Volcanic aerosol layers observed with multiwavelength Raman lidar over central Europe in

Transcription

1 Click Here for Full Article JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 115,, doi: /2009jd013472, 2010 Volcanic aerosol layers observed with multiwavelength Raman lidar over central Europe in Ina Mattis, 1 Patric Siefert, 1 Detlef Müller, 1,2 Matthias Tesche, 1 Anja Hiebsch, 1 Thomas Kanitz, 1 Jörg Schmidt, 1 Fanny Finger, 1 Ulla Wandinger, 1 and Albert Ansmann 1 Received 2 November 2009; revised 9 March 2010; accepted 29 March 2010; published 22 July [1] In the framework of regular European Aerosol Research Lidar Network (EARLINET) observations, aerosol layers have been monitored with a multiwavelength aerosol Raman lidar in the upper troposphere and lower stratosphere over Leipzig (51.4 N, 12.4 E), Germany, since the summer of The origins of these layers are eruptions of different volcanoes on the Aleutian Islands, Kamchatka, Alaska, and on the Kuril Islands. FLEXPART transport simulations show that the volcanic aerosol is advected from Alaska to central Europe within about 7 days. The aerosol layers typically occurred in the upper troposphere above 5 km height and in the lower stratosphere below 25 km height. The optical depths of the volcanic aerosol layers are mostly between and at 532 nm. The wavelength dependence of the backscatter coefficients and extinction coefficients indicate Ångström exponents from Lidar ratios in the stratosphere are found in the range from sr (355 nm) and sr (532 nm). The estimation of the effective radius, surface area, and mass concentrations of a volcanic aerosol layer, observed well within the stratosphere at end of August 2009, reveals values of mm, 5 10 mm 2 cm 3, and mg m 3, respectively. Citation: Mattis, I., P. Siefert, D. Müller, M. Tesche, A. Hiebsch, T. Kanitz, J. Schmidt, F. Finger, U. Wandinger, and A. Ansmann (2010), Volcanic aerosol layers observed with multiwavelength Raman lidar over central Europe in , J. Geophys. Res., 115,, doi: /2009jd Introduction [2] Regular multiwavelength Raman lidar observations of the vertical aerosol distribution have been performed at Leipzig (51.4 N, 12.4 E), Germany, since Our measurements in the past 12 years do not show any major event of volcanic aerosol pollution in the upper tropospherelower stratosphere (UTLS) region [Mattis et al., 2008]. The situation changed due to a series of strong eruptions of volcanoes on the Aleutian Islands, Kamchatka, Alaska, and on the Kuril Islands since the summer of Table 1 lists the corresponding volcanoes, their location, and the estimated maximum height of the emitted gas or ash plumes. All volcanoes are located between 48 N and 61 N. The importance of volcanic aerosols arises from the fact that they can be distributed over the entire Northern Hemisphere within a few months and influence climate [Schneider et al., 2009] and tropospheric cloud formation [Sassen, 1992; Gassó, 2008] over months to years. [3] The occurrence of so many volcanic eruptions of different emission strengths combined with different injection heights provides the unique opportunity to study mixing 1 Leibniz Institute for Tropospheric Research, Leipzig, Germany. 2 Atmospheric Remote Sensing Laboratory, Gwangju Institute of Science and Technology, Gwangju, South Korea. Copyright 2010 by the American Geophysical Union /10/2009JD and vertical layering of aged and young, freshly produced volcanic aerosol particles. Lidar can support atmospheric modeling efforts by monitoring the UTLS region continuously over years with high vertical resolution, and by providing data on aerosol optical depth, volume extinction coefficients, and microphysical properties of the volcanic aerosol particles. [4] The aerosol layers from the recent volcanic outbreaks could immediately be observed (within less than two weeks after the eruption) with lidars at latitudes >40 N and occurred in the upper troposphere as well as in the lower stratosphere. In contrast, most of the volcanic aerosol particles form in the stratosphere within several weeks after the injection of large amounts of SO 2 gas [McCormick et al., 1995] after major eruptions at low latitudes (e.g., El Chichon in 1982 and Pinatubo in 1991). The main volcanic aerosol clouds reach higher northern latitudes (>50 N) typically several months after the eruption, and the vertical transport of volcanic aerosols from the stratospheric reservoir into the upper troposphere occurs slowly over years. [5] Lidar has been utilized to continuously document the decay of volcanically induced stratospheric perturbations since the 1970s [Woods and Osborn, 2001; Jäger, 2005; Deshler et al., 2006]. However, most short term or continuous observations are performed with standard backscatter lidars that do not allow for a direct determination of the vertical profile of the climatically important volume extinction coefficient of particles and of the volcanic aerosol 1of9

2 Table 1. Volcanic Eruptions in the Northern Hemisphere Since July 2008 Until the Beginning of October 2009 a Volcano Latitude Longitude Height Start of Eruptions End of Eruptions Plume Top Okmok N W 1073 m m Cleveland N W 1730 m m m m Kasatochi N W 314 m m Bezymianny N E 2882 m m Shiveluch N E 3283 m m m m m Kliuchevskoi N E 4835 m m Redoubt N W 3108 m m Sarychev N E 1496 m m a The volcanoes are sorted by the date of the first eruption since summer The strongest eruptions are given in bold font. The information was obtained from and from optical depth. Ansmann et al. [1990, 1992] introduced the Raman lidar technique and demonstrated, for the first time, that lidar permits a direct determination of the particle extinction profile and of the aerosol optical depth in the troposphere. Based on Pinatubo aerosol observations since August 1991 [Ansmann et al., 1993] and several observations in December 1991 [Ferrare et al., 1992] it was then shown that this new aerosol lidar technique can be successfully applied to stratospheric aerosols, too. We monitored the decay of the stratosphere aerosol load after the eruption of Mt. Pinatubo with a Raman lidar over several years [Mattis, 1996; Ansmann et al., 1997]. Furthermore, it was shown that the measured set of the particle backscatter and extinction coefficients at one wavelength enables us to retrieve the microphysical properties such as the effective, i.e., cross section weighted radius, and mass and surfacearea concentrations of the volcanic aerosol [Wandinger et al., 1995]. A multiwavelength Raman lidar as used in this study, transmitting laser pulses at 355, 532 nm, and 1064 nm, is an almost ideal tool for stratospheric aerosol studies because the spectral dependence of the particle extinction coefficient (Ångström exponent) is obtained in addition and the microphysical properties can also be determined by means of a more comprehensive inversion method [Müller et al., 1999; Ansmann and Müller, 2005]. [6] The paper is organized as follows. Section 2 contains a brief description of the lidar instrument and observational products. Section 3 briefly explains the Lagrangian particle dispersion model FLEXPART. The Raman lidar observations obtained since July 2008 are presented and discussed in section 4. A summary and an outlook related to future tasks including EARLINET (European Aerosol Research Lidar Network) and spaceborne CALIOP (Cloud and Aerosol Lidar with Orthogonal Polarization) observations of the volcanic aerosol clouds are given in section MARTHA [7] Tropospheric and stratospheric aerosol profiles are routinely measured at Leipzig in the framework of EARLINET. From our observations with the multiwavelength Raman lidar MARTHA (Multiwavelength Atmospheric Raman lidar for Temperature, Humidity, and Aerosol profiling) we obtain vertical profiles of the particle backscatter coefficient at 355, 532, and 1064 nm, extinction coefficient at 355 and 532 nm, the corresponding extinction to backscatter ratios (lidar ratios) at 355 and 532 nm, as well as depolarization ratio profiles at 532 nm. The lidar system and data and error analysis are described by Mattis et al. [2002a, 2002b] and Ansmann et al. [2002]. [8] The complete set of products can be derived for stratospheric layers only under cloud free conditions during night time. Nevertheless, the backscatter coefficient at 1064 nm can be derived very well even at daytime and in the presence of cirrus clouds. The 1064 nm signal is the one with the best signal to noise ratio in the stratosphere among all signals detected with MARTHA. It is almost not affected by daytime background light. [9] It should be mentioned that the extinction coefficient retrieval algorithm (applied to the Raman signal profiles) corrects for Rayleigh extinction and the decrease of air density with height. The correction is based on nearby radiosonde information of temperature and pressure. By applying the radiosonde ascents at Lindenberg (approximately 150 km to the northeast of Leipzig), Meiningen (170 km southwest), Bergen (230 km northwest) in Germany and Prague (220 km southeast), Czech Republic, we found that the variability in the obtained four particle extinction profiles is large with values of ±2 Mm 1 up to about 2 km above the tropopause, and ±1 Mm 1 higher up. Thus the uncertainty in the extinction coefficients and lidar ratios in the volcanic layer from 8 10 km height is of the order of 10% 20%. The statistical error is 25% 45%. [10] By means of the measured 355 nm lidar ratio, the microphysical properties (effective radius, mass, surface area) can be estimated provided that the aerosol mainly consists of sulfuric acid droplets [Wandinger et al., 1995]. Alternatively, the microphysical properties are retrieved by means of an inversion algorithm [Müller et al., 1999; Ansmann and Müller, 2005]. Here the full set of backscatter coefficients at the three wavelengths and of extinction coefficients at two wavelengths is required for the retrieval. [11] Linearly polarized laser light is transmitted at 532 nm. The cross polarized signal component is measured in addition to the total signal (sum of cross polarized and parallelpolarized signal components). We obtain the linear volume depolarization ratio from the ratio of the cross polarized signal to the total signal. The volume depolarization ratio is 2of9

3 Figure 1. (top) Time series of aerosol layers in the upper troposphere lower stratosphere (UTLS) region in terms of 10 day mean profiles of the backscatter coefficient at 1064 nm. Black horizontal bars indicate the tropopause height in the individual observations. Triangles show the plume top heights of the different volcanic eruptions listed in Table 1. (bottom) The particle optical depth at 532 nm is computed for the entire volcanic aerosol layer (solid circles) and for the stratosphere (above the tropopause) only. The date is given in the format YYYY MM. useful to characterize the shape (and thus the aggregate state) of the particles. In the case of non spherical ash particles the depolarization ratio is >0.05 [Müller et al., 2007], in the case of liquid sulfuric acid drops, the depolarization ratio is close to zero. 3. FLEXPART [12] The origin of the aerosol layers in the UTLS region was determined with FLEXPART simulations. FLEXPART is a Lagrangian particle dispersion model [Stohl et al., 2005]. It treats long range transport, dry and wet deposition, turbulent diffusion, and convection. The transport simulations are driven by meteorological analysis data from the Global Forecast Model (GFS) which have a horizontal resolution of 1 1. We used archived GFS data with a temporal resolution of 6 hours [U.S. National Centers for Environmental Prediction, ]. We performed simulations backward in time (starting at location and time of the lidar observation) as well as simulations forward in time (starting at time and location of the volcanic eruptions). [13] In both cases, the simulation result is the residence time of the released air parcels per grid cell (2 (longitude) 1 (latitude)) and time step (4 hours). This measure has no explicit physical meaning because it does not contain any information on the amount of aerosol particles or gas concentrations within the air parcels. Furthermore, the absolute value of the residence time depends on the number of simulated air parcels. We can use the forward simulations to investigate whether there has been an air flow from the volcanoes to the lidar site. If we release the air parcels only during an eruption and in the height range of the eruption plume, it is conclusive to assume that the emitted gases and aerosol particles are transported within this air flow. The more air parcels are detected in a grid cell e.g., above the lidar site the larger is the residence time in this grid cell and the larger is the probability that this grid cell contains volcanic gases and particles. The interpretation of backward simulations is similar. 4. Results [14] Figure 1 shows the temporal development of the aerosol load in the UTLS region from January 2008 to December 2009 in terms of the backscatter coefficients at 1064 nm. The 532 nm particle optical depth, given in Figure 1 (bottom), is computed from the 1064 nm backscatter coefficients integrated from the bottom to the top of the identified UTLS aerosol layers. The integrated backscatter coefficients have been converted to 532 nm by assuming a backscatter related Ångström exponent of Finally, multiplication by a factor of 38 sr (lidar ratio) provides a good estimate of the 532 nm particle optical depth. Both conversion factors (532 nm lidar ratio, and 532 nm to 1064 nm backscatter related Ångström exponent) are mean values of all multiwavelength Raman lidar observations of volcanic layers in the UTLS region in 2008 and of9

4 smaller than the one observed during the first two years after the Pinatubo eruption [Ansmann et al., 1998]. [16] A first volcanic aerosol layer was detected on July 28, According to FLEXPART simulations, the layers observed between end of July and beginning of August most probably originated from the Mt. Okmok eruption (see Figures 2a and 3 (top)). A very strong aerosol layer was observed on August 21, 2008, about 14 days after the Mt. Kasatochi eruption. This case is discussed in detail below. Martinsson et al. [2009] reported in situ aerosol observations over Europe (Austria, Hungary, Romania) at 10 km height on August 15, They found an increase in the particulate sulfur concentration by a factor of 10 compared to the background values and a high ratio of carbonaceous to sulfur aerosol mass of 2.6. Martinsson et al. [2009] concluded that it took the volcanic aerosol roughly 7 days for the 10,000 km travel from Alaska to central Europe. Schmale et al. [2010] found a slightly increased carbonaceous tosulfur ratio still in October [17] Figure 4 illustrates the transport of the Kasatochi plume by means of satellite observations. The top panel shows the OMI observation of the SO 2 column on August 13, SO 2 column values of about 10 DU can be observed southwest of Greenland. CALIOP observed an aerosol layer over Figure 2. Examples for source identification of the aerosol layers with FLEXPART forward simulations. The air parcels were released from the (a) Mt. Okmok eruptions from July 12 to August 19, 2008, (b) Mt. Kasatochi eruption on August 7, 2008, and (c) Redoubt eruption between March 23 and April 4, The thick black vertical lines indicate the time of the investigated lidar measurements that were performed at Leipzig on August 4, 2008 (Figure 2a), August 18, 2008 (Figure 2b), March 30, 2009 (solid lines, Figure 2c), and April 6, 2009 (dotted lines, Figure 2c). The color coding corresponds to the logarithm of the sum over the residence time of all air parcels over Leipzig per height bin and time. [15] An estimation of the plume height of the volcanic eruptions, listed in Table 1, is indicated by a triangle. The values were obtained from different methods, e.g., pilot observations or radar measurements. They give the top height of the visible ash plumes. Gas plumes could have easily reached larger altitudes. Thick horizontal bars indicate lapse rate tropopause heights that were obtained from atmospheric soundings at stations around Leipzig (Lindenberg, Meiningen, Bergen, and Praque). Many strong aerosol layers occurred in the upper troposphere. The particle optical depth typically ranges from , and is sometimes larger than The stratospheric aerosol optical depth is small with values from , and thus a factor of Figure 3. Examples for source identification of the aerosol layers with FLEXPART 10 day backward simulations. The air parcels were released in the altitude range of the observed layers over Leipzig on (top) August 4, 2008, (middle) August 18, 2008, and (bottom) April 6, The color coding corresponds to the logarithm of the sum over the residence time of all air parcels that have passed over a pixel between 0 and 20 km height. 4of9

5 Figure 4. Transport of the Kasatochi eruption plume. (a) The SO 2 column observed with the OMI instrument on August 13, 2008 (from (b) Those parts of the CALIOP footprints, where the corresponding profiles show the volcanic layer. The day of observation is color coded. The blue, cyan, green, orange, red, and magenta lines correspond to August 13 18, 2008, respectively. the same region at about 13 km height. Figure 4b shows those parts of the CALIOP footprints of the following days where the aerosol layer was observed. On August 18, 2008 the plume arrived over central Europe. This coincides with our lidar observation of this day (see Figure 2). [18] The aerosol composition in volcanic plumes changes with time. First, ash particles and SO 2 gas are emitted into the atmosphere. In case of the Kasatochi eruption, the ash particles could still be observed one week after the eruption, but vanished after one month [Martinsson et al., 2009]. On the other hand, SO 2 is converted into sulfuric acid droplets within several weeks. Those droplets remain in the stratosphere for a long time, travel around the globe many times, and get distributed in this way over the entire hemisphere. [19] Another period with strong aerosol backscattering mainly below the tropopause occurred in spring of Figures 2 and 3 illustrate how the measurement of April 6, 2009 can be attributed to the Redoubt eruption of March 23. Again, the volcanic plume reached Europe one week after the eruption. Nevertheless, it might be that aerosol particles from previous eruptions also contribute to the measurement of April 6, [20] On December 8, 2008, a pronounced aerosol layer was detected with a plume top height of up to 30 km. FLEXPART backward simulations show that most air par- 5of9

6 particle optical depth of the volcanic layer on August 21, 2008, is (532 nm) and (355 nm). [22] The volume depolarization ratio in Figure 6 is very low, a clear indication for the presence of spherical aerosol particles. Non spherical ash particles were obviously absent. [23] By applying the method developed by Wandinger et al. [1995] to the 355 nm lidar ratios, we obtain effective radii of around 0.1 mm, and surface area and mass concentrations in the range from mm 2 cm 3 and mg m 3, respectively (see Figure 6). Figure 5. Tropospheric layer of volcanic aerosol from 8 10 km height, observed on August 21, Rangecorrected lidar signal at 1064 nm (arbitrary units) is shown with vertical and temporal resolution of 60 m and 30 s, respectively. Complex vertical layering of aerosols throughout the entire troposphere up to the tropopause is observed. The signals below 1 km height are not trustworthy because of the incomplete overlap of the laser beam with the receiver field of view. cels were transported to Leipzig by westerly winds within the stratosphere above 10 km height. The air parcels did not have contact to aerosol sources in the troposphere. We speculate that the aerosols in the tropical stratospheric reservoir originate from one of the volcanic eruptions in Beside the eruptions in mid latitudes the Global Volcanism Program ( reported also eruptions of tropical volcanoes, e.g., Tungurahua (Ecuador) on February 2, 2008 (14.3 km plume height), ol Doinyo Lengai (Tanzania) on March 5, 2008 (15.2 km plume height) or Soufrire Hills (Montserrat) on July 28, 2008 (top height of the plume 12 km). The meridional stratospheric aerosol transport towards higher northern latitudes strengthens during the autumn months. A similar seasonal pattern was observed after the Pinatubo eruption in December 1991 when the plume height suddenly increased from 20 to 30 km height in the beginning of December, and decreased to values below 25 km height in January 1992 and in the following months [Ansmann et al., 1993]. [21] Two case studies based on our multiwavelength Raman lidar observations are discussed next. The pronounced tropospheric volcanic aerosol layer observed from 8 10 km height on August 21, 2008, is shown in Figure 5. The mean profiles of optical and estimated microphysical properties are presented in Figure 6. Only the trustworthy portions of the profiles are presented. As can be seen, the volume extinction coefficients of the particles reach values of about 10 Mm 1 (532 nm) and 25 Mm 1 (355 nm), respectively. Lidar ratios are sr (532 nm) and sr (355 nm). Backscatter related Ångström exponents range from The extinction related Ångström exponent is >2 and may be caused by high absorption of the UV wavelength by the volcanic particles. As mentioned above, Martinsson et al. [2009] reported a carbonaceous to sulfur aerosol mass of 2.6 at 10 km height on August 15, The Figure 6. The 1 hour mean values of optical and microphysical properties derived from the Raman lidar measurements at Leipzig on August 21, 2008, UTC. (top) Particle backscatter coefficients (355 nm in blue, 532 nm in green, 1064 nm in red), extinction coefficients (at 355 and 532 nm), the corresponding extinction tobackscatter ratios at 355 and 532 nm, the backscatter related Ångström exponents (for the nm wavelength range in blue, for the nm wavelength range in red), and the volume depolarization ratio at 532 nm (green). (bottom) Relative humidity and temperature as measured with radiosonde at Bergen, 200 km northwest of Leipzig, effective radius, surface area concentration, and mass concentration of the volcanic particles. The tropopause is indicated by horizontal lines in all plots, bottom and top of the volcanic layer are shown by horizontal lines in the backscatter plot only. The vertical resolution of the backscatter profiles is 660 m. The extinction coefficients and the microphysical properties have been obtained with a vertical resolution of 2400 m. Error bars (standard deviation, particle optical properties) show the uncertainty introduced by signal noise. 6of9

7 Figure 7. Stratospheric layer of volcanic aerosol from km height, observed on August 31, 2009, 2058 UTC to September 1, 2009, 0201 UTC. Range corrected lidar signal at 1064 nm (arbitrary units) is shown with vertical and temporal resolution of 60 m and 30 s, respectively. Complex vertical layering of aerosols throughout the entire troposphere and in the lower stratosphere up to 20 km is observed. The signals below 1 km height are not trustworthy because of the incomplete overlap of the laser beam with the receiver field of view. optical depth is roughly 0.01 (532 nm) and (355 nm). Lidar ratios range from sr at both wavelengths. Again, the volcanic aerosol consists of spherical particles according to the rather low volume depolarization ratio. [27] Based on balloon borne in situ observations of the stratospheric aerosol size distribution at Laramie, Wyoming, Jäger et al. [1995] computed lidar ratios of sr (532 nm) before the Mt. Pinatubo eruption in 1991 and values around 30 sr (532 nm) in Pinatubo aerosol within the first 2 3 years after the eruption. We found values for the 308 nm lidar ratio of sr in from Raman lidar measurements at 308 nm [Ansmann et al., 1993]. [28] According to the method of Wandinger et al. [1995], 355 nm lidar ratios around sr and sr indicate particle size distributions with effective radii of mm and mm, respectively. We assume that carbonaceous aerosol particles (as observed by Martinsson et al. [2009]) are no longer present at such high altitudes so that the retrieval results are more accurate than for tropospheric volcanic layers. [29] As can be seen in Figure 8, effective radii were as low as mm in the center of the volcanic aerosol layer (15 16 km height). For the surface area and mass [24] In the case of tropospheric aerosol layers the estimation of the microphysical properties must be interpreted with caution. The retrieval method assumes the presence of pure liquid sulfuric acid droplets. We did not make an attempt to check the impact of the presence of carbonaceous aerosols (as observed by Martinsson et al. [2009]) on the retrieval results. However, by applying an inversion scheme [Müller et al., 1999] to the full set of available backscatter and extinction coefficients similar results were obtained. The inversions also clearly indicate the presence of absorbing particles. The imaginary part of the refractive index (for 500 nm) was found to be around Such high numbers point to a single scattering albedo of the particles in the range of [25] On August 31, 2009, a stratospheric volcanic aerosol layer was observed from km height (see Figure 7). Our transport simulations indicate that this layer originated from the Sarychev eruption in June Besides the stratospheric layer, many other aerosol layers are visible above the planetary boundary layer (top height at 2 km). A clean, almost aerosol free free troposphere was rarely observed during the past 15 months. This behavior might not be caused by the volcanic eruptions alone because aerosol layers from long range transport can be observed in about 45% of all measurements at Leipzig [Mattis et al., 2008]. However, these lofted layers above the boundary layer were mostly confined to the lowermost 6 km of the troposphere, whereas the volcanic layers observed since July 2008 occurred mostly above 6 km height. [26] Peak volume extinction values in the stratospheric volcanic aerosol layer are 1.5 Mm 1 (532 nm) and 3 4Mm 1 (355 nm), the respective Ångström exponent is 1.2. The backscatter related Ångström exponents also vary around in the center of the volcanic layer. The aerosol Figure 8. Same as Figure 6, but for the 5 hour mean values of optical and microphysical properties derived from the Raman lidar measurements at Leipzig on August 31, 2009, UTC. The vertical resolution of the backscatter profiles is 1260 m. The extinction coefficients and the microphysical properties have been obtained with a vertical resolution of 4 km. Error bars show the uncertainties introduced by signal noise and due to the uncertainty of the temperature profile. 7of9

8 concentrations we found values of 5 10 mm 2 cm 3 and mg m 3, respectively. [30] After the Pinatubo eruption, the particle size distributions consisted of two modes. The fraction of large volcanic particles caused effective radii around 0.4 mm. Monthly mean values of the surface area and mass concentrations were about 20 mm 2 cm 3 and 6 8 mg m 3 in the center of the volcanic layer for more than a year [Ansmann et al., 1997, 1998]. 5. Summary and Outlook [31] Volcanic aerosol observations at a central European EARLINET lidar site, performed with an advanced multiwavelength Raman lidar from January 2008 to December 2009, were presented. The aerosol layers originated from numerous eruptions of volcanoes on the Aleutian Islands, Kamchatka, Alaska, and the Kuril Islands and typically occurred between 5 and 25 km height. The volcanic particles traveled about a week before reaching central Europe. The optical depths of the volcanic aerosol layers were found to be mostly between and at 532 nm. The wavelength dependence of the backscatter coefficients and extinction coefficients indicate Ångström exponents from Lidar ratios were in the range of sr (355 nm) and sr (532 nm). The effective radius, surface area, and mass concentrations of the stratospheric, volcanic aerosol particles are mm, and mostly <10 mm 2 cm 3, and <0.5 mgm 3, respectively. The surface area and mass concentrations are thus about a factor of lower than the respective values observed after the Mt. Pinatubo eruption in the years 1992 and [32] The monitoring is ongoing and currently subject of intensified EARLINET observational activities. We will analyze the network data covering a latitudinal range from 38 N to 70 N to document the spreading of the volcanic aerosols over a larger latitudinal belt. Observations with the spaceborne two wavelength standard backscatter lidar CALIOP will be included in this effort. Note that lidar ratio information, provided by the EARLINET Raman lidar observations, is required in the CALIOP data analysis to obtain trustworthy results in terms of the volume extinction coefficients and particle optical depth. [33] Acknowledgments. The EARLINET ASOS project is funded by the European Commission (EC) under grant RICA The wind fields for the FLEXPART simulations are from the Research Data Archive (RDA) which is maintained by the Computational and Information Systems Laboratory (CISL) at the National Center for Atmospheric Research (NCAR). NCAR is sponsored by the National Science Foundation (NSF). The original data are available from the RDA ( in data set number ds CALIPSO data were obtained from the NASA Langley Research Center Atmospheric Science Data Center. The OMI data plot was provided by the Sulfur Dioxide Group of the University of Maryland, Baltimore County and the NASA Goddard Space Flight Center. References Ansmann, A., and D. Müller (2005), Lidar and atmospheric aerosol particles, in Lidar: Range Resolved Optical Remote Sensing of the Atmosphere, edited by C. Weitkamp, pp , Springer, New York. Ansmann, A., M. Riebesell, and C. Weitkamp (1990), Measurement of atmospheric aerosol extinction profiles with a Raman lidar, Opt. Lett., 15, Ansmann, A., M. Riebesell, U. Wandinger, C. Weitkamp, E. Voss, W. Lahmann, and W. Michaelis (1992), Combined Raman elasticbackscatter lidar for vertical profiling of moisture, aerosol extinction, backscatter, and lidar ratio, Appl. Phys. B, 55, Ansmann, A., U. Wandinger, and C. Weitkamp (1993), One year observations of Mount Pinatubo aerosol with an advanced Raman lidar over Germany at 53.5 N, Geophys. Res. Lett., 20, Ansmann, A., I. Mattis, U. Wandinger, F. Wagner, J. Reichardt, and T. Deshler (1997), Evolution of the Pinatubo aerosol: Raman lidar observations of particle optical depth, effective radius, mass and surface area over central Europe at 53.4 N, J. Atmos. Sci, 54, Ansmann, A., I. Mattis, H. Jäger, and U. Wandinger (1998), Stratospheric aerosol monitoring with lidar: conventional backscatter versus Raman lidar observations of Pinatubo aerosol, Contrib. Atmos. Phys., 71, Ansmann,A.,F.Wagner,D.Müller,D.Althausen,A.Herber,W.v. Hoyningen Huene, and U. Wandinger (2002), European pollution outbreaks during ACE 2: Optical particle properties inferred from multiwavelength lidar and star Sun photometry, J. Geophys. Res., 107(D15), 4259, doi: /2001jd Deshler, T., et al. (2006), Trends in the nonvolcanic component of stratospheric aerosol over the period , J. Geophys. Res., 111, D01201, doi: /2005jd Ferrare, R. A., S. H. Melfi, D. N. Whitenman, and K. D. Evans (1992), Raman lidar measurements of Pinatubo aerosols over southeastern Kansas during November December 1991, Geophys. Res. Lett., 19, Gassó, S. (2008), Satellite observations of the impact of weak volcanic activity on marine clouds, J. Geophys. Res., 113, D14S19, doi: / 2007JD Jäger, H. (2005), Long term record of lidar observations of the stratospheric aerosol layer at Garmisch Partenkirchen, J. Geophys. Res., 110, D08106, doi: /2004jd Jäger, H., T. Deshler, and D. Hofmann (1995), Midlatitude lidar backscatter conversions based on balloonborne aerosol measurements, Geophys. Res. Lett., 22, Martinsson, B. G., C. A. M. Brenninkmeijer, S. A. Carn, M. Hermann, K. P. Heue, P. F. J. van Velthoven, and A. Zahn (2009), Influence of the 2008 Kasatochi volcanic eruption on sulfurous and carbonaceous aerosol constituents in the lower stratosphere, Geophys. Res. Lett., 36, L12813, doi: /2009gl Mattis, I. (1996), Zeitliche Entwicklung des stratosphärischen Aerosols nach dem Ausbruch des Pinatubo: Analyse von Raman Lidarmessungen (Temporal development of the stratospheric aerosol after the eruption of Mt. Pinatubo: analysis of Raman lidar observations), Diploma thesis, Univ. Leipzig, Leipzig, Germany. Mattis, I., A. Ansmann, D. Althausen, V. Jaenisch, U. Wandinger, D. Müller, Y. F. Arshinov, S. M. Bobrovnikov, and I. B. Serikov (2002a), Relative humidity profiling in the troposphere with a Raman lidar, Appl. Opt., 41, Mattis, I., A. Ansmann, D. Müller, U. Wandinger, and D. Althausen (2002b), Dual wavelength Raman lidar observations of the extinctionto backscatter ratio of Saharan dust, Geophys. Res. Lett., 29(9), 1306, doi: /2002gl Mattis, I., D. Müller, A. Ansmann, U. Wandinger, J. Preißler, P. Seifert, and M. Tesche (2008), Ten years of multiwavelength Raman lidar observations of free tropospheric aerosol layers over central Europe: Geometrical properties and annual cycle, J. Geophys. Res., 113, D20202, doi: /2007jd McCormick, M. P., L. W. Thomason, and C. R. Trepte (1995), Atmospheric effects of the Mt. Pinatubo eruption, Nature, 373, Müller, D., U. Wandinger, and A. Ansmann (1999), Microphysical particle parameters from extinction and backscatter lidar data by inversion with regularization: Simulation, Appl. Opt., 38(12), Müller, D., A. Ansmann, I. Mattis, M. Tesche, U. Wandinger, D. Althausen, and G. Pisani (2007), Aerosol type dependent lidar ratios observed with Raman lidar, J. Geophys. Res., 112, D16202, doi: /2006jd Sassen, K. (1992), Evidence for liquid phase cirrus cloud formation from volcanic aerosols: Climatic implications, Science, 257, Schmale, J., et al. (2010), Aerosol layers from the 2008 eruptions of Mt. Okmok and Mt. Kasatochi: In situ UT/LS measurements of sulfate and organics over Europe, J. Geophys. Res., doi: / 2009JD013628, in press. Schneider, D. P., C. M. Ammann, B. L. Otto Bliesner, and D. S. Kaufmann (2009), Climate response to large high latitude and low latitude volcanic eruptions in the community climate system model, J. Geophys. Res., 114, D15101, doi: /2008jd Stohl, A., C. Forster, A. Frank, P. Seibert, and G. Wotawa (2005), Technical note: The Lagrangian particle dispersion model FLEXPART version 6.2, Atmos. Chem. Phys., 5, of9

9 U.S. National Centers for Environmental Prediction ( ), NCEP FNL Operational Model Global Tropospheric Analyses, Data Set ds083.2, CISL Data Support Sect., Natl. Cent. for Atmos. Res., Boulder, Colo., updated daily. Wandinger, U., A. Ansmann, J. Reichardt, and T. Deshler (1995), Determination of stratospheric aerosol microphysical properties from independent extinction and backscattering measurements with a Raman lidar, Appl. Opt., 34, Woods, D. C., and M. T. Osborn (2001), Twenty six years of lidar monitoring of northern midlatitude stratospheric aerosols, in Remote Sensing of Clouds and the Atmosphere, vol. 4168, edited by J. E. Russell, K. Schaefer, and O. Lado Bordowsky, pp , Soc. of Photo Opt. Instrum. Eng., Bellingham, Wash. A. Ansmann, F. Finger, A. Hiebsch, T. Kanitz, I. Mattis, D. Müller, J. Schmidt, P. Siefert, M. Tesche, and U. Wandinger, Leibniz Institute for Tropospheric Research, Permoserstr. 15, Leipzig, Germany. (albert@tropos.de; finger@tropos.de; hiebsch@tropos.de; kanitz@tropos. de; ina@tropos.de; detlef@tropos.de; joerg.schmidt@tropos.de; seifert@ tropos.de; tesche@tropos.de; ulla@tropos.de) 9of9