Atmospheric distribution and removal of volcanic ash after the eruption of Kasatochi volcano: A regional model study

Transcription

1 JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 115,, doi: /2009jd013298, 2010 Atmospheric distribution and removal of volcanic ash after the eruption of Kasatochi volcano: A regional model study Baerbel Langmann, 1 Klemen Zakšek, 1 and Matthias Hort 1 Received 29 September 2009; revised 3 March 2010; accepted 18 March 2010; published 7 August [1] In August 2008, Kasatochi volcano on the Aleutian Islands erupted without much advance warning. Volcanic ash released during this eruption quickly settled out of the atmosphere, mainly into the NE Pacific Ocean. The amount of volcanic ash, as well as the ash fall area and volume into the NE Pacific Ocean, remains speculative, as only a limited number of measurements is available. We used a three dimensional atmosphere/ chemistry aerosol model to determine the atmospheric distribution of SO 2 and volcanic ash and its fallout after the eruption of Kasatochi volcano. In a first step, modeled atmospheric SO 2 distributions are compared with satellite data, thereby evaluating the model capabilities to reasonably reproduce atmospheric transport patterns. For modeled volcanic ash mass a considerable reduction of the atmospheric content already occurred by 10 August, the second day after the eruption in accordance with satellite observations. Gravitational settling is the most efficient removal process for volcanic ash mass, exceeding dry and wet deposition by far. Assuming an ash volume of 0.3 km 3 released during the eruption of Kasatochi volcano and a median ash particle diameter of 4 mm, the mass of volcanic ash removed at ground within the 0.1 mm isopach covers an area of km 2 over the NE Pacific Ocean and makes up 49% of the removed material out of the atmosphere. The amount of ash and that of iron attached to it is sufficient to explain measured seawater CO 2 decrease at the ocean station Papa in August 2008 induced by iron fertilization and subsequent phytoplankton production. Citation: Langmann, B., K. Zakšek, and M. Hort (2010), Atmospheric distribution and removal of volcanic ash after the eruption of Kasatochi volcano: A regional model study, J. Geophys. Res., 115,, doi: /2009jd Introduction [2] During summer 2008, explosive eruptions of three remote Aleutian arc island volcanoes occurred. The largest explosive event occurred at Kasatochi volcano (52.17 N, W) on 7 and 8 August with three major eruptions that rose to altitudes of about 15 km. The last explosive event on 8 August lasted for about 17 h [Waythomas et al., 2008]. The plume formed a counter clockwise spiral at altitudes between 9 km and 14 km and spread further south eastward across the NE Pacific (Figure 1) ( Kasatochi.php), thereby severely affecting aviation routes [Waythomas et al., 2008]. It has been estimated that the Kasatochi eruption cloud contained about Tg of SO 2 [e.g., S. Carn et al., submitted to Journal of Geophysical Research, 2010] and is one of the largest volcanic SO 2 clouds since Mt. Hudson volcano in Chile erupted in 1991 ( earthobservatory.nasa.gov/iotd/view.php?id=8998). Satellite observations from GOME 2 and OMI reveal large scale transport of SO 2 across the northern latitudes during August 2008 whereas volcanic ash could be detected from MODIS 1 Institute of Geophysics, University of Hamburg, Hamburg, Germany. Copyright 2010 by the American Geophysical Union /10/2009JD and AVHRR instruments only for a few days. Ash particles quickly settle out of the atmosphere, in the case of Kasatochi mainly into the NE Pacific Ocean. [3] One may consider the deposition of volcanic ash into the ocean as the final graveyard for volcanic ash and its environmental impact. However, new evidence reveals that volcanic ash can represent a so far underestimated component of the surface ocean biogeochemical iron cycle and marine primary productivity (MPP). In so called highnutrient low chlorophyll (HNLC) areas (the subarctic and equatorial Pacific and the Southern Ocean, which make up about 20% of the total ocean area) macronutrient concentrations are high, but iron is the key biologically limiting micronutrient for primary productivity [Martin and Fitzwater, 1988]. Laboratory experiments have shown that volcanic ash rapidly releases iron on contact with seawater [Watson, 1997; Duggen et al., 2007; Jones and Gislason, 2008; N. Olgun et al., submitted to Global Biogeochemical Cycles, 2010] emphasizing the fertilization potential for the surface ocean. By analyzing chlorophyll a satellite data from MODIS, Langmann et al. [2010] presented direct evidence that volcanic ash caused natural iron fertilization and MPP increase following the eruption of Kasatochi volcano. However, the spatial distribution of ash deposition has not been explicitly determined so far. 1of10

