Simulation of organics in the atmosphere: evaluation of EMACv2.54 with the Mainz Organic Mechanism (MOM) coupled to the ORACLE (v1.0) submodel

. An updated and expanded representation of organics in the chemistry general circulation model EMAC (ECHAM5/MESSy for Atmospheric Chemistry) has been evaluated. First, the comprehensive Mainz Organic Mechanism (MOM) in the submodel MECCA (Module Efﬁciently Calculating the Chemistry of the Atmosphere) was activated with explicit degradation of organic species up to

2006b). The biogenic emissions of non-methane volatile organic compounds (NMVOC) are calculated on-line using the Model 80 of Emissions of Gases and Aerosol from Nature (MEGANv2.04, Guenther et al., 2006Guenther et al., , 2012. Lightning NO x production is based on the parametrization of Grewe et al. (2001), while the algorithm of Yienger (1995) is used for soil NO x emissions, as described in detail in Jöckel et al. (2016).
Biomass burning emissions are calculated daily online based on dry matter burned from observations and fire type (Kaiser et al., 2012). The emission factors for different tracers and fire types are taken from Andreae (2019) and Akagi et al. (2011). 85 The ONEMIS submodel (Kerkweg et al., 2006b) calculates natural emission fluxes of sea-salt (Guelle et al., 2001) and of dust (Klingmüller et al., 2018;Astitha et al., 2012). Oceanic emissions and deposition are calculated online with the AIRSEA submodel Fischer et al., 2012;Lana et al., 2011) for DMS, CH 3 COCH 3 , CH 3 OH, C 5 H 8 , HCN, and CH 3 CN (a net sink for CH 3 OH, HCN, and CH 3 CN, a net source for the others). For C 2 -C 3 alkanes and alkenes, the oceanic offline emissions from Inness et al. (2012) have been adopted. 90 In Table 1 the total emissions are listed for all primary species emitted in the model. Dry deposition and sedimentation are estimated by DDEP and SEDI submodel (Kerkweg et al., 2006a), while wet deposition is simulated by the SCAV submodel . The dissolution of species in the liquid phase mechanism was augmented by including liquid-gas equilibrium for all additional organics present in the mechanism with a Henry's law constant (solubility) above 10 3 mol L −1 atm −1 . For less soluble species, wet scavenging is not expected to be important (Crutzen and Lawrence, still quite uncertain (Wang et al., 2017). As the hydroxyl (and hydroperoxyl) group affects the solubility in water the most, the Henry's law constants of polyols from the compilation by Sander (2015) are taken as proxies.
In the chemistry submodel MECCA, we used the Mainz Organic Mechanism (MOM, Sander et al., 2019), with roughly 600 species and 1600 reactions. It originates from a reduced isoprene oxidation mechanism (Taraborrelli et al., 2009), which 100 has been updated with recent kinetic data and expanded with efficient mechanisms of OH-recycling under low-NO conditions (Taraborrelli et al., 2012;Nölscher et al., 2014). The oxidation mechanism for α-pinene and β-pinene is based on the MCM (E. Jenkin et al., 2000) with modifications according to the theoretical work of L. Vereecken (Vereecken et al., 2007;Capouet et al., 2008;Nguyen et al., 2009;Vereecken and Peeters, 2012). The degradation of monocyclic aromatics follows the MCM (Jenkin et al., 2003;Bloss et al., 2005) but with some modifications for the chemistry of phenols (Cabrera-Perez et al., 2016). 105 The complete mechanism used in this study is part of the electronic supplement.
Organic Aerosol (OA) formation is simulated by the submodel ORACLE (Tsimpidi et al., 2014, , v.1.0), where logarithmically spaced saturation concentration bins are used to describe the organic aerosol components based on their volatility.
For the formation of primary (POA) and secondary (SOA) organic aerosol from the emissions and photochemical oxidation 115 of semivolatile and intermediate volatility organic compounds the setup of Tsimpidi et al. (2016) is used. However, the SOA formation from the oxidation of the volatile organic compounds (VOC) in ORACLE has been modified to accommodate the photochemical production of components explicitly calculated by the MOM chemical mechanism. An earlier version of MOM has been compared against chamber measurements  and improved by Novelli et al. (2020), while ORA-CLE was derived empirically from chamber experiments (Donahue et al., 2011) and has been evaluated against observations 120 from a field campaign by Janssen et al. (2017). Nevertheless, the MOM + ORACLE combination still has to be fully evaluated.
The effective saturation concentration of all the species present in the MECCA submodel is calculated based on their elemental composition (number of carbon, oxygen, nitrogen, and sulfur atoms), following the molecular corridor approach (Li et al., 2016) and the study by Donahue et al. (2011). In Figure 1, the volatility versus the molar mass is shown for all tracers present in the model, ranging from high volatility -low molar mass to low volatility -high molar mass. These calculations 125 result in 199 tracers which, under atmospheric conditions, can partition to the aerosol phase forming SOA (i.e., with a volatility lower than 3.2 × 10 3 µg m −3 ), while the other 396 tracers are considered too volatile for any condensation on aerosol particles.
Additionally, the enthalpy of vaporization for the same condensing species has been estimated based on the study by Epstein et al. (2010). In addition to the 199 tracers, 11 additional "lumped species" for different oxidation levels of pentanes, higher alkanes, higher alkenes and terpenes have been added. Then, ORACLE calculates the partitioning of these organic species 130 between the gas and particle phases by assuming a bulk equilibrium, and by further assuming that all organic compounds form a pseudo-ideal solution. The aerosol size distribution is determined by distributing the change in aerosol mass after the bulk equilibrium into each size mode using a weighting factor (Tsimpidi et al., 2014).
The aerosol optical properties are calculated with the submodel AEROPT (Dietmüller et al., 2016), which is based on the scheme by Lauer et al. (2007) and makes use of predefined lognormal modes (i.e., the mode width σ and the mode mean radius 135 have to be taken into account). Lookup tables with the extinction coefficient, the single scattering albedo, and the asymmetry factor for the shortwave, and extinction coefficient for the longwave part of the spectrum are pre-calculated with the help of Mie theory-based explicit radiative transfer calculations (see Pozzer et al., 2012a;Dietmüller et al., 2016 Due to the high complexity of the MOM mechanism, we do not expect the model to be capable of representing polluted episodic conditions. Additionally, investigations of these conditions should be accompanied with detailed process studies, as done in previous studies with this mechanism (e.g. Lelieveld et al. (2018); Tadic et al. (2021)). Instead we want to demonstrate 145 its ability to reproduce background conditions in a climatological sense. For this reason three different aircraft databases were chosen, that are based on profiles of several flights and seasons, being over background regions, and providing an extensive set of observed trace gases.

