)
The threshold 70 % is suggested by Altimir et al. (2006).
rac is the resistance representing the turbulent transport
from the reference height of the understorey vegetation to the soil surface. Since the gas transport is explicitly calculated in SOSAA and the bottom layer height is only 0.3 m, the
turbulence resistance between vegetation and ground is expected not to be an important factor for soil deposition, and consequently we have set rac to zero. rbs is the soil boundary layer resistance which is calculated as (Nemitz et al., 2000; Launiainen et al., 2013)
rbs =
Sc ln( 0/z )
u g
. (16)
Here Sc (1.07) is the Schmidt number for O3. is the von Krmn constant (0.41). 0 = DO3/( u g) is the height
above ground where the molecular diffusivity is equal to tur
bulent eddy diffusivity. z (0.1 m) is the height under which
the logarithmic wind prole is assumed. u g is the friction ve
locity near the ground. rsoil is the soil resistance; 400 s m1 is used here according to Ganzeveld and Lelieveld (1995). A
P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model 1367
Figure 2. (a) Modelled (solid line) and measured (dots) time series of air temperature (T , red) and the measured ambient relative humidity (RH, blue) at 23 m above the ground. (b) Measured O3 concentration (blue) at 23 m above the ground. The time period is August 2010.
methods were kept the same as described in Sect. 2.4. In cases RSOIL200, RSOIL600 and RSOIL800 rsoil was altered to 200, 600 and 800 s m1, respectively. In the case FREEO3, the O3 concentration at 23 m was computed from Eq. (17) instead of being set to the measurement data.
2.6 Sensitivity analysis of rsoil
rsoil varied in different studies, ranging from 10 to 180 s m1 for dry soil and 180 to 1100 s m1 for wet soil (Massman, 2004). In this study the dry deposition module was developed on the basis of the model from Ganzeveld and Lelieveld (1995) in which rsoil was 400 s m1. In order to assess the uncertainties involved in estimating rsoil, different values of rsoil ranging from 200 to 800 s m1 were tested in this study (Table 2). As can be expected, the modelled O3 uxes decrease as rsoil increases. The BASE case shows the best performance in general, although it overestimates 16 % night
time O3 uxes. Since the RSOIL200 case overestimates O3 uxes by 17 % in average for the whole month, 12 %
at daytime and 35 % at nighttime, the RSOIL200 sensitiv
ity case indicates that using this lower estimate, a value that might be more appropriate for high organic (and dry) soils, seems not to represent properly the role of soil removal at this site. However, taking higher resistance values, e.g. one of 600 or 800 s m1, seems to result in a better simulation of the role of the soil uptake at nighttime. However, considering the overall performance and better estimation of daytime O3 uxes, we used 400 s m1 as the soil resistance in this study.
Table 1. Table of sensitivity cases. The case names and their short description texts are shown.
Name Description
BASE the same as described in Sect. 2.4 RSOIL200 rsoil = 200 s m
1
RSOIL600 rsoil = 600 s m
1
RSOIL800 rsoil = 800 s m
1
FREEO3 O3 concentration at 23 m was also computed instead of using observed data
3 Results and discussion
3.1 Micrometeorology
The simulated month was warm and dry with little precipitation. Moreover, the temperature decreased dramatically in the middle of the month. In the rst half of month (1 to15 August) the average temperature at 23 m was 19.0 C, while it dropped to 12.1 C in the second half of month (16 to31 August) (Fig. 2a). Analysis of the full temperature record indicated that this transition in the weather conditions at the site was well simulated by the model. RH varied inversely with air temperature. Its average value increased only slightly from 66.0 % in the rst half of the month to 69.3 % in the second half. However, a dramatic increase of daily mean RH values from 49.3 to 73.5 % occurred between 20 and 21 August (Fig. 2a). The combination of the dry weather and the large variation of temperature provided a good sample for verifying the O3 dry deposition module.
Figure 3 shows the comparison results between simulated and measured horizontal wind speed and friction velocity (u ). Both of them are essential for estimating the
www.atmos-chem-phys.net/17/1361/2017/ Atmos. Chem. Phys., 17, 13611379, 2017
1368 P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model
Table 2. The average (mean) and standard deviation (SD) of modelled and measured O3 uxes (g m2 s1) above the canopy during different time periods (All for the whole month, D for daytime, N for nighttime) for different cases (OBS for measurement, BASE for basic settings used in this study, RSOIL200 uses the same settings as in BASE except rsoil = 200 s m
1, similarly, RSOIL600 with
1) are shown. The relative error (RE) of modelled O3 ux compared to the observation (Ft,mod Ft,obs)/Ft,obs is also presented.
Cases All D N
Mean [notdef] SD RE Mean [notdef] SD RE Mean [notdef] SD RE OBS 0.246 [notdef] 0.175 0.334 [notdef] 0.165 0.103 [notdef] 0.073
RSOIL200 0.286 [notdef] 0.173 +16.4 % 0.375 [notdef] 0.162 +12.1 % 0.140 [notdef] 0.067 +35.0 %
BASE 0.250 [notdef] 0.153 +1.77 % 0.329 [notdef] 0.143 1.74 % 0.120 [notdef] 0.059 +16.2 %
RSOIL600 0.231 [notdef] 0.144 6.00 % 0.305 [notdef] 0.134 8.85 % 0.109 [notdef] 0.057 +5.16 %
RSOIL800 0.219 [notdef] 0.139 10.8 % 0.290 [notdef] 0.129 13.2 % 0.101 [notdef] 0.055 2.17 %
rsoil = 600 s m
1 and RSOIL800 with rsoil = 800 s m
Figure 3. Modelled (red solid line for daytime, red dashed line for nighttime) and measured (blue solid circle for daytime, blue empty circle for nighttime) proles of horizontal wind speed (windh) (a) and friction velocity (u ) (b). Nighttime values are shifted by 3 and 1 m s1 for wind and u for clarity of presentation, respectively. The ranges of [notdef]1 SD (standard deviation) of modelled and measured data are marked
as shades and error bars. The height is normalised by canopy height hc. The monthly-mean diurnal cycles of modelled (red) and measured (blue) friction velocity at 23 and 3 m are shown in (c) and (d). The ranges of [notdef]1 SD are marked as shades in the same colours.
turbulent transport above and within the canopy as well as for the calculation of the quasi-laminar boundary layer resistance of leaves (rb) at each canopy layer and the soil boundary layer resistance (rbs). Figure 3a shows the good agreement between modelled and measured monthly-mean horizontal wind speed proles during both daytime and nighttime. The wind speed decreases quickly inside the canopy due to canopy drag, then changes little below 0.5 hc until near the surface where wind speed varies logarithmically to zero on the surface. The model reproduces the diurnal cycle of u but overestimates the nighttime values by 0.05 m s1 in
average above the canopy (Fig. 3c). Below the canopy crown at 3 m, u is underestimated by 0.02 m s1 at nighttime
and 0.05 m s1 at daytime (Fig. 3d). The discrepancy is
likely due to the limitation of representing the real heterogeneous dynamics by a 1-D model with homogeneous canopy conguration.
