2.1 Summary of the POLIPHON method with focus on dust

The POLIPHON method is described in detail by Mamouri and Ansmann (2014, 2015, 2016, 2017) and with respect to the INP concentration retrieval also by Marinou et al. (2019). The main part of the POLIPHON data analysis deals with the conversion of aerosol-type-dependent particle extinction coefficients into respective particle microphysical properties. Table 1 provides an overview of POLIPHON dust products and the respective conversions. Similar conversions for non-dust aerosols such as maritime particles or continental fine-mode aerosol pollution (urban haze, biomass burning smoke) can be found in Mamouri and Ansmann (2016, 2017).

Table 2Lidar ratios for different desert regions from AERONET observations at 675 nm (Shin et al., 2018) and from numerous lidar observations at 532 nm. Recommendations for lidar ratios to be used in the POLIPHON data analysis are given in the right column.

Download Print Version | Download XLSX

In the first part of the POLIPHON data analysis, the polarization lidar observations are analyzed to obtain height profiles of dust and non-dust backscatter coefficients. Here we assume that pure dust causes particle linear depolarization ratios of 0.3–0.35 around the globe, disregarding the dust source region. This is corroborated by numerous studies (see the reviews in Tesche et al., 2009; Mamouri and Ansmann, 2014, 2017) and also during recent field campaigns (Groß et al., 2015; Veselovskii et al., 2016; Haarig et al., 2017; Hofer et al., 2017). Details of the aerosol type separation procedure (including the separation of fine and coarse dust by the use of fine-mode- and coarse-mode-related dust depolarization ratios) can be found in Mamouri and Ansmann (2014, 2017). The derived total, fine-, and coarse-dust backscatter coefficients β_d, β_df, and β_dc are then converted to respective dust extinction values σ_d, σ_df, and σ_dc by means of appropriate dust extinction-to-backscatter ratios or lidar ratios S_d, S_df, and S_dc. As shown in Table 2, the 532 nm dust lidar ratio S_d may vary from about 30 to 60 sr for different mineral dust types (Müller et al., 2007; Tesche et al., 2011; Mamouri et al., 2013; Nisantzi et al., 2015; Groß et al., 2015; Veselovskii et al., 2016; Haarig et al., 2017; Hofer et al., 2017; Shin et al., 2018). However, for most dust regions, except the western Sahara, the typical dust lidar ratio is 40 sr at 532 nm. We therefore recommend using 40 sr as dust lidar ratio and to select 50 sr only in cases with airflow from the western Sahara. The best option is however to use actual Raman lidar observations of the dust lidar ratio. We further assume that $S_{\mathrm{d}}=S_{\mathrm{df}}=S_{\mathrm{dc}}$ (see lines 2–4 in Table 1). A relative uncertainty in the dust lidar ratio assumptions of 10 % is considered in the estimation of the relative uncertainties (error propagation) in Table 1.

In the second part of the POLIPHON data analysis (see Table 1, lines 5–15), the height profile of the dust mass concentration M_d(z) is derived from the dust extinction coefficients σ_d(z), also separately for coarse dust (M_dc considering particles with a radius >500 nm) and fine dust (M_df considering dust particles with radius <500 nm) from respective coarse- and fine-dust extinction coefficients σ_dc and σ_df. The dust extinction coefficients are converted into dust particle volume concentrations v_d, v_dc, and v_df by means of extinction-to-volume conversion factors c_v,d, c_v,df, and c_v,dc, and afterwards multiplied by the dust particle density ρ_d of 2.6 g cm⁻³ (Ansmann et al., 2012) to obtain the respective dust mass concentrations. The required conversion factors are determined from AERONET observations as described in Sects. 2.2 and 3.1.