NASA ATom campaign
We used observational data from the NASA Atmospheric Tomography Mission (ATom). The ATom campaign took place 150 from July 2016 to May 2018 and included measurements in four different seasons (6 flights each). During the flights with the NASA DC-8 aircraft numerous profiles were recorded (see Figure 2), which makes the dataset perfect for the evaluation of atmospheric model simulations. The data used in this evaluation was taken from the "Merged Atmospheric Chemistry, Trace Gases, and Aerosols, Version 2 dataset" (Wofsy et al., 2021) and is based on various instruments: UC-Irvine Whole Air Sampler (WAS; Barletta et al., 2019), Trace Organic Gas Analyzer (TOGA; Apel et al., 2021) Canada/Alaska/Greenland) based on expected homogeneous properties of organic compounds in those regions. A climatological comparison between the simulated year (2010) and the data from the ATom campaign (2016-2018) will be performed in order to further evaluate, if the MOM mechanism is able to reproduce remote conditions. For that purpose we divided the altitude range into 12 bins (linearly between 0.5 and 12.5 km) and defined a maximum of 336 (7 regions × 4 seasons × 12 altitude bins) points of interest (POI). For each POI we compared the mean value of all observations in the specific region, 165 altitude bin and season to the mean value of all accordingly simulated values in 2010 in the range of ± 1 day around the flights conducted in that region.

Emmons database
Additionally, we used the database of aircraft measurements of trace gases produced by Emmons et al. (2000). It is based on observations from numerous aircraft campaigns that took place during the period 1990-2001 to create observational-based 170 climatologies of chemical species relevant to tropospheric chemistry. Although these measurements cover only limited time periods and regions, they provide valuable information about the vertical distribution of the analysed trace gases. Note that the field campaigns used in this evaluation have been extended, also including observations after the year 2000, such as those from the TOPSE and TRACE-P campaigns.

Heald database 175
The aircraft evaluation is completed by aircraft measurements for organic aerosols presented by Heald et al. (2011). This dataset includes organic aerosol measurements from 17 aircraft campaigns during the period 2001-2009. Here, similar to Emmons  (2), East Pacific (3), Southern Ocean (4), South Atlantic (5), North Atlantic (6), and North Canada/Alaska/Greenland (7). et al. (2000), we consider the aircraft campaigns representative for the period and the regions, i.e., as an observational-based climatology. Therefore, we compared these data with the model results for the year 2010, yielding an improved evaluation of the vertical distribution of the organic aerosols in the troposphere. Global Monitoring program were also used in this study. The network consists of 44 background stations from the NOAA Global Greenhouse Gases Reference Network (GGGRN) (Pollmann et al., 2008), where pairs of whole air samples are collected 185 weekly and shipped to a central laboratory for analyses. For this study, we considered measurements of ethane, propane, isoand n-butane from this network.

EPA
The Clean Air Markets Division of the U.S. Environmental Protection Agency (EPA) operates the Clean Air Status and Trends Network (CASTNET), with a total of 97 stations. Here, we include weekly filter pack data from 81 stations for SO 2− 4 , NO − 3 , NH + 4 , Na + , Mg 2+ , Ca 2+ , K + and Cl − for the year 2010. The data can be downloaded at www.epa.gov/castnet, last access on December 10, 2020.

IMPROVE
In order to evaluate the simulated regional background OA concentrations, we use monthly averaged PM 2.5 OA measurements during the year 2010 from the Interagency Monitoring of Protected Visual Environments (IMPROVE) program 195 (http://views.cira.colostate.edu/fed, last access on September 21, 2021). IMPROVE is a cooperative measurement effort in the United States designed to characterize current visibility and aerosol conditions in scenic areas (primarily National Parks and Forests). This network includes 198 monitoring sites that are representative of the regional haze conditions over North America. IMPROVE collocated samplers collect 24 hour samples every three days.

200
The co-operative programme for monitoring and evaluation of the long-range transmission of air pollutants in Europe (EMEP) is a scientifically-based and policy-driven programme under the Convention on Long-range Transboundary Air Pollution for international co-operation to solve transboundary air pollution problems. We use data from 42 stations, although the number of stations providing sufficient data for a certain species ranges from 22 to 42 (see Table 6). As Teflon filters are used predominantly, depending on the respective station, the concentrations of particulate nitrate can be systematically underestimated 205 (Ames and Malm, 2001). The data were obtained from the EBAS database (http://ebas.nilu.no/, last access on September 21, 2021) and stations not providing full coverage of monthly mean values for the year 2010 were excluded from the analysis.

EANET
The Acid Deposition Monitoring Network in East Asia (EANET) regularly monitors acid deposition as from January 2001.
Thirteen countries participate currently in this effort, submitting data from a total of 54 sites. We use monthly averaged aerosol 210 concentrations from 15 to 25 stations for the year 2010, with the data obtained from the Network Center for EANET, at https: //monitoring.eanet.asia/document/public/index, last access on September 21, 2021. It is worth mentioning that EANET does not include OA observations. The simulated PM 2.5 OA over East Asia is evaluated against collected short-term measurement data as summarized by Jo et al. (2013). The MOPITT (Measurement of Pollution in the Troposphere) is a sensor onboard the NASA's Terra satellite. We use the MOPO3JM version 8 product of MOPITT which provides monthly mean gridded column integrated CO, vertical profiles of CO mixing ratios at ten regularly spaced pressure levels from the surface up to 100 hPa as well as the corresponding averaging kernels (Deeter et al., 2019). The MOPO3JM product is based on the joint near-infrared (NIR) and thermal-infrared (TIR) 220 retrievals of CO. We apply the MOPITT averaging kernels on the EMAC CO profiles according to equation 1 to ensure the same level of smoothing as that in the MOPO3JM product.

Remote sensing observations
Here, x true is the EMAC profile, x a is the a priori profile, A is the averaging kernel matrix and I represents an identity matrix.