2 Figure 1. Atmospheric dispersion of the Kasatochi SO 2 cloud from 8 to 12 August Shown are (left) REMOTE model results and (right) GOME 2 satellite data for (a) 2200 UTC on 8 August UTC, (b) 2100 UTC on 9 August 2008, (c) 2100 UTC on 10 August 2008, (d) 2000 UTC on 11 August 2008, and (e) 2000 UTC on 12 August of10

3 [4] Fierstein and Nathenson [1992] describe several ways to determine the deposited volcanic ash fall volume and mass which, however, only work if ash is deposited on land and are unsuitable when volcanic ash is mainly deposited into the ocean. Deep sea sediment traps are only rarely available at the right location and time for in situ measurements of the submarine fallout of volcanic ash, as it was the case after the eruption of Mt. Pinatubo in 1991 in the South China Sea [Wiesner et al., 1995]. As volcanic ash deposited over the ocean is preserved in marine sediments, analysis of sediment cores [Kutterolf et al., 2008; Wiesner et al., 2004] allow to estimate tephra volume and erupted mass, but this information is not available shortly after a volcanic eruption. Another possibility is to use threedimensional atmospheric models which include transport and removal processes of volcanic ash [e.g., Costa et al., 2006; Folch et al., 2008] to estimate the atmospheric distribution and fallout of volcanic ash. Even though such models include several assumptions and simplifications, they offer the possibility to analyze atmospheric distribution and fallout of volcanic ash out of one consistent data set. Here we use the three dimensional Eulerian atmospherechemistry/aerosol model REMOTE [Langmann, 2000; Langmann et al., 2008] to simulate the distribution of atmospheric SO 2 for the evaluation of the transport pattern and volcanic ash as well as deposition of volcanic ash after the eruption of Kasatochi. Section 2 introduces the model and describes the setup for the simulation of the eruption of Kasatochi volcano. Model results and discussions are presented in section 3 and section 4 provides conclusions. 2. Model Description and Setup [5] We use the regional scale three dimensional on line atmosphere chemistry/aerosol model REMOTE ( mpimet.mpg.de/en/wissenschaft/modelle/remote.html) [Langmann, 2000; Langmann et al., 2008] which is one of the few regional climate models that determines the physical, photochemical and aerosol state of the model atmosphere at every model time step simultaneously thus offering the possibility to consider trace species effects on climate as well [e.g., Langmann, 2007]. Besides various studies on the dispersion and chemical transformation of anthropogenic and natural emissions, REMOTE has also been applied to study the dispersion of volcanic emissions in Indonesia [Pfeffer et al., 2006] and Nicaragua [Langmann et al., 2009]. The dynamical part of the model is based on the former regional weather forecast system of the German Weather Service [Majewski, 1991] which is using a hydrostatic assumption for the vertical pressure gradient. In addition to the German Weather Service physical parameterizations, those of the global climate model ECHAM 4 [Roeckner et al., 1996] have been implemented in REMOTE and are used for the current study. [6] REMOTE is applied with 31 vertical layers of increasing thickness between Earth s surface and the 10 hpa pressure level using terrain following hybrid pressure sigma coordinates. Between 9 and 17 km height, the vertical model resolution is about 1 km. The model domain covers large parts of the NE Pacific Ocean, Kamchatka, the Aleutian Islands, Alaska, Canada and the western United States (Figure 1). The domain is subdivided into grid boxes of 0.5 resolution (approximately 55 km) on a rotated latitude/longitude grid. A model time step of 5 min is chosen. REMOTE is initialized using meteorological analysis data of the European Centre for Medium Range Weather Forecast (ECMWF), which are updated at the lateral model boundaries every 6 h with linear interpolation in between. Every 30 h, ECMWF data are used for an update over the whole model domain to force the model to stay close to the observed weather situation. The prognostic equations for surface pressure, temperature, specific humidity, cloud water, horizontal wind components and aerosol mass mixing ratios are written on an Arakawa C grid [Mesinger and Arakawa, 1976]. [7] In the current model setup, SO 2 and volcanic ash are included as prognostic but passive species. They are subject of transport and removal processes, but chemical and microphysical conversions; for example, oxidation of SO 2 to sulphate are not considered here. The aerosol dynamic and