3.2 PAR above and below the canopy crown
PAR plays an important role in the stomatal exchange which determines to a large extent the daytime vegetation uptake. The PAR above the canopy is calculated directly from the measured incoming short-wave radiation serving as input to the model and shows a daytime maximum of about 250 300 W m2 during the simulation month. The PAR inside the canopy is calculated by considering the absorption, reection and scattering effects of canopy leaves in the model (Sogachev et al., 2002). The comparison between modelled and observed PAR at 0.6 m below the canopy crown is shown
in Fig. 4. The monthly-mean diurnal cycle of attenuated PAR below the canopy crown in the model is consistent with the observation except two missing peaks at daytime (Fig. 4b). These two peaks in the measurement are the consequence of direct exposure of PAR sensors to incoming solar radiation.
Atmos. Chem. Phys., 17, 13611379, 2017 www.atmos-chem-phys.net/17/1361/2017/
P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model 1369
Figure 4. (a) Time series of PAR at 0.6 m from model (red) and measurement (blue) in August 2010. (b) The monthly averaged diurnal cycle of time series in (a) for model (red) and measurement (blue). The range of [notdef]1 SD is marked by the shade with the same colour.
Such situation always occurs when point-wise measurement is compared with a model assuming a homogeneous forest canopy.
3.3 Energy balance
The monthly-mean diurnal cycles of sensible heat ux, latent heat ux, net radiation and soil heat ux are shown in Fig. 5 in order to verify the simulated energy balance above the canopy. The upward energy ux or the loss of surface energy is represented by positive values. During daytime, the soil and canopy loses energy by heat uxes and gains energy mainly from net incoming solar radiation. At night, the surface loses energy by net upward long-wave radiation with an average rate of 33 W m2, which is partly compensated by 20 W m2 downward sensible heat ux.
During the simulation period the modelled diurnal cycles of energy uxes agree well with the observation, although, for example, the latent heat ux is slightly underestimated by 30 W m2 during daytime. In the afternoon from
14:00 to 20:00 LT, the sensible heat ux is underestimated by
20 W m2. This could be explained by the underestimation in net radiation. However, the modelled values are generally within the 1 standard deviation range of the observations. The agreement between modelled and measured latent heat ux also indicates that the stomatal exchange, which controls the latent heat ux and is directly related to the stomatal resistance of O3 and many other gaseous compounds, is realistically simulated as a function of the meteorological drivers.
Figure 5. The monthly averaged diurnal cycle of different energy ux terms at 23 m above the ground for model (dashed lines) and measurement (solid lines), including sensible heat ux (H, red line), soil heat ux (Gsoil, green line), upward net radiation (Rnet, purple line, note the observed Rnet is at 67 m) and latent heat ux (LE, blue line). The range of [notdef]1 SD for measurement data is plotted for every
term by the shade with the same colour.
3.4 O3 uxes
The modelled time series of O3 turbulent ux and its diurnal cycle are compared with the measurement data above the canopy (Fig. 6). In general, the modelled ux shows a good agreement with the observations especially in the second half of month (Fig. 6a). Large discrepancies mostly occur in the rst half of month, which is warm and dry. On the rst 3 days
www.atmos-chem-phys.net/17/1361/2017/ Atmos. Chem. Phys., 17, 13611379, 2017
1370 P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model
Figure 6. (a) Time series of the simulated (red) and measured (blue) O3 turbulent uxes above the canopy in August 2010. (b) The monthly averaged diurnal cycles of time series presented in (a) for the model (red) and measurement (blue). The ranges of [notdef]1 SD are marked by the
shades with the same colours. Positive values represent downward uxes.
of the month, the O3 turbulent ux is overestimated by the model. On some days at noon (e.g. 9, 12, 13, 14, 27, 30 August), the model is not able to predict the observed high peaks of O3 turbulent uxes. In an average diurnal cycle of O3 turbulent ux the model does not capture the rapid increase of downward O3 turbulent ux in the morning, but it follows the measurement well after 10:00 LT. In general the agreement between the simulated and measured monthly-mean diurnal cycles of O3 turbulent uxes is promising.
Figure 7 shows the correlation between the simulated and measured O3 turbulent uxes above the canopy for different humidity conditions at daytime and nighttime separately. The overall R2 between the modelled and measured O3 turbulent uxes for the whole dataset is 0.47. Among the four individual datasets under different conditions, the best prediction by the model occurs for the NH data points with R2 of 0.37, followed by the results reecting the daytime high humidity conditions (R2 = 0.19). Note that these conditions
with highest correlations are also the conditions with high
RH, especially at nighttime. All the correlations are significant (p < 0.001) except the condition NL for which R2 is
only 0.02 (Fig. 7). This indicates the difculty of simulating the O3 turbulent ux in weak turbulence at nighttime. Usually at nighttime RH is larger than 70 % (Fig. 2); under this condition (NH condition), the wet skin uptake contributes more than 50 % (Table 3) to the deposition ux. Therefore, the turbulent mixing above the ground which affects the deposition ux onto soil only plays a minor role on the deposition ux above the canopy. However, in NL condition, which does not happen frequently, nearly all the deposition inside the canopy is caused by soil deposition. Hence, the difculty of simulating the exchange processes near the sur-
Figure 7. Scatter plots of modelled versus measured O3 turbulent uxes above the canopy. The data points are plotted separately for different groups (DH, DL, NH and NL) with their R2 values shown in the legend. R2 of the whole dataset is shown below the legend.
face may cause more uncertainty of simulating the deposition ux onto soil surface under NL condition than NH condition. Moreover, the vertical advection of O3 could also affect the turbulent ux at nighttime (Rannik et al., 2009), which complicates the analysis. In contrast, there are only 69 ob-served data points in the condition NL, which implies larger
Atmos. Chem. Phys., 17, 13611379, 2017 www.atmos-chem-phys.net/17/1361/2017/
P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model 1371
Figure 8. Measured average vertical proles of O3 concentration for the whole month (dark grey, the horizontal bars are [notdef]1 SD) and
individual conditions (daytime under high humidity condition, labelled as DH with red lled circle; daytime under low humidity condition, labelled as DL with blue lled circle; nighttime under high humidity condition, labelled as NH with red empty circle; nighttime under low humidity condition, labelled as NL with blue empty circle). Modelled results are plotted as solid lines (daytime) and dashed lines (nighttime) with the same colour as measurement. The height is normalised by the canopy height hc.
Table 3. The rst four columns are the contribution fractions of different deposition pathways (stm as stomatal uptake, wet as wet skin uptake, cut as cuticle uptake, soil as soil surface uptake) in the integrated O3 deposition ux inside the canopy in the model.
The last column is the sub-canopy (below 4.2 m) O3 turbulent ux (Ft,mod (4.2 m)) compared to the O3 turbulent ux above the canopy (Ft,mod) in the model. Different conditions are listed along the row. D and N represent daytime and nighttime and H and L represent high and low humidity, respectively. ALL is for the whole dataset.
Stm Wet Cut Soil Ft,mod (4.2 m)/Ft,mod
D 63.0 % 3.79 % 1.12 % 32.1 % 38.0 % N 3.70 % 40.5 % 1.87 % 53.9 % 59.5 % DH 47.2 % 18.5 % 0.94 % 33.4 % 39.6 % DL 67.1 % 0.00 % 1.17 % 31.8 % 37.6 % NH 3.28 % 51.0 % 1.04 % 44.7 % 51.4 % NL 5.42 % 1.78 % 4.73 % 88.1 % 89.5 % ALL 52.5 % 10.4 % 1.25 % 35.8 % 41.7 %
random uncertainty. When the surface is wetter, the simulated nocturnal O3 turbulent uxes correlate much better with the measurement. In addition, the measurement data show a larger range of variation (about 1.20.0 g m2 s1) com
pared to the range in the modelled O3 turbulent ux (about
0.80.0 g m2 s1), which implies that the model does not capture the O3 turbulent ux peaks or the measurements are more scattered due to random errors. Regarding the low R2 values here, we should consider the uncertainty of measured uxes. Such uncertainty contributes to the data scattering when comparing the modelled and measured uxes, such as in Fig. 7, and reduces the correlation statistics.
