We developed a novel measurement channel that utilizes Raman scattering from silicon dioxide (SiO2) quartz at an ultraviolet wavelength (361 nm). The excitation of the Raman signals is done at the primary wavelength of 355 nm emitted from a lidar instrument. In combination with Raman signals from scattering from nitrogen molecules, we may infer the mineral-quartz-related backscatter coefficient. This technique thus allows us to identify in a comparably direct way the mineral quartz content in mixed pollution plumes that consist, e.g., of a mix of desert dust and urban pollution. We tested the channel for the complex situation of East Asian pollution. We find good agreement of the inferred mineral-quartz-related backscatter coefficient to results obtained with another mineral quartz channel which was operated at 546 nm (primary emission wavelength at 532 nm), the functionality of which has already been shown for a lidar system in Tsukuba (Japan). The advantage of the novel channel is that it provides a better signal-to-noise ratio because of the shorter measurement wavelength.
© 2011 OSA
Aerosols play a significant role in the radiative and climate balance of the Earth's atmosphere, and have significant impact on environment and ecology. There exist already a great number of studies that deal with the possible effect of aerosols (direct and indirect effects) on radiative forcing. However, the impact of tropospheric aerosols on the radiation budget still is not well understood, because of the large temporal, horizontal and vertical variability of aerosol number concentration and highly diverse aerosol characteristics such as particle shape, size distribution and refractive index.
Particularly the radiative properties of mineral dust are only poorly understood. For example the latest report of the Intergovernmental Panel for Climate Research  provides only a rough estimate for the magnitude of the direct mineral dust forcing of about −0.3 W m2 to + 0.1 W m2. This low understanding is particularly critical as mineral dust presents the largest natural source for particles in the atmosphere. The current estimate of the global source strength is in the range of several ten million tons – 1000-3000 Tg per year . Most of this particle flux is subject to long-range transport.
Dust may alter cloud optical properties by changing the number of cloud condensation nuclei and ice nucleus concentration [3–8]. Desert aerosols may mix with sulfur and nitrogen compounds, soot from combustion sources, and particulate matter in maritime environments. These mixtures may generate even more complex aerosol properties than pure mineral dust. The efficiency of dust particles to form liquid water drops and/or ice particles may alter during transport due to mixing and coating with soluble aerosol species [9–12]. A change of water drop and ice particle properties may influence the brightness of clouds and the formation of rain, and therefore the lifetime of clouds.
A large number of physical, chemical and optical measurement techniques are used for studying mineral dust by in situ sampling, and passive and active remote sensing techniques. In recent years, lidar techniques have evolved into effective tools for investigating properties of tropospheric dust. Lidar networks like the European Aerosol Research Lidar Network (EARLINET)  and the Asian Dust Network (AD-Net)  in recent year collected valuable height-resolved optical information on atmospheric mineral dust [15–17]. For these obervations these networks employ various instruments like elastic-backscatter lidars, polarization lidars, multi-wavelength lidars, high-spectral-resolution lidars (HSRLs), Raman lidars, and combinations of these lidar techniques.
Apart from such network activities intensive field activities are another cornerstone necessary to enhance our understanding of mineral dust. In recent years many field campaigns were carried out with this goal like the Puerto Rico Dust Experiment (PRIDE) . the Saharan Dust Experiment (SHADE) , the United Arab Emirates Unifies Aerosol Experiment (UAE2) , the Dust and Biomass-burning Experiment (DABEX) , the Dust Outflow and Deposition to the Ocean (DODO) , field studies in the frame of the international multi-year African Monsoon Multidisciplinary Analysis (AMMA) program . Most recently the Saharan Mineral Dust Experiment (SAMUM)  focused on a comprehensive characterization of pure mineral dust properties on the basis of a sophisticated interplay between airborne and ground-based lidar/in situ instrumentation, augmented by ground-based sun photometers.
Despite the success of these dust field investigations, large deficits remain in our understanding of dust radiation interactions. One of the reasons lies in the fact that pure mineral dust is only rarely observed. In general, the pollution plumes consist of a mix of different aerosol types which makes an assessment of the contribution of the various components to radiative forcing considerably complicated. The modeling of transport of aerosol pollution also suffers from the fact that the various aerosol components, and thus the impact of source regions on, e.g., free-tropospheric pollution  cannot be fully assessed. For that reason, we need novel techniques that allow us to separate, e.g., mineral dust from other pollution types, if dust is immersed in such plumes.
In recent years a novel concept has emerged in lidar technology that will allow us for separating different aerosol components . In this contribution we focus on identifying mineral dust immersed in urban/industrial haze plumes. In a previous study we proposed a method to estimate the concentration of mineral dust by using lidar return signals from Raman scattering of quartz (silicon dioxide, silica), which is the major constituent of mineral dust . The method combines a Raman lidar that detects radiation from the quartz line at 466 cm−1 (detection wavelength at 546 nm for a primary emission wavelength at 532 nm) with a high-spectral-resolution lidar (HSRL). In this way, we can directly measure the quartz-related backscatter coefficient, from which we may estimate the quartz concentration  in atmospheric aerosols. In this paper, we modify and develop the method for the case of a two wavelength quartz Raman and multiwavelength nitrogen Raman lidar systems .
For first time we a used UV spectral range for the detection of mineral quartz, namely the laser wavelength λL = 355 nm. The excitation of the silicon dioxide molecules by this wavelength yields a scattered radiation with a wavelength λR = 361 nm for the Raman line at 466 cm−1. The main advantage of detecting in the UV wavelength range is the scattering cross section of mineral quartz which is considerably higher compared to the visible wavelength range. Another advantage is that the UV spectral region is eye safe. The use of two quartz channels (361 nm and 546 nm) allow us to compare at the one hand the results at two different wavelength regions, and, on the other hand, to observe experimentally the Ångström exponent of quartz Raman scattering. In addition, the combination of quartz Raman observations with the multiwavelength Raman lidar technique gives us a unique chance for characterizing a complex situation of mixed mineral dust/urban haze pollution plumes. High-performance multiwavelength aerosol Raman/depolarization lidars provide elastic signals at 355, 532, 1064 nm, nitrogen signals at 355 nm and 532 nm, water vapor signals at 407 nm, and particle depolarization signals at 532 nm. An example of this kind of characterization is presented by Müller at al .
The basic idea of the methodology is using quartz as a tracer for detecting mineral dust. For this purpose we measure signals from Raman scattering from quartz grains . We choose quartz for two main reasons: First, the concentration of quartz is relatively high in mineral aerosols. Approximately 46-77% of the particles generated in Asian deserts are rich in silicon (Si). These particles are primarily composed of quartz and aluminosilicate. The second reason is that the quartz Raman line at 466 cm−1 is suitable for lidar measurements. Commercially available filters can easily separate this line from the Mie scattering signal, the Doppler-broadened molecular signal, and Raman lines of other atmospheric components, such as N2 at 2330 cm−1, O2 at 1556 cm−1, O3 at 1103 cm−1, CO2 at 1388 cm−1, and water vapor at 3652 cm−1.
In section 2, we describe the methodology, the lidar system used for our study, and the error sources. In section 3 we present and discuss a measurement example. A summary and conclusions are presented in section 4.
2. Methods and apparatus
2.1 Raman backscatter coefficient of quartz
The quartz Raman backscatter signal is described by the so-called Raman lidar equation and can be written as :
The Raman backscatter coefficient βR(r,λL,λR) of the scatterers can be determined from the received Raman scattering signals, if we know the system parameters (B and F(r)) and the optical depth of the atmosphere. The geometrical form factor and the system dectection efficiency of the system can be calculated or determined by calibration measurements. Optical depth at the emission wavelength and the detection wavelength can independently measured, or inferred on the basis of additional assumptions. Aerosol extinction and backscatter coefficients at the laser wavelength are obtained by simultaneous measurements of Raman scattering from nitrogen molecules.
To solve the quartz Raman scattering equation (Eq. (1)), we must know the profiles of the extinction αm(r,λL) and backscatter βm(r,λL) coefficients of the air molecules as well as the profiles of the particle extinction coefficient at the Raman wavelength. Profiles of scattering by the atmospheric molecules can be obtained from model data or from temperature and pressure profiles derived from standard meteorological measurements. The aerosol extinction profile at the quartz Raman wavelength can be obtained from the aerosol extinction profile at the laser wavelength by assuming a power-law wavelength dependence .
In this manner we obtain the Raman backscatter coefficient for quartz from the quartz Raman equation (Eq. (1)) as follows:Eq. (2)) is obtained on the basis of the following three steps. 1) We use factory data of the filters and beam-splitters. 2) We carried out calibration measurements without any filters. 3) We interchanged the two Raman channels.
On the basis of Eq. (2) we can give a definition of the Raman quartz-backscatter-related Angström exponent (åQ) that can be computed from the Raman backscatter coefficients βR(r,λL,λR1) and βR(r,λL,λR2) which are measured at two different wavelengths,λR1 and λR2:31], this cross-section in a first approximation is proportional to (v-vs)4, where vs is the Raman shift. The value of åQ can be affected by additional factors, as for example resonance absorption. One of our future aims is a detailed experimental study of the Raman quartz-backscatter-related Ångström exponent for the case of mineral dust
2.2 Mass concentration of mineral quartz
The mass concentration of silica in the atmosphere can be estimated from the Raman backscatter coefficient of quartz. We use the relationship that connects the Raman backscatter coefficient with the Raman backscatter differential cross section dσ(λL,λR,π)/dΩ and the number density of quartz molecules Nq . This relation is defined as:31] by assuming a power-law 4 wavelength dependence of the Raman-scattering cross section.
2.3. Lidar instrument
The GIST multiwavelength Raman lidar [32,28] was used for measurements of quartz concentrations of mineral dust . For the Raman quartz measurements we modified the instrument set-up. We equipped the system with two additional receiver channels which operate at 361 nm and 546 nm.
Figure 1 shows the design of the GIST lidar the way it was used during our Raman quartz measurements.The transmitter is a Nd:YAG laser (Surelite III-10). The emitted energy is 640 mJ at 1064 nm, 154 mJ at 532 nm and 140 mJ at 355 nm. The pulse repetition rate is 10 Hz. We use a frequency doubler and a frequency tripler for generating light at 532 and 355 nm. All three laser beams are aligned on the same optical axis. We expand the beam 5 fold from 9 mm to 45 mm before it is emitted into the atmosphere. The laser beam divergence is less than 0.2 mrad. We use a co-axial configuration for the infrared and visible spectral range (1064, 532, 546, and 607 nm) channels and a bi-axial configuration for the UV spectral range (355, 361, 387, and 407 nm) channels.
The return signals are collected with a Schmidt-Cassegrain telescope (CELESTRON). The focal length is 3910 mm. The beam is separated according to wavelength with beam splitters and transmitted to photomultipliers. We detect the signals elastically backscattered at 355, 532 and 1064 nm. In addition we collect signals from Raman scattering from nitrogen molecules (387 and 607 nm), from water vapor (407 nm), and from quartz (361 and 546 nm). The signals at 532 nm are measured at the parallel and perpendicular-polarized component. All components of the signal detection unit are placed on one optical table.
We chose the third and the second harmonic of the primary emission wavelength of the Nd:YAG laser (λL = 355 nm and λL = 532 nm) as pumping wavelengths for the quartz Raman measurements. The excitation of the silicon dioxide molecules by these wavelengths results in a scattered radiation with a wavelength of λR = 360.77 nm and λR = 545.8 nm for the Raman line at 466 cm−1. We use three custom designed interference filters (BARR) to detect the Raman-shifted return signals at 360.77. Two of the filters are identical, they are characterized by a transmission of more than 92% at λR = 360.77 nm. The optical density for each filter is more than 4.1 at the pumping wavelength λL = 355 nm as well as for wavelengths longer than 367 nm. The third interference filter is characterized by a transmission of 53% at the central wavelength 359.61 nm and a bandwidth of 2.6 nm (full width at half maximum (FWHM)). The optical density of the third filter is more than 4 for wavelengths <355 nm as well as >366 nm. The combination of those filters secures that the elastic return signals at 355 nm are suppressed by approximately 13 orders of magnitude compared to the transmission for the 360.77 nm quartz Raman signals.
Two commercially available filters were used to detect the Raman-shifted return signal at 545.8 nm - a “Raman edge filter” (CVI, REF-532.0-A) which is characterized by an optical density of about 10 for the laser line (λR = 532.24 nm) and an interference filter (CVI, F03-546.1-4-2.00) with a center wavelength of 546.1 nm and a bandwidth of 3 nm (full width at half maximum (FWHM)). The blocking optical density at the laser wavelength is close to 5. Using these filters the transmission at 532 nm is about 15 orders of magnitude smaller compared to the transmission for the quartz Raman signals.
We also carried out tests regarding the optical rejection. We used an additional interference filter with a center wavelength of 532 nm. We did not observe elastic scattering signals in the Raman channels above the noise level. Our tests also included scattering from very dense water clouds. The ratios of the system parameters were obtained using filter factory data and calibration measurements.
From the detected signals we may infer the following optical aerosol parameters: particle volume backscatter coefficients at 355, 532, and 1064 nm, particle volume extinction coefficients at 355 and 532 nm, Raman quartz backscatter coefficients at 361 and 546 nm, linear particle polarization ratios at 532 nm, particle extinction-to-backscatter (lidar) ratios at 355 and 532 nm, extinction-related Ångström exponents for the wavelength pair at 355/532-nm, backscatter-related Ångström exponents for the wavelength pairs at 355/532 nm and 532/1064 nm, and Raman-quartz backscatter-related Ångström exponents for the wavelength pair at 361/546 nm. Data inversion algorithms [33–35] can be applied to infer the following microphysical particle parameters : real and imaginary part of the complex refractive index, number, volume and surface-area concentration, particle volume size distribution, and particle effective radius. From these microphysical parameters we may infer single-scattering albedo. The profiles of the Raman quartz backscatter coefficients allow us to derive the mineral quartz concentration.
A detailed description of all parameters listed above will be given in a future contribution in which we present the final design stage of our multiwavelength Raman lidar. This instrument stage will also include particle depolarization ratio measurements at multiple wavelengths. In the context of this paper we focus on the parameters that deal with the Raman quartz signals.
2.4. Error sources
In the following we discuss the specific error sources our quartz Raman measurements are affected with. A detailed description of the error analysis of particle optical properties measured by nitrogen Raman and multiwavelength Raman lidars can be found in Ref [30,37].
Random and systematic components contribute to the error of lidar measurements of the quartz mass concentration. The random component of the error may be calculated from standard error propagation formulas . It is in general difficult to evaluate the systematic part of the error because the knowledge of the absolute value of the quantity being measured is required. The specific errors in the quartz mass concentration measurements can be divided into the following main categories:
- 1. Errors due to photon counting statistics and background subtraction.
- 2. Errors due to system restrictions and system calibration.
- 3. Errors due to the absolute value of the Raman backscatter differential cross section of quartz.
- 4. Errors in the molecular/nitrogen channel.
- 5. Errors due to other scattering processes.
The errors due to system restrictions and system calibration are caused by a non-ideal signal rejection by the interference filters, the system detection efficiency, and system misalignments. We use a combination of interference filters that provide a high signal rejection of elastic aerosol scattering, which makes the separation between quartz Raman scattering and all other scattering signals relative easy. However, this rejection is strongly dependent on the mechanical alignment of the filters and the angular dependence of the incoming light. Therefore, a slight misalignment of the filters could cause significant error in the detection of the quartz Raman signals, as for example filter leaking. The system detection efficiency is a combination of the system transmission efficiency and the photomultiplier quantum efficiency. The system detection efficiency affects the amount of the detected signals and therefore, it is directly related to the photon counting statistics.
As we comment above (section 2.2) the absolute value of the Raman differential cross section of natural quartz in the backscatter direction is significantly dependent on the structure of the crystal, the crystallization process, and the conditions of admixtures. The value of the Raman scattering cross section of quartz also depends on the temperature. All these factors can cause errors in the quartz mass concentration measurements.
The errors that originate from the molecular/nitrogen channels are mainly caused by errors of the overlap function and an inaccurate determination of the aerosol Ångström exponent.
Additional errors in quartz mass concentration measurements can be caused by the presence of scattering processes such as fluorescence, multiple scattering in clouds, and pure Rotational Raman scattering from air molecules. We assume that the influence of these scattering processes is of secondary order compared to the previously mentioned error sources.
3. Example of measurements
In this section, we present the example of a measurement that illustrates the capability of the methodology and the operation of the new UV quartz Raman channel in combination with a state-of-the-art multiwavelength Raman lidar. Examples that focus on complex aerosol situations for the case of mixed mineral dust/urban haze pollution plumes are given by Müller et al. .
The observations were done during the night of March 21, 2010. The system is located at the Gwangju Institute of Science and Technology (GIST; 35.23° N, 126.84° E), Republic of Korea (South Korea). The optical profiles from the molecules were obtained on the basis of routine radiosonde observations carried out at the Gwangju airport (35.12° N, 126.82° E) which is located approximately 13 km away from GIST. The measurement situation was characterized by a large mineral dust plume that had been transported over the lidar site on this day. Independent satellite pictures, PM10 measurements and back-trajectory analysis confirm the presence of the dust plume over the lidar site on that day.
Figure 1 shows the evolution of the dust layers on this day in terms of time-height cross sections of the attenuated backscatter (range-corrected lidar signal) and the linear volume (particle plus molecular) depolarization ratio (VDR) at the wavelength 532 nm. Several aerosol layers with relatively high particle number concentrations and high VDR (VDR>6%) were detected up to altitudes of around 6 km height. The plots show that cirrus clouds were episodically present. The cloud base varied between 5 km to 11 km. According to the lidar signals and depolarization ratios, we can conclude that in the altitude range of the cirrus clouds a mixture of clouds and mineral dust was present. For example, the comparably low lidar return signal and a depolarization ratio of about 12% in the 6-8-km altitude range (observed from15:40 to 16:00 UTC) are typical for scattering from mineral dust. In addition, we detected an optically dense water cloud, characterized by a VDR smaller than 5%, in about 3km altitude from 18:40 to 19:20 UTC.
The vertical profiles in Fig. 3 present the details of the quartz Raman lidar signals at 361 nm and 546 nm, respectively, as well as the elastically backscattered signals at 532 nm and the linear volume depolarization ratio for the period from 15:00 to 18:00 UTC.
In the profile of the range-corrected elastically backscattered lidar signals at 532 nm (Fig. 3b), one can see the presence of two significant aerosol layers (high aerosol concentration) – one layer extends from the ground up to around 2 km altitude and the second layer extends from 2 km to 5 km above ground. We find increased levels of lidar signals between 5 km and 10 km, which are caused by cirrus clouds probably mixed with dust.
The Raman quartz lidar signals (Fig. 3a) clearly indicate the presence of quartz for altitudes up to 7 km according to the 361 nm channel and up to 5 km according to the 546 nm channel. The difference in maximum height between the two channels is caused by the lower signal-to-noise ratio in the visible channel. Values larger than 0 in this channel indicate the presence of mineral quartz. Both quartz-Raman lidar profiles have a maximum value at the altitude 3.5 km. In this height we also observe a maximum value of the elastic backscatter profile (532 nm). This fact indicates that the quartz most probably contributed significantly to the total aerosol in that aerosol layer.
The vertical profile of the linear volume depolarization ratio (Fig. 3c) has relatively high values near the ground (10% up to 1 km). Values of up to 15% are found between 1.75 km and 5 km, as well as from 6 km to 10.5 km. These comparably high values correspond to scattering from particles with non-spherical shape. The higher values of the depolarization ratio are observed in the same altitudes where quartz Raman scattering is detected. However, the profile of the volume depolarization ratio and the quartz Raman signals show different behavior in the dust layers between 2 km and 5 km. The maximum of the VDR is detected at around 4 km height. In contrast the maximum value of the quartz Raman profiles is at 3.5 km. The reason for this difference most probably is the fact that the depolarization ratio describes non-sphericity, whereas Raman profiles are proportional to the concentration of the mineral quartz.
We also see a difference in the behavior of the 361 nm profile compared to the elastic signal at 532 nm in the altitude range from 5 km to 7 km. The elastic backscatter signal has a maximum value at around 5.5 – 6 km height. This maximum corresponds to scattering from clouds. In the same altitude (5.5 – 6 km) the 351 nm profile shows a minimum. This profile has a maximum at 6.5 km height. This fact indicates the presence of a mixture of dust with clouds.
Figure 4 shows the profiles of the quartz backscatter coefficient at 361 nm and 546 nm, the Raman-backscatter-related Angstrom exponent and profiles of the mineral quartz concentration. We obtained these profiles from lidar signals applying the retrieval methodology described in Section 2. The Raman backscatter coefficient for quartz varies from 2x10−12 m−1sr−1 to 5x10−12 m−1sr−1 at 361 nm. The peak value is 5x10−12 m−1sr−1 at the altitude 3.4 km. The values of the same coefficient measured at 546 nm are less than 1.1x10−12 m−1sr−1. The peak value also is detected at 3.4 km height.
The Raman-backscatter-related Ångström exponent follows from applying Eq. (3). We find a mean value of 3.78 ± 0.48 for the altitude range from 2 - 4.3 km. This mean value is slightly lower than the expected theoretical value of 4. We point out that Müller et al.  report on several cases of mineral quartz measurements carried out with this new experimental set-up. That study focused on extracting a first overview of mineral quartz concentrations in dust observed over Korea. The authors also report on Raman-backscatter-related Ångström exponents and again find values less than the theoretical value 4. It is not clear yet if these lower values are caused by specific properties of the silicates (e.g. internal structure), or if our measurements of the quartz backscatter coefficients are affected by a systematic error source. Further tests of the experimental setup will be carried out during the coming dust seasons in fall 2010 and spring 2011 to clarify this question.
The profiles of the mineral dust concentration were obtained from the profiles of the quartz backscatter coefficients using Eq. (2) and Eq. (4). The mineral quartz concentration is the same for both profiles within the error bounds. The peak value of the concentration of mineral quartz is 7.14 ± 0.92 μg/m3 (wavelength 355nm) at 3360 m height. The mean values of the mineral quartz concentration were found to be 4.15 ± 1.21 μg/m3 at 361nm and 4.43 ± 1.50 μg/m3 at 546 nm in the altitude range from 2040 m to 4320 m.
We presented the technical details of a novel instrument channel that can be used for detecting quartz in mineral dust plumes. The channel detects Raman return signals from Silicon which is one of the main components of atmospheric mineral dust. The channel operates at an ultraviolet wavelength (Raman signals around 361 nm) and uses 355 nm of a frequency-tripled Nd:YAG laser as excitation wavelength. Advantage of such a channel is the higher signal intensity of the ultraviolet Raman return signals compared to return signals at 546 nm which may be detected if an excitation wavelength of 532 nm (frequency doubled Nd:YAG laser) is used. We use an optical setup in which a high suppression (13 orders of magnitude) of the elastic return signals at 355 nm is achieved. In this way we make sure that the Raman-quartz channel (at 361 nm) detects only the Raman return signals from mineral quartz.
We tested the novel channel during several episodes of transport of Asian dust from China across South Korea. Such Asian dust plumes in most cases do not occur as pure mineral dust plumes. Because of the high number and the emission intensity of anthropogenic pollution sources in East Asia (traffic, industry, biomass burning) mineral dust plume occur as mixture with urban/industrial haze. For that reason it is highly important to separate the mineral dust content from other aerosol pollution. Only in this way we may be able to make a correct assessment of the impact of mineral dust on climate.
We presented one case study for such an outbreak of a mixed Asian dust/urban pollution plume. This case was characterized by dust up to 10 km height above ground. Cirrus clouds were periodically embedded in this dust layer. We could detect Raman-quartz signals up to 7 km, before signals intensity dropped to background levels. For comparison we also operated a well-established Raman-quartz channel at 546 nm. This channel also indicated the presence of mineral quartz. However, due to the weaker Raman-quartz signals at visible wavelength, signals could only be detected up to 5 km height. We infer the mineral quartz mass concentration on the basis of a well-established data analysis technique. We obtain approximately 7 µg/m3 in the center of the dust plume at 3.5 km height.
This work was funded by the Korea Meteorological Administration Research and Development Program under grant CATER 2009-3112.
References and links
1. P. Forster, V. Ramaswamy, P. Artaxo, T. Berntsen, R. Betts, J. Haywood, P. Lean, D. C. Lowe, G. Myhre, J. Nganga, R. Prinn, G. Raga, M. Schulz, and R. Van Dorland, “Lidar and atmospheric aerosol particles”, in Climate Change2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the International Panel on Climate Change, edited by S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyth, M. Tigner and H. L. Miller (Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA., 996pp). [PubMed]
2. M. O. Andreae, “Climatic effects of changing atmospheric aerosol levels”, in World Survey of Climatology Vol. 16, edited by A. Henderson-Sellers (Future Climates of the World, p. 347–398. Elsevier, New York, 1995).
3. H. R. Pruppacher and J. D. Klett, Microphysics of clouds and precipitation (C. Kluwer Academic Publishers, Dordrecht/Boston/London, 954pp., 1997).
4. P. J. DeMott, K. Sassen, M. R. Poellet, D. Baumgardner, D. C. Rogers, S. D. Brooks, A. J. Prenni, and S. M. Kreidenweis, “„African dust aerosols as atmospheric ice nuclei,” Geophys. Res. Lett. 30(14), 1732 (2003). [CrossRef]
5. P. R. Field, O. Möhler, P. Connolly, M. Krämer, R. Cotton, A. J. Heymsfield, H. Saathoff, and M. Schnaiter, “Some ice nucleation characteristics of Asian and Saharan desert dust,” Atmos. Chem. Phys. 6(10), 2991–3006 (2006). [CrossRef]
6. A. Ansmann, M. Tesche, D. Althausen, D. Müller, P. Seifert, V. Freudenthaler, B. Heese, M. Wiegner, G. Pisani, P. Knippertz, and O. Dubovik, “„Influence of Saharan dust on cloud glaciation in southern Morocco during the Saharan Mineral Dust Experiment,” J. Geophys. Res. 113(D4), D04210 (2008). [CrossRef]
7. A. Ansmann, M. Tesche, P. Seifert, D. Althausen, R. Engelmann, J. Fruntke, U. Wandinger, I. Mattis, and D. Müller, “Evolution of the ice phase in tropical altocumulus: SAMUM lidar observations over Cape Verde,” J. Geophys. Res. 114(D17), D17208 (2009). [CrossRef]
8. P. Seifert, A. Ansmann, I. Mattis, U. Wandinger, M. Tesche, R. Engelmann, D. Müller, C. Pérez, and K. Haustein, “Saharan dust and heterogeneous ice formation: Eleven years of cloud observations at a central European EARLINET site,” J. Geophys. Res. 115(D20), D20201 (2010). [CrossRef]
9. Z. Levin, E. Ganor, and V. Gladstein, “The effects of desert particles coated with sulfate on rain formation in the eastern Mediterranean,” J. Appl. Meteorol. 35(9), 1511–1523 (1996). [CrossRef]
10. S. Wurzler, T. G. Reisin, and Z. Levin, “Modification of mineral dust particles by cloud processing and subsequent effects on drop size distributions,” J. Geophys. Res. 105(D4), 4501–4512 (2000). [CrossRef]
11. O. Möhler, S. Benz, H. Saathoff, M. Schnaiter, R. Wagner, J. Schneider, S. Walter, V. Ebert, and S. Wagner, “The effect of organic coating on the heterogeneous ice nucleation efficiency of mineral dust aerosols,” Environ. Res. Lett. 3(2), 025007 (2008). [CrossRef]
12. D. J. Cziczo, K. D. Froyd, S. J. Gallavardin, O. Moehler, S. Benz, H. Saathoff, and D. M. Murphy, “Deactivation of ice nuclei due to atmospherically relevant surface coating,” Environ. Res. Lett. 4(4), 044013 (2009). [CrossRef]
13. J. Bösenberg and E. Volker, EARLINET: A European Aerosol Research Lidar Network to establish an aerosol climatology, Rep. 348, Max-Planck-Institute, Hamburg, Germany, 2003.
14. N. Sugimoto, I. Matsui, A. Shimizu, T. Nishizawa, Y. Hara, C. Xie, I. Uno, K. Yumimoto, Z. Wang, and S.-C. Yoon, “Lidar network observations of tropospheric aerosols,” Proc. SPIE 7153, 71530A, 71530A-13 (2008), doi:. [CrossRef]
15. A. Shimizu, N. Sugimoto, I. Matsui, K. Arao, I. Uno, T. Murayama, N. Kagawa, K. Aoki, A. Uchiyama, and A. Yamazaki, “Continuous observations of Asian dust and other aerosols by polarization lidar in China and Japan during ACE-Asia,” J. Geophys. Res. 109(D19), S17 (2004). [CrossRef]
16. T. Murayama, D. Müller, K. Wada, A. Shimizu, M. Sekiguchi, and T. Tsukamoto, “Characterization of Asian dust and Siberian smoke with multi-wavelength Raman lidar over Tokyo, Japan in spring 2003,” Geophys. Res. Lett. 31(23), L23103 (2004). [CrossRef]
17. L. Mona, A. Amodeo, M. Pandolfi, and G. Pappalardo, “Saharan dust intrusions in the Mediterranean area: Three years of Raman lidar measurements,” J. Geophys. Res. 111(D16), D16203 (2006). [CrossRef]
18. J. S. Reid and H. B. Maring, “Foreword to special section on the Puerto Rico Dust Experiment (PRIDE),” J. Geophys. Res. 108(D19), 8585 (2003), doi:. [CrossRef]
19. D. Tanré, J. Haywood, J. Pelon, J.-F. Léon, B. Chatenet, P. Formenti, P. Francis, P. Goloub, E. Highwood, and G. Myhre, “Measurements and modeling of the Saharan dust radiative impact: Overview of the Saharan Dust Experiment (SHADE),” J. Geophys. Res. 108(D18), 8574 (2003). [CrossRef]
20. J. S. Reid, S. J. Piketh, A. L. Walker, R. P. Burger, K. E. Ross, D. L. Westphal, R. T. Bruintjes, B. N. Holben, C. Hsu, T. L. Jensen, R. A. Kahn, A. P. Kuciauskas, A. Al Mandoos, A. Al Mangoosh, S. D. Miller, J. N. Porter, E. A. Reid, and S.-C. Tsay, “An overview of UAE flight operations: Observations of summertime atmospheric thermodynamics and aerosol profiles of the southern Arabian Gulf,” J. Geophys. Res. 113(D14), D14213 (2008). [CrossRef]
21. J. M. Haywood, J. Pelon, P. Formenti, N. Bharmal, M. Brooks, G. Capes, P. Chazette, C. Chou, S. Christopher, H. Coe, J. Cuesta, Y. Derimian, K. Desboeufs, G. Greed, M. Harrison, B. Heese, E. J. Highwood, B. Johnson, M. Mallet, B. Marticorena, J. Marsham, S. Milton, G. Myhre, S. R. Osborne, D. J. Parker, J.-L. Rajot, M. Schulz, A. Slingo, D. Tanré, and P. Tulet, “Overview of the dust and biomass-burning experiment and African Monsoon Multidisciplinary Analysis Special Observing Period-0,” J. Geophys. Res. 113, D00C17 (2008). [CrossRef]
22. C. L. McConnell, E. J. Highwood, H. Coe, P. Formenti, B. Anderson, S. Osborne, S. Nava, K. Desboeufs, G. Chen, and M. A. Harrison, “Seasonal variations of the physical and optical characteristics of Saharan dust: Results from the Dust Outflow and Deposition to the Ocean (DODO) experiment,” J. Geophys. Res. 113(D14), D14S05 (2008). [CrossRef]
23. J.-L. Redelsperger, C. D. Thorncroft, A. Diedhiou, T. Lebel, D. J. Parker, and J. Polcher, “„African Monsoon Multi-disciplinary Analysis: An international research project and field campaign,” Bull. Am. Meteorol. Soc. 87(12), 1739–1746 (2006). [CrossRef]
24. “Results of the Saharan Mineral Dust Experiment,” Tellus B Chem. Phys. Meterol. 61, 1–353 (2009).