thermodynamic module M7, described in detail in the work of Vignati et al. [2004] and Stier et al. [2005], provides the framework for the volcanic ash size determination with volcanic ash as the only component considered as soluble coarse particle. Various mathematical distribution functions can be used to describe the ash size spectrum; for example, Gaussian distribution [Folch et al., 2008], Weibull distribution [Weibull, 1951] or the log normal distribution [Seinfeld and Pandis, 1998] which has the advantage of a constant standard deviation for number, surface and volume distribution. The widely used ash size classification in volcanology is connected to the ash diameter d via the following equation: = log 2 (d/d 0 ) with d 0 = 1 mm and d in mm. We represent the ash size spectrum by a log normal distribution described by the three moments: particle number N, median radius r, and standard deviation s. Standard deviations is prescribed (s = 2) so that the median radius can be calculated from the corresponding particle number and mass. The transport of SO 2, volcanic ash mass and number concentration is determined by horizontal and vertical advection according to the algorithm of Smolarkiewicz [1983], convective updraft and downdraft by a modified scheme of Tiedtke [1989] and vertical diffusion after Mellor and Yamada [1974]. Dry deposition fluxes for SO 2 are determined after Wesely [1989]. For volcanic ash, the dry deposition calculation is based on the work of Stier et al. [2005]. Dry deposition represents the removal process close to the surface which is more important for atmospheric trace gases than for volcanic ash, as gases are not subject to gravitational settling. Wet deposition is computed according to Walcek and Taylor [1986] by integrating the product of the grid averaged precipitation rate and the mean cloud water concentration which is determined from cloud base (first layer above the surface containing more than g/kg liquid water) to cloud top (highest level exceeding an amount of g/kg liquid water) for fair weather clouds and from the surface to cloud top for raining clouds. Sedimentation of ash particles is calculated throughout the atmospheric column. The calculation of the sedimentation velocity is based on the Stokes velocity with the Cunningham slip flow correction factor accounting for non continuum effects [e.g., Seinfeld and Pandis, 1998]. [8] Emissions from the Kasatochi eruption are released into the model at altitudes between 3 14 km above ground 3of10

4 (50% at 14 km; the maximum eruption column height, the other 50% equally distributed between 3 and 14 km, it is assumed that emissions in the planetary boundary layer up to about 3 km are negligible) following the available observations ( php), and into the grid cell where Kasatochi volcano is located. Starting on 8 August 2008 at 0435 UTC, 1.4 Tg of SO 2 [S. Carn et al., submitted to Journal of Geophysical Research, 2010] and 600 Tg of volcanic ash are released at a constant rate during the following 17 h. The initial ash diameter is set to 4.0 mm, as it is intended to determine the distribution and fallout of fine volcanic ash. (In the lognormal distribution 67% of the ash particles are in the size range between 2.0 and 8.0 mm, and 95% between 1.0 and 16.0 mm). The amount of volcanic ash released during the eruption of Kasatochi volcano is based on a backward estimate by Langmann et al. [2010], who calculated the amount of volcanic ash that would have been necessary to generate the observed phytoplankton increase in the NE Pacific during August Langmann et al. [2010] confirmed this estimate by applying a 1 D eruption column model [Mastin, 2007] which determines the mass flow rate dependent on plume height and meteorological conditions. In section 3 sensitivity studies are discussed which assume a smaller amount of ash released from Kasatochi. This is done as the fine ash content during explosive volcanic eruptions can vary considerably. According to Rose and Durant [2009] very fine ash particles with diameters less than 30 mm make up only a few percent during mafic eruptions whereas they can contribute 30 50% to the total ash content during silicic eruptions like Kasatochi. Pyroclastic flows which occurred during the eruption of Kasatochi [Waythomas et al., 2008] are known to produce high proportions of very fine ash. In addition, phreatomagmatic processes, interaction between magma and water (abundantly available in the crater lake of Kasatochi), are also likely to enhance the very fine ash content [e.g., Schmincke, 2004]. Further sensitivity tests assume different vertical distributions of the erupting material and bigger initial ash diameters. 3. Model Results and Discussion [9] For the evaluation of the model capabilities to reasonably reproduce atmospheric transport patterns in the lower stratosphere/upper troposphere, Figure 1 compares REMOTE simulations results and GOME 2 satellite data of the atmospheric distribution of SO 2 column concentration in [DU] over the NE Pacific Ocean shortly after the eruption of Kasatochi volcano. Using Differential Optical Absorption Spectroscopy (DOAS), SO 2 levels can be well estimated on the basis of its strong and structured absorption in UV spectra. The GOME 2 instrument aboard MetOp is the latest operational UV instrument (next to OMI and SCIAMACHY). The raw data we used were processed by German Aerospace Centre [Loyola et al., 2008; Rix et al., 2009]. Overall, the timing and location of the volcanic SO 2 cloud can be realistically reproduced by REMOTE, taking into account that due to the chosen horizontal resolution of 0.5 a smoothing of SO 2 concentrations occurs with transport. The counterclockwise spiral which formed on 8 August when the eruption of Kasatochi volcano was still ongoing cannot be fully reproduced by the model. This can be either due to the coarse horizontal resolution or the initialization process, which prescribes the vertical distribution of the erupting material and focuses on the third eruption period of Kasatochi, thereby neglecting in particular SO 2 which has already been released some hours before in the two preceding eruption pulses, which are assumed to be less loaded with ash [Waythomas et al., 2008]. However, SO 2 concentrations remain higher during the simulation as oxidation to sulphate is not considered. Assuming an e folding time of 25 days and linear decrease during the first 4 days, a maximum loss of 10% of the initial SO 2 mass is neglected. [10] For volcanic ash in the atmosphere, it is difficult to use a universal detection because the ash particle radius in the cloud usually varies from 1 to 15 mm and the chemical properties may vary from one volcano to another. Remote sensing data of volcanic ash after the eruption of Kasatochi have been analyzed using different methods and are published in this special issue [S. Corradini et al., submitted to Journal of Geophysical Research, 2010; Karagulian et al., 2010; Kristiansen et al., 2010; Prata et al., 2010]. In order to track the pathway of the Kasatochi ash cloud in the atmosphere we used MODIS level 1b images and applied Brightness Temperature Difference (BTD) at 11 mm and 12 mm for a qualitative analysis [Prata, 1989a, 1989b]. If the differences are negative, there is a high probability that the cloud contains ash. The ash cloud is tracked using BTD for 4 days (Figure 2), but already on 11 August, the detection is no longer reliable. We note here that the BTD signal does not correlate with the ash content, however, comparing modeled ash column concentration with Figure 2 shows qualitative agreement in the travelling routes. From comparison of Figure 1 and Figure 2 it is obvious that the ash cloud travelled along the same pathways as SO 2, as also shown by Prata et al. [2010]. Therefore, the assumption of the same initial percentage vertical distribution for volcanic ash and SO 2 is justified for the model simulations. [11] REMOTE model results of the vertical distribution of SO 2, volcanic ash mass and particle number concentration (Figure 3) give further insight into the atmospheric dispersion and removal processes after the eruption of Kasatochi volcano. The vertical distribution of the total SO 2 mass within the model domain shows a maximum in 14 km height, the major injection level. After 11 August, SO 2 maximum vertical concentration is reduced because the volcanic cloud is transported across the lateral model boundaries (Figure 1) resulting in a loss of mass inside the model domain. The highest concentrations of SO 2 remain located in the upper troposphere/lower stratosphere, getting even transported slightly higher up throughout the first days after the eruption. The vertical distribution of volcanic ash mass shows a complete different behavior. Volcanic ash mass is quickly transported to the ground so that already by 10 August, a considerable reduction of the atmospheric volcanic ash content has occurred in accordance with observation data from satellite (Figure 2). As a small portion of the larger ash particles carries the bulk of the mass, the vertical particle number distribution remains largely unaffected by sedimentation and dry and wet deposition during the first few days after the eruption, so that it resembles much more to that of SO 2. [12] Dry and wet deposition fluxes, and sedimentation fluxes of volcanic ash mass (see section 2 for more expla- 4of10

5 Figure 2. Atmospheric dispersion of the Kasatochi ash cloud from 8 to 10 August 2008, based on MODIS satellite data. nation) show the contribution of the individual processes in removing volcanic ash mass out of the atmosphere and at which locations this occurs (Figure 4a). Sedimentation is the most efficient removal process for volcanic ash mass with 70% of the total mass removed out of the atmosphere at ground level, followed by wet deposition (23%) and dry deposition (7%). The dominance of the removal of ash mass by sedimentation lasted for the first 3 days after the eruption Figure 3. REMOTE model results for the area averaged vertical profiles of SO 2, volcanic ash mass, and particle number as function of time from 8 to 16 August 2008 at 0000 UTC. The averaging is done over the whole model area, as shown in Figure 4. 5of10

6 Figure 4. REMOTE model results for dry and wet deposition, and sedimentation of (a) volcanic ash mass (mm/8 d), (b) particle number concentration (1.E 12/ha/8 d), and (c) mean radius (mm). (Table 1), during which 80% of the ejected ash mass (the total amount of ash released is g) is removed from the atmosphere, with 47% being removed on 9 August. After day three after the eruption, wet deposition becomes dominant. Generally, sedimentation fluxes decrease with increasing distance from the volcano taking into account the nonlinear dispersion pattern. The mass of volcanic ash removed at ground within the 0.1 mm isopach covers an area of km 2 and makes up 49% of the removed material from the atmosphere. The fallout into the NE Pacific Ocean makes up 92% of the total ash mass removed from the atmosphere. In comparison with the eruption of Chaiten volcano in Chile in May 2008 ( km 2 and 0.07 km 3 dense rock equivalent (DRE); see Watt et al. [2009]), our estimate for Kasatochi is the same order of magnitude and both, 0.1 mm isopach area and DRE, about a factor of four higher. [13] The different behavior of the removal processes of volcanic ash mass (Figure 4a) and particle number concentration (Figure 4b) is a consequence of the assumption of a log normal size distribution. Heavier particles, those with greater radii, sink to ground fastest removing a considerable amount of mass, but only a small number of particles by sedimentation (Figure 5). Wet deposition collects particles within and below the clouds by falling raindrops removing also smaller particles out of the higher atmosphere which are subject to slow sedimentation only (Figure 5). Dry deposition represents the removal process close to the surface which is more important for atmospheric trace gases than for volcanic ash, as gases are not subject to gravitational set- 6of10

7 Table 1. Temporal Development of the Contribution of the Removal Processes of Volcanic Ash Mass During the First 5 Days After the Eruption of Kasatochi Volcano 8 August 9 August 10 August 11 August 12 August Total removal (g/d) Sedimentation (%) Wet deposition (%) Dry deposition (%) tling. The distribution of the particle radius in the removal fluxes (Figure 4c) emphasizes this explanation as well as the time series of atmospheric concentration and removal fluxes (Figure 5). Until 2300 UTC on 8 August, when the eruption of Kasatochi volcano stopped, an increase in volcanic mass and particle number is visible in Figure 5. The total mass of volcanic ash removed from the atmosphere exceeds the amount of ash remaining in the atmosphere already by 10 August whereas the particle number concentration changes only slightly during these first days after the eruption as the large particles carrying most of the mass are removed first. For volcanic ash particle number concentration, wet deposition represents the dominant removal process (>99%). [14] The model results presented above are determined with an initial volcanic ash median diameter of 4.0 mm. Therefore, more proximal thickening of tephra is likely to occur resulting from ashes with larger grain size settling close to the volcano. This can be seen in the results of a sensitivity study, which assumes an initial ash diameter of 40 mm (Figure 6). In this case, the 0.1 mm isopach is reduced to an area of km 2 and the area within the 1 mm isopach is considerably increased to km 2 because the majority of the volcanic ash mass is removed from the atmosphere nearly immediately, before 9 August The Kasatochi grid cell receives 3.5 mm deposits when an initial diameter of 40 mm is assumed (only 0.29 mm with an initial diameter of 4 mm). Only fine ash is matter of transport over longer distances. To take into account the effect of quick settling of larger ash particles, two sensitivity studies are carried out with a reduced amount of volcanic ash released with an initial diameter of 4.0 mm (400 and 200 Tg, respectively). The amount of tephra produced in the sensitivity studies shows a linear relationship in comparison with the standard experiment as no interactions between the ash particles are assumed to occur. [15] In another set of sensitivity experiments the impact of the vertical distribution of the injection of volcanic material into the atmosphere on atmospheric SO 2 concentrations has been analyzed. If a maximum emission height of 15 or 16 km is assumed (with 50% of the material injected into the maximum height level), the distribution of atmospheric SO 2 is less perfectly reproduced when compared with satellite data from GOME 2 (Figure 1). In these cases, the injection penetrates into the lower stratosphere, where westerly wind speed was weaker than in the upper troposphere leading to slower advection into eastern direction but an increased advection toward the south during the first few days after the eruption. [16] Measurements of the seawater CO 2 partial pressure at the ocean station Papa (50.0 N, W) showed a considerable decrease of about 40 ppm ( 1 mol/m 3 ) after 7 August 2008 ( papa/data_145w_all.htm). Such a sharp decrease has not been observed in 2007, when these measurements started, nor in It is generally an indicator for the production of biomass; for example, phytoplankton. Langmann et al. [2010] presented strong evidence for the fertilization of the iron limited NE Pacific Ocean by volcanic ash from the eruption of Kasatochi volcano in August 2008 which stimulated considerable phytoplankton production. Based on the modeled tephra fallout, we estimate here if sufficient iron was available at the ocean station Papa for phytoplankton production and the associated reduction of seawater CO 2. Assuming an ocean mixed layer depth of 30 m and a release of 200 nmol Fe per gram volcanic ash (a typical value for subduction zone volcanoes; see Olgun et al., submitted manuscript, 2010) the resulting surface ocean iron concentration at Papa (Figure 7) is between mmol/m 3. The amount of CO 2 transferred into phyto- Figure 5. REMOTE model results for the temporal development of the total area atmospheric (left) volcanic ash mass, (middle) particle number, and (right) mean radius and its dry and wet deposition and sedimentation from 8 August at 0500 UTC to 11 August at 0500 UTC. 7of10

8 atmospheric distribution of SO 2 and volcanic ash and its fallout after the eruption of Kasatochi volcano in August We find that during the first days after the eruption, atmospheric SO 2 column concentrations as observed by GOME 2 can be reproduced very satisfactory by the model. Furthermore, our analysis of MODIS data shows that volcanic ash is transported along the same pathways as SO 2 thereby emphasizing that both, volcanic ash and SO 2 are initially released into the atmosphere with the same vertical distribution. Given these two observations, we feel confident that our model reasonably simulates the atmospheric distribution of volcanic ash and its removal from the atmosphere. The majority of volcanic ash was removed into the NE Pacific Ocean by gravitational settling with 80% being removed from the atmosphere before 11 August and 47% being removed on 9 August. The amount of ash and that of iron attached to it is sufficient to explain the measured seawater CO 2 decrease at the ocean station Papa in August Therefore the current model simulation study further emphasizes that volcanic ash from the eruption of Kasatochi volcano in August 2008 stimulated considerable phytoplankton production in the NE Pacific Ocean. It is complementary to the study of Langmann et al. [2010] which describes the detection of a large scale phytoplankton bloom in the NE Pacific after the eruption of Kasatochi by analyzing chlorophyll a satellite data from MODIS. [18] For future simulations of volcanic eruption processes, further improvements of the model used for this study are planned. These include the implementation of a physically based approach to determine the plume height of the eruptive column and the vertical distribution of the eruption material following, for example, Folch et al. [2008]. If additionally, volcanic ash would be treated as radiatively active tracer, modifications in atmospheric heating rates during the early phase of a volcanic eruption would allow investigating the effects on SO 2 transport. Another aspect is Figure 6. REMOTE model results for a mode radius of 40 mm. (a) Total tephra thickness (mm/8 d) and (b) averaged vertical profile of volcanic ash mass as function of time (mg/m 2 ) from 8 to 16 August 2008 at 0000 UTC. plankton can be constrained from the C/Fe ratio at which phytoplankton from Fe limited oceanic areas incorporate C and Fe in their tissue. When using the widely applied Fe/C ratio of [Watson, 1997], only a reduction of mol/m 3 CO 2 in seawater can be explained at Papa. However, recent measurements point to a Fe/C ratio of [Cassar et al., 2007] thereby explaining a loss of mol/m 3 CO 2 in seawater, covering the range of seawater CO 2 reduction observed at Papa. 4. Conclusions [17] Applying a regional scale atmosphere/chemistryaerosol model [Langmann et al., 2008], we determined the Figure 7. REMOTE model results for the seawater iron concentration (mmol/m 3 ) as a consequence of volcanic ash fallout. The location of the ocean station Papa is indicated. 8of10