In general, the parametrisation of wet skin fraction (Eq. 15) and its impact on O3 non-stomatal removal seems to represent the O3 deposition mechanisms inside the canopy well considering the good performance under high humidity conditions. Although the prediction of O3 turbulent ux with weak turbulence at night under low humidity condition still has large uncertainties (Fig. 7), the simulated average nocturnal O3 turbulent ux above the canopy shows a good agreement with the observation (Fig. 6b).
3.5 O3 concentration prole
In order to assess whether the good agreement between the observed and simulated O3 turbulent uxes above the canopy also implies a realistic representation of the O3 concentration inside the canopy, we have conducted an evaluation of the simulated in-canopy O3 concentration prole. The 1-month averaged O3 concentration proles from model results and measurements are shown in Fig. 8. The large variation range results from the meteorological variations in this month, especially the abrupt transition in the middle of the month (Fig. 2). The average O3 concentration of the whole month is 60.4 g m3 at 23 m, then decreases gradually inside the canopy to 54.1 g m3 at 4.2 m due to the in-canopy sinks.Similar vertical gradients are also found for the four different conditions. At night, the turbulent mixing is weaker compared to daytime, which inhibits the downward transport of air with larger concentration of O3 into the canopy. Hence the
O3 removal by canopy and especially by soil surface results in larger gradient of O3 inside the canopy during nighttime (Fig. 8).
The model results of O3 concentration proles show a good agreement with the observations except the slight overestimation for the DH condition below 8 m (0.45 hc) and
the apparent underestimation for the NL condition through-
www.atmos-chem-phys.net/17/1361/2017/ Atmos. Chem. Phys., 17, 13611379, 2017
1372 P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model
the cumulative uptake on leaf surfaces increases with height due to dense leaves in the plant crown area. Above 0.8 hc the remaining small portion of biomass ( 7 %) provides less
than 2 % O3 uptake compared to the total O3 deposition.
The soil uptake contributes to the total O3 deposition ux at both daytime and nighttime (Fig. 9b and c) with a percentage of 32 and 54 %, respectively. At daytime, 63 %
of the O3 deposition ux is due to stomatal uptake, while at nighttime, when RH is larger than 70 % at most of the time, the cumulative wet skin uptake contributes 41 % to
the total O3 deposition. At nighttime under high humidity conditions, the wet skin uptake even contributes 51 % to
the total O3 deposition uxes (Table 3). This indicates that wet skin uptake plays a crucial role at night, which is consistent with the results in Rannik et al. (2012). As a result, the simulated averaged non-stomatal contribution to the integrated O3 deposition ux above the canopy is 37 % dur
ing daytime and 96 % during nighttime (Table 3). It should
be noted that the stomata are not completely closed at night (Caird et al., 2007) and the minimum stomatal conductance at nighttime is about 5 % of its maximum value at daytime (Kolari et al., 2007), which is similar with the simulation result here (3.7/63.0 % 6 %, Table 3).
Above 0.2 hc, the stomatal uptake (DL, Fig. 9b), wet skin uptake (NH, Fig. 9c) or both of them (DH, Fig. 9a; NL, Fig. 9d) start to play a signicant role in the cumulative O3 deposition uxes. Hence at 0.8 hc the cumulative contribution of soil deposition is less than 50 % except in the NL condition when both the cumulative stomatal uptake and wet skin uptake are limited. In all four conditions the dry cuticle
Atmos. Chem. Phys., 17, 13611379, 2017 www.atmos-chem-phys.net/17/1361/2017/
Figure 9. Simulated vertical proles of cumulative O3 deposition ux normalised by the integrated O3 deposition ux above the canopy (cum, solid black line) for four conditions: DH (a), DL (b), NH (c) and NL (d). D and N represent daytime and nighttime and H and L represent high and low humidity, respectively. Shaded areas are the cumulative contribution fractions for different deposition pathways, including stomatal uptake (stm, red), cuticle uptake (cut, green), wet skin uptake (ws, blue) and soil uptake (soil, pale brown). The all-sided LAI prole for each layer and LAD is plotted again here (e). The height is normalised by the canopy height hc.
out the whole canopy. This is consistent with the model results of the O3 turbulent uxes, which show 20 % under
estimation for the DH condition and 38 % overestimation
for the NL condition. In addition, the modelled vertical gradient of O3 concentration during nighttime at drier conditions (NL) is much larger inside the canopy compared to the measured gradient, which implies that the soil deposition is largely overestimated when the soil and dry vegetation surface uptake dominates the overall removal inside the canopy.This also indicates that further investigation is needed for the more accurate representation of ground surface deposition at different humidity conditions, including possibly the roles of uptake by the moss layer and soil humus layer.
3.6 O3 ux prole
The normalised cumulative O3 deposition ux at layer i can be obtained as
Fc,i =
i
Pk=1 Fk N
Pk=1 Fk
, (19)
where Fk is the O3 deposition ux at layer k and N is the layer index just above the canopy. The proles of Fc and the contributions of different deposition pathways for four different conditions are shown in Fig. 9. For the whole month, the O3 uptake is dominated by soil deposition below 0.2 hc ( 3.6 m) with only 8 % contribution from the understorey
vegetation via stomatal uptake. From 0.2 to 0.8 hc ( 14.4 m)
P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model 1373
uptake is minor with a maximum contribution of about 5.0 % for the NL condition.
During daytime the sub-canopy layer, including soil surface, contributes about 38 % to the integrated O3 deposition, which is consistent with the results from Launiainen et al. (2013) in which the sub-canopy (lower than 4.2 m) contribution was 3545 % at daytime. At night the contribution increases to around 60 % due to the closed stomata in crown layers. This is much higher than that (2530 %) in Launiainen et al. (2013) (Table 3). The overestimation could result from the underestimation of the soil resistance, which is dif-cult to determine in such a complex ground ecosystem. However, among these four different conditions with the same constant soil uptake efciency, only under the nocturnal dry conditions (NL) there is apparently an overestimation in O3 uptake and consequently underestimation of the O3 concentration inside the canopy (Fig. 8). Therefore, we expect that the poor performance for the NL condition also results from the limited data amount under this condition (only 69 data points), which leads to larger ratio of random uncertainty and thus smaller R2.
Moreover, the assumption that the resistance rac between the understorey vegetation and ground is not a limiting factor for soil deposition might not hold under certain conditions. However, Launiainen et al. (2013) studied a period (1 July to 4 August 2010) 1-month earlier than the time period (1 to 31 August 2010) in this study, so the difference between these two studies could also be due to the meteorological and biological variations during the two summer months.However, the daytime contribution of the sub-canopy layer is consistent, so the difference between the 2 months could only play a minor effect.