Further POLIPHON conversion products listed in Table 1 (lines 8–15) are needed in the estimation of the cloud-relevant aerosol parameters such as the cloud condensation nucleus concentration (CCNC) and ice-nucleating particle concentration (INPC). The number concentration n_100,d (considering particles with radius >100 nm) is a good proxy for the dust CCNC (Mamouri and Ansmann, 2016; Lv et al., 2018). However, CCNC depends on the water supersaturation at cloud base where aerosol particles mainly enter the cloud and serve as CCN. A typical water supersaturation value is 0.2 % (Siebert and Shaw, 2017) and occurs when air parcels are lifted into the base of a liquid water cloud by weak updrafts, e.g., in the case of fair weather cumuli. Water supersaturation values may exceed even 1 % in strong updrafts. For the conversion of σ_d into number concentration n_100,d, the conversion parameters c_100,d and exponent x_d as shown in Table 1 are used and obtained from the AERONET observations (see Sects. 2.2 and 3.2).

We introduce the factor f_ss,d to consider the water supersaturation dependence. With increasing supersaturation at cloud base an increasing number of dust particles (i.e., particles with lower radius) can be activated as CCN. For a supersaturation value of 0.4 % even dust particles with radius of 70–80 nm become activated. According to Shinozuka et al. (2015), $f_{\mathrm{ss},\mathrm{d}}=\mathrm{2}$ is appropriate when using n_100,d as the basic aerosol parameter in the CCNC estimation but the supersaturation is 0.4 % (see Mamouri and Ansmann, 2016, for more details). For completeness, in Table 1, f_ss,d is 1.0, and the respective equation holds for a liquid-water supersaturation level of 0.2 %.

The particle number concentration n_250,d (considering particles with radius >250 nm, Table 1, line 9) and the dust particle surface area concentration s_d (line 10) and s_100,d (considering only particles with radius >100 nm, line 11) are input in the estimation of height profiles of dust INPC when using the INPC parameterization for immersion freezing of DeMott et al. (2015) (Table 1, line 13, D15) and Ullrich et al. (2017) (line 14, U17-I) and for deposition nucleation of Ullrich et al. (2017) (line 15, U17-D). For the conversion of σ_d into number concentration n_250,d and surface area concentrations s_d and s_100,d the conversion factors c_250,d, c_s,d, and $c_{\mathrm{s},\mathrm{100},\mathrm{d}}$ are required. In addition to aerosol number and surface concentrations, the temperature profile T(z) and an assumed ice supersaturation value S_ice (in the case of deposition-freezing INPC, U17-D) are input in the INPC estimation. The ice supersaturation S_ice is set to a typical value of 1.15.

We introduce a new parameter (not considered in Mamouri and Ansmann, 2016), namely the surface area s_100,d as an alternative input parameter in the estimation of immersion freezing INPC. In the case of immersion freezing, liquid droplets form first before freezing occurs. As discussed above, appropriate dust CCN for typical water supersaturation values of 0.2 % have a radius >100 nm. Only these particles (immersed in the liquid droplets) can then serve as INPs so that the surface area s_100,d may be a more appropriate aerosol proxy in the INP estimation by using the immersion freezing parameterization U17-I (Ullrich et al., 2017) than the total surface area concentration s_d. However, both parameters (s_d, s_100,d) are required in the INP parameterization. For deposition nucleation (heterogeneous ice nucleation by water vapor deposition directly on dust particles, without any liquid phase formation), s_d is the relevant aerosol input parameter. All this is described in detail in Mamouri and Ansmann (2016). More details on the INPC retrieval are also given in Sect. 4.

Table 1 also provides an overview of the uncertainties in the POLIPHON products (Mamouri and Ansmann, 2016, 2017). The very large uncertainties in the estimation of n_100,d, n_CCN,d, and n_INP,d (factor of 2–5) are obtained when taking all potential error sources into consideration. INPC parameterizations developed from field observations (for aerosol types that are sometimes not well characterized) and from laboratory experiments with fresh rather than aged dust particles, i.e., chemically and cloud-processed dust particles, must always be handled with care and may not be fully applicable to atmospheric conditions with predominantly aged dust so that uncertainties of the order of a magnitude can not be excluded. However, meanwhile a variety of studies indicate that uncertainties in the n_CCN,d and n_INP,d values of the order of 50 % are more realistic to characterize the errors in lidar-based CCNC and INPC estimations (Düsing et al., 2018; Marinou et al., 2019; Haarig et al., 2019). Uncertainties of 50 % are acceptable in process studies of aerosol–cloud interaction performed to investigate the role of dust in cloud evolution processes. Even uncertainties of a factor of 2–5 are acceptable in attempts to establish a vertically resolved tropospheric climatology for CCNC and INPC. Upper tropospheric long-term observations of INPC are not available in the literature, but strongly required (even if the uncertainties are high) to support weather and climate modeling.