9 to describe volcanic ash by several size distribution functions with different mode radii, thereby including the whole size range from coarse to fine ashes. The model used by Costa et al. [2006] and Folch et al. [2008] takes into account a range of different size distributions, however, as no coagulation or other interaction processes between the different modes are considered until now, this approach equals sequential model applications using different mode radii. Taking into account the interaction of volcanic ash with cloud microphysics in addition, would also allow to study processes like those observed after the eruption of Mt. St. Helens [Durant et al., 2009] where ice nucleation occurs on volcanic ash particles which act as ice nuclei and lead to subsequent ice crystal growth directly from the vapor phase and the formation of Mammatus clouds. According to Durant et al. [2008], early ice formation is ubiquitous in volcanic clouds leading to rapid ash fallout. [19] Acknowledgments. This work is funded within the Excellence Cluster CLiSAP of the University of Hamburg. K.Z. is funded by the Geotechnology project Exupery. GOME 2 satellite data were provided by DLR Oberpfaffenhofen. References Cassar, N., M. L. Bender, B. A. Barnett, S. Fan, W. J. Moxim, H. Levy II, and B. Tilbrook (2007), The Southern Ocean biological response to aeolian iron deposition, Science, 317, , doi: /science Costa, A., G. Macedonio, and A. Folch (2006), A three dimensional Eulerian model for transport and deposition of volcanic ashes, Earth Planet. Sci. Lett., 241, , doi: /j.epsl Duggen, S., P. Croot, U. Schacht, and L. Hoffmann (2007), Subduction zone volcanic ash can fertilize the surface ocean and stimulate phytoplankton growth: Evidence from biogeochemical experiments and satellite data, Geophys. Res. Lett., 34, L01612, doi: /2006gl Durant, A. J., R. A. Shaw, W. I. Rose, Y. Mi, and G. G. J. Ernst (2008), Ice nucleation and overseeding of ice in volcanic clouds, J. Geophys. Res., 113, D09206, doi: /2007jd Durant, A. J., W. I. Rose, A. M. Sarna Wojcicki, S. Carey, and A. C. M. Volentik (2009), Hydrometeor enhanced tephra sedimentation: Constraints from the 18 May 1980 eruption of Mount St. Helens, J. Geophys. Res., 114, B03204, doi: /2008jb Fierstein, J., and M. Nathenson (1992), Another look at the calculation of fallout tephra volumes, Bull. Volcanol., 54, , doi: / BF Folch, A., O. Jorba, and J. Viramonte (2008), Volcanic ash forecast Application to the May 2008 Chaiten eruption, Nat. Hazards, 8, , doi: /nhess Jones, M. T., and S. R. Gislason (2008), Rapid releases of metal salts and nutrients following the deposition of volcanic ash into aqueous environments, Geochim. Cosmochim. Acta, 72, , doi: / j.gca Karagulian, F., L. Clarisse, C. Clerbaux, A. J. Prata, D. Hurtmans, and P. F. Coheur (2010), Detection of volcanic SO 2,ash,andH 2 SO 4 using the Infrared Atmospheric Sounding Interferometer (IASI), J. Geophys. Res., 115, D00L02, doi: /2009jd Kristiansen, N. I., et al. (2010), Remote sensing and inverse transport modeling of the Kasatochi eruption cloud, J. Geophys. Res., doi: /2009jd013286, in press. Kutterolf, S., A. Freundt, and W. Peréz (2008), Pacific offshore record of plinian arc volcanism in Central America: 2. Tephra volumes and erupted masses, Geochem. Geophys. Geosyst., 9, Q02S02, doi: / 2007GC Langmann, B. (2000), Numerical modelling of regional scale transport and photochemistry directly together with meteorological processes, Atmos. Environ., 34, , doi: /s (00)00114-x. Langmann, B. (2007), A model study of the smoke haze influence on clouds and warm precipitation formation in Indonesia 1997/1998, Atmos. Environ., 41, , doi: /j.atmosenv Langmann, B., S. Varghese, E. Marmer, E. Vignati, J. Wilson, P. Stier, and C. O Dowd (2008), Aerosol distribution over Europe: A model evaluation study with detailed aerosol microphysics, Atmos. Chem. Phys., 8, , doi: /acp Langmann, B., M. Hort, and T. Hansteen (2009), Meteorological influence on seasonal and diurnal variability of Nicaraguan volcanic emission dispersion: A numerical model study, J. Volcanol. Geotherm. Res., 182,34 44, doi: /j.jvolgeores Langmann, B., K. Zaksek, M. Hort, and S. Duggen (2010), Volcanic ash as fertiliser for the surface, Atmos. Chem. Phys., 10, , doi: /acpd Loyola, D., J. van Geffen, P. Valks, T. Erbertseder, M. van Roozendael, W. Thomas, W. Zimmer, and K. Wißkirchen (2008), Satellite based detection of sulfur dioxide from recent eruptions in Central and South America, Adv. Geosci., 14, 35 40, doi: /adgeo Majewski, D. (1991), The Europa Model of the Deutscher Wetterdienst, Seminar Proc. Eur. Cent. for Medium Range Weather Forecast., vol.2, pp , Reading, U.K. Martin, J. H., and S. E. Fitzwater (1988), Iron deficiency limits phytoplanktongrowthinthenorth east Pacific subarctic, Nature, 331, , doi: /331341a0. Mastin, L. G. (2007), A user friendly one dimensional model for wet volcanic plumes, Geochem. Geophys. Geosyst., 8, Q03014, doi: / 2006GC Mellor, B., and T. Yamada (1974), A hierarchy of turbulence closure models for planetary boundary layers, J. Atmos. Sci., 31, , doi: / (1974)031<1791:ahotcm>2.0.co;2. Mesinger, F., and A. Arakawa (1976), Numerical methods used in atmospheric models, GARP Publ. Ser., 17, Pfeffer, M. A., B. Langmann, and H. F. Graf (2006), Atmospheric transport and deposition of Indonesian volcanic emissions, Atmos. Chem. Phys., 6, , doi: /acp Prata, A. J. (1989a), Observations of volcanic ash clouds using AVHRR 2 radiances, Int. J. Remote Sens., 10, , doi: / Prata, A. J. (1989b), Radiative transfer calculations for volcanic ash clouds, Geophys. Res. Lett., 16, , doi: /gl016i011p Prata, A. J., G. Gangale, L. Clarisse, and F. Karagulian (2010), Ash and sulphur dioxide in the 2008 eruptions of Okmok and Kasatochi: Insights from high spectral resolution satellite measurements, J. Geophys. Res., doi: /2009jd013556, in press. Rix, M., et al. (2009), Satellite monitoring of volcanic sulfur dioxide emissions for early warning of volcanic hazards, IEEE, Journal of selected topics in applied earth observations and remote sensing, 2, Roeckner, E., K. Arpe, L. Bengtsson, M. Christoph, M. Claussen, L. Dümenil, M. Esch, M. Giorgetta, U. Schlese, and U. Schulzweida (1996), The atmospheric general circulation model ECHAM 4: Model description and simulation of present day climate, MPI Rep. 218, Max Planck Inst. for Meteorol., Hamburg, Germany. Rose, W. I., and A. J. Durant (2009), Fine ash content of explosive eruptions, J. Volcanol. Geotherm. Res., 186, 32 39, doi: /j. jvolgeores Schmincke, H. U. (2004), Volcanology, Springer Verlag, Berlin. Seinfeld, J. H., and S. N. Pandis (1998), Atmospheric Chemistry and Physics: From Air Pollution to Climate Change, Wiley Interscience, London. Smolarkiewicz, P. K. (1983), A simple positive definite advection scheme with small implicit diffusion, Mon. Weather Rev., 111, , doi: / (1983)111<0479:aspdas>2.0.co;2. Stier, P., et al. (2005), The aerosol climate model ECHAM5_HAM, Atmos. Chem. Phys., 5, , doi: /acp Tiedtke, M. (1989), A comprehensive mass flux scheme for cumulus parameterisation in large scale models, Mon. Weather Rev., 117, , doi: / (1989)117<1779:acmfsf>2.0.co;2. Vignati, E., J. Wilson, and P. Stier (2004), M7: An efficient size resolved aerosol microphysics module for large scale aerosol transport models, J. Geophys. Res., 109, D22202, doi: /2003jd Walcek, C. J., and G. R. Taylor (1986), A theoretical method for computing vertical distributions of acidity and sulfate production within cumulus clouds, J. Atmos. Sci., 43, , doi: / (1986)043< 0339:ATMFCV>2.0.CO;2. Watson, A. J. (1997), Volcanic iron, CO2, ocean productivity and climate, Nature, 385, , doi: /385587b0. Watt, S. F. L., D. M. Pyle, T. A. Mather, R. S. Martin, and N. E. Matthews (2009), Fallout and distribution of volcanic ash over Argentina following the May 2008 explosive eruption of Chaitén, Chile, J. Geophys. Res., 114, B04207, doi: /2008jb Waythomas, C. F., S. G. Prejean, and D. J. Schneider (2008), Small volcano, big eruption: Scientists rescued just in time, U.S. Dep. of the Interior, 9of10

10 Washington, D. C. (Available at Kasatochi08/Kasatochi2008PLW.php.) Weibull, W. (1951), A statistical distribution function of wide applicability, J. Appl. Mech., 18, Wesely, M. L. (1989), Parameterization of surface resistances to gaseous dry deposition inregional scale numerical models, Atmos. Environ., 23, , doi: / (89) Wiesner, M. G., Y. Wang, and L. Zheng (1995), Fallout of volcanic ash to the deep South China Sea induced by the 119 eruption of Mount Pinatubo (Philippines), Geology, 23, , doi: / (1995) 023<0885:FOVATT>2.3.CO;2. Wiesner, M. G., A. Wetzel, S. G. Catane, E. L. Listanco, and H. T. Mirabueno (2004), Grain size, areal thickness and controls on sedimentation of the 1991 Mount Pinatubo tephra layer in the South China Sea, Bull. Volcanol., 66, , doi: /s x. M. Hort, B. Langmann, and K. Zakšek, Institute of Geophysics, University of Hamburg, Klimacampus, Bundestrasse 55, D Hamburg, Germany. (baerbel.langmann@zmaw.de) 10 of 10