3.7 Contribution of air chemistry
The role of chemical processes in explaining the O3 removal inside the forest canopy has been discussed in previous studies (e.g. Altimir et al., 2006; Wolfe et al., 2011; Rannik et al., 2012; Launiainen et al., 2013). A study by Wolfe et al. (2011) found that the non-stomatal uptake over a Ponderosa pine stand in the US was associated with additional very reactive BVOCs being present besides the identied ones. In contrast, Rannik et al. (2012) suggested that the air chemistry provided only minor contribution at SMEAR II. In order to estimate the contribution of chemical removal at SMEAR II, two different studies applied multi-layer models (Rannik et al., 2012; Launiainen et al., 2013) to simulate the O3 uxes and concentration inside the boreal forest canopy. However, both of them showed their limitations on estimating the chemical contribution. Rannik et al. (2012) only considered one chemical reaction of O3 with -caryophyllene. In Launiainen et al. (2013), they simplied the chemical production and loss of O3 with only two parameters to represent the rst-order kinetic sink and photochemical production. In this study, we implemented a chemistry module with a detailed
list of chemical reactions (see Sect. 2.4.1), which was able to provide a more accurate estimation of chemical removal of O3 inside the canopy.
In order to get rid of the effect of synoptic-scale transport of O3 and only focus on the local sinks and sources, we applied the simulation case FREEO3. In this simulation case we ignored the role of advection and only considered the role of local sources and sinks inside the canopy, i.e. dry deposition, chemical production and loss, and turbulent transport. Here the time period from 5 to 14 August were selected from the simulation results to analyse the local chemical contribution, because the modelled O3 concentration t to the measurement the best during this period out of the whole month for the case FREEO3, which indicated that the advection also did not have an apparent effect on the local observed O3 variation. The daily averaged (from 5 to 14 August) production and loss of total O3 inside the canopy per square metre caused by dry deposition (Fdepo) and chemistry (Fchem) are plotted in Fig. 10. Positive values correspond to O3 production and negative values represent O3 loss. Here the chemistry production is a net effect of O3 loss reactions and photochemical production. Fdepo (obviously negative) shows a maximum O3 loss rate at about 14:00 LT. The chemistry produces O3 from morning at 06:00 LT to the afternoon
at 15:00 LT and destroys it throughout the other time of
the day, especially at nighttime (Fig. 10). The ratio between Fchem and Fdepo shows that chemical removal has its largest contribution of 9 % of the dry deposition sink in average
at nighttime from 20:00 to 04:00 LT. At daytime, our model simulation indicates that the O3 production caused by chemistry can compensate up to 4 % of dry deposition loss in
average. However, during the selected period, the chemical contribution and compensation can reach up to 24 and 20 % at most. This indicates that in general chemistry has minor impact on O3 alteration, but at some specic time the chemical production and removal of O3 can still play a signicant role.
As a comparison, we also calculated the timescales of different removal processes to estimate the contribution of air chemistry. The average value of measured O3 ux (FO3,avg) in August 2010 above the canopy was0.33 g m2 s1 at daytime and 0.10 g m2 s1 at nighttime whereas the O3 concentration ([O3]) inside the canopy was about 61.6 g m3 during daytime and 50.5 g m3 at night.So the timescale of total O3 ux (O3) could be obtained from
O3 = [O3]hc/FO3,avg, (20) which was 3400 s ( 1 h) for daytime and 9100 s
( 2.5 h) for nighttime. However, the total O3 reactivity (y) at
18 m during a similar time period and at the same boreal forest station was calculated by Mogensen et al. (2015), which was 1.58 [notdef] 105 and 1.67 [notdef] 105 s1 for noon and 02:00 LT
at night. If the same values were assumed to be applicable also inside the canopy, the timescale of the O3 removal by chemistry (c,O3),
www.atmos-chem-phys.net/17/1361/2017/ Atmos. Chem. Phys., 17, 13611379, 2017
1374 P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model
Figure 10. (a) The daily averaged (from 5 to 14 August) production and loss caused by chemistry (Fchem, red) and dry deposition (Fdepo, blue). (b) The ratio between Fchem and Fdepo. Zero lines for Fchem and the ratio are plotted as dashed lines. Shaded areas show the range of
1 SD.
c,O3 = y1, (21)
was 63 300 s ( 18 h) for daytime and 59 900 s ( 17 h)
for nighttime. These estimates showed that the chemical removal accounted for about 5 % (3400/63 300) and 15 % (9100/59 900) of the total O3 removal within the canopy at daytime and nighttime, respectively.
Compared to the simulation results, the timescale analysis could not reect the photochemical production of O3 during daytime, and hence the estimation of net chemical effects is not possible with this method. For nighttime, the timescale analysis overestimates the average contribution of chemical removal by about 88 % (15 % compared to 8 %, 8 % is obtained from 9 %/(100 % + 9 %)). The comparison result
could act as a proof of the statement in Wolfe et al. (2011), which argued that the timescale might not be a good criteria of chemical inuence.
4 Summary
A detailed multi-layer O3 dry deposition model has been implemented into SOSAA to investigate the O3 uptake by canopy and soil surface at a boreal forest station SMEAR II. The presented detailed analysis of the O3 deposition processes for this site also quantied various removal processes, e.g. by the dry and wet cuticle, by stomatal uptake and by the soil surface.
In this model the fraction of wet skin on canopy leaves was parametrised according to RH values to analyse the potential role of canopy wetness on O3 deposition for both high and low humidity conditions. Moreover, the multi-layer model also enabled the study of deposition processes inside the canopy and the partitioning of O3 deposition uxes between the canopy crown and sub-canopy. In this study, the
model has been validated by comparing the modelled and measured O3 turbulent ux above the canopy and its concentration prole inside the canopy.
Further investigation has been done through a more in-depth correlation analysis on O3 turbulent uxes for nighttime and daytime under high and low humidity conditions. The simulated O3 turbulent uxes above the canopy correlated reasonably well with the measurement for the whole month with R2 of 0.47 (p < 0.001), which was also consistent with the plausible prediction of O3 concentration prole inside the canopy. The signicant correlation (p < 0.001) also applied to the daytime humid and dry as well as nighttime humid conditions (DH, DL and NH) with R2 of 0.19, 0.16 and 0.37. However, the model was not able to predict high peaks with O3 turbulent uxes larger than0.8 g m2 s1. The model also did not capture well the measured O3 removal for the nocturnal dry condition (NL), in which R2 was only 0.02 and the O3 concentration inside the canopy was largely underestimated (Figs. 7 and 8). The main reason could be the uncertainty of simulating the exchange processes near the ground in weak turbulent condition at nighttime when the soil deposition dominated the deposition ux inside the canopy.
Nearly all of the O3 uptake occurred below 0.8 hc inside the canopy. During daytime, the contributions of stomatal uptake ( 47 %), wet skin uptake ( 19 %) and soil uptake
( 33 %) were signicant for the total O3 uptake under high
humidity conditions, while under low humidity conditions the stomatal ( 67 %) and soil uptake (32 %) contributed
dominantly the overall canopy deposition. During nighttime, the stomatal uptake contribution ( 3 %) was not zero but
was much smaller compared to the wet skin uptake ( 51 %)
under high humidity conditions. For the low humidity condition at night, nearly all the deposition ( 88 %) was due to
soil uptake. Since RH was larger than 70 % at most of the
Atmos. Chem. Phys., 17, 13611379, 2017 www.atmos-chem-phys.net/17/1361/2017/
P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model 1375
time during night, the uptake by wet canopy could be a dominant factor for the nocturnal O3 removal. In addition, the simulated non-stomatal contributions to the integrated O3 deposition uxes were estimated as about 53, 33, 97 and 95 % for conditions DH, DL, NH and NL, respectively (Table 3).