2.2 POLIPHON dust conversion parameters

Trustworthy and climatologically robust conversion parameters obtained from AERONET observations are of central importance for the applicability and attractiveness of the POLIPHON method. For our study, we downloaded the following data sets of AERONET products (single measurements, inversion products, version 3, level 2.0) (AERONET, 2019): (1) the particle volume size distribution resolved in 22 size classes from 50 nm (bin 1) to 15 µm (bin 22), (2) the corresponding data sets of total, fine-mode-, and coarse-mode-related volume concentrations and effective radii (from which surface area concentrations can also be calculated), and (3) the corresponding data set of aerosol optical thickness (AOT) for eight wavelengths (denoted as extinction AOT in the AERONET database) together with respective Ångström exponents (AEs) for the 440–870 nm wavelength range. Details of the AERONET data-processing steps are given in Mamouri and Ansmann (2014, 2015, 2016, 2017).

To obtain climatologically representative dust conversion factors for a given AERONET station, we filtered out all AERONET data sets fulfilling the constraints of an Ångström exponent AE <0.3 and a 532 nm AOT >0.1. The AOT for 532 nm (in the following equations simply denoted as τ_d) is obtained from the 500 nm AOT τ₅₀₀ and the Ångström exponent a, stored in the AERONET database, by

More information on the dust selection criteria is given in Sect. 3.

It is noteworthy to mention that recent airborne in situ observations of dust size distributions over the Sahara and remote dust outflow regions by Ryder et al. (2019) indirectly corroborate the applicability of the AERONET data analysis and inversion concept and therefore the high quality and consistency of the AERONET optical and microphysical data sets used in our study (AERONET, 2019). The AERONET inversion method required to obtain the dust microphysical properties from the measured optical properties assumes that only particles with a radius ≤15 µm are responsible for the observed dust-related optical effects. The presence of larger dust particles is ignored. Ryder et al. (2019) now show that dust particles with a radius >15 µm contribute only 1 %–3 % to the particle extinction coefficient at 550 nm. This means that this size cutoff effect in the AERONET data inversion procedure has practically no impact on the AERONET inversion products and thus a negligible influence on the derived POLIPHON conversion factors.

In the following, we use the example of the POLIPHON dust mass concentration retrieval to explain the role of the conversion factors and how we derived them from the AERONET database. The mass concentration of dust (index d) is given by

with the dust particle density ρ_d of 2.6 g cm⁻³ and the dust volume concentration v_d. The required dust volume concentration in Eq. (2) can be obtained from the following conversion:

with the extinction-to-volume conversion factor $c_{\mathrm{v},\mathrm{d},\mathit{\lambda }}$ (derived from the AERONET long-term observations) and the particle extinction coefficient σ_d,λ measured with lidar at wavelength λ. We concentrate on lidar observations at 532 nm in this study and omit the wavelength index λ in the following. The conversion factor is obtained from the AERONET observations of the vertically integrated particle volume concentration V_d (denoted also as column volume concentration) and the aerosol optical thickness τ_d (AOT at 532 nm, see Eq. 1),

To provide a link to the lidar-derived height profile of σ_d(z) (see Eq. 3), we introduce an aerosol layer depth with an arbitrarily chosen vertical extent D. With D, Eq. (4) can be written as

with the layer mean volume concentration v_d and the layer mean particle extinction coefficient σ_d. For simplicity, we assume that all aerosol is confined to the introduced layer with vertical depth D. We may interpret this layer as the dust-containing boundary layer or as a lofted dust layer with a vertical extent D. The introduced layer depth D has no impact on the further retrieval of the conversion factors and is only required to move from column-integrated values and AOT to more lidar-relevant quantities like concentrations and extinction coefficients.

To obtain a climatologically representative dust conversion factor for a given AERONET station, we selected all individual dust observations (from number j=1 to J_d collected over many years), as mentioned defined by an Ångström exponent AE <0.3 and 532 nm AOT >0.1. For each dust observation j we computed $c_{\mathrm{v},\mathrm{d},j}$ and then determined the mean value, which we interpret as the climatologically representative POLIPHON conversion factor,

In the same way, all other conversion parameters in Table 1 are computed:

As before, indices df and dc denote fine-mode and coarse-mode dust fractions, respectively. In Mamouri and Ansmann (2015, 2016), we explain how we calculate $n_{\mathrm{250},\mathrm{d},j}$, s_d,j, and $s_{\mathrm{100},\mathrm{d},j}$ (discussed below) from the downloaded AERONET size distribution data sets.

In the retrieval of the conversion parameters required to obtain n_100,d (Table 1, line 8), we used a different approach. Following the procedure suggested by Shinozuka et al. (2015), we applied a log–log regression analysis to the log(n_100,d)–log(σ_d) data field for each of the considered 20 AERONET stations and determined in this way representative values for c_100,d and x_d that best fulfill the relationship,

as will be shown in the next section.

Figure 1Overview of the 20 AERONET stations used in this study.

Table 3Overview of AERONET stations, selected observational periods for which version 3 level 2.0 data are available (Giles et al., 2019), total number of observations (inversion products), 532 nm AOT (mean and SD), dust-related inversion cases J_d (AE <0.3, AOT >0.1), and 532 nm AOT (mean and SD) for the dust observations only.

Download Print Version | Download XLSX

3.1 Correlations between n_250,d, s_d, s_100,d, and v_d with dust extinction coefficient σ_d

In order to illustrate the variability in the POLIPHON conversion factors (summarized in Sect. 3.3 and Table 4), we start with basic correlations between the dust microphysical properties and the dust extinction coefficient. Figure 2 provides an overview of the relationship between the dust particle number concentration of larger particles n_250,d and the dust extinction coefficient σ_d in (a), dust particle surface area concentration s_d and σ_d in (b), and between the dust volume concentration v_d and σ_d in (c). A total of 12 different AERONET stations are considered in the figure. The mean conversion factors c_250,d (Eq. 9), c_s,d (Eq. 10), and c_v,d (Eq. 6) for the Saharan dust stations of Sal, Cabo Verde (a); Dakar, Senegal (b); and Tamanrasset, Algeria (c), are indicated as straight lines (regression lines). These stations are exclusively influenced by Saharan dust.

We simply set the layer depth D in Eq. (5) to 1000 m so that σ_d (in Mm⁻¹ in Fig. 2) divided by 1000 yields the basic AERONET 532 nm AOT value. We selected different colors to distinguish Saharan dust observations (green), Middle Eastern measurements (orange), and data collected in central and East Asia (red). We used bluish colors (blue, cyan) for the American and Australian stations (respectively) and blue-green for the African site (in the Southern Hemisphere) of Gobabeb.

As can be seen in Fig. 2, there are no large differences in the correlation features for the different AERONET stations. The given Saharan dust conversion factors in (a) for Cabo Verde (based on 2982 data points), in (b) for Dakar, Senegal (3823 data points), and in (c) for Tamanrasset, Algeria (3542 data points), characterize the main relationship between the shown microphysical and optical parameters for the different dust regions very well. However, some contrasting features are visible, especially when comparing the Saharan with the central and East Asian stations and thus for clearly separated dust regions. The Middle Eastern AERONET sites are influenced by both Saharan and Middle Eastern dust.

The spread in the data mainly reflects variations in the dust aerosol characteristics (size distribution, refractive index) as a function of varying mixtures of freshly emitted local dust and long-range-transported aged dust. Fresh and aged dust mixtures may have occurred in different dust layers above each other (as in the case study in Sect. 4). Uncertainties in the AERONET data inversion procedure applied to obtain the microphysical properties from the measured AOT and sky radiance observations may have also contributed to the scatter in the data. The scatter provides an impression of the variability in the relationship between dust microphysical and optical properties and thus indicates the uncertainty in the determined conversion factors. However, it should also be mentioned that dust extinction coefficients in lofted layers above the boundary layer (in the free troposphere) seldom exceed 200–300 Mm⁻¹. For σ_d<500 Mm⁻¹ the scatter in the data is comparably low in Fig. 2.

Figure 3 indicates that the relationship between the surface area concentration s_100,d, i.e., the CCN-related particle surface concentration, and the particle extinction coefficient σ_d at 532 nm is much more robust (less variable) than the one for s_d vs. σ_d in Fig. 2b. The reason for this less noisy relationship is probably that the AERONET inversion analysis (for coarse-mode-dominated particle ensembles) is not very accurate for the small-particle fraction (radius classes from 50 to 100 nm) and this inversion-related uncertainty is then reflected in the variability of the s_d values considering all particle classes. With increasing minimum particle radius in the surface area computation the variability in the relationship between respective surface area concentration and extinction coefficient decreases.

However, as will be shown in the next section, in contrast to the s_100,d vs. σ_d relationship, the correlation between n_100,d and σ_d is strongly variable. One of the reasons for this difference is that particles with large geometrical cross sections (coarse-mode particles) have a higher weight in the surface area computation (integral over all sizes classes) and thus control the s_100,d values. In the n_100,d calculation, on the other hand, the size classes with the highest particle number concentration (fine-mode classes) dominate the n_100,d values.

3.2 Relationship between n_100,d and dust extinction coefficient σ_d

A different method of data analysis is used for n_100,d. As suggested by Shinozuka et al. (2015) we correlated log(n_100,d) vs. log(σ_d). Figure 4 shows the relationship between particle number concentration n_100,d and the dust extinction coefficient σ_d at 532 nm for two stations (Mezaira'a, Dushanbe) on a logarithmic scale. As outlined in Sect. 2, the particle number concentration n_100,d, considering only the particles with dry radius >100 nm, represents the CCN reservoir in the case of dust particles for a typical water supersaturation of 0.2 % very well (Mamouri and Ansmann, 2016; Lv et al., 2018).

In Fig. 4, we highlight the difference in the correlation when using all available data (532 nm dust AOT from 0.1 to 3.0 or σ_d from 100 to 3000 Mm⁻¹) and when using only observations with AOT <0.6. By detailed inspection of all data sets (station by station), we observed that the correlation strength significantly decreases with increasing AOT and is no longer clearly visible for all measurements with AOT from 1.0 to 3.0. The Dushanbe data set shown in Fig. 4b is a good example for this observation.

We can only speculate about the reason for the weak relationship for AOT >0.6. When the AOT is too large, the coarse-mode dust fraction may control the measured optical properties and respective inversion results so much that a trustworthy retrieval of the particle fraction with radii from 100 to 200 nm is no longer possible. Another explanation is related to the observational procedure. Most inversion computations are based on AERONET observations in the early morning and evening hours when the effective impact of aerosols is strongest (so that the effective dust AOT is even higher by a factor of 2 and more compared to that for the vertical column stored in the AERONET database). At these low-visibility conditions, the short-wavelength AERONET channels (340 and 380 nm) may have problems correctly measuring the overall AOT (Rayleigh AOT plus particle AOT). The short-wavelength AOT values are, however, especially important in the inversion retrieval of small dust particles and thus have a strong influence on the n_100,d retrieval results.

As a consequence of the low correlation between log(n_100,d) and log(σ_d) for large AOT, we restricted the determination of the conversion parameters c_100,d and x_d (see Eq. 12) by means of a regression analysis to AOT values from 0.1 to 0.6 (or respective σ_d from 100 to 600 Mm⁻¹).

Figure 5 provides further insight into the correlation between log(n_100,d) and log(σ_d). Observations for different stations influenced by Saharan, Middle Eastern, central Asian, American, and Australian dust are shown. The Saharan dust data set collected at Cabo Verde belongs to the few data sets (out of the 20 AERONET stations) with a likewise good correlation between log(n_100,d) and log(σ_d) even for large extinction values >600 Mm⁻¹ and corresponding AOT values >0.6. In Fig. 5, regression analysis results (in accordance with Eq. 12) for Mezaira'a (numbers in orange) and Cabo Verde (numbers in green) are compared. Furthermore, the relationship between n_100,d and σ_d as found by Shinozuka et al. (2015) for dusty field sites is presented.

Table 4Dust conversion parameters required in the conversion of particle extinction coefficients σ_d at 532 nm into particle number, surface area, and volume concentration (index d for total dust, index df for fine dust, index dc for coarse dust) as described in Table 1. The mean values and SD for c_v,d, c_v,df, and c_v,dc (10⁻¹² Mm), of c_250,d (Mm cm⁻³), and c_s,d and $c_{\mathrm{s},\mathrm{100},\mathrm{d}}$ (10⁻¹² Mm m² cm⁻³) are derived from the extended AERONET data analysis described in Sects. 2 and 3.1 for all sites listed in Table 3. c_100,d (cm⁻³ for σ_d=1 Mm⁻¹), and x_d and respective standard deviations (SD) are obtained in the way described in Sect. 3.2 by considering only AOT from 0.1 to 0.6, except for Ilorin (all AOTs are used because only 12 % of AOT <0.6). No data (c_100,d, x_d) are listed when the regression coefficient <0.6. The regional–continental mean values (for North Africa, Middle East, Asia, America–Australia) are obtained by observation-weighted (J_d-weighted) averaging of the given station mean and SD values.

Download Print Version | Download XLSX

As mentioned above, most of the dust-related lidar observations in the free troposphere show dust extinction coefficients (σ_d)<200–300 Mm⁻¹. For a moderate dust extinction value of 100 Mm⁻¹, the POLIPHON retrieval yields $n_{\mathrm{100},\mathrm{d}}\approx \mathrm{150}$ and 250 cm⁻³ when using c_100,d and x_d numbers as derived from the Cabo Verde and Mezaira'a AERONET observations (AOT <0.6), respectively. Thus, a maximum overall error of a factor of 2 in Table 1 (for n_100,d and n_CCN,d) also concluded by Shinozuka et al. (2015) and corroborated by Mamouri and Ansmann (2016) is justified.

3.3 Overview of AERONET-derived conversion parameters

In Table 4, the AERONET-based conversion parameters for all stations are presented. Regional mean sets of conversion parameters are given as well. Figures 6 and 7 provide a station-by-station overview of the conversion parameters (mean and SD values). Systematic differences from region to region are visible in the case of c_250,d and also weakly for c_v,d. The conversion parameters for the Americas, Australia, and southern Africa need to be handled with caution because the number of available observations is relatively low and the mean 532 nm AOT of these observations was low as well with values from 0.15 to 0.25. A decrease in c_250,d and a slight increase in c_v,d (and c_v,dc) from African to East Asian AERONET stations suggests that, for the same measured extinction coefficient (σ_d), the accumulation mode particle number concentration (in our case particles with radius from 250 to 500 nm) is slightly larger and the coarse-mode dust particle number concentration, dominating the dust volume concentration, is lower in the case of Saharan dust compared to East Asian dust. This behavior may indicate that the African AERONET stations, e.g., in Cabo Verde, Izana, and Dakar predominantly observe dust after long-range transport (which leads to a bit enhanced fine-dust fraction because of size-dependent sedimentation and removal of particles), whereas the East Asian AERONET stations may be influenced more frequently by the occurrence of local, freshly emitted dust with the relatively strong contribution of coarse-mode particles. Similar conditions as suggested for central and East Asia may hold for the American and Australian stations.

Figure 6Overview of POLIPHON conversion factors (a) c_250,d, (b) c_s,d, and (c) c_v,d (mean and SD) derived from AERONET dust data sets collected at 20 stations around the world. The stations (and abbreviations) are given in Table 3. Total numbers of observations (considered in the statistical analyses for each stations) are given above the figure frame (a) followed by two lines with respective mean 532 nm dust AOTs for all data sets (considering only the dust cases with AOT $>\mathrm{0}.$1). In (b), open circles show values of the surface-related conversion factor $c_{\mathrm{s},\mathrm{100},\mathrm{d}}$ considering particles with a radius >100 nm, only. In (c), volume-related conversion factors are separately determined for total (index d), fine (index df, open symbols), and coarse dust (index dc, open symbols). The uncertainty bars for c_v,dc are not shown, but are similar to the ones for c_v,d. All statistical results are also summarized in Table 4.

Download

Figure 7POLIPHON conversion parameters (a) c_100,d and (b) x_d derived from AERONET dust observations at 15 stations in northern Africa (green), the Middle East (orange), central and East Asia (red), North America (blue), and Australia (light blue).

Download

The smooth but steady changes in c_250,d and c_v,d from the Saharan, over the Middle East to the central and eastern Asian AERONET stations indicate that the Middle Eastern stations are influenced by both local western Asian dust sources (mostly Arabian dust) and Saharan dust (advected with the prevailing westerly winds). Only the African stations and the East Asian stations (Lanzhou, Dalanzadgad) are clearly separated and allow us to contrast dust properties of African and Asian deserts. We did not make an attempt to separate Saharan from Arabian dust observations in the case of the Middle Eastern data sets by using backward trajectory analysis because of the relatively small differences in the Saharan and Middle Eastern conversion factors and the likewise large variability bars.

As discussed in Sect. 3.1 and corroborated by Fig. 3 the conversion factor $c_{\mathrm{s},\mathrm{100},\mathrm{d}}$ is almost constant and mostly in the range from 1.55 to $\mathrm{1.60}\times {\mathrm{10}}^{-\mathrm{12}}$ Mm m² cm⁻³ for all stations (see also Table 4). In contrast, c_s,d with values from 2.24 to $\mathrm{3.11}\times {\mathrm{10}}^{-\mathrm{12}}$ Mm m² cm⁻³ in Table 4 is significantly more variable. All in all, the observed regional differences in the dust conversion parameters in Fig. 6 are of the order of ±15 %–20 % for most of the parameters and stations.

Figure 7 provides a summarizing overview of the final results for c_100,d and x_d. Because of the large scatter in the log–log data fields expressed in the large uncertainty bars in the figure we can only give recommendations regarding the selection of the most reasonable set of dust conversion parameters. For the extinction exponent x_d a value of 0.80 seems to be appropriate. This exponent is then linked to c_100,d values of 5–6 cm⁻³ (at σ_d=1 Mm⁻¹).

Figure 8Dust layering over the central Asian AERONET site of Dushanbe, Tajikistan, on 13 April 2015 observed with Polly lidar at 1064 nm (range-corrected signal). The densest layer from 2 to 5 km height a.g.l. (above ground level) contained dust particles from Iran, Afghanistan, and Oman according to the backward trajectories in Fig. 9. With increasing height, dust was advected from the Arabian peninsula and the Sahara. The polluted boundary layer reached up to about 2 km height and contained traces of local dust and dust from Kazakhstan. Above 6.5 km height (and temperatures $<-\mathrm{20}$ ^∘C) ice clouds developed triggered by dust particles, which are favorable ice-nucleating particles. POLIPHON results in Figs. 10–12 are derived for the height–time range indicated by the white rectangle.

Download

Figure 9The 6 d backward trajectories computed with the HYSPLIT (Hybrid Single Particle Lagrangian Integrated Trajectory) model (HYSPLIT, 2019; Stein et al., 2015; Rolph et al., 2017) for Dushanbe, Tajikistan, on 13 April 2015, 16:00 UTC. The computation is based on GDAS0.5 meteorological fields (GDAS, 2019). Arrival heights are at 1000 m (red, in the boundary layer with central Asian dust), 3500 m (blue, in a dense layer with dust from several western Asian deserts), and 6000 m (green, in dusty air from the Arabian peninsula and the Sahara).