225
The MODerate resolution Imaging Spectroradiometer (MODIS) sensor is also located on the Terra satellite. Here, AOD550 data (i.e., Aerosol Optical Depth at 550 nm) from the MODIS Level 3 (Col. 6.11) gridded product are used at a spatial

IASI
Global observations of a suite of VOCs are retrieved from the thermal infrared measurements of the nadir-viewing IASI (Infrared Atmospheric Sounding Interferometer) instruments (Clerbaux et al., 2009). The VOC dataset exploited here consists of total column densities of methanol, acetone, formic and acetic acids, and PAN, retrieved near-globally on a daily basis from the IASI/Metop-A observations with the aid of the ANNI (Artificial Neural Network for IASI) v3 retrieval framework. This 235 neural network-based retrieval approach does not rely on a priori information of the total column densities and, hence, the products can be used directly for unbiased comparisons with model data (see Franco et al., 2018;Whitburn et al., 2016). The satellite products are filtered for measurements affected by clouds and poor retrieval performance. The uncertainties on the individual retrieved column densities can be large, but are considerably reduced by averaging numerous observations in space and time, as done in this study by comparing annual averages on the model grid. A full description of the ANNI framework, 240 the characterization of the VOC products, and comparisons of the satellite data with independent measurements can be found in Franco et al. (2018Franco et al. ( , 2019Franco et al. ( , 2020; Mahieu et al. (2021), and references therein. These studies, thanks to comparisons with ground-based column density measurements of VOCs, as well as with column densities derived from aircraft profiles, indicate no large systematic biases of the IASI data and no dependence on the latitude, although an underestimation (locally up to 30%) of the highest column densities over tropical source regions (e.g., the Amazon Basin) has been identified for all the IASI 245 VOC products, except for PAN. The accuracy of the IASI measurements is therefore sufficient to provide a global evaluation of the EMAC performance, considering the large uncertainties that still affect the emissions and atmospheric modelling of the VOCs. For both, IASI and the model, daily gridded averages were constructed at the spatial resolution of the model grid. To have similar temporal coverage over the year between model and satellite observations, the EMAC daily averages were masked when the corresponding IASA data were missing for the same day and location.

AERONET
The AOD observations have been obtained from the global AErosol RObotic NETwork (AERONET Holben et al., 1998;Dubovik et al., 2000). The cloud-screened quality-assured Level 2 AOD data used in this study were obtained from the website http://aeronet.gsfc.nasa.gov/cgi-bin/combined_data_access_new (last access on September 21, 2021) including AOD daily averages.

Pseudo observations
Global annual averages of fine particulate matter with an aerodynamic diameter below 2.5 µm (PM 2.5 ) have been obtained from Hammer et al. (2020). In this dataset, a combination of satellite observed AOD, numerical model, and ground-based PM 2.5 measurements were used to estimate PM 2.5 on a global scale at high resolution.

260
In Table 2 the evaluation of organic tracers using observations from the ATom campaign is presented, while Table 3 shows the comparison of the model results with the aircraft campaign data from Emmons et al. (2000) for selected tracers. There are no measurements of acetic acid (CH 3 COOH) from the ATom campaign and the observations of Ethene (C 2 H 4 ) and Propene (C 3 H 6 ) are very limited. Thus, they will not be considered for discussion. In the Emmons database, there are no observations of i-Butane (iC 4 H 10 ) and n-Butane (nC 4 H 10 ). with former results ). An interesting feature can be observed for methyl hydroperoxide (CH 3 OOH) and 270 formaldehyde (HCHO). Both species are strongly underestimated by the simulation compared to the ATom observations (see Table 2; MM/OM of 0.47 and 0.61, respectively), while they are highly correlated with the observations (≈ 0.9). Thus, the vertical shape is very well reproduced, but the sources or the sinks seem to be systematically underestimated or overestimated, respectively. This feature is depicted in Figure 3. The evaluation results for selected species will be analysed in more detail in the following. December, January and February) mean modified EMAC and MOPITT CO vertical column densities (VCD). Here, we term it as "modified" EMAC VCDs because MOPITT averaging kernels are applied on the simulated CO profiles to account for the  Table 2. Summary of simulated and observed mixing ratios of different tracers from the ATom campaign (Wofsy et al., 2021). We note that the spatial patterns are in good agreement between simulation and observations. In particular, the elevated CO background in the outflow regions (over oceans) in the northern hemisphere is well represented during boreal winter (DJF) and spring (MAM). The outflow region around south America in SON is also simulated well. Over land, a good agreement is found  Table 3. Same as Table 2, but for the comparison of our simulation results to the dataset of Emmons et al. (2000) . not find any underestimation of CO in Indonesia in SON (see Fig. 4 Africa. It must however be stressed that the uncertainties on biomass burning emissions are substantial, depending on regions and species (Carter et al., 2020). In contrast, in South America the overprediction of EMAC's CO column densities can be     Alkanes are VOC ubiquitously present in the atmosphere, mainly emitted from anthropogenic activities. In Figure 6 the scatter plots of model results and in-situ observations from the NOAA/INSTAR database are presented. The statistic for each plot is summarized in Table 4. The model seems to reproduce satisfactorily the mixing ratios of the C 2 -C 4 alkanes with more than 98, 91, 84 and 79 % of the simulated values lying within a factor of two to the observations for C 2 H 6 , C 3 H 8 , i−C 4 H 10 and n−C 4 H 10 , respectively. Furthermore, the average ratios show that a slight overestimation is present for i-butane and n-  (Table 3), which shows that both, ethane and propane, are underestimated by the model, also with respect to the vertical profile (column MM/OM). The comparison with ATom observations (Table 2) shows an underestimation of ethane and propane (albeit smaller) as well. However, for i-butane we see a slight underestimation in the evaluation using ATom observations in contrast to the NOAA/INSTAR database. 315 Figure 6 shows also a noticeable underestimation of the mixing ratios in the region between 30 • N and 0 • N, for propane and the butanes. This is clearly due to the underestimation of the oceanic emissions, as these stations (i.e., at Mahe Island, Sey-   Table 3 .

C 2 -C 3 Alkenes
Differently from alkanes, the alkenes are mostly emitted by the biosphere. The comparison to the ATom campaign can only be based on a very limited number of observations, and thus will not be subject of this discussion. As shown in Table 3 in the comparisons with the Emmons et al. (2000) database, the numerical simulation of alkenes diverges more from the observations 325 than the simulated alkanes do. Both, C 2 H 4 and C 3 H 6 , are underestimated compared to the observations (see model mean against observed mean in Tab. 3). Nevertheless, while C 2 H 4 is overall reasonably reproduced (average of the ratios equal to 1.06, see Table 3), for C 3 H 6 a strong underestimation is present (average of the ratios equal to 0.4, see Table 3). This is shown in Fig. 7, where examples of the vertical distribution of C 2 H 4 are shown. The issues in simulating C 3 H 6 have been already pointed out by Pozzer et al. (2006), where similar results were obtained with large underestimation of this tracer. As alkenes 330 are mostly removed by reaction with OH, there are strong indications that these reactions are too fast, in addition, a substantial lack of emissions could be present, for instance from natural sources as suggested by (Li et al., 2021).

CH 3 OH
CH 3 OH is the most abundant oxygenated VOC in the Earth's atmosphere (Singh et al., 2001), primarily emitted by the terrestrial vegetation . Other sources include biomass burning, the oceans and secondary formation in the 335 troposphere (e.g., Millet et al., 2008;Stavrakou et al., 2011). The annual global distribution of CH 3 OH total colum densities obtained from the IASI measurements and EMAC of the year 2010 as well as their differences are presented in Fig. 8. Over land, the satellite and model spatial distributions of CH 3 OH are relatively consistent, with the main source regions observed within the tropics (mainly over the Amazon Basin and central Africa), in Southeast Asia, and at Northern Hemisphere mid-and high latitudes. However, important differences in magnitude exist. Over the tropical forests, EMAC indeed simulates annual CH 3 OH 340 colum densities up to two times larger than those retrieved from the IASI observations (3.0-3.5 × 10 16 molecules cm −2 ). This can be partly ascribed to the high temperature bias that exists in the model over these areas (Hagemann and Stacke, 2015), which induces an excess of biogenic VOC emissions from the MEGAN submodel due to its high temperature sensitivity (Guenther et al., 2006). This is confirmed by a comparison of the simulated isoprene mixing ratios in the Amazon rainforest with observations from the AMAZE-08 campaign (Martin et al., 2016) and from the ATTO tower (Yáñez Serrano et al., 2015).

345
In comparison to the AMAZE-08 campaign, the simulated isoprene measurements are overestimated by more than a factor of 3 (6.1 ± 1.2 nmol/mol simulated and 1.9 ± 1.4 nmol/mol observed), while the overestimation factor for the ATTO tower is on average approximately 1.6, being higher in February/March and lower in October/November.
Furthermore, dry deposition of methanol in tropical rainforests is likely under-represented by the Wesely (1989) approach used here (Karl et al., 2004). The potential contribution of an additional non-stomatal pathway favourable under humid con- while larger column densities are measured by IASI during the same season further northwards in the boreal regions. Over 355 the oceans, EMAC underestimates the observed CH 3 OH column densities by ∼5 × 10 15 molecules cm −2 , which might result from an insufficient transport from the continental source regions and/or from missing secondary source(s) (see e.g., Bates et al., 2021;Müller et al., 2016). This is confirmed by the comparison of model results with the aircraft observations of the ATom campaign and the Emmons et al. (2000) database, where the methanol model mean is considerably lower than the observed one. This is most prominent over Pacific regions in the Northern Hemisphere at lower altitudes as depicted in Figure 9.

360
However, with respect to the vertical profile CH 3 OH is not generally underestimated in comparison with the ATom campaign, as it is rather overestimated at higher altitudes and for lower absolute values.

CH 3 COCH 3
Acetone is the one of the most abundant oxygenated VOCs in the Earth's atmosphere after CH 3 OH (Singh et al., 2001). Its sources include the terrestrial vegetation, the oxidation of hydrocarbon precursors of biogenic and anthropogenic origin, the 365 oceans and biomass burning (e.g., Jacob et al., 2002;Pozzer et al., 2010;Fischer et al., 2012). The acetone column densities obtained from the EMAC simulation and the IASI observations exhibit major discrepancies in terms of global distribution ( Figure 11). The satellite instrument detects the largest acetone column densities (up to 2 × 10 16 molecules cm −2 ) at mid-and high latitudes of the boreal hemisphere, ascribed to important emissions of biogenic precursors during summertime (Franco et al., 2019), but a moderate burden at low latitudes and over the tropical forests. It is worth noting that this pattern is in 370 agreement with the global acetone measurements obtained with the ACE-FTS satellite limb sounder (Dufour et al., 2016).
Conversely, EMAC simulates strong hotspots of acetone within the tropics, especially over South America and Africa, with annually averaged column densities larger than 2 × 10 16 molecules cm −2 , but underestimates the acetone abundance by up to a factor of 2 in the Northern Hemisphere. Such a mismatch between satellite and model distributions points to major deficiencies in the current emission inventories of acetone and its precursors. For example, the extremely large acetone column densities 375 simulated by EMAC over the Persian Gulf and the Indo Gangetic Plain can be attributed to the highly elevated emissions CH 3 COCH 3 -IASI 0 0.5 1 1.5 2 Total column (molecules/cm 2 ) #10 16 CH 3 COCH 3 -EMAC 0 0.5 1 1.5 2 Total column (molecules/cm 2 ) #10 16 CH 3 COCH 3 -EMAC vs. IASI -1 -0.5 0 0.5 1 Total column (molecules/cm 2 ) #10 16 Figure 11. Same as Fig. 8, but for CH3COCH3.
of propane -an important precursor of acetone -in these regions. In vegetated regions, an underestimation of acetone dry deposition, accounting globally for 20 % of the total loss (Khan et al., 2015), likely contributes to the mismatch. According to measurements, significant amounts of acetone are removed during night through non-stomatal uptake (Karl et al., 2004;Müller et al., 2018). However, over remote areas and the oceans, the simulated acetone abundance is consistent with the 380 IASI measurements, with vertical column densities in the range of 0.5-1.0 × 10 16 molecules cm −2 . The model, nevertheless, simulates a slight overestimation over the oceans at low latitudes compared to IASI, especially in the outflows of the tropical hotspots.
Compared to the ATom observations the model overestimates the mixing ratios, especially at high altitudes leading to a large value of MM/OM, while the simulatedd and observed mean are of a comparable magnitude. However, there are still 385 74.3% of the simulated values within a factor of 2 of the observations (see Table 2). Figure 10 depicts the vertical column densities of observations and simulations over the northern part of North America and Greenland, a region where CH 3 COCH 3 is largely underestimated by the model in comparison to IASI. In Northern Hemisphere summer (July/August; ATom-1) this strong underestimation can be observed as well for the ATom observations. However, for other seasons, this is not the case.
Overall, simulation results show only a small underestimation for this region. Compared to the Emmons et al. (2000) database, 390 the model results present a much better agreement, with only a somewhat small bias (see Tab. 3). This apparent agreement with the observations is due to the uneven distribution of the field campaign present in the observational dataset (see Emmons et al. (2000), Fig.1 and Huijnen et al. (2019) , Fig.1). In fact, most of the campaigns in which acetone was measured took place over the ocean, where the model is correct or slightly overestimating the total column densities observed by IASI. In contrast, only few campaigns in the Northern Hemisphere are present in the dataset (such as ABLE-3B, and POLINAT-2) for which 395 the model strongly underestimates the observed values. As a result of the sparsity of these observations, Table 3 presents a quite fair agreement between model and aircraft observations, an agreement which is, however, not corroborated by the more spatially complete comparison with the IASI total column densities.

HCOOH
Formic acid is the dominant organic acid in the troposphere and a product of the degradation of a large suite of VOC pre-400 cursors, but its observed abundance is generally severely underestimated by state-of-the-art global models (e.g., Millet et al., 2015;Paulot et al., 2011;Stavrakou et al., 2012). Similarly, in our simulation the EMAC model largely underestimates the HCOOH total column densities derived from the IASI observations by up to a factor of 4, in particular in remote environments (see Fig. 12). Although the two main tropical source regions identified by IASI -the Amazon Basin and central Africa -are reproduced by EMAC, the magnitude of the simulated HCOOH column densities are too low in comparison to the satellite 405 measurements. It also has to be mentioned that the apparent agreement over Amazonia is possibly due to the high temperature bias in the region and the subsequent excess of simulated isoprene emissions during the dry season. Over the other source regions (e.g., Southeast Asia and Southeast US), the model underestimation is more pronounced, in particular in the Northern Hemisphere. The large enhancement of HCOOH column densities observed with IASI over western Russia (∼8 × 10 15 molecules cm −2 ), attributed to the August 2010 wildfires, is not reproduced by EMAC (<2 × 10 15 molecules cm −2 ) and 410 suggests that the biomass burning emissions of VOCs are underestimated.
The general underestimation of simulated versus observational data is also confirmed by the comparison with aircraft observations. Especially at low altitudes and for high absolute values the model strongly underestimates the observations by the ATom campaign (see Table 2, the observed mean is higher by more than a factor of 4 compared to the model mean). In general, there is very low agreement between simulated values and ATom observations as only around 35% of the simulated values are 415 within a factor of 2 of the observations and the Pearson correlation coefficient is only 0.3. Although the underestimation of the model results to the Emmons et al. (2000) database is less apparent it must be noted that only a limited number of merged data (53, see Table 3) is available for the comparison with aircraft observations. It must be stressed that Franco et al. (2021) showed the importance of in-cloud chemistry for this tracer, which is missing in this study. The lack of adequate in-cloud chemistry would therefore explain the almost ubiquitous underestimation of this tracer.  Figure 13. Same as Fig. 8, but for CH3COOH. In this figure area above the ocean were excluded.

CH 3 COOH
Acetic acid is the second most abundant carboxylic acid in the troposphere and, based on the IASI retrievals, presents a spatial pattern, regional seasonality, and vertical abundance, which resemble those of HCOOH (Franco et al., 2020). Like the latter, CH 3 COOH is produced from the oxidation of various tropospheric precursors, but has emission factors from biomass burning that are 3 to 10 times larger than those of HCOOH (Akagi et al., 2011;Andreae, 2019). CH 3 COOH is more difficult to detect 425 in the infrared IASI spectra and its retrievals are subject to larger uncertainties, in particular over the oceans (see Franco et al., 2020). Therefore, we limit here the comparison with EMAC to the continents, excluding measurements over desert areas that are altered by surface emissivity artefacts (Fig. 13). From the comparison, conclusions similar as for HCOOH can be drawn for CH 3 COOH: the two main tropical source regions are relatively well reproduced by EMAC (with a better agreement over Africa in this case), whereas the observed CH 3 COOH levels are underpredicted in the Northern Hemisphere by up to a factor 430 of 4. The comparison with aircraft observations (see Table 3) confirms again this strong underestimation also over the ocean, as the only campaign in the dataset including CH 3 COOH measurements is the PEM-Tropics-A, which was performed over the Pacific ocean.
These results confirm that the VOC emissions and oxidation pathways leading to the formation of CH 3 COOH in the troposphere are still poorly understood and constrained (e.g., Khan et al., 2018;Paulot et al., 2011). For example, acetaldehyde -a 435 major CH 3 COOH precursor via its reaction with OH (Lei et al., 2018) -is well known to be largely underestimated by global models (Millet et al., 2010;Wang et al., 2019). On the other hand, the large model versus observations differences in Southeast Asia also point to missing emissions from biomass burning. Finally, in-cloud chemistry could be important in the CH 3 COOH formation, analogously to formic acid (Franco et al., 2021), and this process could bring model results and observations to a closer agreement.

PAN
Owing to its complex photochemical sources and its thermal instability, PAN (Peroxyacetylnitrate), the main tropospheric reservoir species of NO x , is a very challenging tracer to simulate. The comparison with the IASI data reveals that the model reproduces correctly the main spatial patterns of PAN, with the source regions and main outflows over the oceans being correct. However, the model constantly underestimates the observed PAN column densities over the globe. The satellite column densi-445 ties are indeed between two and four times the simulated ones, with the most pronounced negative bias observed at Northern Hemisphere mid-and high latitudes. The same conclusion can be drawn from the comparison with aircraft observations (Table   2 and 3), for which the model clearly underestimates the observed mixing ratios consistently using the database from Emmons et al. (2000) and the measurements of the ATom campaign. A closer inspection reveals that PAN is especially underestimated in the middle and upper-troposphere. The model-to-observations discrepancies can mostly be attributed to the unsatisfactory 450 representation of different VOCs. For example, we have seen that the model results do not agree with observations for acetone, which is an important precursor of the peroxyacetyl radical in the free troposphere (Fischer et al., 2014). Furthermore, the strongest model's underestimation appears to be exactly where CH 3 COCH 3 is at most underestimated, confirming that deficiencies of simulated precursor patterns are the main cause of the PAN biases. Additionally, sensitivity simulations suggested that the PAN formation in global models is more sensitive to the representation of VOCs (Fischer et al., 2014). Finally, 455 an insufficient vertical transport of VOCs from the planetary boundary layer to the free troposphere in the model might also reduce the amount of PAN, because PAN is more stable at the lower temperatures of the free troposphere.

OH
The simulated distribution of the hydroxyl radical (OH) is in line with the findings by Lelieveld et al. (2016) (further denoted as L16), who earlier thoroughly analysed the MOM performance in EMAC with a setup featuring a simplified treatment of 460 aerosol microphysics and gas/aerosol partitioning. The total OH turnover (composition of annual production and loss terms, shown in Fig. 15) of 234 Tmol yr −1 in 2010 is 7 % less than that found by L16 and is consistent with the addition of the explicit treatment of secondary organic formation processes of aerosols and changes in trace gas emission fluxes. The OH production in the VOC-and ROOH-initiated reactions is 3 % lower in both, the free troposphere (FT), and the boundary layer (BL). This is compensated by the increased share of primary production (P , via the O( 1 D)+H 2 O pathway), especially in the 465 BL (4 % increase vs. 2 % in the FT). Regarding the partitioning of the OH sinks, however, no substantial changes can be noted (see Fig. 15, right panel). Together with the reduction of the secondary sources (S), the resulting OH recycling probability r OH (defined as 1 − P/(S + P ), see details in L16) establishes at 59 % and 62 % in the BL and FT, respectively. Being about 5 % lower than the estimate reckoned by L16, such high r OH values signify still highly stable (buffered) OH concentrations and therefore a likewise stable tropospheric oxidative capacity obtained with this model set-up.   Overall, the simulated tropospheric OH distribution is comparable to that thoroughly analysed by L16 with a minor increase in reactivity of up to 5 % when calculated from the changes to the CH 4 and MCF (CH 3 CCl 3 ) lifetimes against removal by OH (τ OH ). The here estimated τ OH (CH 4 ) values are 8.4 and 4.4 years in the FT and BL, respectively; the tropospheric τ OH (MCF) value is obtained to be 4.9 years.

Aerosol Optical Depth
In order to evaluate the overall performance of the model in reproducing aerosol distribution, the AOD at 550 nm simulated by the model is compared to satellite based and station observations. Figure 16 shows the model results compared to the AERONET measured AOD in the case of the daily and monthly averages, for the annual 500 averages the model seems to perform equally well, with a similar coefficient of determination and a fraction of values within a factor of 2 to the observations. Due to the intrinsic characteristics of the model (such as its relative coarse resolution) it is not expected to have a good representation of the short term variability.

Fine particulate mater (PM 2.5 )
The comparison of model results for PM 2.5 (i.e. fine particles with an aerodynamic diameter smaller than 2.5 µm) was per-505 formed against the dataset of Hammer et al. (2020). Although this dataset is not a pure observational dataset, it has a global coverage and has been constrained from in-situ measurements. The comparison is done for PM 2.5 at 35 % relative humidity, i.e. dry fine particulate mass (in µg m −3 ). In Figure 18, the maps of the annual average of the model results and the pseudo-observation dataset are shown. Although the distribution patterns appear to be similar, the model underestimates the fine particle concentration over northern India and eastern China, on the one hand, while the opposite happens over east Africa 510 and the Middle East. On the other hand side, the model seems to reproduce PM 2.5 quite accurately over both Europe and North In Figure 19, the scatter plot of the two datasets is shown, while in Table 5 the statistics for the different regions presented in the figures are listed. Due to the large grid resolution difference, the pseudo-observations have been aggregated to the grid 515 resolution of the model. As already noted from the global map, the model underestimate PM 2.5 over South Asia and East Asia by ∼ 40 % and ∼ 20 %, respectively. Nevertheless, in both regions, more than 70 % of the model results are within a factor of 2 of the pseudo-observations. As shown by Pozzer et al. (2012aPozzer et al. ( , 2017, and more recently by Miao et al. (2020), BC and OC are very important for the PM 2.5 budget in East and South Asia. As shown by Crippa et al. (2018) and Saikawa et al. (2017) the emissions of these tracers are associated with large uncertanties in these regions, which could strongly affect our results.

520
The model agrees well, however, with data from Europe and North America, with more than 95 % of the model results within a factor of two. The overall comparison indicates that the model agrees well with the pseudo-observations, as the spatial-temporal averages of the two datasets are very close (17.1 µg m −3 for the model and 19.4 µg m −3 for the pseudo-observations), with more than 70 % of the model simulated PM 2.5 within a factor of two of the pseudo-observed PM 2.5 .

525
For the evaluation of the simulated mass concentrations of sulfate, nitrate, ammonium, sodium and five species related to sea spray and organic aerosols, we use in-situ measurements from different monitoring networks, as described in Sect. 3.2.
Co-located time series of simulated and observed quantities were obtained by bilinear interpolation of the gridded model data at ground level to the respective site location. The analysis is based on monthly mean concentrations, which are derived from daily (model data, EMEP observations) and weekly (EPA) data, with the exception of EANET, which directly provides monthly 530 averages. Note that some monitoring stations are located relatively close to each other, such that the reported concentrations may not be independent from each other, which can lead to an overestimation of the degrees of freedom in the calculation of quantities such as the root mean square error.  Light blue, red, green, and orange colors depict points located in Europe, North America, East Asia, and South Asia, respectively. Grey points indicate the remaining parts of the world.

Sulfate (SO 2− 4 )
Observed particulate sulfate concentrations are reproduced well by the model. The monthly mean concentration is matched 535 closely for North America (EPA) and slightly underestimated for Europe (EMEP) and East Asia (EANET) (see Table 6).  Table 5. Summary for simulated and pseudo-observed, annually averaged PM2.5. MM and OM denote the arithmetic mean of the simulated and observed concentrations, respectively, in µgm −3 . PF2 is the percentage of simulated points within a factor of 2 with respect to the observations.
For EPA and EMEP data, more than 80 % of the simulated monthly mean concentrations lie within a factor of two of the observations, and more than 70 % for EANET. The standard deviation of observed monthly mean values is lower than the root mean square error (RMSE) for all monitoring networks, which is an indication for a good quality of the model results (Barna and Lamb, 2000). Also, RMSE is lower in the present study compared to a similar analysis by Pozzer et al. (2012a) in which,  (Figure 21). This feature also lead to a good agreement between observed and simulated mean concentrations in Table 6. The North-South gradient over Europe is also captured by the model.
Despite the less dense spatial distribution of monitoring stations providing data for particulate sulfate, one can observe a correspondence between simulated and observed annual mean concentrations, especially over Japan. One outlier in Japan with a very low observed annual mean concentration is located at a relatively high altitude ("Happo", 1850 m), so that the 555 comparison to model data on ground level may not be appropriate for this station. The same holds for two EMEP stations, namely Jungfraujoch (Switzerland) at 3578 m, and Chopok (Slovakia) at 2008 m, where annual mean concentrations are largely overestimated, possibly due to the coarse resolution of the model, which cannot reproduce correctly the orography (and hence the altitude) of these stations. We therefore compared SO 2− 4 concentrations with observations from aircraft campaigns, as compiled by Heald et al. (2011). The results are presented in Fig.22 Table 6. Summary of simulated and observed monthly averaged aerosol concentrations for SO 2− 4 , NO − 3 , NH + 4 , Na + , Mg 2+ , Ca 2+ , K + , Cl − and organic aerosols (OA) and different observational networks (second column). Nstation is the number of stations providing data for the respective species and network; note that this is not the sample size for the shown statistics. The next columns provide the arithmetic mean of the simulated (MM) and observed concentrations (OM), and the respective standard deviations (MSTD, OSTD) in µgm −3 . PF2 denotes the percentage of simulated points within a factor of 2 with respect to the observations, and RMSE represents the root mean square error between simulated and observed points. the aircraft observations in the free and lower troposphere, with an underestimation in a few cases (see Fig.22, ITCT-2K4, ADIENT or IMPEX campaigns), but always within the measurement standard deviations, confirming that the vertical profile of sulfate is generally well reproduced in the lower troposphere.   of the simulated monthly mean concentrations exceed a factor of 2 with respect to the observations. For EANET and EMEP, the number of monitoring stations is substantially lower than for EPA, which may partly explain the larger spread in both, the model results and observations (OSTD, MSTD). The RMSE is larger than the observed standard deviations for all considered 570 networks, suggesting a less faithful representation of this species. This overestimation may be attributed to the usage of Teflon filters, as nitrate can evaporate from the filters under warm and dry conditions (Ames and Malm, 2001). Schaap (2003), Schaap et al. (2004) andde Meij et al. (2006) showed that particulate ammonium and nitrate partially evaporate from the filters at temperatures between 15 to 20 • C and can evaporate completely at temperatures above 20 • C. Therefore, this effect predominantly affects measurements during the warmer seasons, when nitrate concentrations are close to the annual minimum.

575
An indication for overestimation due to evaporation would then be a closer agreement between model and observation in the colder winter months. However, both in relative and absolute terms, the overestimation of monthly mean nitrate concentrations is more pronounced in winter than in summer for the three considered networks (Figure 23). All regions show the same seasonal cycle with respect to averaged monthly mean concentrations; a maximum in winter and a minimum in summer, both for model results and observations. The spatial gradient of annual mean concentrations is captured well for North America 580 (EPA), although the concentrations towards the East coast are generally overestimated. The low number of monitoring stations for EANET and EMEP does not allow us to draw solid conclusions; yet we can report that the North-South gradient of nitrate over Europe is reproduced by the model, except for an outlier at high altitude in Central Europe ( Figure 24). The region with highest simulated annual mean concentrations and strong spatial gradients is East Asia, however it is scarcely covered by monitoring stations reporting full coverage for the considered year 2010.

Ammonium (NH + 4 )
Considering the ratio of averaged monthly mean ammonium concentrations, the model overestimates the observed concen- trations, yet to a lower degree than for nitrate (Table 6). This feature is also apparent in the fraction of simulated monthly The North-South gradient of observations over Europe is also captured by the model, although the number of stations is again comparably low. The simulated spatial patterns in East Asia also match the observations arguably well.

Dust and sea spray 600
This section provides an overview for species frequently found in dust and sea spray aerosol: sodium, calcium, magnesium, potassium and chloride. As the ocean is a large source of these water-soluble species, concentrations over the sea and coastal areas are large, as opposed to lower concentrations over the continents. Sodium, potassium, magnesium and calcium can also be found in desert dust, for instance in the Sahara, as reported by Reid et al. (2003) and Moreno et al. (2006). However, most of the monitoring locations except for those in coastal areas are sampling quite remote sites far from the source regions, and 605 thus can not be expected to display the full range of concentrations. In the following, we present the results for sodium in more detail, followed by a brief overview of the remaining species.
Sodium is represented well by the model, as averaged monthly mean concentrations from model and observations agree in general, particularly for EPA (Table 6). The mean ratio MM/OM, however, shows an average overestimation by a factor of at least 2. The observational standard deviation OSTD for all regions is smaller than the RMSE, the latter being reduced 610 compared to the results from a previous study by Pozzer et al. (2012a). For EMEP, more than 3 out of 5 simulated monthly mean values lie within a factor of 2 of the observations and more than 53% and 46% for EMEP and EANET, respectively, which is a substantial improvement compared to Pozzer et al. (2012a). The latter two networks offer again only a small number of stations. Considering the average of monthly mean concentrations separated by season, one can observe an underestimation of summer concentrations for all networks ( Figure 27).
As mentioned before, the regional distribution of stations does not cover those regions with high concentrations. However, including the information of all networks at once, one can observe that continental stations exhibit low annual mean concentrations, while high concentrations can be observed for instance at stations close to the coast (i.e. Japan, Florida, Denmark, Sweden), indicating that spatial features of the global sodium distribution are qualitatively captured ( Figure 28).
Potassium, magnesium and calcium ions share spatial features with sodium in the model, exhibiting elevated values over the 620 ocean, as well as over the Sahara and Gobi desert, and reduced concentrations over other continental areas.
The number of stations in EANET is quite low for all sea spray species which, on the one hand, allows for a few outliers to  Figure 30 shows the comparison of model calculated OA concentrations with measurements from the EMEP observational network over Europe, the IMPROVE network over rural North American locations, as well as short-term measurement data collected over East Asia as summarized by Jo et al. (2013). As we used only measurements taken in the year 2010, a low number of observations are present for East Asia, and the comparison must be taken cautiously for this region. The comparison statistic is presented in Tab. 6. The model captures relatively well the monthly average concentrations of OA over these highly 635 populated regions of the Northern Hemisphere. This is rather encouraging, given the expected uncertainties of the emission inventory and the complex chemistry involved to simulate the secondary organic aerosol formation. It is worth emphasizing that the model considers the formation of SOA solely from the homogeneous gas-phase photochemical oxidation of its precursors.
Therefore, the omission of other SOA formation pathways (e.g., from aqueous-phase and heterogeneous reactions) can add to the model bias. This is mostly evident during winter (e.g., over Europe), when the relative importance of these processes on 640 SOA formation is increasing due to the lower photochemical activity and the limited conversion of gas phase organic precursors to SOA. In addition, recent studies have provided strong evidence that the uptake of water-soluble gas-phase oxidation products (even small carbonyls like formaldehyde and acetic acid) can be the main driver of SOA pollution during haze events over Eastern Asia (Gkatzelis et al., 2020). Given the coarse resolution of the model and its inability to simulate these SOA formation pathways, the observed total OA concentrations are expected to be systematically underestimated by the model over Eastern Despite the agreement of the surface OA concentrations between model results and observations, the OA are strongly underestimated in the free troposphere. In Fig. 31, the model is compared with 12 field campaign measurements (Heald et al., 2011).
Our results show similar agreement to the one obtained by Heald et al. (2011), with the simulated values below the observed one.

Conclusions
We have presented an evaluation of EMAC with a comprehensive degradation scheme for organics (MOM) in the MECCA chemistry submodel, combined with an explicit estimation of the organic tracers' condensation on aerosols with a VBS approach (ORACLE submodel). We have compared the model results with a large number of in-situ and satellite based remote sensing observations. The evaluation focuses on carbon monoxide (CO), simple organics and aerosols, which are most in-665 Figure 31. As Fig.22, but for organic aerosol.
fluenced by the new scheme. CO is correctly reproduced compared with station measurements, with the correct north-south gradient. Nevertheless, once the model results are compared with the satellite observations from MOPITT, an overestimation of CO over the Amazon basin is apparent, especially in autumn, possibly due to an overestimation of the biomass burning emissions simulated by the model. For alkanes the comparison shows a good agreement, although an oceanic source for the butanes is missing. In contrast, the model underestimates alkenes, and especially C 3 H 6 , which shows large differences com-670 pared to the observations. Model results of oxygenated species show an indifferent picture when compared to observations, depending on the tracer: while methanol is well reproduced, simulated acetone shows large spatial discrepancies compared to the remotely sensed observations. In fact, for the latter, the model misses strong emissions over the boreal forests, while it predicts instead too high total columns over tropical forests (i.e., Amazonia and Central Africa). Formic and acetic acid show similar agreements to the observations, with the main tropical sources well reproduced, but with a strong underestimation 675 elsewhere.
The aerosols at the surface are well reproduced by the model, both with respect to composition and total amount. The annually averaged simulated AODs show a very good agreement with the AERONET station observations, but with some deficiencies in the representation of the short term variability. The agreement of the fine particulate mass (PM 2.5 ) near the surface is also quite good, with most of the simulated values within a factor of 2 to the observations. The evaluation of the 680 aerosol chemical composition shows a high level of agreement for near-surface sulphate and some more deviations for NH + 4 and nitrate, which are both overestimated by the model, especially the latter (even though the observations are characterized with a high level of uncertainty). The mechanically produced primary aerosol species (dust and seasalt) show a reasonable agreement with the observations with some substantial overestimation of the simulated chloride concentrations indicating a too weak acid displacement process over the continents.

685
Finally, the comparison of OA shows a good agreement with station observations, while the vertical distribution of the simulated values are largely underestimated, when compared with aircraft measurements, analogously to other global atmospheric chemistry models of similar complexity, although significant improvement in this direction has been recently published (Pai et al., 2020).

690
The presented model evaluation is useful for identifying areas for further model improvements. A few possible directions are discussed next.
Emissions have always been a critical point for simulating tropospheric chemistry. It has already been shown in many studies that even anthropogenic emissions of alkanes need to be improved as severely underestimated in many inventories (e.g. Emmons et al., 2015;Tilmes et al., 2016), mostly coming from erroneous speciation of the total VOC emissions (Dalsøren Residential wood burning and/or biofuel use (e.g., for heating) that often contribute significantly to both POA and SOA formation over urban areas are strongly underestimated in the emission inventories (Tsimpidi et al., 2016;Denier Van Der Gon et al., 2015). In addition, global models lack OA emissions from residential and commercial cooking activities that can be an important source of OA (Mohr et al., 2012;Sun et al., 2011;Ge et al., 2012;Hayes et al., 2013). Underestimation of cold-start 700 vehicle emissions (Weilenmann et al., 2009) can also lead to significant underestimation of OA formation from the transport sector (e.g., during wintertime).
As seen in this study, the model results underestimate selected VOCs in boreal regions. The thawing of permafrost is potentially an additional emission source of organic compounds at high latitudes in the northern hemisphere (Li et al., 2020), which has not been considered yet. However, an estimate with global data sets is still missing. Furthermore, the emission of VOCs 705 from biomass burning is likely strongly underestimated. In fact, overwintering fires in boreal forest (Scholten et al., 2021), which smoulder through the non-fire season, are normally not detected by satellite observations and therefore are missing in our used emissions dataset. Moreover, peat fires are not easily detected from space, but are characterized by larger emission factors for many VOCs. They are important for simulating air composition during peat fires in Indonesia (Rosanka et al., 2021a) and likely critical for resolving the model biases at high latitudes, where most of peatland is located and could be fur-710 ther released by permafrost thawing. Ramo et al. (2021) also found that small undetected fires enhance the estimated amount of emitted carbon from biomass burning in sub-Saharan Africa by 31-101% . Finally, soil temperature and soil wetness play an important role in the estimation of the online emissions of biogenic VOCs, and it is therefore important to have a correct simulation of the surface properties by the underlying climate model.
Furthermore, dry deposition is a relevant process also affecting VOCs in the lower troposphere (Khan et al., 2015). The 715 inclusion of the additional uptake at the plant cuticle leads to a better representation (Karl et al., 2004;Müller et al., 2018;Emmerichs et al., 2021) and would likely reduce the overestimation of species like acetaldehyde in the Tropics, as shown here.
In the Amazon forest, the missing storage capacity of soil water leading to too high temperatures (Hagemann and Stacke, 2015) additionally causes an underestimation of dry deposition (Emmerichs et al., 2021), while temperature-dependent processes like VOC emissions are overestimated.

720
Another source of uncertainty is related to the scavenging efficiency of gas-phase OA precursors. The water solubility of these oxidized organic vapors is largely unknown and is typically considered uniform for all organic compounds in modeling studies even though they become increasingly more hydrophilic during their atmospheric lifetime (Hodzic et al., 2014;. The extension of MOM for the oxidation of additional VOCs has the potential to reduce the negative model biases for some 725 oxygenated volatile organic compounds (OVOCs) and enhance the predicted OA levels. For instance, with an emission strength of more than 40 Tg yr −1 , anthropogenic C 5 -and higher hydrocarbons can be a significant source of OVOCs like acetaldehyde and acetone . Further improvements are expected from an extended representation of the emission and oxidation of known biogenic VOCs with more than 5 carbon atoms. Moreover, the emissions of aromatic compounds from biogenic sources could be as large as 40 Tg yr −1 (Misztal et al., 2015). So far only a small emission of toluene is considered 730 and a more comprehensive representation of biogenic aromatics would likely lead to further improvements.
The formation of SOA in most models relies on the reaction of its gas-phase precursors with reactive atmospheric radicals (e.g., OH, O 3 , NO 3 ) and their chemical aging is usually considered by the further photo-oxidation of the semivolatile products solely in the gas phase. However, SOA is also subject to oxidation in the aqueous phase through superficial and bulk interactions with gas-phase oxidants (George and Abbatt, 2010). Therefore, fundamental processes of SOA formation in the aqueous 735 phase (e.g., from isoprene epoxydiols (Nguyen et al., 2009) and glyoxal (Fu et al., 2008)) are typically missing from the conventionally used parameterizations. In addition, the uptake of small carbonyls (e.g., aldehydes and acids) to the aqueous phase and their subsequent oxidation and oligomerization has been recently linked to significant increases of SOA mass during pollution events (Gkatzelis et al., 2021).
Recently, MOM has been coupled to the detailed Jülich Aqueous-phase Mechanism of Organic Chemistry (JAMOC, Rosanka et al., 2021c) for cloud droplets. An explicit treatment of multiphase chemistry of OVOC has already been used for assessing the global role of clouds as a sink of ozone (Rosanka et al., 2021b). By comparing EMAC's prediction of total methanol columns to IASI satellite retrievals (comparable to Sect. 4.2.3 and Fig. 8), they find that EMAC's tendency to overestimate methanol is partially reduced by the additional cloud sink. This suggests that the missing representation of in-cloud OVOC chemistry introduces a significant bias in the present study, and therefore a more detailed multi-phase chemistry should be 745 included for future studies. Furthermore, JAMOC is limited to the oxidation of OVOCs containing up to four carbon atoms (Rosanka et al., 2021c), and an expansion to larger species (i.e. containing more than four carbon atoms) is thus desirable, in order to improve the representation of these OVOCs in EMAC when using MOM. Based on the results of Franco et al. (2021), more detailed in-cloud chemistry could improve the representation of acids (such as acetic and formic acid), which is clearly underestimated here where cloud chemistry is important.