The modelled contribution of sub-canopy deposition during daytime ( 38 %) was consistent with that (3545 %) in
Launiainen et al. (2013), but it was much higher at nighttime ( 60 %) compared to that in the same study (2530 %) (Ta
ble 3). This discrepancy at nighttime was most likely due to the overestimation of soil uptake.
The contribution of O3 removal by chemical reactions with currently identied BVOCs has also been evaluated. In general the air chemistry played a minor role in O3 uptake inside the canopy. In the simulated averaged diurnal cycle, the air chemistry produced O3 during daytime from about 06:00 to 15:00 LT, compensating up to 4 % of dry deposition sinks, while at nighttime the chemical loss enhanced O3 removal by
9 % of that by dry deposition. A qualitative estimation of chemical contribution with timescale analysis was also conducted as a comparison. However, this method overestimated the air chemical removal by about 88 % for nighttime and it was not able to reect the O3 production at daytime.
This study is the rst step to establish a detailed gas dry deposition model in SOSAA. Further analysis of dry deposition will be done for other chemical compounds, especially for BVOCs. This will improve not only the ability to simulate air chemistry and aerosol processes but also our understanding of the mechanisms involved in the removal processes at boreal forest. In addition, it is also of scientic interest to investigate how future climate change might ultimately affect the removal processes of compounds like O3 and BVOCs for boreal forests.
5 Data availability
The model code of SOSAA and the output data of the simulations can be obtained by emailing Zhou Putian (putian.zhou@helsinki.).
www.atmos-chem-phys.net/17/1361/2017/ Atmos. Chem. Phys., 17, 13611379, 2017
1376 P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model
Appendix A
Table A1. Table of symbols.
Symbol Value Unit Description
hc 18 m canopy heightLAI m2 m2 integral all-sided leaf area index, it can also represent the LAI at each layer in the context
T K air temperature qv kg m3 specic humidity
RH relative humidity X scalar quantity u m s
1 friction velocity u g m s
1 friction velocity near the ground H W m2 sensible heat ux
LE W m2 latent heat ux Ft,X turbulent ux of X
Ft g m2 s1 O3 turbulent uxKt m2 s1 turbulent eddy diffusivity
Kh m2 s1 turbulent eddy diffusivity for heat uxes TKE m2 s2 turbulent kinetic energy
" m2 s3 dissipation rate of TKE! s1 specic dissipation of TKE
Cp,air 1009.0 J kg1 K1 latent heat uxair 1.205 kg m3 air density d 0.0098 K m1 lapse rate of dry air
Lv 2.256 [notdef] 10
6 J kg1 latent heat of vapourisation for water C[notdef] 0.0436 closure constant in calculating KtAs m2 m3 soil area index
Qchem g m3 s1 chemical production and loss F g m2 s1 O3 deposition ux
O3
[bracketrightbig]
g m3 O3 concentrationVd m s1 layer-specic conductance for O3
Vdveg m s1 layer-specic leaf surface conductanceVdsoil m s1 soil conductancerveg s m1 leaf surface resistancerveg1 s m1 leaf surface resistance to the side without stomatarveg2 s m1 leaf surface resistance to the side with stomatarb s m1 quasi-laminar boundary layer resistance over leaf surface
rac 0 s m1 resistance of turbulent transport from the reference height of the understorey vegetation to
the soil surfacerbs s m1 soil boundary layer resistancersoil 400 s m1 soil resistancerstm s m1 stomatal resistancerstm,H2O s m1 stomatal resistance for water vapour rmes 0 s m1 mesophyllic resistancercut 105 s m1 cuticle resistancerws 2000 s m1 wet skin resistancefwet fraction of wet skinDH2O 2.12 [notdef] 10
5 m2 s1 molecular diffusivity of water vapour DO3 1.33 [notdef] 10
5 m2 s1 molecular diffusivity of O3 0.41 von Krmn constant 0 m the height above ground where the molecular diffusivity is equal to turbulent eddy diffusivity z 0.1 m the height under which the logarithmic wind prole is assumed
Sc 1.07 Schmidt number for O3
Atmos. Chem. Phys., 17, 13611379, 2017 www.atmos-chem-phys.net/17/1361/2017/
P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model 1377
Author contributions. Putian Zhou implemented the deposition code into SOSAA, made the simulation runs, analysed the results and wrote the main part of this manuscript. Laurens Ganzeveld provided and developed the deposition code, suggested the concepts of manuscript structure, contributed to the micrometeorology part and the discussions related to O3 uxes. llar Rannik contributed to the micrometeorology part, the discussions related to O3 ux measurements and the discussions in chemical removal processes.
Luxi Zhou contributed to implementing the deposition code into SOSAA and conguration of simulation runs. Rosa Gierens contributed to the conguration of meteorology part in SOSAA and conguration of simulation runs. Ditte Taipale contributed to the discussions related to air chemistry and site description. Ivan Mammarella contributed to discussions related to O3 ux measurements.
Michael Boy provided SOSAA code and the main concept and structure of this manuscript.
Competing interests. The authors declare that they have no conict of interest.
Acknowledgements. This work was supported by Maj ja Tor Nessling funding, the Academy of Finland (projects 1118615 and 272041), CRAICC (Cryosphere-atmosphere interactions in a changing Arctic climate), eSTICC (eScience tools for investigating Climate Change in Northern High Latitudes) and FCoE (The Centre of Excellence in Atmospheric Science From Molecular and Biological processes to The Global Climate). This work was also supported by institutional research funding (IUT20-11) of the Estonian Ministry of Education and Research, and the European Regional Development Fund (Centre of Excellence EcolChange). The authors also wish to acknowledge CSC IT Center for Science, Finland, for computational resources.
Edited by: S. M. NoeReviewed by: three anonymous referees
Bck, J., Aalto, J., Henriksson, M., Hakola, H., He, Q., and Boy, M.: Chemodiversity of a Scots pine stand and implications for terpene air concentrations, Biogeosciences, 9, 689702, doi:http://dx.doi.org/10.5194/bg-9-689-2012
Web End =10.5194/bg-9-689-2012 http://dx.doi.org/10.5194/bg-9-689-2012
Web End = , 2012.
Caird, M. A., Richards, J. H., and Donovan, L. A.: Nighttime Stomatal Conductance and Transpiration in C3 and C4 Plants, Plant
Physiol., 143, 410, 2007.
Dee, D. P., Uppala, S. M., Simmons, A. J., Berrisford, P., Poli,P., Kobayashi, S., Andrae, U., Balmaseda, M. A., Balsamo, G., Bauer, P., Bechtold, P., Beljaars, A. C. M., van de Berg, L., Bidlot, J., Bormann, N., Delsol, C., Dragani, R., Fuentes, M., Geer,A. J., Haimberger, L., Healy, S. B., Hersbach, H., Hlm, E. V., Isaksen, L., Kllberg, P., Khler, M., Matricardi, M., McNally,A. P., Monge-Sanz, B. M., Morcrette, J.-J., Park, B.-K., Peubey,C., de Rosnay, P., Tavolato, C., Thpaut, J.-N., and Vitart, F.: The ERA-Interim reanalysis: conguration and performance of the data assimilation system, Q. J. Roy. Meteorol. Soc., 137, 553 597, doi:http://dx.doi.org/10.1002/qj.828
Web End =10.1002/qj.828 http://dx.doi.org/10.1002/qj.828
Web End = , 2011.