25. I. Mattis, D. Müller, A. Ansmann, U. Wandinger, J. Preißler, P. Seifert, and M. Tesche, “„Ten years of multiwavelength Raman lidar observations of free-tropospheric aerosol layers over central Europe: Geometrical properties and annual cycle,” J. Geophys. Res. 113(D20), D20202 (2008). [CrossRef]
26. B. Tatarov, N. Sugimoto, I. Matsui, D.-H. Shin, and D. Mueller, “Multi-channel lidar spectrometer for atmospheric aerosol typing on the basis of chemical signature in Raman spectra”, 25th International Laser Radar Conference, 5–9 July 2010, St. Petersburg, Russia, (ISBN 978–5-94458–109–9), pp.47–50.
27. B. Tatarov and N. Sugimoto, “Estimation of quartz concentration in the tropospheric mineral aerosols using combined Raman and high-spectral-resolution lidars,” Opt. Lett. 30(24), 3407–3409 (2005). [CrossRef]
28. D. H. Shin, Y. M. Noh, B. Tatarov, S. K. Shin, Y. J. Kim, and D. Müller, “Multiwavelength Aerosol Raman Lidar for Optical and Microphysical Aerosol Typing over East Asia”, 25th International Laser Radar Conference, 5–9 July 2010, St. Petersburg, Russia, (ISBN 978–5-94458–109–9), pp.239–243.
29. D. Müller, I. Mattis, B. Tatarov, Y. M. Noh, D. H. Shin, S. K. Shin, K. H. Lee, Y. J. Kim, and N. Sugimoto, “Mineral Quartz concentration measurements of mixed mineral dust/urban haze pollution plumes over Korea with multiwavelength/aerosol/Raman-quartz lidar,” Geophys. Res. Lett. 37(20), L20810 (2010), doi:. [CrossRef]
31. P. E. Schoen, and H. Z. Cummins, Absolute cross sections for Raman and Brillouin light scattering in quartz, in Proceedings of Second International Conference on Light Scattering in Solids, edited by M. Balkanski (p. 460, Flammarion, Paris, France, 1971).
32. Y. M. Noh, D. Müller, D. H. Shin, H. Lee, J. S. Jung, K. H. Lee, M. Cribb, Z. Li, and Y. J. Kim, “Optical and microphysical properties of severe haze and smoke aerosol measured by integrated remote sensing techniques in Gwangju, Korea,” Atmos. Environ. 43(4), 879–888 (2009). [CrossRef]
33. D. Müller, U. Wandinger, and A. Ansmann, “Microphysical particle parameters from extinction and backscatter lidar data by inversion with regularization: theory,” Appl. Opt. 38(12), 2346–2357 (1999). [CrossRef]
34. I. Veselovskii, A. Kolgotin, V. Griaznov, D. Müller, U. Wandinger, and D. N. Whiteman, “Inversion with regularization for the retrieval of tropospheric aerosol parameters from multiwavelength lidar sounding,” Appl. Opt. 41(18), 3685–3699 (2002). [CrossRef] [PubMed]
35. C. Böckmann, I. Mironova, D. Müller, L. Schneidenbach, and R. Nessler, “Microphysical aerosol parameters from multiwavelength lidar,” J. Opt. Soc. Am. A 22(3), 518–528 (2005). [CrossRef]
36. I. Veselovskii, O. Dubovik, A. Kolgotin, T. Lapyonok, P. Di Girolamo, D. Summa, D. N. Whiteman, M. Mishchenko, and D. Tanré, “Application of randomly oriented spheroids for retrieval of dust particle parameters from multi-wavelength lidar measurements,”J. Geophys. Res. 115, 21203 (2010). [CrossRef]
38. P. R. Bevington and D. K. Robinson, Data Reduction and Error Analysis for the Physical Sciences, 2nd ed. McGraw-Hill (1992).