Fares, S., McKay, M., Holzinger, R., and Goldstein, A. H.: Ozone uxes in a Pinus ponderosa ecosystem are dominated by nonstomatal processes: Evidence from long-term continuous measurements, Agr. Forest Meteorol., 150, 420431, 2010.
Felzer, B. S., Cronin, T., Reilly, J. M., Melillo, J. M., and Wang,X.: Impacts of ozone on trees and crops, Comptes Rendus Geo-science, 339, 784798, 2007.
Ganzeveld, L. and Lelieveld, J.: Dry deposition parameterization in a chemistry general circulation model and its inuence on the distribution of reactive trace gases, J. Geophy. Res., 100, 20999 21012, 1995.
Ganzeveld, L., Lelieveld, J., and Roelofs, G.-J.: A dry deposition parameterization for sulfur oxides in a chemistry and general circulation model, J. Geophys. Res.-Atmos., 103, 56795694, doi:http://dx.doi.org/10.1029/97JD03077
Web End =10.1029/97JD03077 http://dx.doi.org/10.1029/97JD03077
Web End = , 1998.
Ganzeveld, L., Bouwman, L., Stehfest, E., van Vuuren, D.P., Eickhout, B., and Lelieveld, J.: Impact of future land use and land cover changes on atmospheric chemistry-climate interactions, J. Geophys. Res.-Atmos., 115, d23301, doi:http://dx.doi.org/10.1029/2010JD014041
Web End =10.1029/2010JD014041 http://dx.doi.org/10.1029/2010JD014041
Web End = , 2010.
Ganzeveld, L. N., Lelieveld, J., Dentener, F. J., Krol, M. C., Bouwman, A. J., and Roelofs, G.-J.: Global soil-biogenic NOx emissions and the role of canopy processes, J. Geophys. Res., 107, ACH 9-1ACH 9-17, doi:http://dx.doi.org/10.1029/2001JD001289
Web End =10.1029/2001JD001289 http://dx.doi.org/10.1029/2001JD001289
Web End = , 2002a.Ganzeveld, L. N., Lelieveld, J., Dentener, F. J., Krol, M. C., and
Roelofs, G.-J.: Atmosphere-biosphere trace gas exchanges simulated with a single-column model, J. Geophys. Res., 107, ACH 8-1ACH 8-21, doi:http://dx.doi.org/10.1029/2001JD000684
Web End =10.1029/2001JD000684 http://dx.doi.org/10.1029/2001JD000684
Web End = , 2002b.
Gierens, R. T., Laakso, L., Mogensen, D., Vakkari, V., Beukes,J. P., Van Zyl, P. G., Hakola, H., Guenther, A., Pienaar, J. J., and Boy, M.: Modelling new particle formation events in the South African savannah, S. Afr. J. Sci., 110, doi:http://dx.doi.org/10.1590/sajs.2014/20130108
Web End =10.1590/sajs.2014/20130108 http://dx.doi.org/10.1590/sajs.2014/20130108
Web End = , 2014.
Goldstein, A. H., McKay, M., Kurpius, M. R., Schade, G. W., Lee,A., Holzinger, R., and Rasmussen, R. A.: Forest thinning experiment conrms ozone deposition to forest canopy is dominated by reaction with biogenic VOCs, Geophys. Res. Lett., 31, l22106, doi:http://dx.doi.org/10.1029/2004GL021259
Web End =10.1029/2004GL021259 http://dx.doi.org/10.1029/2004GL021259
Web End = , 2004.
Guenther, A., Karl, T., Harley, P., Wiedinmyer, C., Palmer, P. I., and Geron, C.: Estimates of global terrestrial isoprene emissions using MEGAN (Model of Emissions of Gases and Aerosols from
www.atmos-chem-phys.net/17/1361/2017/ Atmos. Chem. Phys., 17, 13611379, 2017
References
Altimir, N., Kolari, P., Tuovinen, J.-P., Vesala, T., Bck, J., Suni,T., Kulmala, M., and Hari, P.: Foliage surface ozone deposition: a role for surface moisture?, Biogeosciences, 3, 209228, doi:http://dx.doi.org/10.5194/bg-3-209-2006
Web End =10.5194/bg-3-209-2006 http://dx.doi.org/10.5194/bg-3-209-2006
Web End = , 2006.
Boy, M., Sogachev, A., Lauros, J., Zhou, L., Guenther, A., and Smolander, S.: SOSA a new model to simulate the concentrations of organic vapours and sulphuric acid inside the ABL Part 1: Model description and initial evaluation, Atmos. Chem. Phys., 11, 4351, doi:http://dx.doi.org/10.5194/acp-11-43-2011
Web End =10.5194/acp-11-43-2011 http://dx.doi.org/10.5194/acp-11-43-2011
Web End = , 2011.
Boy, M., Mogensen, D., Smolander, S., Zhou, L., Nieminen, T., Paasonen, P., Plass-Dlmer, C., Sipil, M., Petj, T., Mauldin, L., Berresheim, H., and Kulmala, M.: Oxidation of SO2 by stabilized Criegee intermediate (sCI) radicals as a crucial source for atmospheric sulfuric acid concentrations, Atmos. Chem. Phys., 13, 38653879, doi:http://dx.doi.org/10.5194/acp-13-3865-2013
Web End =10.5194/acp-13-3865-2013 http://dx.doi.org/10.5194/acp-13-3865-2013
Web End = , 2013.
1378 P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model
Nature), Atmos. Chem. Phys., 6, 31813210, doi:http://dx.doi.org/10.5194/acp-6-3181-2006
Web End =10.5194/acp-6- http://dx.doi.org/10.5194/acp-6-3181-2006
Web End =3181-2006 , 2006.
Hardacre, C., Wild, O., and Emberson, L.: An evaluation of ozone dry deposition in global scale chemistry climate models, Atmos. Chem. Phys., 15, 64196436, doi:http://dx.doi.org/10.5194/acp-15-6419-2015
Web End =10.5194/acp-15-6419- http://dx.doi.org/10.5194/acp-15-6419-2015
Web End =2015 , 2015.
Hari, P. and Kulmala, M.: Station for Measuring Ecosystem-Atmosphere Relations (SMEAR II), Boreal. Environ. Res., 10, 315322, 2005.
Junninen, H., Lauri, A., Keronen, P., Aalto, P., Hiltunen, V., Hari, P., and Kulmala, M.: Smart-SMEAR: on-line data exploration and visualization tool for SMEAR stations, Boreal Environ. Res., 14, 447457, 2009.
Kampa, M. and Castanas, E.: Human health effects of air pollution,Environ. Poll., 151, 362367, 2008.
Keronen, P., Reissell, A., Rannik, ., Pohja, T., Siivola, E.,
Hiltunen, V., Hari, P., Kulmala, M., and Vesala, T.: Ozone ux measurements over a Scots pine forest using eddy covariance method: performance evaluation and comparison with uxprole method, Boreal. Environ. Res., 8, 425443, 2003.Kolari, P., Pumpanen, J., Kulmala, L., Ilvesniemi, H., Nikinmaa, E.,
Grnholm, T., and Hari, P.: Forest oor vegetation plays an important role in photosynthetic production of boreal forests, Forest Ecol. Manage., 221, 241248, 2006.
Kolari, P., Lappalainen, H. K., Hnninen, H., and Hari, P.: Relationship between temperature and the seasonal course of photosynthesis in Scots pine at northern timberline and in southern boreal zone, Tellus B, 59, 542552, doi:http://dx.doi.org/10.1111/j.1600-0889.2007.00262.x
Web End =10.1111/j.1600- http://dx.doi.org/10.1111/j.1600-0889.2007.00262.x
Web End =0889.2007.00262.x , 2007.
Korhonen, H., Lehtinen, K. E. J., and Kulmala, M.: Multicomponent aerosol dynamics model UHMA: model development and validation, Atmos. Chem. Phys., 4, 757771, doi:http://dx.doi.org/10.5194/acp-4-757-2004
Web End =10.5194/acp-4- http://dx.doi.org/10.5194/acp-4-757-2004
Web End =757-2004 , 2004.
Kulmala, L., Launiainen, S., Pumpanen, J., Lankreijer, H., Lindroth,A., Hari, P., and Vesala, T.: H2O and CO2 uxes at the oor of a boreal pine forest, Tellus B, 60, 167178, doi:http://dx.doi.org/10.1111/j.1600-0889.2007.00327.x
Web End =10.1111/j.1600- http://dx.doi.org/10.1111/j.1600-0889.2007.00327.x
Web End =0889.2007.00327.x , 2008.
Kurpius, M. R. and Goldstein, A. H.: Gas-phase chemistry dominates O3 loss to a forest, implying a source of aerosols and hydroxyl radicals to the atmosphere, Geophys. Res. Lett., 30, 1371, doi:http://dx.doi.org/10.1029/2002GL016785
Web End =10.1029/2002GL016785 http://dx.doi.org/10.1029/2002GL016785
Web End = , 2003.
Kurtn, T., Zhou, L., Makkonen, R., Merikanto, J., Risnen,P., Boy, M., Richards, N., Rap, A., Smolander, S., Sogachev,A., Guenther, A., Mann, G. W., Carslaw, K., and Kulmala,M.: Large methane releases lead to strong aerosol forcing and reduced cloudiness, Atmo. Chem. Phys., 11, 69616969, doi:http://dx.doi.org/10.5194/acp-11-6961-2011
Web End =10.5194/acp-11-6961-2011 http://dx.doi.org/10.5194/acp-11-6961-2011
Web End = , 2011.
Lamaud, E., Carrara, A., Brunet, Y., Lopez, A., and Druilhet, A.: Ozone uxes above and within a pine forest canopy in dry and wet conditions, Atmos. Environ., 36, 7788, 2002.
Lammel, G.: Formation of nitrous acid: parameterisation and comparison with observations, Tech. Rep. REPORT No. 286, Max-Planck-Institut fr Meteorologie, Hamburg, 1999.
Launiainen, S., Katul, G. G., Grnholm, T., and Vesala, T.: Partitioning ozone uxes between canopy and forest oor by measurements and a multi-layer model, Agr. Forest Meteorol., 173, 8599, 2013.
Mammarella, I., Peltola, O., Nordbo, A., Jrvi, L., and Rannik, ..: Quantifying the uncertainty of eddy covariance uxes due to the
use of different software packages and combinations of processing steps in two contrasting ecosystems, Atmo. Meas. Tech., 9, 49154933, doi:http://dx.doi.org/10.5194/amt-9-4915-2016
Web End =10.5194/amt-9-4915-2016 http://dx.doi.org/10.5194/amt-9-4915-2016
Web End = , 2016.
Massman, W. J.: Toward an ozone standard to protect vegetation based on effective dose: a review of deposition resistances and a possible metric, Atmos. Environ., 38, 23232337, 2004. Meyers, T. P.: The sensitivity of modeled SO2 uxes and proles to stomatal and boundary layer resistances, Water Air Soil Poll., 35, 261278, doi:http://dx.doi.org/10.1007/BF00290935
Web End =10.1007/BF00290935 http://dx.doi.org/10.1007/BF00290935
Web End = , 1987.
Mogensen, D., Smolander, S., Sogachev, A., Zhou, L., Sinha, V., Guenther, A., Williams, J., Nieminen, T., Kajos, M. K., Rinne, J., Kulmala, M., and Boy, M.: Modelling atmospheric OH-reactivity in a boreal forest ecosystem, Atmos. Chem. Phys., 11, 9709 9719, doi:http://dx.doi.org/10.5194/acp-11-9709-2011
Web End =10.5194/acp-11-9709-2011 http://dx.doi.org/10.5194/acp-11-9709-2011
Web End = , 2011.
Mogensen, D., Gierens, R., Crowley, J. N., Keronen, P., Smolander,S., Sogachev, A., Nlscher, A. C., Zhou, L., Kulmala, M., Tang,M. J., Williams, J., and Boy, M.: Simulations of atmospheric OH, O3 and NO3 reactivities within and above the boreal forest,
Atmos. Chem. Phys., 15, 39093932, doi:http://dx.doi.org/10.5194/acp-15-3909-2015
Web End =10.5194/acp-15-3909- http://dx.doi.org/10.5194/acp-15-3909-2015
Web End =2015 , 2015.
Nemitz, E., Sutton, M. A., Schjoerring, J. K., Husted, S., and Paul,W. G.: Resistance modelling of ammonia exchange over oilseed rape, Agr. Forest Meteorol., 105, 405425, 2000.
Rannik, ., Kolari, P., Vesala, T., and Hari, P.: Uncertainties in measurement and modelling of net ecosystem exchange of a forest, Agr. Forest Meteorol., 138, 244257, 2006.
Rannik, U., Mammarella, I., Keronen, P., and Vesala, T.: Vertical advection and nocturnal deposition of ozone over a boreal pine forest, Atmos. Chem. Phys., 9, 20892095, doi:http://dx.doi.org/10.5194/acp-9-2089-2009
Web End =10.5194/acp-9- http://dx.doi.org/10.5194/acp-9-2089-2009
Web End =2089-2009 , 2009.
Rannik, U., Altimir, N., Mammarella, I., Bck, J., Rinne, J., Ruuskanen, T. M., Hari, P., Vesala, T., and Kulmala, M.: Ozone deposition into a boreal forest over a decade of observations: evaluating deposition partitioning and driving variables, Atmos. Chem. Phys., 12, 1216512182, doi:http://dx.doi.org/10.5194/acp-12-12165-2012
Web End =10.5194/acp-12- http://dx.doi.org/10.5194/acp-12-12165-2012
Web End =12165-2012 , 2012.
Rinne, J., Bck, J., and Hakola, H.: Biogenic volatile organic compound emissions from the Eurasian taiga: current knowledge and future directions, Boreal Environ. Res., 14, 807826, 2009. Ruckstuhl, K. E., Johnson, E. A., and Miyanishi, K.: Introduction. The boreal forest and global change, Philos. T. Roy. Soc. Lond. B, 363, 22432247, doi:http://dx.doi.org/10.1098/rstb.2007.2196
Web End =10.1098/rstb.2007.2196 http://dx.doi.org/10.1098/rstb.2007.2196
Web End = , 2008. Seok, B., Helmig, D., Ganzeveld, L., Williams, M. W., and Vogel,C. S.: Dynamics of nitrogen oxides and ozone above and within a mixed hardwood forest in northern Michigan, Atmos. Chem.d Phys., 13, 73017320, doi:http://dx.doi.org/10.5194/acp-13-7301-2013
Web End =10.5194/acp-13-7301-2013 http://dx.doi.org/10.5194/acp-13-7301-2013
Web End = , 2013. Smolander, S., He, Q., Mogensen, D., Zhou, L., Bck, J., Ruuskanen, T., Noe, S., Guenther, A., Aaltonen, H., Kulmala, M., and Boy, M.: Comparing three vegetation monoterpene emission models to measured gas concentrations with a model of meteorology, air chemistry and chemical transport, Biogeosciences, 11, 54255443, doi:http://dx.doi.org/10.5194/bg-11-5425-2014
Web End =10.5194/bg-11-5425-2014 http://dx.doi.org/10.5194/bg-11-5425-2014
Web End = , 2014. Sogachev, A.: A note on two-equation closure modelling of canopy ow, Bound.-Lay. Meteorol., 130, 423435, 2009.
Sogachev, A., Menzhulin, G., Heimannn, M., and Lloyd, J.: A simple three dimensional canopy planetary boundary layer simulation model for scalar concentrations and uxes, Tellus B, 54, 784819, 2002.
Atmos. Chem. Phys., 17, 13611379, 2017 www.atmos-chem-phys.net/17/1361/2017/
P. Zhou et al.: Simulating ozone dry deposition at a boreal forest with a multi-layer canopy deposition model 1379
Stocker, T. F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S. K., Boschung, J., Nauels, A., Xia, Y., Bex, V., and Midgley, P. M.: IPCC, 2013: Climate Change 2013: The Physical Science Basis, in: Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, UK and New York, NY, USA, 2013.
Wesely, M. L.: Parameterization of surface resistances to gaseous dry deposition in regional-scale numerical models, Atmos. Environ., 23, 12931304, 1989.
Williams, J., Crowley, J., Fischer, H., Harder, H., Martinez, M., Petj, T., Rinne, J., Bck, J., Boy, M., Dal Maso, M., Hakala,J., Kajos, M., Keronen, P., Rantala, P., Aalto, J., Aaltonen,H., Paatero, J., Vesala, T., Hakola, H., Levula, J., Pohja, T., Herrmann, F., Auld, J., Mesarchaki, E., Song, W., Yassaa, N., Nlscher, A., Johnson, A. M., Custer, T., Sinha, V., Thieser,J., Pouvesle, N., Taraborrelli, D., Tang, M. J., Bozem, H., Hosaynali-Beygi, Z., Axinte, R., Oswald, R., Novelli, A., Kubistin, D., Hens, K., Javed, U., Trawny, K., Breitenberger, C., Hidalgo, P. J., Ebben, C. J., Geiger, F. M., Corrigan, A. L., Russell, L. M., Ouwersloot, H. G., Vil-Guerau de Arellano,J., Ganzeveld, L., Vogel, A., Beck, M., Bayerle, A., Kampf,C. J., Bertelmann, M., Kllner, F., Hoffmann, T., Valverde, J., Gonzlez, D., Riekkola, M.-L., Kulmala, M., and Lelieveld,J.: The summertime Boreal forest eld measurement intensive (HUMPPA-COPEC-2010): an overview of meteorological and chemical inuences, Atmos. Chem. Phys., 11, 1059910618, doi:http://dx.doi.org/10.5194/acp-11-10599-2011
Web End =10.5194/acp-11-10599-2011 http://dx.doi.org/10.5194/acp-11-10599-2011
Web End = , 2011.
Wolfe, G. M. and Thornton, J. A.: The Chemistry of Atmosphere-Forest Exchange (CAFE) Model Part 1: Model description and characterization, Atmos. Chem. Phys., 11, 77101, doi:http://dx.doi.org/10.5194/acp-11-77-2011
Web End =10.5194/acp-11-77-2011 http://dx.doi.org/10.5194/acp-11-77-2011
Web End = , 2011.
Wolfe, G. M., Thornton, J. A., McKay, M., and Goldstein, A.H.: Forest-atmosphere exchange of ozone: sensitivity to very reactive biogenic VOC emissions and implications for in-canopy photochemistry, Atmos. Chem. Phys., 11, 78757891, doi:http://dx.doi.org/10.5194/acp-11-7875-2011
Web End =10.5194/acp-11-7875-2011 http://dx.doi.org/10.5194/acp-11-7875-2011
Web End = , 2011.
Wu, Y., Brashers, B., Finkelstein, P. L., and Pleim, J. E.: A multilayer biochemical dry deposition model 1. Model formulation, J. Geophys. Res.-Atmos., 108, ACH 1-1ACH 1-12, doi:http://dx.doi.org/10.1029/2002JD002306
Web End =10.1029/2002JD002306 http://dx.doi.org/10.1029/2002JD002306
Web End = , 2003.
Zhou, L., Nieminen, T., Mogensen, D., Smolander, S., Rusanen,A., Kulmala, M., and Boy, M.: SOSAA a new model to simulate the concentrations of organic vapours, sulphuric acid and aerosols inside the ABL Part 2: Aerosol dynamics and one case study at a boreal forest site, Boreal Environ. Res., 19, 237256, 2014.
Zhou, L., Gierens, R., Sogachev, A., Mogensen, D., Ortega,J., Smith, J. N., Harley, P. C., Prenni, A. J., Levin, E. J.T., Turnipseed, A., Rusanen, A., Smolander, S., Guenther,A. B., Kulmala, M., Karl, T., and Boy, M.: Contribution from biogenic organic compounds to particle growth during the 2010 BEACHON-ROCS campaign in a Colorado temperate needleleaf forest, Atmos. Chem. Phys., 15, 86438656, doi:http://dx.doi.org/10.5194/acp-15-8643-2015
Web End =10.5194/acp-15-8643-2015 http://dx.doi.org/10.5194/acp-15-8643-2015
Web End = , 2015.
www.atmos-chem-phys.net/17/1361/2017/ Atmos. Chem. Phys., 17, 13611379, 2017
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Copyright Copernicus GmbH 2017
Abstract
A multi-layer ozone (O<sub>3</sub>) dry deposition model has been implemented into SOSAA (a model to Simulate the concentrations of Organic vapours, Sulphuric Acid and Aerosols) to improve the representation of O<sub>3</sub> concentration and flux within and above the forest canopy in the planetary boundary layer. We aim to predict the O<sub>3</sub> uptake by a boreal forest canopy under varying environmental conditions and analyse the influence of different factors on total O<sub>3</sub> uptake by the canopy as well as the vertical distribution of deposition sinks inside the canopy. The newly implemented dry deposition model was validated by an extensive comparison of simulated and observed O<sub>3</sub> turbulent fluxes and concentration profiles within and above the boreal forest canopy at SMEAR II (Station to Measure Ecosystem-Atmosphere Relations II) in Hyytiälä, Finland, in August 2010. In this model, the fraction of wet surface on vegetation leaves was parametrised according to the ambient relative humidity (RH). Model results showed that when RH was larger than 70% the O<sub>3</sub> uptake onto wet skin contributed ∼ 51% to the total deposition during nighttime and ∼ 19% during daytime. The overall contribution of soil uptake was estimated about 36%. The contribution of sub-canopy deposition below 4.2m was modelled to be ∼ 38% of the total O<sub>3</sub> deposition during daytime, which was similar to the contribution reported in previous studies. The chemical contribution to O<sub>3</sub> removal was evaluated directly in the model simulations. According to the simulated averaged diurnal cycle the net chemical production of O<sub>3</sub> compensated up to ∼ 4% of dry deposition loss from about 06:00 to 15:00LT. During nighttime, the net chemical loss of O<sub>3</sub> further enhanced removal by dry deposition by a maximum ∼ 9%. Thus the results indicated an overall relatively small contribution of airborne chemical processes to O<sub>3</sub> removal at this site.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer