Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Retrieval of sulphur dioxide from the infrared atmospheric sounding interferometer (IASI)

Retrieval of sulphur dioxide from the infrared atmospheric sounding interferometer (IASI) Atmos. Meas. Tech., 5, 581–594, 2012 Atmospheric www.atmos-meas-tech.net/5/581/2012/ Measurement doi:10.5194/amt-5-581-2012 © Author(s) 2012. CC Attribution 3.0 License. Techniques Retrieval of sulphur dioxide from the infrared atmospheric sounding interferometer (IASI) 1 1 1,2 2 1 1 L. Clarisse , D. Hurtmans , C. Clerbaux , J. Hadji-Lazaro , Y. Ngadi , and P.-F. Coheur Spectroscopie de l’Atmosphere, ` Service de Chimie Quantique et Photophysique, Universite ´ Libre de Bruxelles, Brussels, Belgium UPMC Univ. Paris 6, Universite ´ Versailles St.-Quentin, CNRS/INSU, LATMOS-IPSL, Paris, France Correspondence to: L. Clarisse ([email protected]) Received: 15 November 2011 – Published in Atmos. Meas. Tech. Discuss.: 7 December 2011 Revised: 23 February 2012 – Accepted: 6 March 2012 – Published: 13 March 2012 Abstract. Thermal infrared sounding of sulphur dioxide fraction of the emissions makes it to the upper troposphere (SO ) from space has gained appreciation as a valuable com- and lower stratosphere (UTLS), a large volcanic eruption plement to ultraviolet sounding. There are several strong reaching the UTLS can impact the climate significantly as absorption bands of SO in the infrared, and atmospheric the lifetime of sulfate aerosol is proportional to the injection sounders, such as AIRS (Atmospheric Infrared Sounder), altitude. Bottom up approaches are well suited to determine TES (Tropospheric Emission Spectrometer) and IASI (In- total emissions of anthropogenic SO and emissions of some frared Atmospheric Sounding Interferometer) have the abil- degassing volcanoes, but quantifying UTLS SO emissions ity to globally monitor SO abundances. Most of the ob- is best done directly via satellite measurements (Bluth et al., served SO is found in volcanic plumes. In this paper we 1993). In this paper we detail a novel algorithm for calculat- outline a novel algorithm for the sounding of SO above ing SO columns above the mid troposphere (500 hPa) from 2 2 ∼5 km altitude using high resolution infrared sounders and infrared (IR) satellite measurements. apply it to measurements of IASI. The main features of the Apart from climatological relevance, measuring high al- algorithm are a wide applicable total column range (over 4 titude SO is also important for studying uplift of anthro- orders of magnitude, from 0.5 to 5000 dobson units), a low pogenic pollution (e.g. Clarisse et al., 2011b), for analyz- theoretical uncertainty (3–5 %) and near real time applicabil- ing explosive volcanic eruptions (e.g. Carn and Prata, 2010), ity. We make an error analysis and demonstrate the algorithm and, when data are available in near real time, for monitoring on the recent eruptions of Sarychev, Kasatochi, Grimsvotn, ¨ volcanic activity (e.g. Surono et al., 2012) and tracking of Puyehue-Cordon ´ Caulle and Nabro. volcanic clouds for the mitigation of aviation hazards (Prata, 2008; Rix et al., 2009; Carn et al., 2009). Since 1978, the Total Ozone Mapping Spectrometer (TOMS) (Krueger et al., 1995) and subsequent follow-up 1 Introduction ozone monitoring instruments have been measuring SO Prodigious amounts of sulphur dioxide (SO ) are released through solar backscattered ultraviolet (BUV) measurements every year in the atmosphere. Anthropogenic emissions, (see e.g. Yang et al., 2007, and references therein). BUV mostly coming from combustion of sulphur-rich biomass measurements have a good sensitivity to SO , even in the −1 such as coal and petroleum, add up to 50–65 Tg S yr lowest atmospheric layers. The record of IR sounding of SO (Smith et al., 2011; Lee et al., 2011). Volcanoes are the also goes back to 1978 with the High-Resolution Infrared largest natural source of sulphur dioxide and account for Sounder (HIRS/2) (Prata et al., 2003). One clear advantage −1 7.5–10.5 Tg S yr on average (Andres and Kasgnoc, 1998; of thermal infrared (TIR) instruments is that they can mea- Halmer et al., 2002). These emissions lead to acid deposi- sure in the absence of sunlight (thus also at night and at high tion and can affect air quality and climate through the for- latitudes in the winter) and often have a higher spatial res- mation of sulfate aerosols (Longhurst et al., 1993; Chin and olution. For an overview of satellite instruments capable of Jacob, 1996; Graf et al., 1997; Haywood and Boucher, 2000; measuring SO and their characteristics and limitations, we Robock, 2000; Zhang et al., 2007). While in general only a refer to Thomas and Watson (2010). Here we give a short Published by Copernicus Publications on behalf of the European Geosciences Union. 582 L. Clarisse et al.: Retrieving SO from IASI measurements overview of TIR sounding of SO without going into instru- lookup table. Our algorithm is akin to some of the meth- mental specifics. ods applied for broadband sensors. The advantage, however, Sulphur dioxide has three absorption bands in the mid in- is that we can select specific channels, making the algorithm frared, see Fig. 1. The ν is by far the strongest band. Com- simpler and less sensitive to changes of other atmospheric peting water vapor absorption limits its vertical sensitivity to variables (water vapour, clouds). SO above 3–5 km, depending on the humidity profile and We outline the algorithm for observations of the high res- SO abundance. Higher altitude SO is also affected, di- olution infrared sounder IASI (Clerbaux et al., 2009), but it 2 2 rectly, by water vapor in and above the SO layer, but also can easily be transferred to other high resolution sounders. indirectly by variable radiation coming from below. The ν Instrumental specifics of the IASI instrument are a continu- −1 band is situated in an atmospheric window, and can pen- ous spectral coverage between 645 and 2760 cm , a spectral −1 −1 etrate the lower troposphere. While water vapor is not as resolution of 0.5 cm (which is apodized at 0.25 cm ) and −1 important here, the 800–1200 cm region is very sensitive a noise equivalent delta temperature at 280 K around 0.05 K to the surface temperature, surface emissivity and volcanic for the ν band and 0.12 K for the ν band. It has a global 3 1 ash (Clarisse et al., 2010a,b), and for young volcanic plumes coverage twice a day with a footprint ranging from circular from explosive eruptions, SO and ash often need to be re- (12 km diameter at nadir) to elliptical (up to 20 by 39 km at trieved simultaneously. The combination band ν + ν can the end of the swath) and a mean local equatorial overpass 1 3 only be used when there is reflected solar light. It is weak, time at 09:30 LT and 21:30 LT. but has been applied for the study of major volcanic erup- In the next section we outline the theoretical basis of the tions as an alternative to a saturating ν band (Karagulian algorithm. In Sect. 3 we give an overview of the most impor- et al., 2010; Prata et al., 2010). Note that all TIR measure- tant sources in the error budget. Examples are presented in ments require thermal contrast between the SO plume and Sect. 4 and we conclude in Sect. 5. the underlying source of radiation. Broadband instruments typically have a handful of chan- −1 2 The algorithm nels (each covering 50–100 cm ) which can be used to re- trieve SO . Most retrieval algorithms are based on approx- In what follows, we assume an atmosphere with a SO cloud imating the SO affected bands from the other bands as- present at a given altitude. We adopt the notations from Wat- suming the absence of SO . The difference between these son et al. (2004). When the plume is at sufficient altitude reconstructed background radiances and the observed radi- (where the absorption of other species can be ignored) the ances can then be used to infer abundances. In the case of measured radiance L at a wavenumber ν (and correspond- the ν band this can be done by first estimating the surface ing measured brightness temperature at the sensor T ) can be temperature (Realmuto et al., 1994, 1997) or by assuming a approximated as linear correlation with another band (Prata and Kerkmann, 2007). For the ν band it has been shown that it is possi- 3 L (ν)= L (ν)t +L (ν)(1−t ), (1) s ucb c c c ble to estimate the relevant unperturbed band radiance from with L (ν) = B (ν,T ) the ambient radiance coming from c c a linear interpolation of two other bands (Prata et al., 2003; the cloud at temperature T and specified by Planck’s law, Doutriaux-Boucher and Dubuisson, 2008). Other schemes L (ν) the upwelling radiance at the cloud base and t the ucb c rely on the use of a large series of simulated radiances (see transmission of the cloud, given by the Bouguer-Lambert- e.g. Corradini et al., 2010). For retrievals using the ν band, Beer law explicit (Corradini et al., 2009) or implicit (Campion et al., 2010) corrections for aerosols can be made. −cu t = e , (2) Retrievals using high spectral resolution instruments typi- cally use (optimal) least square procedures (Carn et al., 2005; with c an absorption coefficient dependent on pressure and Prata and Bernardo, 2007; Clerbaux et al., 2008; Clarisse temperature and u the column abundance. While Eq. (1) et al., 2008), preceded by a SO detection routine. These is valid under the mentioned assumptions, a subtlety arises are time consuming, but have the advantage of fully exploit- when applying it to real measurements. Real radiance ing the spectral resolution by simultaneously retrieving com- measurements are always integrated (convolved) over a peting species (e.g. H O) and potentially extracting plume wavenumber interval and are altered by the instrumental line altitude information. It was shown (Karagulian et al., 2010; shape. To check to what extent Eq. (1) holds at the level Haywood et al., 2010) that for the ν band it often suffices of finite microwindows (here IASI channels), we have sim- to perform optimal estimation on a selected number of pixels ulated the radiative transfer of a standard atmosphere and and exploit the empirical correlation between these retrieved introduced a SO layer at a fixed altitude, but with varying total columns and brightness temperature differences. It is abundances. The results are shown in Fig. 2 in brightness −1 this scheme we generalize and put on a more solid theoretical temperature space at wavenumber ν = 1371.75 cm . The footing. Instead of relying on optimal estimation retrievals, simulations are shown as black squares and the best fit with however, we use elementary radiative transfer and a large Eq. (1) (best choice of the absorption coefficient c) is shown Atmos. Meas. Tech., 5, 581–594, 2012 www.atmos-meas-tech.net/5/581/2012/ L. Clarisse et al.: Retrieving SO from IASI measurements 583 Fig. 1. Top panel: example IASI spectrum measured over the plume of the August 2008 eruption of Kasatochi. Bottom panel: line positions and intensities of SO from HITRAN (see Rothman et al., 2009, and references therein). Band centers and inte- −1 grated band intensities of SO are (see Flaud et al., 2009, and references therein): the ν symmetric stretch (∼1152 cm = 8.7 μm at 2 1 −17 −1 −2 −1 −17 −1 −2 0.35× 10 cm /(molecule cm )), the ν asymmetric stretch (∼1362 cm = 7.3 μm at 2.72× 10 cm /(molecule cm )) and the −1 −17 −1 −2 ν + ν combination band (∼2500 cm = 4 μm at 0.054× 10 cm /(molecule cm )). 1 3 −1 Fig. 2. Brightness temperature at 1371.75 cm as a function of SO mass loading for a low (left, plume at 247 K and 450 hPa ∼ 5 km) and high (right, plume at 230 K and 10 hPa ∼25 km) altitude plume. The colored black squares were calculated from simulated IASI spectra, while the red full line is a best fit of these simulations with Eq. (1). www.atmos-meas-tech.net/5/581/2012/ Atmos. Meas. Tech., 5, 581–594, 2012 584 L. Clarisse et al.: Retrieving SO from IASI measurements in red. For a plume at high pressure (left panel, 450 hPa), an almost perfect fit can be obtained. The asymptotic behavior for increasingly large abundances can also be observed (L (ν)→ B (ν, T ) or T → T ). This saturation is slower for c s c lower pressure (right panel, 10 hPa). At very low pressure, spectral lines saturate at a lower concentration at their line centers than their wings. In contrast, at a higher pressure, pressure broadening of the individual lines is important and will distribute absorption over a wider spectral range, result- ing in a net larger absorption and thus a quicker saturation over the complete band when taking into account all spectral lines. For the low pressure test case, a good fit with Eq. (1) and a constant absorption coefficient c is not possible. Be- cause of the lower pressure broadening, the instrumental line shape and apodisation become relatively more important, and these effects are not taken into account in Eq. (1). One way Fig. 3. Brightness temperature of the two sets of absorption chan- to resolve this is to introduce an explicit column dependence −1 −1 nels (at ∼1371.5 cm and at ∼1385 cm ) as a function of SO in the coefficient c, so that c = c (T , P , u). These coefficients abundance for a plume located at 150 hPa and 207 K. can be estimated from forward simulations as outlined below. To determine the SO abundance from Eq. (1), all that is altitude, some residual water absorption can still affect ob- left is to estimate L (ν). This can be done from channels ucb served channels. Assuming that water vapour above is colder not affected by SO , but for which the channel ν responds than the SO plume (so disregarding significant water vapor similarly to H O and other atmospheric parameters than the above lower stratospheric plumes), we have for a saturating channels sensitive to SO . It is here easier to work in bright- cloud T < T . We therefore introduce a virtual cloud temper- s c ness temperature space, where Eq. (1) reads ∗ 21 ature T = T − [H O]/10 , with [H O] the partial column c 2 2 −2 B(T ,ν)= B(T ,ν)t +B(T ,ν)(1−t ). (3) of water (in molecules cm ) above the SO layer . The fac- s ucb c c c tor 10 was determined empirically, and while this is a first Now T can be estimated from another channel ν when for ucb order correction, it is largely sufficient as we will see below. background concentrations of SO To calculate the absorption coefficients c (T ,P ,u) we have −1 −1 0 0 T = B (L (ν),ν)≈ B (L (ν ),ν )= T . (4) used representative atmospheric profiles (temperature, pres- s s s ucb sure, humidity and ozone) from the ECMWF 40-yr reanaly- The critical part is to choose these channels ν and ν to make sis, ERA-40 (Chevallier, 2001). The total set contains 13 495 this estimate as good as possible. We have used combina- well sampled profiles. Pressure and temperature (PT) pairs tions of 4 channels: two to estimate T , representing the ab- between 5 and 30 km altitude are plotted in Fig. 4. The vis- sorption in the ν band and two reference channels to es- ible pressure bands are an artifact caused by the specific 60- timate T . Table 1 lists two sets of such parameters to- ucb level coordinate system in the data set, and these disappear gether with their bias and standard deviation (estimated from when working with the interpolated data. We have calculated a full day of IASI measurements with no detectable volcanic c (T ,P ,u) on a subgrid of this PT diagram, indicated by the SO ). Note that this doubling of channels allows to reduce black dots. the standard deviation significantly and also that the bias can For each PT pair in the subgrid, we selected 10 atmo- be subtracted in the calculation of the brightness tempera- spheres from ERA-40 with the closest match in the PT pro- ture difference. Figure 3 illustrates the sensitivity range of file. A variable SO cloud (from 0 to 10 000 DU) was then in- both sets for a plume at 150 hPa. The absorption channels in serted at the altitude corresponding to the PT pair and the re- the ν band of the first set are chosen close to the region of sulting IASI spectrum was simulated. Based on these simula- −1 maximum absorption, around 1371.75 cm . It is sensitive tions a best value for c (T ,P ,u) was obtained from minimiz- to mass loadings as low as 0.5 DU, but saturates at around ing the relative error between the real and the calculated SO 200 DU, above which differences in the observed channels abundance. Each c (T ,P ,u) is obtained from 10 independent become too small. The second set has its absorption chan- simulations and determining the best value is therefore an −1 nels further away from the band center, at 1385 cm . It has over-constrained problem. The solution however is guaran- a lower sensitivity of about 10 DU, but can measure columns teed not to be overly dependent on an individual atmosphere, up to 5000 DU. The combined use of both sets therefore en- and the average relative error is a good indication for the the- ables to retrieve columns of SO from about 0.5 to 5000 DU oretical error (caused by the variability of other atmospheric at 150 hPa. parameters) which can be achieved with this algorithm. Equation (1) is only valid when no absorption above the SO plume takes place. Even at altitudes above ∼500 hPa Atmos. Meas. Tech., 5, 581–594, 2012 www.atmos-meas-tech.net/5/581/2012/ L. Clarisse et al.: Retrieving SO from IASI measurements 585 Table 1. Two sets of absorption and background channels used in the calculation of SO abundances. The mean and standard deviation of their brightness temperature differences were calculated on one day with no detectable quantities of SO . ν Absorption channels Background channels Mean Std −1 −1 Set 1 1371.50, 1371.75 cm 1407.25, 1408.75 cm −0.05 K 0.15 K −1 −1 Set 2 1384.75, 1385.00 cm 1407.50, 1408.00 cm 0.05 K 0.25 K Fig. 4. Pressure and temperature correlations of the ERA-40 data set between 5 and 30 km. The black dots are the PT pairs for which the lookup tables were built. The top panel in Fig. 5 shows the absorption coefficients (due to the smooth and monotonous behavior of the c coeffi- for the two sets of channels at 10 and 750 DU respectively. cients). Also note that we find two estimates u and u for u, 1 2 For 4 PT pairs, T was very close or inferior to T for all 10 for each set of absorption and background channels. Theo- ucb c profiles. These sets of low thermal contrast or temperature retically, these two estimates should only agree when the as- inversion were excluded. These are situated at the very edge sumed altitude corresponds to the real altitude (because the of the PT space and are uncommon. The bottom panel shows corresponding brightness temperature differences have a dif- the mean relative error between the input SO abundance and ferent pressure and temperature dependence). From looking the retrieved for the ten different profiles. Errors are less than at a few test cases, the two estimates generally agree well be- 3 % and 5 % for the first and second set respectively, except tween 25 DU and 75 DU (with a standard deviation of around again at some points at the edge of the PT space. 10 %). On either side of this range, differences increase, with the u estimate obviously superior for lower total col- We end this section with a practical consideration, which umn amounts and the u estimate by construction superior is important in the implementation of the above retrieval al- for large column total amounts. When either u or u exceed 1 2 gorithm. The use of c (T ,P ,u) to calculate the column abun- 100 DU, we used the u estimate, otherwise u was used. Fi- 2 1 dance u is inherently a recursive problem. It is therefore nec- nally, the retrieval is also preceded by a detection criterion, essary to start with a first guess c (T ,P ) and iteratively calcu- here taken to be T − T > 0.4 K. ucb s late u and c (T ,P ,u) until convergence is achieved. We have verified numerically that this convergence is always achieved www.atmos-meas-tech.net/5/581/2012/ Atmos. Meas. Tech., 5, 581–594, 2012 586 L. Clarisse et al.: Retrieving SO from IASI measurements 4 4 x 10 x 10 Absorption coefficients c Absorption coefficients c 6 6 2 5 5 −3 x 10 0.03 4 4 3 3 0.025 2 2 0.02 1 1 0.015 0 0 180 200 220 240 260 280 180 200 220 240 260 280 4 4 x 10 x 10 Relative errors set 1 Relative errors set 2 6 6 5 5 15 15 4 4 3 3 10 10 2 2 5 5 1 1 0 0 0 0 180 200 220 240 260 280 180 200 220 240 260 280 Temperature (K) Temperature (K) Fig. 5. Absorption coefficients for the two sets of IASI channels (top) and their corresponding average errors in percentage (bottom). Here the absorption coefficients and errors are shown for a SO cloud of 10 DU (set 1) and 750 DU (set 2) respectively. 3 Sources of error 3.1 Measurements errors A good description of typical sources of error can be found We call measurements errors any errors that affect the dif- in Prata et al. (2003). Most of these are inherent to any re- ference of T and T beyond the contribution of SO . This s ucb 2 trieval which uses the ν band. There are broadly speaking 3 includes the instrumental noise, but also contributions from five main sources of error. The first category is related to the fact that the background channels are only a best-effort propagation of errors in the measurements, in our case in the estimate of the absorption channels in the absence of SO . measurements of T and T . The second category includes s ucb Following Table 1 we estimate the error to be of the order errors related to the assumed or measured altitude or cloud 0.15 K and 0.25 K for the first and second set of channels. temperature T . A third source of errors becomes important c From Fig. 3 it is easily seen that the influence of these errors when Eq. (1) is no longer a good approximation for the radia- will be largest for very thin or very thick SO clouds. For tive transfer due to presence of aerosols above the SO layer. 2 very thin clouds the contribution of SO on T will be of the 2 s There is the modeling error related to Eq. (1), which was es- same order of magnitude as the measurement error and hence timated above to be in the range 3–5 %, and finally there are relatively important. For very thick clouds, we are close to errors related to spectroscopy and radiative transfer. In this saturation regime and a small error on the observed temper- section we will discuss the first three types of error. atures will lead to large differences in the SO estimates. As an example of how this type of error translates in errors on the abundance, an error of 0.15 K and 0.25 K was introduced Atmos. Meas. Tech., 5, 581–594, 2012 www.atmos-meas-tech.net/5/581/2012/ Pressure (Pa) Pressure (Pa) L. Clarisse et al.: Retrieving SO from IASI measurements 587 Fig. 6. Illustration of the measurement error. Relative errors in Fig. 7. Effect of the assumed altitude on retrieved abundances; il- the retrieved abundances, made from introducing 0.15 K and 0.25 K lustrated for different eruptions. error in the data of Fig. 3. the temperature contrast is highest and the minimum amount in the data of Fig. 3 and the relative differences are plotted in of SO is required to account for the observed absorption. Fig. 6. It illustrates the increase of errors near the extremes. For instance the Merapi (Java/Indonesia) and the Nabro (Er- The errors between 0.5 DU and 5000 DU are in this exam- itrea) plumes have their minimum retrieved mass at a higher ple below 30 % (and below 6 % for loadings above 3 DU). altitude (17 km) than e.g. Sarychev (Kuril Islands, Russia) It should be stressed though that this type of error is a ran- or Kasatochi (Aleutian Islands, Alaska) plumes, which have dom error and averages out when calculating the total mass their minimum at 10–12 km. In the stratosphere the SO re- of plumes much larger than the footprint of the instrument. trievals increase as T approaches T , with the rate of in- c ucb Related to this, there is the situation where the SO cloud crease controlled by the stratospheric temperature gradient. at T has little or no thermal contrast with the radiation from As can be seen from Fig. 7, the effect of altitude is gener- below T . In this case (see again Fig. 3) the regime of low ucb ally within 10–20 % between 10 and 20 km. For low altitude sensitivity and the regime of saturation overlap and errors are plumes, the assumed altitude is more critical with differences naturally very large. This dependence on thermal contrast is up to 500 % between a plume at 5 and 10 km due to the steep inherent to IR sounding. temperature gradient in the troposphere. 3.2 Altitude 3.3 Aerosols As the present algorithm does not retrieve altitude, a cloud altitude (and therefore pressure and temperature) must be as- Large eruption plumes contain typically a large amount of sumed. This affects the estimated loading through the as- various particles (ash, ice, sulfate aerosols and aggregates). sumed water vapour absorption above the plume, c (T ,P ,u) All these absorb and scatter IR radiation. The wavenum- and T . The latter is the most important, especially close to ber dependence is most pronounced for ash and ice as illus- saturation or when considering large temperature differences. trated in Fig. 8 for the 2008 Kasatochi eruption (ash) and To assess their combined effect it is best to look at some ex- 2011 Nabro eruption (ice). Extinction of IR radiation by ash −1 amples. Figure 7 is a plot of retrieved total masses (as a is strongest in the 800–1200 cm range (see also Clarisse percentage of the maximum measured total mass for a given et al., 2010b), but almost uniform throughout the ν band of altitude) for different eruptive plumes (young and aged) as a SO . Note that the specific extinction depends on the total function of the assumed altitude. ash loading but also on the particle size distribution and the To understand the effect of the assumed altitude it is use- mineral composition. Ice particles have their largest extinc- −1 ful to look at the thermal contrast between the cloud and the tion feature in the 800–1000 cm range (see also Clarisse background (T − T ). Water vapour is the main source of et al., 2008). The retrieval algorithm is not sensitive to what ucb c the upwelling radiance in the vicinity of the ν band and happens below the SO cloud as long as the radiation com- 3 2 T therefore corresponds to an altitude of 3–6 km. At ing from below has sufficient thermal contrast with the SO ucb 2 cloud altitudes between 5 and 7 km the temperature contrast plume and as long as the radiation at background and absorp- is low and maximum amounts of SO are required to pro- tion channels extinguishes uniformly. Low-altitude aerosol duce the observed absorption. For clouds at the tropopause layers of low-to-medium optical thickness which are located www.atmos-meas-tech.net/5/581/2012/ Atmos. Meas. Tech., 5, 581–594, 2012 588 L. Clarisse et al.: Retrieving SO from IASI measurements Fig. 8. Observed IASI spectra from the 2008 Kasatochi (top) and 2011 Nabro eruptions (bottom) illustrating the effect of volcanic aerosol (ash and ice). Different degrees of aerosol extinction demonstrate the damping of the SO signature. Note that some of the differences are also due to the different total column of water vapour. well below the SO layer have therefore limited or no impact aerosol loading and its altitude, and while our tests point to on our retrieval. Opaque aerosol layers just below the SO an overestimation of the SO loading, pixels with completely 2 2 plume impede the sensitivity of the algorithm as is apparent opaque ash in or above the SO layer will go undetected and when comparing the black and the blue spectra in the top this will lead to an underestimation of the total measured SO panel of Fig. 8. mass. An example of such a spectrum is shown in pink in the top of Fig. 8. A little SO can be detected at ∼225 K above Aerosols above or at the same altitude as SO will have 2 the ash cloud at ∼220 K, but everything below the ash cloud an impact on the retrieved abundance. As a test case, we is not measurable. have simulated the radiative transfer (following the methods Note finally that for fresh plumes, it is not uncommon for described in Clarisse et al., 2010a) of a thick aerosol layer a portion of the erupted SO to be sequestered on ice, only to located below, above and in a 25 DU upper tropospheric SO 2 be later released in the volcanic cloud by sublimation (Rose plume. The aerosol abundance was in the three cases chosen −1 et al., 2004). This could account for some of the increases as to cause a drop of 20 K in the spectrum at ∼1362 cm . in SO total mass timeseries observed in ice rich volcanic As expected and explained above, aerosol below the SO 2 plumes (Krueger et al., 2008; Clarisse et al., 2008). layer had limited impact (2 %) on the retrieved abundance. Ash in the SO layer caused a 20 % overestimation, while aerosol above gave rise to a 45 % overestimation. It is clear that the effect of aerosol depends very much on the specific Atmos. Meas. Tech., 5, 581–594, 2012 www.atmos-meas-tech.net/5/581/2012/ L. Clarisse et al.: Retrieving SO from IASI measurements 589 −180 −175 −170 −165 −160 −155 −150 −180 −175 −170 −165 −160 −155 −150 55 55 50 50 7 August evening 8 August morning 45 45 Total / maximum Total / maximum @5 km : 655 kT / 2884 DU @5 km : 7688 kT / 3842 DU @7 km : 380 kT / 1676 DU @7 km : 3731 kT / 1845 DU SO (DU) @10 km : 135 kT / 586 DU @10 km : 1208 kT / 658 DU SO (DU) @13 km : 105 kT / 429 DU @13 km : 1164 kT / 581 DU @16 km : 92 kT / 403 DU @16 km : 1163 kT / 587 DU 0 100 200 300 0 100 200 300 400 500 @25 km : 186 kT / 1181 DU @25 km : 2139 kT / 1304 DU 40 40 −180 −175 −170 −165 −160 −155 −150 −180 −175 −170 −165 −160 −155 −150 55 55 50 50 8 August evening 9 August morning 45 45 Total / maximum Total / maximum @5 km : 6871 kT / 2085 DU @5 km : 6924 kT / 1802 DU @7 km : 3626 kT / 1066 DU @7 km : 3763 kT / 926 DU @10 km : 1163 kT / 353 DU SO (DU) @10 km : 1340 kT / 329 DU SO (DU) @13 km : 1065 kT / 335 DU @13 km : 1148 kT / 298 DU @16 km : 1082 kT / 369 DU @16 km : 1236 kT / 338 DU 0 100 200 300 0 50 100 150 200 @25 km : 1824 kT / 758 DU @25 km : 2009 kT / 678 DU 40 40 Fig. 9. The eruption of Kasatochi (Aleutian islands) on 7 and 8 August as seen by IASI, with a 5–20 km altitude SO plume drifting to North America. For the displayed columns an altitude of 10 km was assumed. 4 Examples to 2.2 Tg (Kristiansen et al., 2010; Krotkov et al., 2010) from OMI (Ozone Monitoring Instrument). In the IR estimates are 1.2 to 1.4 Tg (Prata et al., 2010) from AIRS (Atmo- 4.1 Kasatochi – large columns spheric Infrared Sounder), 1.7 Tg (Karagulian et al., 2010) from IASI and 0.94 to 2.65 Tg (Corradini et al., 2010) from Kasatochi volcano (part of the Aleutian Islands) erupted on MODIS (Moderate Resolution Imaging Spectroradiometer). 7 and 8 August 2008 five times (Waythomas et al., 2010) The main difficulty in comparing the respective retrievals is and ejected the largest amount of SO in the UTLS since understanding the impact of the different assumed or cal- the eruption of Cerro Hudson in 1991 (Krotkov et al., 2010). culated heights coupled with the different responses in the There are several aspects which complicate the SO retrieval. IR/UV absorption bands. Also important are the different The five eruptions occurred in quick succession, and these strategies applied to cope with non-linear effects associated were different in nature (phreatomagmatic and magmatic with very large columns as also reflected in the large vari- Waythomas et al., 2010) and altitude (5–20 km Kristiansen ance in reported maximum columns, ranging from 100 to et al., 2010). The resulting plume was therefore highly het- 700 DU: OMI 280 DU (Kristiansen et al., 2010), GOME2 erogeneous in SO , H S (Clarisse et al., 2011a), H O, ash 2 2 2 100 DU (operational retrieval)–700 DU (Richter et al., 2009; (Corradini et al., 2010) and ice content and likely multilay- Bobrowski et al., 2010) and IASI 300 DU (Karagulian et al., ered within a typical operational satellite’s footprint (>10 km 2010). diameter). In terms of total ejected SO mass, estimates from satel- Retrieval results using the new algorithm are shown in lites vary widely, from 1 to 3 Tg, in part due to the fact that Fig. 9 for the first 4 IASI overpasses (the first overpass on most retrievals are based on a single cloud altitude. Es- 7 August happened after 3 of the 5 explosive events). In timates in the UV are 2.5 Tg (Richter et al., 2009) from terms of maximum columns, the first overpass on the 8th GOME2 (Global Ozone Monitoring Experiment-2) and 1.4 measured columns in excess of 500 DU (depending on the www.atmos-meas-tech.net/5/581/2012/ Atmos. Meas. Tech., 5, 581–594, 2012 590 L. Clarisse et al.: Retrieving SO from IASI measurements Fig. 10. Time series of SO measured with IASI: original study Fig. 11. Time series of SO measured with IASI: original study (Karagulian et al., 2010) in blue, current reanalysis in red and OMI (Haywood et al., 2010) in black, current reanalysis in red and (Krotkov et al., 2010) in black. HadGEM2 model in blue. injection altitude). This is higher than any other retrieval 4.2 Sarychev – aging plume reported using ν measurements, and of the same order as the maximum columns measured by GOME2 and shows the Another large eruption took place in 2009, namely Sarychev ability of our retrieval algorithm to deal efficiently with band Peak (Kuril Islands, Russia) on 11–16 June (Matoza et al., saturation. Retrieved total masses vary from 3.7 Tg (7 km) 2011; Rybin et al., 2011; Haywood et al., 2010). There were over 1.3 Tg (10–13 km) to 2 Tg (25 km). As the plume was several explosive events, but the majority of the high altitude spread over altitudes ranging from 5 to 20 km, this is again SO was injected on 15 and 16 June at an altitude of 10– consistent with data from other sounders. 16 km. An earlier study (Haywood et al., 2010) using IASI The retrieved total mass at 10 km (Krotkov et al., 2010; data estimated the sulphur dioxide emissions for those two Karagulian et al., 2010) as a function of date is displayed in days to be of the order of 1.2± 0.2 Tg and this figure is com- Fig. 10. We find that the retrieved values increase after the mensurate with OMI measurements (Carn and Lopez, 2011). 9th to about 1.6 Tg on the 11 August. As the presented al- Like the Kasatochi eruption, the eruption of Sarychev peak gorithm is able to retrieve very large SO columns, we do presented a nice validation opportunity for modeling and not except saturation problems to be an issue. However, we measuring lower stratospheric injections of SO and grad- made the assumption that the plume is concentrated at a sin- ual oxidation to sulfate (Haywood et al., 2010; Kravitz et al., gle fixed altitude and so one possible reason for this post- 2011; Vernier et al., 2011). eruptive increase of the total mass is that part of the plume Figure 11 shows the measured total mass in the Northern was vertically stratified, with the ν band mostly sensitive to Hemisphere at 13 km (the total mass does not vary a lot for the upper part (Corradini et al., 2010; Krotkov et al., 2010). height assumptions between 10 and 16 km) as a function of By the 11th vertical wind shear probably dispersed the mul- time in June 2009. The difference with the previous time- tilayered cloud sufficiently for it to be exposed completely. series (shown in black and reported in Haywood et al., 2010) For the overall time-series we find substantial differences is minimal, except for the maximum retrieved value, which is (up to 50 %) with an earlier IASI analysis (Karagulian et al., now determined at 0.9 Tg. The current reanlysis is likely to 2010) which used the ν + ν combination band for the re- 1 3 be more accurate, given the fact that the original retrieved trievals of the plume on the 8th, 9th and 10th. This analy- mass was quite noisy for the first week after the eruption sis did not exhibit an increase in retrieved total masses after (with differences up to 100 % for consecutive overpasses). the 8th (likely due to the weaker altitude/temperature depen- The current retrieval gives a very smooth timeseries. Note dency of the retrievals). However, a remarkable discrepancy also that for the 2009 Sarychev eruption as a whole, the total of this earlier analysis with OMI retrievals was found for re- released SO is likely higher then 0.9 Tg as prior to 15 June trievals of the aged plume, with differences over 100 % and there were several smaller eruptions. leading to factor two in the estimated e-folding lifetime of SO . In the present study measurements of IASI are compat- ible with the OMI retrievals (Krotkov et al., 2010), and the timeseries clearly fits the expected exponential decay better than the earlier analysis. Atmos. Meas. Tech., 5, 581–594, 2012 www.atmos-meas-tech.net/5/581/2012/ L. Clarisse et al.: Retrieving SO from IASI measurements 591 −150 −100 −50 0 50 100 150 −20 −40 −60 −80 0 2 4 6 8 SO (DU) Fig. 12. Composite image of maximum observed SO columns for the period 20 May to 30 June 2011. The value for each grid cell equals the maximum observed SO columns in that grid cell for the given time period during which three major volcanic eruptions took ◦ ◦ ◦ ◦ place. Grimsvotn ¨ (−17.33 , 64.42 ) erupted first on 21 May, then Puyehue-Cordon ´ Caulle (−40.59 , −72.12 ) on 3 June and finally Nabro ◦ ◦ (13.37 , 41.70 ) on 12 June. A plume altitude of 10 km was assumed. 150 160 170 150 160 170 150 160 170 150 160 170 65 65 65 65 60 60 60 60 1.8 55 55 55 55 07.01.2011 09.01.2011 23.01.2011 24.01.2011 1.6 50 50 50 50 150 160 170 150 160 170 150 160 170 150 160 170 65 65 65 65 1.4 60 60 60 60 55 55 55 55 1.2 26.01.2011 29.01.2011 30.01.2011 31.01.2011 50 50 50 50 150 160 170 150 160 170 150 160 170 150 160 170 65 65 65 65 60 60 60 60 0.8 55 55 55 55 0.6 31.01.2011 01.02.2011 01.02.2011 03.02.2011 50 50 50 50 150 160 170 150 160 170 150 160 170 150 160 170 65 65 65 65 0.4 60 60 60 60 0.2 55 55 55 55 03.02.2011 04.02.2011 07.02.2011 17.02.2011 50 50 50 50 Fig. 13. Snapshots of volcanic SO plumes detected in the first part of 2011 over the Kamchatka Peninsula. www.atmos-meas-tech.net/5/581/2012/ Atmos. Meas. Tech., 5, 581–594, 2012 SO (DU) 2 592 L. Clarisse et al.: Retrieving SO from IASI measurements 4.3 Grimsvotn, ¨ Puyehue-Cordon ´ Caulle, Nabro – exploit IASI’s high spectral resolution to the fullest (for esti- global retrievals mating plume altitude see Clarisse et al., 2008; for improved detection see Walker et al., 2011). Ideally therefore, these In May and June 2011, three volcanoes erupted, each releas- algorithms should be used in combination with each other. ing large amounts of SO (Fig. 12). Near real time retrieval Apart from this intrinsic uncertainty associated to the al- using the outlined algorithm illustrates its operational use- gorithm, the accuracy will be determined by knowledge of fulness and robustness for a variety of very different atmo- the plume altitude. This is especially the case in the mid tro- spheric conditions and eruptive plumes. The retrieved total posphere where we have a large temperature gradient. An- masses agree well with those retrieved from other sensors other source of error is the presence of (volcanic) aerosols (AIRS and OMI) as reported on various forums and news and while the magnitude of the associated errors in the re- sites. A full analysis taking into account precise altitude es- trieval is hard to quantify, thin ash clouds will in general timates is out of the scope of this paper and we report total lead to slightly overestimated loadings, while thick opaque masses here assuming an altitude of 10 km. aerosol layers can cover up part or all SO and will give rise Grimsvotn ¨ (Iceland) erupted first on 21 May, with about to underestimates. Although a validation or comparison of this algorithm is 350–400 kT of SO . Last traces of the initial plume were ob- served until the 15 June. SO from Puyehue-Cordon ´ Caulle out of the scope of this paper, we have illustrated the algo- (Chile) was detected first on 5 June; and a fast westerly jet rithm on a number of examples and found that the results stream carried the plume of about 250 kT SO round the were in agreement with the literature. world in 9–10 days. The third eruption was the one of the Acknowledgements. IASI has been developed and built under the volcano Nabro (Eritrea), which was prior to this event be- responsibility of the Centre National d’Etudes Spatiales (CNES, lieved to be totally extinct and is not monitored so actively France). It is flown onboard the Metop satellites as part of the as other volcanoes. First SO was measured on 12 June and EUMETSAT Polar System. The IASI L1 data are received through continued emissions were observed in the days and weeks the EUMETCast near real time data distribution service. L. Clarisse which followed. Total masses of the order 1.5 Tg were mea- and P.-F. Coheur are respectively Postdoctoral Researcher (Charge ´ sured. Water and ice rich plumes and low altitude filaments de Recherches) and Research Associate (Chercheur Qualifie) ´ with F.R.S.-FNRS. C. Clerbaux is grateful to CNES for scientific hampered retrieval on several occasions, and we therefore be- collaboration and financial support. The research in Belgium was lieve this to be a lower bound. By the end of June all traces of funded by the F.R.S.-FNRS (M.I.S. nF.4511.08), the Belgian State Nabro plumes disappeared, which indicates a shorter lifetime Federal Office for Scientific, Technical and Cultural Affairs and of SO compared to Kasatochi or Sarychev. This is possibly the European Space Agency (ESA-Prodex arrangements and the due larger H O and OH concentrations at tropical latitudes. Support to Aviation Control Service (SACS+) project). We also Apart from large volcanic eruptions, IASI regularly picks acknowledge financial support from the EVOSS project (2nd Space up smaller puffs from world’s most active volcanoes such Call of the 7th Framework Program of the European Commission, as Etna. As an example, Fig. 13 shows some snap- with grant agreement 242535) and the “Actions de Recherche shots of volcanic plumes detected in the first part of 2011 Concertees” ´ (Communaute ´ Franc ¸aise de Belgique). We would over the Kamchatka Peninsula (originating from volcanoes like to thank the referees for the many useful corrections and improvements. such as Bezymianny, Kizimen, Karymsky, Kliuchevskoi and Shiveluch). Edited by: F. Prata 5 Conclusions References In this paper we have presented an algorithm for retrieving Andres, R. and Kasgnoc, A.: A time-averaged inventory of sub- SO abundances from IASI, although the algorithm can in aerial volcanic sulfur emissions, J. Geophys. Res., 103, 25251– principle be applied to any high resolution TIR sounder with 25261, 1998. Bluth, G. J. S., Schnetzler, C. C., Krueger, A. J., and Walker, L. S.: sufficient spectral resolution. It was specifically designed for The contribution of explosive volcanism to global atmospheric quantifying high altitude SO plumes from volcanic erup- sulphur dioxide concentrations, Nature, 366, 327–329, 1993. tions. A first attractive feature of the algorithm is its robust- Bobrowski, N., Kern, C., Platt, U., Hormann, ¨ C., and Wag- ness, simplicity and near real time applicability. With just a ner, T.: Novel SO spectral evaluation scheme using the 360– few lines of code this algorithm could for instance be imple- 390 nm wavelength range, Atmos. Meas. Tech., 3, 879–891, mented by volcanic ash advisory centers. Its second strong doi:10.5194/amt-3-879-2010, 2010. point is its very low theoretical uncertainty (3 % uncertainty Campion, R., Salerno, G. G., Coheur, P.-F., Hurtmans, D., Clarisse, for 0.5–100 DU and 6 % for 100–5000 DU for assumed alti- L., Kazahaya, K., Burton, M., Caltabiano, T., Clerbaux, C., and tudes above 500 hPa) coupled with a large applicable range Bernard, A.: Measuring volcanic degassing of SO in the lower (4 orders of magnitude of SO columns 0.5 to 5000 DU). By 2 troposphere with ASTER band ratios, J. Volcanol. Geoth. Res., not using all IASI’s channels the outlined algorithm does not 194, 42–54, doi:10.1016/j.jvolgeores.2010.04.010, 2010. Atmos. Meas. Tech., 5, 581–594, 2012 www.atmos-meas-tech.net/5/581/2012/ L. Clarisse et al.: Retrieving SO from IASI measurements 593 Carn, S. A. and Lopez, T. M.: Opportunistic validation of sul- Corradini, S., Merucci, L., Prata, A. J., and Piscini, A.: Volcanic fur dioxide in the Sarychev Peak volcanic eruption cloud, At- ash and SO in the 2008 Kasatochi eruption: Retrievals compar- mos. Meas. Tech., 4, 1705–1712, doi:10.5194/amt-4-1705-2011, ison from different IR satellite sensors, J. Geophys. Res., 115, 2011. D00L21, doi:10.1029/2009JD013634, 2010. Carn, S. A. and Prata, F.: Satellite-based constraints on explosive Doutriaux-Boucher, M. and Dubuisson, P.: Detection of volcanic SO release from Soufriere ` Hills Volcano, Montserrat, Geophys. SO by spaceborne infrared radiometers, Atmos. Res., 92, 69– 2 2 Res. Lett., 37, L00E22, doi:10.1029/2010GL044971, 2010. 79, doi:10.1016/j.atmosres.2008.08.009, 2008. Carn, S. A., Strow, L., de Souza-Machado, S., Edmonds, Y., and Flaud, J.-M., Lafferty, W., and Sams, R.: Line intensities for the ν , Hannon, S.: Quantifying tropospheric volcanic emissions with ν , and ν + ν bands of SO , J. Quant. Spectrosc. Ra., 110, 3 1 3 2 AIRS: The 2002 eruption of Mt. Etna (Italy), Geophys. Res. 669–674, 2009. Lett., 32, L02301, doi:10.1029/2004GL021034, 2005. Graf, H., Feichter, J., and Langmann, B.: Volcanic sulfur emissions: Carn, S. A., Krueger, A. J., Krotkov, N. A., Yang, K., and Evans, Estimates of source strength and its contribution to the global K.: Tracking volcanic sulfur dioxide clouds for aviation hazard sulfate distribution, J. Geophys. Res., 102, 10727–10738, 1997. mitigation, Nat. Hazards, 51, 325–343, 2009. Halmer, M., Schmincke, H.-U., and Graf, H.-F.: The annual vol- Chevallier, F.: Sampled databases of 60-level atmospheric profiles canic gas input into the atmoshpere, in particular into the strato- from the ECMWF analyses, Tech. rep., Eumetsat/ECMWF SAF sphere: a global data set for the past 100 years, J. Volcanol. Programme, Research Report No. 4, 2001. Geoth. Res., 115, 511–528, 2002. Chin, M. and Jacob, D.: Anthropogenic and natural contributions to Haywood, J. and Boucher, O.: Estimates of the direct and indi- tropospheric sulfate: A global model analysis, J. Geophys. Res., rect radiative forcing due to tropospheric aerosols: a review, Rev. 101, 18691–18699, 1996. Geophys., 38, 513–543, 2000. Clarisse, L., Coheur, P. F., Prata, A. J., Hurtmans, D., Razavi, Haywood, J. M., Jones, A., Clarisse, L., Bourassa, A., Barnes, A., Phulpin, T., Hadji-Lazaro, J., and Clerbaux, C.: Tracking J., Telford, P., Bellouin, N., Boucher, O., Agnew, P., Clerbaux, and quantifying volcanic SO with IASI, the September 2007 C., Coheur, P., Degenstein, D., and Braesicke, P.: Observations eruption at Jebel at Tair, Atmos. Chem. Phys., 8, 7723–7734, of the eruption of the Sarychev volcano and simulations using doi:10.5194/acp-8-7723-2008, 2008. the HadGEM2 climate model, J. Geophys. Res., 115, D21212, Clarisse, L., Hurtmans, D., Prata, A. J., Karagulian, F., Clerbaux, doi:10.1029/2010JD014447, 2010. C., Maziere, ` M. D., and Coheur, P.-F.: Retrieving radius, con- Karagulian, F., Clarisse, L., Clerbaux, C., Prata, A. J., Hurtmans, centration, optical depth, and mass of different types of aerosols D., and Coheur, P. F.: Detection of volcanic SO , ash and from high-resolution infrared nadir spectra, Appl. Optics, 49, H SO using the IASI sounder, J. Geophys. Res., 115, D00L02, 2 4 3713–3722, doi:10.1364/AO.49.003713, 2010a. doi:10.1029/2009JD012786, 2010. Clarisse, L., Prata, F., Lacour, J.-L., Hurtmans, D., Clerbaux, C., Kravitz, B., Robock, A., Bourassa, A., Deshler, T., Wu, D., Mattis, and Coheur, P.-F.: A correlation method for volcanic ash detec- I., Finger, F., Hoffmann, A., Ritter, C., Bitar, L., Duck, T. J., tion using hyperspectral infrared measurements, Geophys. Res. and Barnes, J. E.: Simulation and observations of stratospheric Lett., 37, L19806, doi:10.1029/2010GL044828, 2010b. aerosols from the 2009 Sarychev volcanic eruption, J. Geophys. Clarisse, L., Coheur, P.-F., Chefdeville, S., Lacour, J.-L., Hurt- Res., 116, D18211, doi:10.1029/2010JD015501, 2011. mans, D., and Clerbaux, C.: Infrared satellite observations Kristiansen, N. I., Stohl, A., Prata, A. J., Richter, A., Eckhardt, S., of hydrogen sulfide in the volcanic plume of the August Seibert, P., Hoffmann, A., Ritter, C., Bitar, L., Duck, T. J., and 2008 Kasatochi eruption, Geophys. Res. Lett., 38, L10804, Stebel, K.: Remote sensing and inverse transport modeling of the doi:10.1029/2011GL047402, 2011a. Kasatochi eruption sulfur dioxide cloud, J. Geophys. Res., 115, Clarisse, L., Fromm, M., Ngadi, Y., Emmons, L., Clerbaux, C., D00L16, doi:10.1029/2009JD013286, 2010. Hurtmans, D., and Coheur, P.-F.: Intercontinental transport of Krotkov, N., Schoeberl, M., Morris, G., Carn, S., and Yang, anthropogenic sulfur dioxide and other pollutants; an infrared K.: Dispersion and lifetime of the SO cloud from the Au- remote sensing case study, Geophys. Res. Lett., 38, L19806, gust 2008 Kasatochi eruption, J. Geophys. Res., 115, D00L20, doi:10.1029/2011GL048976, 2011b. doi:10.1029/2010JD013984, 2010. Clerbaux, C., Coheur, P.-F., Clarisse, L., Hadji-Lazaro, J., Hurt- Krueger, A., Walter, L., Bhartia, P., Schnetzler, C., Krotkov, N., mans, D., Turquety, S., Bowman, K., Worden, H., and Carn, Sprod, I., and Bluth, G.: Volcanic sulfur dioxide measurements S.: Measurements of SO profiles in volcanic plumes from the from the Total Ozone Mapping Spectrometer (TOMS) Instru- NASA Tropospheric Emission Spectrometer (TES), Geophys. ments., J. Geophys. Res., 100, 14057–14076, 1995. Res. Lett., 35, L22807, doi:10.1029/2008GL035566, 2008. Krueger, A., Krotkov, N., and Carn, S.: El Chichon: Clerbaux, C., Boynard, A., Clarisse, L., George, M., Hadji-Lazaro, The genesis of volcanic sulfur dioxide monitoring J., Herbin, H., Hurtmans, D., Pommier, M., Razavi, A., Turquety, from space, J. Volcanol. Geoth. Res., 175, 408–414, S., Wespes, C., and Coheur, P.-F.: Monitoring of atmospheric doi:10.1016/j.jvolgeores.2008.02.026, 2008. composition using the thermal infrared IASI/MetOp sounder, At- Lee, C., Martin, R. V., van Donkelaar, A., Lee, H., Dickerson, mos. Chem. Phys., 9, 6041–6054, doi:10.5194/acp-9-6041-2009, R. R., Hains, J. C., Krotkov, N., Richter, A., Vinnikov, K., 2009. and Schwab, J. J.: SO emissions and lifetimes: Estimates Corradini, S., Merucci, L., and Prata, A. J.: Retrieval of SO from from inverse modeling using in situ and global, space-based thermal infrared satellite measurements: correction procedures (SCIAMACHY and OMI) observations, J. Geophys. Res., 116, for the effects of volcanic ash, Atmos. Meas. Tech., 2, 177–191, D06304, doi:10.1029/2010JD014758, 2011. doi:10.5194/amt-2-177-2009, 2009. Longhurst, J. W., Raper, D. W., Lee, D. S., Heath, B. A., Con- www.atmos-meas-tech.net/5/581/2012/ Atmos. Meas. Tech., 5, 581–594, 2012 594 L. Clarisse et al.: Retrieving SO from IASI measurements lan, B., and King, H. J.: Acid deposition: a select review 1852– come, N., Lafferty, W., Mandin, J.-Y., Massie, S., Mikhailenko, 1990: 1. Emissions, transport, deposition, effects on freshwa- S., Miller, C., Moazzen-Ahmadi, N., Naumenko, O., Nikitin, A., ter systems and forests, Fuel, 72, 1261–1280, doi:10.1016/0016- Orphal, J., Perevalov, V., Perrin, A., Predoi-Cross, A., Rinsland, 2361(93)90125-L, 1993. C., Rotger, M., Simeck ˇ ova, ´ M., Smith, M., Sung, K., Tashkun, Matoza, R. S., Pichon, A. L., Vergoz, J., Herry, P., Lalande, J.- S., Tennyson, J., Toth, R., Vandaele, A., and Auwera, J. V.: M., Lee, H.-I., Che, I.-Y., and Rybin, A.: Infrasonic obser- The HITRAN 2008 molecular spectroscopic database, Journal vations of the June 2009 Sarychev Peak eruption, Kuril Is- of Quantitative Spectroscopy and Radiative Transfer, 110, 533– lands: Implications for infrasonic monitoring of remote ex- 572, doi:10.1016/j.jqsrt.2009.02.013, 2009. plosive volcanism, J. Volcanol. Geoth. Res., 200, 35–48, Rybin, A., Chibisova, M., Webley, P., Steensen, T., Izbekov, P., doi:10.1016/j.jvolgeores.2010.11.022, 2011. Neal, C., and Realmuto, V.: Satellite and ground observations Prata, A.: Satellite detection of hazardous volcanic clouds and of the June 2009 eruption of Sarychev Peak volcano, Matua Is- the risk to global air traffic, Nat. Hazards, 51, 303–324, land, Central Kuriles, B. Volcanol., 1–16, doi:10.1007/s00445- doi:10.1007/s11069-008-9273-z, 2008. 011-0481-0, 2011. Prata, A. and Bernardo, C.: Retrieval of volcanic SO column Smith, S. J., van Aardenne, J., Klimont, Z., Andres, R. J., abundance from Atmospheric Infrared Sounder data, J. Geophys. Volke, A., and Delgado Arias, S.: Anthropogenic sulfur diox- Res., 112, D20204, doi:10.1029/2006JD007955, 2007. ide emissions: 1850–2005, Atmos. Chem. Phys., 11, 1101–1116, Prata, A. and Kerkmann, J.: Simultaneous retrieval of volcanic doi:10.5194/acp-11-1101-2011, 2011. ash and SO using MSG-SEVIRI measurements, Geophys. Res. Surono, Jousset, P., Pallister, J., Boichu, M., Buongiorno, M., Bud- Lett., 34, L05813, doi:10.1029/2006GL028691, 2007. isantoso, A., Costa, F., Andreastuti, S., Prata, F., Schneider, D., Prata, A., Rose, W., Self, S., and O’Brien, D.: Global, long-term Clarisse, L., Humaida, H., Sumarti, S., Bignami, C., Griswold, J., sulphur dioxide measurements from TOVS data: A new tool for Carn, S., and Oppenheimer, C.: The 2010 explosive eruption of studying explosive volcanism and climate, in: Volcanism and the Javas Merapi volcano a 100-year event, submitted to J. Volcanol. Earth’s Atmosphere, edited by: Robock, A. and Oppenheimer, Geoth. Res., 2012. C., vol. 139 of Geophys. Monogr., 75–92, AGU, Washington, Thomas, H. and Watson, I.: Observations of volcanic emissions DC, 2003. from space: current and future perspectives, Nat. Hazards, 54, Prata, A., Gangale, G., Clarisse, L., and Karagulian, F.: Ash and 323–354, 2010. sulfur dioxide in the 2008 eruptions of Okmok and Kasatochi: Vernier, J.-P., Pommereau, J.-P., Thomason, L. W., Pelon, J., Gar- Insights from high spectral resolution satellite measurements, nier, A., Deshler, T., Jumelet, J., and Nielsen, J. K.: Overshoot- J. Geophys. Res., 115, D00L18, doi:10.1029/2009JD013556, ing of clean tropospheric air in the tropical lower stratosphere 2010. as seen by the CALIPSO lidar, Atmos. Chem. Phys., 11, 9683– Realmuto, V., Abrams, M., Buongiorno, M., and Pieri, D.: The 9696, doi:10.5194/acp-11-9683-2011, 2011. use of multispectral thermal infrared image data to estimate the Walker, J. C., Dudhia, A., and Carboni, E.: An effective method sulfur dioxide flux from volcanoes: A case study from Mount for the detection of trace species demonstrated using the MetOp Etna, Sicily, 29 July 1986, J. Geophys. Res., 99, 481–488, 1994. Infrared Atmospheric Sounding Interferometer, Atmos. Meas. Realmuto, V., Sutton, A., and Elias, T.: Multispectral thermal in- Tech., 4, 1567–1580, doi:10.5194/amt-4-1567-2011, 2011. frared mapping of sulfur dioxide plumes: A case study from the Watson, I., Realmuto, V., Rose, W., Prata, A., Bluth, G., Gu, Y., East Rift Zone of Kilauea Volcano, Hawaii, J. Geophys. Res., Bader, C., and Yu, T.: Thermal infrared remote sensing of vol- 102, 15057–15072, 1997. canic emissions using the moderate resolution imaging spectro- Richter, A., Wittrock, F., Schonhardt, A., and Burrows, J.: Quan- radiometer, J. Volcanol. Geoth. Res., 135, 75–89, 2004. tifying volcanic SO emissions using GOME2 measurements, Waythomas, C. F., Scott, W. E., Prejean, S. G., Schneider, D. J., Geophys. Res. Abstr., EGU2009-7679, EGU General Assembly Izbekov, P., and Nye, C. J.: The 7–8 August 2008 eruption of 2009, Vienna, Austria, 2009. Kasatochi Volcano, central Aleutian Islands, Alaska, J. Geophys. Rix, M., Valks, P., Hao, N., van Geffen, J., Clerbaux, C., Clarisse, Res., 115, B00B06, doi:10.1029/2010JB007437, 2010. L., Coheur, P.-F., Loyola R, D. G., Erbertseder, T., Zimmer, W., Yang, K., Krotkov, N., Krueger, A., Carn, S., Bhartia, P., and Levelt, and Emmadi, S.: Satellite Monitoring of Volcanic Sulfur Dioxide P.: Retrieval of large volcanic SO columns from the Aura Ozone Emissions for Early Warning of Volcanic Hazards, IEEE J. Sel. Monitoring Instrument: Comparison and limitations, J. Geophys. Top. Appl., 2, 196–206, doi:10.1109/JSTARS.2009.2031120, Res., 112, D24S43, doi:10.1029/2007JD008825, 2007. 2009. Zhang, Q., Jimenez, J. L., Canagaratna, M. R., Allan, J. D., Robock, A.: Volcanic eruptions and climate, Rev. Geophys., 38, Coe, H., Ulbrich, I., Alfarra, M. R., Takami, A., Middlebrook, 191–219, 2000. A. M., Sun, Y. L., Dzepina, K., Dunlea, E., Docherty, K., De- Rose, W. I., Bluth, G., and Watson, I.: Ice in volcanic clouds: When Carlo, P. F., Salcedo, D., Onasch, T., Jayne, J. T., Miyoshi, and where?, in: Proc. of the 2nd Int. Conf. on Volcanic Ash and T., Shimono, A., Hatakeyama, S., Takegawa, N., Kondo, Y., Aviation Safety, OFCM, Washington, DC, Session 3, 61, Ex- Schneider, J., Drewnick, F., Borrmann, S., Weimer, S., Demer- tended observations of volcanic SO and sulfate aerosol in the jian, K., Williams, P., Bower, K., Bahreini, R., Cottrell, L., stratosphere, 2004. Griffin, R. J., Rautiainen, J., Sun, J. Y., Zhang, Y. M., and Rothman, L., Gordon, I., Barbe, A., Benner, D., Bernath, P., Birk, Worsnop, D. R.: Ubiquity and dominance of oxygenated species M., Boudon, V., Brown, L., Campargue, A., Champion, J.-P., in organic aerosols in anthropogenically-influenced Northern Chance, K., Coudert, L., Dana, V., Devi, V., Fally, S., Flaud, J.- Hemisphere midlatitudes, Geophys. Res. Lett., 34, L13801, M., Gamache, R., Goldman, A., Jacquemart, D., Kleiner, I., La- doi:10.1029/2007GL029979, 2007. Atmos. Meas. Tech., 5, 581–594, 2012 www.atmos-meas-tech.net/5/581/2012/ http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Atmospheric measurement techniques Unpaywall

Retrieval of sulphur dioxide from the infrared atmospheric sounding interferometer (IASI)

Clarisse, L.; Hurtmans, D.; Clerbaux, C.; Hadji-Lazaro, J.; Ngadi, Y.; Coheur, P.-F.
Atmospheric measurement techniques – Mar 13, 2012
Download PDF
14 pages

Loading next page...
 
/lp/unpaywall/retrieval-of-sulphur-dioxide-from-the-infrared-atmospheric-sounding-DnTM8WwzI5

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
1867-1381
DOI
10.5194/amt-5-581-2012
Publisher site
See Article on Publisher Site

Abstract

Atmos. Meas. Tech., 5, 581–594, 2012 Atmospheric www.atmos-meas-tech.net/5/581/2012/ Measurement doi:10.5194/amt-5-581-2012 © Author(s) 2012. CC Attribution 3.0 License. Techniques Retrieval of sulphur dioxide from the infrared atmospheric sounding interferometer (IASI) 1 1 1,2 2 1 1 L. Clarisse , D. Hurtmans , C. Clerbaux , J. Hadji-Lazaro , Y. Ngadi , and P.-F. Coheur Spectroscopie de l’Atmosphere, ` Service de Chimie Quantique et Photophysique, Universite ´ Libre de Bruxelles, Brussels, Belgium UPMC Univ. Paris 6, Universite ´ Versailles St.-Quentin, CNRS/INSU, LATMOS-IPSL, Paris, France Correspondence to: L. Clarisse ([email protected]) Received: 15 November 2011 – Published in Atmos. Meas. Tech. Discuss.: 7 December 2011 Revised: 23 February 2012 – Accepted: 6 March 2012 – Published: 13 March 2012 Abstract. Thermal infrared sounding of sulphur dioxide fraction of the emissions makes it to the upper troposphere (SO ) from space has gained appreciation as a valuable com- and lower stratosphere (UTLS), a large volcanic eruption plement to ultraviolet sounding. There are several strong reaching the UTLS can impact the climate significantly as absorption bands of SO in the infrared, and atmospheric the lifetime of sulfate aerosol is proportional to the injection sounders, such as AIRS (Atmospheric Infrared Sounder), altitude. Bottom up approaches are well suited to determine TES (Tropospheric Emission Spectrometer) and IASI (In- total emissions of anthropogenic SO and emissions of some frared Atmospheric Sounding Interferometer) have the abil- degassing volcanoes, but quantifying UTLS SO emissions ity to globally monitor SO abundances. Most of the ob- is best done directly via satellite measurements (Bluth et al., served SO is found in volcanic plumes. In this paper we 1993). In this paper we detail a novel algorithm for calculat- outline a novel algorithm for the sounding of SO above ing SO columns above the mid troposphere (500 hPa) from 2 2 ∼5 km altitude using high resolution infrared sounders and infrared (IR) satellite measurements. apply it to measurements of IASI. The main features of the Apart from climatological relevance, measuring high al- algorithm are a wide applicable total column range (over 4 titude SO is also important for studying uplift of anthro- orders of magnitude, from 0.5 to 5000 dobson units), a low pogenic pollution (e.g. Clarisse et al., 2011b), for analyz- theoretical uncertainty (3–5 %) and near real time applicabil- ing explosive volcanic eruptions (e.g. Carn and Prata, 2010), ity. We make an error analysis and demonstrate the algorithm and, when data are available in near real time, for monitoring on the recent eruptions of Sarychev, Kasatochi, Grimsvotn, ¨ volcanic activity (e.g. Surono et al., 2012) and tracking of Puyehue-Cordon ´ Caulle and Nabro. volcanic clouds for the mitigation of aviation hazards (Prata, 2008; Rix et al., 2009; Carn et al., 2009). Since 1978, the Total Ozone Mapping Spectrometer (TOMS) (Krueger et al., 1995) and subsequent follow-up 1 Introduction ozone monitoring instruments have been measuring SO Prodigious amounts of sulphur dioxide (SO ) are released through solar backscattered ultraviolet (BUV) measurements every year in the atmosphere. Anthropogenic emissions, (see e.g. Yang et al., 2007, and references therein). BUV mostly coming from combustion of sulphur-rich biomass measurements have a good sensitivity to SO , even in the −1 such as coal and petroleum, add up to 50–65 Tg S yr lowest atmospheric layers. The record of IR sounding of SO (Smith et al., 2011; Lee et al., 2011). Volcanoes are the also goes back to 1978 with the High-Resolution Infrared largest natural source of sulphur dioxide and account for Sounder (HIRS/2) (Prata et al., 2003). One clear advantage −1 7.5–10.5 Tg S yr on average (Andres and Kasgnoc, 1998; of thermal infrared (TIR) instruments is that they can mea- Halmer et al., 2002). These emissions lead to acid deposi- sure in the absence of sunlight (thus also at night and at high tion and can affect air quality and climate through the for- latitudes in the winter) and often have a higher spatial res- mation of sulfate aerosols (Longhurst et al., 1993; Chin and olution. For an overview of satellite instruments capable of Jacob, 1996; Graf et al., 1997; Haywood and Boucher, 2000; measuring SO and their characteristics and limitations, we Robock, 2000; Zhang et al., 2007). While in general only a refer to Thomas and Watson (2010). Here we give a short Published by Copernicus Publications on behalf of the European Geosciences Union. 582 L. Clarisse et al.: Retrieving SO from IASI measurements overview of TIR sounding of SO without going into instru- lookup table. Our algorithm is akin to some of the meth- mental specifics. ods applied for broadband sensors. The advantage, however, Sulphur dioxide has three absorption bands in the mid in- is that we can select specific channels, making the algorithm frared, see Fig. 1. The ν is by far the strongest band. Com- simpler and less sensitive to changes of other atmospheric peting water vapor absorption limits its vertical sensitivity to variables (water vapour, clouds). SO above 3–5 km, depending on the humidity profile and We outline the algorithm for observations of the high res- SO abundance. Higher altitude SO is also affected, di- olution infrared sounder IASI (Clerbaux et al., 2009), but it 2 2 rectly, by water vapor in and above the SO layer, but also can easily be transferred to other high resolution sounders. indirectly by variable radiation coming from below. The ν Instrumental specifics of the IASI instrument are a continu- −1 band is situated in an atmospheric window, and can pen- ous spectral coverage between 645 and 2760 cm , a spectral −1 −1 etrate the lower troposphere. While water vapor is not as resolution of 0.5 cm (which is apodized at 0.25 cm ) and −1 important here, the 800–1200 cm region is very sensitive a noise equivalent delta temperature at 280 K around 0.05 K to the surface temperature, surface emissivity and volcanic for the ν band and 0.12 K for the ν band. It has a global 3 1 ash (Clarisse et al., 2010a,b), and for young volcanic plumes coverage twice a day with a footprint ranging from circular from explosive eruptions, SO and ash often need to be re- (12 km diameter at nadir) to elliptical (up to 20 by 39 km at trieved simultaneously. The combination band ν + ν can the end of the swath) and a mean local equatorial overpass 1 3 only be used when there is reflected solar light. It is weak, time at 09:30 LT and 21:30 LT. but has been applied for the study of major volcanic erup- In the next section we outline the theoretical basis of the tions as an alternative to a saturating ν band (Karagulian algorithm. In Sect. 3 we give an overview of the most impor- et al., 2010; Prata et al., 2010). Note that all TIR measure- tant sources in the error budget. Examples are presented in ments require thermal contrast between the SO plume and Sect. 4 and we conclude in Sect. 5. the underlying source of radiation. Broadband instruments typically have a handful of chan- −1 2 The algorithm nels (each covering 50–100 cm ) which can be used to re- trieve SO . Most retrieval algorithms are based on approx- In what follows, we assume an atmosphere with a SO cloud imating the SO affected bands from the other bands as- present at a given altitude. We adopt the notations from Wat- suming the absence of SO . The difference between these son et al. (2004). When the plume is at sufficient altitude reconstructed background radiances and the observed radi- (where the absorption of other species can be ignored) the ances can then be used to infer abundances. In the case of measured radiance L at a wavenumber ν (and correspond- the ν band this can be done by first estimating the surface ing measured brightness temperature at the sensor T ) can be temperature (Realmuto et al., 1994, 1997) or by assuming a approximated as linear correlation with another band (Prata and Kerkmann, 2007). For the ν band it has been shown that it is possi- 3 L (ν)= L (ν)t +L (ν)(1−t ), (1) s ucb c c c ble to estimate the relevant unperturbed band radiance from with L (ν) = B (ν,T ) the ambient radiance coming from c c a linear interpolation of two other bands (Prata et al., 2003; the cloud at temperature T and specified by Planck’s law, Doutriaux-Boucher and Dubuisson, 2008). Other schemes L (ν) the upwelling radiance at the cloud base and t the ucb c rely on the use of a large series of simulated radiances (see transmission of the cloud, given by the Bouguer-Lambert- e.g. Corradini et al., 2010). For retrievals using the ν band, Beer law explicit (Corradini et al., 2009) or implicit (Campion et al., 2010) corrections for aerosols can be made. −cu t = e , (2) Retrievals using high spectral resolution instruments typi- cally use (optimal) least square procedures (Carn et al., 2005; with c an absorption coefficient dependent on pressure and Prata and Bernardo, 2007; Clerbaux et al., 2008; Clarisse temperature and u the column abundance. While Eq. (1) et al., 2008), preceded by a SO detection routine. These is valid under the mentioned assumptions, a subtlety arises are time consuming, but have the advantage of fully exploit- when applying it to real measurements. Real radiance ing the spectral resolution by simultaneously retrieving com- measurements are always integrated (convolved) over a peting species (e.g. H O) and potentially extracting plume wavenumber interval and are altered by the instrumental line altitude information. It was shown (Karagulian et al., 2010; shape. To check to what extent Eq. (1) holds at the level Haywood et al., 2010) that for the ν band it often suffices of finite microwindows (here IASI channels), we have sim- to perform optimal estimation on a selected number of pixels ulated the radiative transfer of a standard atmosphere and and exploit the empirical correlation between these retrieved introduced a SO layer at a fixed altitude, but with varying total columns and brightness temperature differences. It is abundances. The results are shown in Fig. 2 in brightness −1 this scheme we generalize and put on a more solid theoretical temperature space at wavenumber ν = 1371.75 cm . The footing. Instead of relying on optimal estimation retrievals, simulations are shown as black squares and the best fit with however, we use elementary radiative transfer and a large Eq. (1) (best choice of the absorption coefficient c) is shown Atmos. Meas. Tech., 5, 581–594, 2012 www.atmos-meas-tech.net/5/581/2012/ L. Clarisse et al.: Retrieving SO from IASI measurements 583 Fig. 1. Top panel: example IASI spectrum measured over the plume of the August 2008 eruption of Kasatochi. Bottom panel: line positions and intensities of SO from HITRAN (see Rothman et al., 2009, and references therein). Band centers and inte- −1 grated band intensities of SO are (see Flaud et al., 2009, and references therein): the ν symmetric stretch (∼1152 cm = 8.7 μm at 2 1 −17 −1 −2 −1 −17 −1 −2 0.35× 10 cm /(molecule cm )), the ν asymmetric stretch (∼1362 cm = 7.3 μm at 2.72× 10 cm /(molecule cm )) and the −1 −17 −1 −2 ν + ν combination band (∼2500 cm = 4 μm at 0.054× 10 cm /(molecule cm )). 1 3 −1 Fig. 2. Brightness temperature at 1371.75 cm as a function of SO mass loading for a low (left, plume at 247 K and 450 hPa ∼ 5 km) and high (right, plume at 230 K and 10 hPa ∼25 km) altitude plume. The colored black squares were calculated from simulated IASI spectra, while the red full line is a best fit of these simulations with Eq. (1). www.atmos-meas-tech.net/5/581/2012/ Atmos. Meas. Tech., 5, 581–594, 2012 584 L. Clarisse et al.: Retrieving SO from IASI measurements in red. For a plume at high pressure (left panel, 450 hPa), an almost perfect fit can be obtained. The asymptotic behavior for increasingly large abundances can also be observed (L (ν)→ B (ν, T ) or T → T ). This saturation is slower for c s c lower pressure (right panel, 10 hPa). At very low pressure, spectral lines saturate at a lower concentration at their line centers than their wings. In contrast, at a higher pressure, pressure broadening of the individual lines is important and will distribute absorption over a wider spectral range, result- ing in a net larger absorption and thus a quicker saturation over the complete band when taking into account all spectral lines. For the low pressure test case, a good fit with Eq. (1) and a constant absorption coefficient c is not possible. Be- cause of the lower pressure broadening, the instrumental line shape and apodisation become relatively more important, and these effects are not taken into account in Eq. (1). One way Fig. 3. Brightness temperature of the two sets of absorption chan- to resolve this is to introduce an explicit column dependence −1 −1 nels (at ∼1371.5 cm and at ∼1385 cm ) as a function of SO in the coefficient c, so that c = c (T , P , u). These coefficients abundance for a plume located at 150 hPa and 207 K. can be estimated from forward simulations as outlined below. To determine the SO abundance from Eq. (1), all that is altitude, some residual water absorption can still affect ob- left is to estimate L (ν). This can be done from channels ucb served channels. Assuming that water vapour above is colder not affected by SO , but for which the channel ν responds than the SO plume (so disregarding significant water vapor similarly to H O and other atmospheric parameters than the above lower stratospheric plumes), we have for a saturating channels sensitive to SO . It is here easier to work in bright- cloud T < T . We therefore introduce a virtual cloud temper- s c ness temperature space, where Eq. (1) reads ∗ 21 ature T = T − [H O]/10 , with [H O] the partial column c 2 2 −2 B(T ,ν)= B(T ,ν)t +B(T ,ν)(1−t ). (3) of water (in molecules cm ) above the SO layer . The fac- s ucb c c c tor 10 was determined empirically, and while this is a first Now T can be estimated from another channel ν when for ucb order correction, it is largely sufficient as we will see below. background concentrations of SO To calculate the absorption coefficients c (T ,P ,u) we have −1 −1 0 0 T = B (L (ν),ν)≈ B (L (ν ),ν )= T . (4) used representative atmospheric profiles (temperature, pres- s s s ucb sure, humidity and ozone) from the ECMWF 40-yr reanaly- The critical part is to choose these channels ν and ν to make sis, ERA-40 (Chevallier, 2001). The total set contains 13 495 this estimate as good as possible. We have used combina- well sampled profiles. Pressure and temperature (PT) pairs tions of 4 channels: two to estimate T , representing the ab- between 5 and 30 km altitude are plotted in Fig. 4. The vis- sorption in the ν band and two reference channels to es- ible pressure bands are an artifact caused by the specific 60- timate T . Table 1 lists two sets of such parameters to- ucb level coordinate system in the data set, and these disappear gether with their bias and standard deviation (estimated from when working with the interpolated data. We have calculated a full day of IASI measurements with no detectable volcanic c (T ,P ,u) on a subgrid of this PT diagram, indicated by the SO ). Note that this doubling of channels allows to reduce black dots. the standard deviation significantly and also that the bias can For each PT pair in the subgrid, we selected 10 atmo- be subtracted in the calculation of the brightness tempera- spheres from ERA-40 with the closest match in the PT pro- ture difference. Figure 3 illustrates the sensitivity range of file. A variable SO cloud (from 0 to 10 000 DU) was then in- both sets for a plume at 150 hPa. The absorption channels in serted at the altitude corresponding to the PT pair and the re- the ν band of the first set are chosen close to the region of sulting IASI spectrum was simulated. Based on these simula- −1 maximum absorption, around 1371.75 cm . It is sensitive tions a best value for c (T ,P ,u) was obtained from minimiz- to mass loadings as low as 0.5 DU, but saturates at around ing the relative error between the real and the calculated SO 200 DU, above which differences in the observed channels abundance. Each c (T ,P ,u) is obtained from 10 independent become too small. The second set has its absorption chan- simulations and determining the best value is therefore an −1 nels further away from the band center, at 1385 cm . It has over-constrained problem. The solution however is guaran- a lower sensitivity of about 10 DU, but can measure columns teed not to be overly dependent on an individual atmosphere, up to 5000 DU. The combined use of both sets therefore en- and the average relative error is a good indication for the the- ables to retrieve columns of SO from about 0.5 to 5000 DU oretical error (caused by the variability of other atmospheric at 150 hPa. parameters) which can be achieved with this algorithm. Equation (1) is only valid when no absorption above the SO plume takes place. Even at altitudes above ∼500 hPa Atmos. Meas. Tech., 5, 581–594, 2012 www.atmos-meas-tech.net/5/581/2012/ L. Clarisse et al.: Retrieving SO from IASI measurements 585 Table 1. Two sets of absorption and background channels used in the calculation of SO abundances. The mean and standard deviation of their brightness temperature differences were calculated on one day with no detectable quantities of SO . ν Absorption channels Background channels Mean Std −1 −1 Set 1 1371.50, 1371.75 cm 1407.25, 1408.75 cm −0.05 K 0.15 K −1 −1 Set 2 1384.75, 1385.00 cm 1407.50, 1408.00 cm 0.05 K 0.25 K Fig. 4. Pressure and temperature correlations of the ERA-40 data set between 5 and 30 km. The black dots are the PT pairs for which the lookup tables were built. The top panel in Fig. 5 shows the absorption coefficients (due to the smooth and monotonous behavior of the c coeffi- for the two sets of channels at 10 and 750 DU respectively. cients). Also note that we find two estimates u and u for u, 1 2 For 4 PT pairs, T was very close or inferior to T for all 10 for each set of absorption and background channels. Theo- ucb c profiles. These sets of low thermal contrast or temperature retically, these two estimates should only agree when the as- inversion were excluded. These are situated at the very edge sumed altitude corresponds to the real altitude (because the of the PT space and are uncommon. The bottom panel shows corresponding brightness temperature differences have a dif- the mean relative error between the input SO abundance and ferent pressure and temperature dependence). From looking the retrieved for the ten different profiles. Errors are less than at a few test cases, the two estimates generally agree well be- 3 % and 5 % for the first and second set respectively, except tween 25 DU and 75 DU (with a standard deviation of around again at some points at the edge of the PT space. 10 %). On either side of this range, differences increase, with the u estimate obviously superior for lower total col- We end this section with a practical consideration, which umn amounts and the u estimate by construction superior is important in the implementation of the above retrieval al- for large column total amounts. When either u or u exceed 1 2 gorithm. The use of c (T ,P ,u) to calculate the column abun- 100 DU, we used the u estimate, otherwise u was used. Fi- 2 1 dance u is inherently a recursive problem. It is therefore nec- nally, the retrieval is also preceded by a detection criterion, essary to start with a first guess c (T ,P ) and iteratively calcu- here taken to be T − T > 0.4 K. ucb s late u and c (T ,P ,u) until convergence is achieved. We have verified numerically that this convergence is always achieved www.atmos-meas-tech.net/5/581/2012/ Atmos. Meas. Tech., 5, 581–594, 2012 586 L. Clarisse et al.: Retrieving SO from IASI measurements 4 4 x 10 x 10 Absorption coefficients c Absorption coefficients c 6 6 2 5 5 −3 x 10 0.03 4 4 3 3 0.025 2 2 0.02 1 1 0.015 0 0 180 200 220 240 260 280 180 200 220 240 260 280 4 4 x 10 x 10 Relative errors set 1 Relative errors set 2 6 6 5 5 15 15 4 4 3 3 10 10 2 2 5 5 1 1 0 0 0 0 180 200 220 240 260 280 180 200 220 240 260 280 Temperature (K) Temperature (K) Fig. 5. Absorption coefficients for the two sets of IASI channels (top) and their corresponding average errors in percentage (bottom). Here the absorption coefficients and errors are shown for a SO cloud of 10 DU (set 1) and 750 DU (set 2) respectively. 3 Sources of error 3.1 Measurements errors A good description of typical sources of error can be found We call measurements errors any errors that affect the dif- in Prata et al. (2003). Most of these are inherent to any re- ference of T and T beyond the contribution of SO . This s ucb 2 trieval which uses the ν band. There are broadly speaking 3 includes the instrumental noise, but also contributions from five main sources of error. The first category is related to the fact that the background channels are only a best-effort propagation of errors in the measurements, in our case in the estimate of the absorption channels in the absence of SO . measurements of T and T . The second category includes s ucb Following Table 1 we estimate the error to be of the order errors related to the assumed or measured altitude or cloud 0.15 K and 0.25 K for the first and second set of channels. temperature T . A third source of errors becomes important c From Fig. 3 it is easily seen that the influence of these errors when Eq. (1) is no longer a good approximation for the radia- will be largest for very thin or very thick SO clouds. For tive transfer due to presence of aerosols above the SO layer. 2 very thin clouds the contribution of SO on T will be of the 2 s There is the modeling error related to Eq. (1), which was es- same order of magnitude as the measurement error and hence timated above to be in the range 3–5 %, and finally there are relatively important. For very thick clouds, we are close to errors related to spectroscopy and radiative transfer. In this saturation regime and a small error on the observed temper- section we will discuss the first three types of error. atures will lead to large differences in the SO estimates. As an example of how this type of error translates in errors on the abundance, an error of 0.15 K and 0.25 K was introduced Atmos. Meas. Tech., 5, 581–594, 2012 www.atmos-meas-tech.net/5/581/2012/ Pressure (Pa) Pressure (Pa) L. Clarisse et al.: Retrieving SO from IASI measurements 587 Fig. 6. Illustration of the measurement error. Relative errors in Fig. 7. Effect of the assumed altitude on retrieved abundances; il- the retrieved abundances, made from introducing 0.15 K and 0.25 K lustrated for different eruptions. error in the data of Fig. 3. the temperature contrast is highest and the minimum amount in the data of Fig. 3 and the relative differences are plotted in of SO is required to account for the observed absorption. Fig. 6. It illustrates the increase of errors near the extremes. For instance the Merapi (Java/Indonesia) and the Nabro (Er- The errors between 0.5 DU and 5000 DU are in this exam- itrea) plumes have their minimum retrieved mass at a higher ple below 30 % (and below 6 % for loadings above 3 DU). altitude (17 km) than e.g. Sarychev (Kuril Islands, Russia) It should be stressed though that this type of error is a ran- or Kasatochi (Aleutian Islands, Alaska) plumes, which have dom error and averages out when calculating the total mass their minimum at 10–12 km. In the stratosphere the SO re- of plumes much larger than the footprint of the instrument. trievals increase as T approaches T , with the rate of in- c ucb Related to this, there is the situation where the SO cloud crease controlled by the stratospheric temperature gradient. at T has little or no thermal contrast with the radiation from As can be seen from Fig. 7, the effect of altitude is gener- below T . In this case (see again Fig. 3) the regime of low ucb ally within 10–20 % between 10 and 20 km. For low altitude sensitivity and the regime of saturation overlap and errors are plumes, the assumed altitude is more critical with differences naturally very large. This dependence on thermal contrast is up to 500 % between a plume at 5 and 10 km due to the steep inherent to IR sounding. temperature gradient in the troposphere. 3.2 Altitude 3.3 Aerosols As the present algorithm does not retrieve altitude, a cloud altitude (and therefore pressure and temperature) must be as- Large eruption plumes contain typically a large amount of sumed. This affects the estimated loading through the as- various particles (ash, ice, sulfate aerosols and aggregates). sumed water vapour absorption above the plume, c (T ,P ,u) All these absorb and scatter IR radiation. The wavenum- and T . The latter is the most important, especially close to ber dependence is most pronounced for ash and ice as illus- saturation or when considering large temperature differences. trated in Fig. 8 for the 2008 Kasatochi eruption (ash) and To assess their combined effect it is best to look at some ex- 2011 Nabro eruption (ice). Extinction of IR radiation by ash −1 amples. Figure 7 is a plot of retrieved total masses (as a is strongest in the 800–1200 cm range (see also Clarisse percentage of the maximum measured total mass for a given et al., 2010b), but almost uniform throughout the ν band of altitude) for different eruptive plumes (young and aged) as a SO . Note that the specific extinction depends on the total function of the assumed altitude. ash loading but also on the particle size distribution and the To understand the effect of the assumed altitude it is use- mineral composition. Ice particles have their largest extinc- −1 ful to look at the thermal contrast between the cloud and the tion feature in the 800–1000 cm range (see also Clarisse background (T − T ). Water vapour is the main source of et al., 2008). The retrieval algorithm is not sensitive to what ucb c the upwelling radiance in the vicinity of the ν band and happens below the SO cloud as long as the radiation com- 3 2 T therefore corresponds to an altitude of 3–6 km. At ing from below has sufficient thermal contrast with the SO ucb 2 cloud altitudes between 5 and 7 km the temperature contrast plume and as long as the radiation at background and absorp- is low and maximum amounts of SO are required to pro- tion channels extinguishes uniformly. Low-altitude aerosol duce the observed absorption. For clouds at the tropopause layers of low-to-medium optical thickness which are located www.atmos-meas-tech.net/5/581/2012/ Atmos. Meas. Tech., 5, 581–594, 2012 588 L. Clarisse et al.: Retrieving SO from IASI measurements Fig. 8. Observed IASI spectra from the 2008 Kasatochi (top) and 2011 Nabro eruptions (bottom) illustrating the effect of volcanic aerosol (ash and ice). Different degrees of aerosol extinction demonstrate the damping of the SO signature. Note that some of the differences are also due to the different total column of water vapour. well below the SO layer have therefore limited or no impact aerosol loading and its altitude, and while our tests point to on our retrieval. Opaque aerosol layers just below the SO an overestimation of the SO loading, pixels with completely 2 2 plume impede the sensitivity of the algorithm as is apparent opaque ash in or above the SO layer will go undetected and when comparing the black and the blue spectra in the top this will lead to an underestimation of the total measured SO panel of Fig. 8. mass. An example of such a spectrum is shown in pink in the top of Fig. 8. A little SO can be detected at ∼225 K above Aerosols above or at the same altitude as SO will have 2 the ash cloud at ∼220 K, but everything below the ash cloud an impact on the retrieved abundance. As a test case, we is not measurable. have simulated the radiative transfer (following the methods Note finally that for fresh plumes, it is not uncommon for described in Clarisse et al., 2010a) of a thick aerosol layer a portion of the erupted SO to be sequestered on ice, only to located below, above and in a 25 DU upper tropospheric SO 2 be later released in the volcanic cloud by sublimation (Rose plume. The aerosol abundance was in the three cases chosen −1 et al., 2004). This could account for some of the increases as to cause a drop of 20 K in the spectrum at ∼1362 cm . in SO total mass timeseries observed in ice rich volcanic As expected and explained above, aerosol below the SO 2 plumes (Krueger et al., 2008; Clarisse et al., 2008). layer had limited impact (2 %) on the retrieved abundance. Ash in the SO layer caused a 20 % overestimation, while aerosol above gave rise to a 45 % overestimation. It is clear that the effect of aerosol depends very much on the specific Atmos. Meas. Tech., 5, 581–594, 2012 www.atmos-meas-tech.net/5/581/2012/ L. Clarisse et al.: Retrieving SO from IASI measurements 589 −180 −175 −170 −165 −160 −155 −150 −180 −175 −170 −165 −160 −155 −150 55 55 50 50 7 August evening 8 August morning 45 45 Total / maximum Total / maximum @5 km : 655 kT / 2884 DU @5 km : 7688 kT / 3842 DU @7 km : 380 kT / 1676 DU @7 km : 3731 kT / 1845 DU SO (DU) @10 km : 135 kT / 586 DU @10 km : 1208 kT / 658 DU SO (DU) @13 km : 105 kT / 429 DU @13 km : 1164 kT / 581 DU @16 km : 92 kT / 403 DU @16 km : 1163 kT / 587 DU 0 100 200 300 0 100 200 300 400 500 @25 km : 186 kT / 1181 DU @25 km : 2139 kT / 1304 DU 40 40 −180 −175 −170 −165 −160 −155 −150 −180 −175 −170 −165 −160 −155 −150 55 55 50 50 8 August evening 9 August morning 45 45 Total / maximum Total / maximum @5 km : 6871 kT / 2085 DU @5 km : 6924 kT / 1802 DU @7 km : 3626 kT / 1066 DU @7 km : 3763 kT / 926 DU @10 km : 1163 kT / 353 DU SO (DU) @10 km : 1340 kT / 329 DU SO (DU) @13 km : 1065 kT / 335 DU @13 km : 1148 kT / 298 DU @16 km : 1082 kT / 369 DU @16 km : 1236 kT / 338 DU 0 100 200 300 0 50 100 150 200 @25 km : 1824 kT / 758 DU @25 km : 2009 kT / 678 DU 40 40 Fig. 9. The eruption of Kasatochi (Aleutian islands) on 7 and 8 August as seen by IASI, with a 5–20 km altitude SO plume drifting to North America. For the displayed columns an altitude of 10 km was assumed. 4 Examples to 2.2 Tg (Kristiansen et al., 2010; Krotkov et al., 2010) from OMI (Ozone Monitoring Instrument). In the IR estimates are 1.2 to 1.4 Tg (Prata et al., 2010) from AIRS (Atmo- 4.1 Kasatochi – large columns spheric Infrared Sounder), 1.7 Tg (Karagulian et al., 2010) from IASI and 0.94 to 2.65 Tg (Corradini et al., 2010) from Kasatochi volcano (part of the Aleutian Islands) erupted on MODIS (Moderate Resolution Imaging Spectroradiometer). 7 and 8 August 2008 five times (Waythomas et al., 2010) The main difficulty in comparing the respective retrievals is and ejected the largest amount of SO in the UTLS since understanding the impact of the different assumed or cal- the eruption of Cerro Hudson in 1991 (Krotkov et al., 2010). culated heights coupled with the different responses in the There are several aspects which complicate the SO retrieval. IR/UV absorption bands. Also important are the different The five eruptions occurred in quick succession, and these strategies applied to cope with non-linear effects associated were different in nature (phreatomagmatic and magmatic with very large columns as also reflected in the large vari- Waythomas et al., 2010) and altitude (5–20 km Kristiansen ance in reported maximum columns, ranging from 100 to et al., 2010). The resulting plume was therefore highly het- 700 DU: OMI 280 DU (Kristiansen et al., 2010), GOME2 erogeneous in SO , H S (Clarisse et al., 2011a), H O, ash 2 2 2 100 DU (operational retrieval)–700 DU (Richter et al., 2009; (Corradini et al., 2010) and ice content and likely multilay- Bobrowski et al., 2010) and IASI 300 DU (Karagulian et al., ered within a typical operational satellite’s footprint (>10 km 2010). diameter). In terms of total ejected SO mass, estimates from satel- Retrieval results using the new algorithm are shown in lites vary widely, from 1 to 3 Tg, in part due to the fact that Fig. 9 for the first 4 IASI overpasses (the first overpass on most retrievals are based on a single cloud altitude. Es- 7 August happened after 3 of the 5 explosive events). In timates in the UV are 2.5 Tg (Richter et al., 2009) from terms of maximum columns, the first overpass on the 8th GOME2 (Global Ozone Monitoring Experiment-2) and 1.4 measured columns in excess of 500 DU (depending on the www.atmos-meas-tech.net/5/581/2012/ Atmos. Meas. Tech., 5, 581–594, 2012 590 L. Clarisse et al.: Retrieving SO from IASI measurements Fig. 10. Time series of SO measured with IASI: original study Fig. 11. Time series of SO measured with IASI: original study (Karagulian et al., 2010) in blue, current reanalysis in red and OMI (Haywood et al., 2010) in black, current reanalysis in red and (Krotkov et al., 2010) in black. HadGEM2 model in blue. injection altitude). This is higher than any other retrieval 4.2 Sarychev – aging plume reported using ν measurements, and of the same order as the maximum columns measured by GOME2 and shows the Another large eruption took place in 2009, namely Sarychev ability of our retrieval algorithm to deal efficiently with band Peak (Kuril Islands, Russia) on 11–16 June (Matoza et al., saturation. Retrieved total masses vary from 3.7 Tg (7 km) 2011; Rybin et al., 2011; Haywood et al., 2010). There were over 1.3 Tg (10–13 km) to 2 Tg (25 km). As the plume was several explosive events, but the majority of the high altitude spread over altitudes ranging from 5 to 20 km, this is again SO was injected on 15 and 16 June at an altitude of 10– consistent with data from other sounders. 16 km. An earlier study (Haywood et al., 2010) using IASI The retrieved total mass at 10 km (Krotkov et al., 2010; data estimated the sulphur dioxide emissions for those two Karagulian et al., 2010) as a function of date is displayed in days to be of the order of 1.2± 0.2 Tg and this figure is com- Fig. 10. We find that the retrieved values increase after the mensurate with OMI measurements (Carn and Lopez, 2011). 9th to about 1.6 Tg on the 11 August. As the presented al- Like the Kasatochi eruption, the eruption of Sarychev peak gorithm is able to retrieve very large SO columns, we do presented a nice validation opportunity for modeling and not except saturation problems to be an issue. However, we measuring lower stratospheric injections of SO and grad- made the assumption that the plume is concentrated at a sin- ual oxidation to sulfate (Haywood et al., 2010; Kravitz et al., gle fixed altitude and so one possible reason for this post- 2011; Vernier et al., 2011). eruptive increase of the total mass is that part of the plume Figure 11 shows the measured total mass in the Northern was vertically stratified, with the ν band mostly sensitive to Hemisphere at 13 km (the total mass does not vary a lot for the upper part (Corradini et al., 2010; Krotkov et al., 2010). height assumptions between 10 and 16 km) as a function of By the 11th vertical wind shear probably dispersed the mul- time in June 2009. The difference with the previous time- tilayered cloud sufficiently for it to be exposed completely. series (shown in black and reported in Haywood et al., 2010) For the overall time-series we find substantial differences is minimal, except for the maximum retrieved value, which is (up to 50 %) with an earlier IASI analysis (Karagulian et al., now determined at 0.9 Tg. The current reanlysis is likely to 2010) which used the ν + ν combination band for the re- 1 3 be more accurate, given the fact that the original retrieved trievals of the plume on the 8th, 9th and 10th. This analy- mass was quite noisy for the first week after the eruption sis did not exhibit an increase in retrieved total masses after (with differences up to 100 % for consecutive overpasses). the 8th (likely due to the weaker altitude/temperature depen- The current retrieval gives a very smooth timeseries. Note dency of the retrievals). However, a remarkable discrepancy also that for the 2009 Sarychev eruption as a whole, the total of this earlier analysis with OMI retrievals was found for re- released SO is likely higher then 0.9 Tg as prior to 15 June trievals of the aged plume, with differences over 100 % and there were several smaller eruptions. leading to factor two in the estimated e-folding lifetime of SO . In the present study measurements of IASI are compat- ible with the OMI retrievals (Krotkov et al., 2010), and the timeseries clearly fits the expected exponential decay better than the earlier analysis. Atmos. Meas. Tech., 5, 581–594, 2012 www.atmos-meas-tech.net/5/581/2012/ L. Clarisse et al.: Retrieving SO from IASI measurements 591 −150 −100 −50 0 50 100 150 −20 −40 −60 −80 0 2 4 6 8 SO (DU) Fig. 12. Composite image of maximum observed SO columns for the period 20 May to 30 June 2011. The value for each grid cell equals the maximum observed SO columns in that grid cell for the given time period during which three major volcanic eruptions took ◦ ◦ ◦ ◦ place. Grimsvotn ¨ (−17.33 , 64.42 ) erupted first on 21 May, then Puyehue-Cordon ´ Caulle (−40.59 , −72.12 ) on 3 June and finally Nabro ◦ ◦ (13.37 , 41.70 ) on 12 June. A plume altitude of 10 km was assumed. 150 160 170 150 160 170 150 160 170 150 160 170 65 65 65 65 60 60 60 60 1.8 55 55 55 55 07.01.2011 09.01.2011 23.01.2011 24.01.2011 1.6 50 50 50 50 150 160 170 150 160 170 150 160 170 150 160 170 65 65 65 65 1.4 60 60 60 60 55 55 55 55 1.2 26.01.2011 29.01.2011 30.01.2011 31.01.2011 50 50 50 50 150 160 170 150 160 170 150 160 170 150 160 170 65 65 65 65 60 60 60 60 0.8 55 55 55 55 0.6 31.01.2011 01.02.2011 01.02.2011 03.02.2011 50 50 50 50 150 160 170 150 160 170 150 160 170 150 160 170 65 65 65 65 0.4 60 60 60 60 0.2 55 55 55 55 03.02.2011 04.02.2011 07.02.2011 17.02.2011 50 50 50 50 Fig. 13. Snapshots of volcanic SO plumes detected in the first part of 2011 over the Kamchatka Peninsula. www.atmos-meas-tech.net/5/581/2012/ Atmos. Meas. Tech., 5, 581–594, 2012 SO (DU) 2 592 L. Clarisse et al.: Retrieving SO from IASI measurements 4.3 Grimsvotn, ¨ Puyehue-Cordon ´ Caulle, Nabro – exploit IASI’s high spectral resolution to the fullest (for esti- global retrievals mating plume altitude see Clarisse et al., 2008; for improved detection see Walker et al., 2011). Ideally therefore, these In May and June 2011, three volcanoes erupted, each releas- algorithms should be used in combination with each other. ing large amounts of SO (Fig. 12). Near real time retrieval Apart from this intrinsic uncertainty associated to the al- using the outlined algorithm illustrates its operational use- gorithm, the accuracy will be determined by knowledge of fulness and robustness for a variety of very different atmo- the plume altitude. This is especially the case in the mid tro- spheric conditions and eruptive plumes. The retrieved total posphere where we have a large temperature gradient. An- masses agree well with those retrieved from other sensors other source of error is the presence of (volcanic) aerosols (AIRS and OMI) as reported on various forums and news and while the magnitude of the associated errors in the re- sites. A full analysis taking into account precise altitude es- trieval is hard to quantify, thin ash clouds will in general timates is out of the scope of this paper and we report total lead to slightly overestimated loadings, while thick opaque masses here assuming an altitude of 10 km. aerosol layers can cover up part or all SO and will give rise Grimsvotn ¨ (Iceland) erupted first on 21 May, with about to underestimates. Although a validation or comparison of this algorithm is 350–400 kT of SO . Last traces of the initial plume were ob- served until the 15 June. SO from Puyehue-Cordon ´ Caulle out of the scope of this paper, we have illustrated the algo- (Chile) was detected first on 5 June; and a fast westerly jet rithm on a number of examples and found that the results stream carried the plume of about 250 kT SO round the were in agreement with the literature. world in 9–10 days. The third eruption was the one of the Acknowledgements. IASI has been developed and built under the volcano Nabro (Eritrea), which was prior to this event be- responsibility of the Centre National d’Etudes Spatiales (CNES, lieved to be totally extinct and is not monitored so actively France). It is flown onboard the Metop satellites as part of the as other volcanoes. First SO was measured on 12 June and EUMETSAT Polar System. The IASI L1 data are received through continued emissions were observed in the days and weeks the EUMETCast near real time data distribution service. L. Clarisse which followed. Total masses of the order 1.5 Tg were mea- and P.-F. Coheur are respectively Postdoctoral Researcher (Charge ´ sured. Water and ice rich plumes and low altitude filaments de Recherches) and Research Associate (Chercheur Qualifie) ´ with F.R.S.-FNRS. C. Clerbaux is grateful to CNES for scientific hampered retrieval on several occasions, and we therefore be- collaboration and financial support. The research in Belgium was lieve this to be a lower bound. By the end of June all traces of funded by the F.R.S.-FNRS (M.I.S. nF.4511.08), the Belgian State Nabro plumes disappeared, which indicates a shorter lifetime Federal Office for Scientific, Technical and Cultural Affairs and of SO compared to Kasatochi or Sarychev. This is possibly the European Space Agency (ESA-Prodex arrangements and the due larger H O and OH concentrations at tropical latitudes. Support to Aviation Control Service (SACS+) project). We also Apart from large volcanic eruptions, IASI regularly picks acknowledge financial support from the EVOSS project (2nd Space up smaller puffs from world’s most active volcanoes such Call of the 7th Framework Program of the European Commission, as Etna. As an example, Fig. 13 shows some snap- with grant agreement 242535) and the “Actions de Recherche shots of volcanic plumes detected in the first part of 2011 Concertees” ´ (Communaute ´ Franc ¸aise de Belgique). We would over the Kamchatka Peninsula (originating from volcanoes like to thank the referees for the many useful corrections and improvements. such as Bezymianny, Kizimen, Karymsky, Kliuchevskoi and Shiveluch). Edited by: F. Prata 5 Conclusions References In this paper we have presented an algorithm for retrieving Andres, R. and Kasgnoc, A.: A time-averaged inventory of sub- SO abundances from IASI, although the algorithm can in aerial volcanic sulfur emissions, J. Geophys. Res., 103, 25251– principle be applied to any high resolution TIR sounder with 25261, 1998. Bluth, G. J. S., Schnetzler, C. C., Krueger, A. J., and Walker, L. S.: sufficient spectral resolution. It was specifically designed for The contribution of explosive volcanism to global atmospheric quantifying high altitude SO plumes from volcanic erup- sulphur dioxide concentrations, Nature, 366, 327–329, 1993. tions. A first attractive feature of the algorithm is its robust- Bobrowski, N., Kern, C., Platt, U., Hormann, ¨ C., and Wag- ness, simplicity and near real time applicability. With just a ner, T.: Novel SO spectral evaluation scheme using the 360– few lines of code this algorithm could for instance be imple- 390 nm wavelength range, Atmos. Meas. Tech., 3, 879–891, mented by volcanic ash advisory centers. Its second strong doi:10.5194/amt-3-879-2010, 2010. point is its very low theoretical uncertainty (3 % uncertainty Campion, R., Salerno, G. G., Coheur, P.-F., Hurtmans, D., Clarisse, for 0.5–100 DU and 6 % for 100–5000 DU for assumed alti- L., Kazahaya, K., Burton, M., Caltabiano, T., Clerbaux, C., and tudes above 500 hPa) coupled with a large applicable range Bernard, A.: Measuring volcanic degassing of SO in the lower (4 orders of magnitude of SO columns 0.5 to 5000 DU). By 2 troposphere with ASTER band ratios, J. Volcanol. Geoth. Res., not using all IASI’s channels the outlined algorithm does not 194, 42–54, doi:10.1016/j.jvolgeores.2010.04.010, 2010. Atmos. Meas. Tech., 5, 581–594, 2012 www.atmos-meas-tech.net/5/581/2012/ L. Clarisse et al.: Retrieving SO from IASI measurements 593 Carn, S. A. and Lopez, T. M.: Opportunistic validation of sul- Corradini, S., Merucci, L., Prata, A. J., and Piscini, A.: Volcanic fur dioxide in the Sarychev Peak volcanic eruption cloud, At- ash and SO in the 2008 Kasatochi eruption: Retrievals compar- mos. Meas. Tech., 4, 1705–1712, doi:10.5194/amt-4-1705-2011, ison from different IR satellite sensors, J. Geophys. Res., 115, 2011. D00L21, doi:10.1029/2009JD013634, 2010. Carn, S. A. and Prata, F.: Satellite-based constraints on explosive Doutriaux-Boucher, M. and Dubuisson, P.: Detection of volcanic SO release from Soufriere ` Hills Volcano, Montserrat, Geophys. SO by spaceborne infrared radiometers, Atmos. Res., 92, 69– 2 2 Res. Lett., 37, L00E22, doi:10.1029/2010GL044971, 2010. 79, doi:10.1016/j.atmosres.2008.08.009, 2008. Carn, S. A., Strow, L., de Souza-Machado, S., Edmonds, Y., and Flaud, J.-M., Lafferty, W., and Sams, R.: Line intensities for the ν , Hannon, S.: Quantifying tropospheric volcanic emissions with ν , and ν + ν bands of SO , J. Quant. Spectrosc. Ra., 110, 3 1 3 2 AIRS: The 2002 eruption of Mt. Etna (Italy), Geophys. Res. 669–674, 2009. Lett., 32, L02301, doi:10.1029/2004GL021034, 2005. Graf, H., Feichter, J., and Langmann, B.: Volcanic sulfur emissions: Carn, S. A., Krueger, A. J., Krotkov, N. A., Yang, K., and Evans, Estimates of source strength and its contribution to the global K.: Tracking volcanic sulfur dioxide clouds for aviation hazard sulfate distribution, J. Geophys. Res., 102, 10727–10738, 1997. mitigation, Nat. Hazards, 51, 325–343, 2009. Halmer, M., Schmincke, H.-U., and Graf, H.-F.: The annual vol- Chevallier, F.: Sampled databases of 60-level atmospheric profiles canic gas input into the atmoshpere, in particular into the strato- from the ECMWF analyses, Tech. rep., Eumetsat/ECMWF SAF sphere: a global data set for the past 100 years, J. Volcanol. Programme, Research Report No. 4, 2001. Geoth. Res., 115, 511–528, 2002. Chin, M. and Jacob, D.: Anthropogenic and natural contributions to Haywood, J. and Boucher, O.: Estimates of the direct and indi- tropospheric sulfate: A global model analysis, J. Geophys. Res., rect radiative forcing due to tropospheric aerosols: a review, Rev. 101, 18691–18699, 1996. Geophys., 38, 513–543, 2000. Clarisse, L., Coheur, P. F., Prata, A. J., Hurtmans, D., Razavi, Haywood, J. M., Jones, A., Clarisse, L., Bourassa, A., Barnes, A., Phulpin, T., Hadji-Lazaro, J., and Clerbaux, C.: Tracking J., Telford, P., Bellouin, N., Boucher, O., Agnew, P., Clerbaux, and quantifying volcanic SO with IASI, the September 2007 C., Coheur, P., Degenstein, D., and Braesicke, P.: Observations eruption at Jebel at Tair, Atmos. Chem. Phys., 8, 7723–7734, of the eruption of the Sarychev volcano and simulations using doi:10.5194/acp-8-7723-2008, 2008. the HadGEM2 climate model, J. Geophys. Res., 115, D21212, Clarisse, L., Hurtmans, D., Prata, A. J., Karagulian, F., Clerbaux, doi:10.1029/2010JD014447, 2010. C., Maziere, ` M. D., and Coheur, P.-F.: Retrieving radius, con- Karagulian, F., Clarisse, L., Clerbaux, C., Prata, A. J., Hurtmans, centration, optical depth, and mass of different types of aerosols D., and Coheur, P. F.: Detection of volcanic SO , ash and from high-resolution infrared nadir spectra, Appl. Optics, 49, H SO using the IASI sounder, J. Geophys. Res., 115, D00L02, 2 4 3713–3722, doi:10.1364/AO.49.003713, 2010a. doi:10.1029/2009JD012786, 2010. Clarisse, L., Prata, F., Lacour, J.-L., Hurtmans, D., Clerbaux, C., Kravitz, B., Robock, A., Bourassa, A., Deshler, T., Wu, D., Mattis, and Coheur, P.-F.: A correlation method for volcanic ash detec- I., Finger, F., Hoffmann, A., Ritter, C., Bitar, L., Duck, T. J., tion using hyperspectral infrared measurements, Geophys. Res. and Barnes, J. E.: Simulation and observations of stratospheric Lett., 37, L19806, doi:10.1029/2010GL044828, 2010b. aerosols from the 2009 Sarychev volcanic eruption, J. Geophys. Clarisse, L., Coheur, P.-F., Chefdeville, S., Lacour, J.-L., Hurt- Res., 116, D18211, doi:10.1029/2010JD015501, 2011. mans, D., and Clerbaux, C.: Infrared satellite observations Kristiansen, N. I., Stohl, A., Prata, A. J., Richter, A., Eckhardt, S., of hydrogen sulfide in the volcanic plume of the August Seibert, P., Hoffmann, A., Ritter, C., Bitar, L., Duck, T. J., and 2008 Kasatochi eruption, Geophys. Res. Lett., 38, L10804, Stebel, K.: Remote sensing and inverse transport modeling of the doi:10.1029/2011GL047402, 2011a. Kasatochi eruption sulfur dioxide cloud, J. Geophys. Res., 115, Clarisse, L., Fromm, M., Ngadi, Y., Emmons, L., Clerbaux, C., D00L16, doi:10.1029/2009JD013286, 2010. Hurtmans, D., and Coheur, P.-F.: Intercontinental transport of Krotkov, N., Schoeberl, M., Morris, G., Carn, S., and Yang, anthropogenic sulfur dioxide and other pollutants; an infrared K.: Dispersion and lifetime of the SO cloud from the Au- remote sensing case study, Geophys. Res. Lett., 38, L19806, gust 2008 Kasatochi eruption, J. Geophys. Res., 115, D00L20, doi:10.1029/2011GL048976, 2011b. doi:10.1029/2010JD013984, 2010. Clerbaux, C., Coheur, P.-F., Clarisse, L., Hadji-Lazaro, J., Hurt- Krueger, A., Walter, L., Bhartia, P., Schnetzler, C., Krotkov, N., mans, D., Turquety, S., Bowman, K., Worden, H., and Carn, Sprod, I., and Bluth, G.: Volcanic sulfur dioxide measurements S.: Measurements of SO profiles in volcanic plumes from the from the Total Ozone Mapping Spectrometer (TOMS) Instru- NASA Tropospheric Emission Spectrometer (TES), Geophys. ments., J. Geophys. Res., 100, 14057–14076, 1995. Res. Lett., 35, L22807, doi:10.1029/2008GL035566, 2008. Krueger, A., Krotkov, N., and Carn, S.: El Chichon: Clerbaux, C., Boynard, A., Clarisse, L., George, M., Hadji-Lazaro, The genesis of volcanic sulfur dioxide monitoring J., Herbin, H., Hurtmans, D., Pommier, M., Razavi, A., Turquety, from space, J. Volcanol. Geoth. Res., 175, 408–414, S., Wespes, C., and Coheur, P.-F.: Monitoring of atmospheric doi:10.1016/j.jvolgeores.2008.02.026, 2008. composition using the thermal infrared IASI/MetOp sounder, At- Lee, C., Martin, R. V., van Donkelaar, A., Lee, H., Dickerson, mos. Chem. Phys., 9, 6041–6054, doi:10.5194/acp-9-6041-2009, R. R., Hains, J. C., Krotkov, N., Richter, A., Vinnikov, K., 2009. and Schwab, J. J.: SO emissions and lifetimes: Estimates Corradini, S., Merucci, L., and Prata, A. J.: Retrieval of SO from from inverse modeling using in situ and global, space-based thermal infrared satellite measurements: correction procedures (SCIAMACHY and OMI) observations, J. Geophys. Res., 116, for the effects of volcanic ash, Atmos. Meas. Tech., 2, 177–191, D06304, doi:10.1029/2010JD014758, 2011. doi:10.5194/amt-2-177-2009, 2009. Longhurst, J. W., Raper, D. W., Lee, D. S., Heath, B. A., Con- www.atmos-meas-tech.net/5/581/2012/ Atmos. Meas. Tech., 5, 581–594, 2012 594 L. Clarisse et al.: Retrieving SO from IASI measurements lan, B., and King, H. J.: Acid deposition: a select review 1852– come, N., Lafferty, W., Mandin, J.-Y., Massie, S., Mikhailenko, 1990: 1. Emissions, transport, deposition, effects on freshwa- S., Miller, C., Moazzen-Ahmadi, N., Naumenko, O., Nikitin, A., ter systems and forests, Fuel, 72, 1261–1280, doi:10.1016/0016- Orphal, J., Perevalov, V., Perrin, A., Predoi-Cross, A., Rinsland, 2361(93)90125-L, 1993. C., Rotger, M., Simeck ˇ ova, ´ M., Smith, M., Sung, K., Tashkun, Matoza, R. S., Pichon, A. L., Vergoz, J., Herry, P., Lalande, J.- S., Tennyson, J., Toth, R., Vandaele, A., and Auwera, J. V.: M., Lee, H.-I., Che, I.-Y., and Rybin, A.: Infrasonic obser- The HITRAN 2008 molecular spectroscopic database, Journal vations of the June 2009 Sarychev Peak eruption, Kuril Is- of Quantitative Spectroscopy and Radiative Transfer, 110, 533– lands: Implications for infrasonic monitoring of remote ex- 572, doi:10.1016/j.jqsrt.2009.02.013, 2009. plosive volcanism, J. Volcanol. Geoth. Res., 200, 35–48, Rybin, A., Chibisova, M., Webley, P., Steensen, T., Izbekov, P., doi:10.1016/j.jvolgeores.2010.11.022, 2011. Neal, C., and Realmuto, V.: Satellite and ground observations Prata, A.: Satellite detection of hazardous volcanic clouds and of the June 2009 eruption of Sarychev Peak volcano, Matua Is- the risk to global air traffic, Nat. Hazards, 51, 303–324, land, Central Kuriles, B. Volcanol., 1–16, doi:10.1007/s00445- doi:10.1007/s11069-008-9273-z, 2008. 011-0481-0, 2011. Prata, A. and Bernardo, C.: Retrieval of volcanic SO column Smith, S. J., van Aardenne, J., Klimont, Z., Andres, R. J., abundance from Atmospheric Infrared Sounder data, J. Geophys. Volke, A., and Delgado Arias, S.: Anthropogenic sulfur diox- Res., 112, D20204, doi:10.1029/2006JD007955, 2007. ide emissions: 1850–2005, Atmos. Chem. Phys., 11, 1101–1116, Prata, A. and Kerkmann, J.: Simultaneous retrieval of volcanic doi:10.5194/acp-11-1101-2011, 2011. ash and SO using MSG-SEVIRI measurements, Geophys. Res. Surono, Jousset, P., Pallister, J., Boichu, M., Buongiorno, M., Bud- Lett., 34, L05813, doi:10.1029/2006GL028691, 2007. isantoso, A., Costa, F., Andreastuti, S., Prata, F., Schneider, D., Prata, A., Rose, W., Self, S., and O’Brien, D.: Global, long-term Clarisse, L., Humaida, H., Sumarti, S., Bignami, C., Griswold, J., sulphur dioxide measurements from TOVS data: A new tool for Carn, S., and Oppenheimer, C.: The 2010 explosive eruption of studying explosive volcanism and climate, in: Volcanism and the Javas Merapi volcano a 100-year event, submitted to J. Volcanol. Earth’s Atmosphere, edited by: Robock, A. and Oppenheimer, Geoth. Res., 2012. C., vol. 139 of Geophys. Monogr., 75–92, AGU, Washington, Thomas, H. and Watson, I.: Observations of volcanic emissions DC, 2003. from space: current and future perspectives, Nat. Hazards, 54, Prata, A., Gangale, G., Clarisse, L., and Karagulian, F.: Ash and 323–354, 2010. sulfur dioxide in the 2008 eruptions of Okmok and Kasatochi: Vernier, J.-P., Pommereau, J.-P., Thomason, L. W., Pelon, J., Gar- Insights from high spectral resolution satellite measurements, nier, A., Deshler, T., Jumelet, J., and Nielsen, J. K.: Overshoot- J. Geophys. Res., 115, D00L18, doi:10.1029/2009JD013556, ing of clean tropospheric air in the tropical lower stratosphere 2010. as seen by the CALIPSO lidar, Atmos. Chem. Phys., 11, 9683– Realmuto, V., Abrams, M., Buongiorno, M., and Pieri, D.: The 9696, doi:10.5194/acp-11-9683-2011, 2011. use of multispectral thermal infrared image data to estimate the Walker, J. C., Dudhia, A., and Carboni, E.: An effective method sulfur dioxide flux from volcanoes: A case study from Mount for the detection of trace species demonstrated using the MetOp Etna, Sicily, 29 July 1986, J. Geophys. Res., 99, 481–488, 1994. Infrared Atmospheric Sounding Interferometer, Atmos. Meas. Realmuto, V., Sutton, A., and Elias, T.: Multispectral thermal in- Tech., 4, 1567–1580, doi:10.5194/amt-4-1567-2011, 2011. frared mapping of sulfur dioxide plumes: A case study from the Watson, I., Realmuto, V., Rose, W., Prata, A., Bluth, G., Gu, Y., East Rift Zone of Kilauea Volcano, Hawaii, J. Geophys. Res., Bader, C., and Yu, T.: Thermal infrared remote sensing of vol- 102, 15057–15072, 1997. canic emissions using the moderate resolution imaging spectro- Richter, A., Wittrock, F., Schonhardt, A., and Burrows, J.: Quan- radiometer, J. Volcanol. Geoth. Res., 135, 75–89, 2004. tifying volcanic SO emissions using GOME2 measurements, Waythomas, C. F., Scott, W. E., Prejean, S. G., Schneider, D. J., Geophys. Res. Abstr., EGU2009-7679, EGU General Assembly Izbekov, P., and Nye, C. J.: The 7–8 August 2008 eruption of 2009, Vienna, Austria, 2009. Kasatochi Volcano, central Aleutian Islands, Alaska, J. Geophys. Rix, M., Valks, P., Hao, N., van Geffen, J., Clerbaux, C., Clarisse, Res., 115, B00B06, doi:10.1029/2010JB007437, 2010. L., Coheur, P.-F., Loyola R, D. G., Erbertseder, T., Zimmer, W., Yang, K., Krotkov, N., Krueger, A., Carn, S., Bhartia, P., and Levelt, and Emmadi, S.: Satellite Monitoring of Volcanic Sulfur Dioxide P.: Retrieval of large volcanic SO columns from the Aura Ozone Emissions for Early Warning of Volcanic Hazards, IEEE J. Sel. Monitoring Instrument: Comparison and limitations, J. Geophys. Top. Appl., 2, 196–206, doi:10.1109/JSTARS.2009.2031120, Res., 112, D24S43, doi:10.1029/2007JD008825, 2007. 2009. Zhang, Q., Jimenez, J. L., Canagaratna, M. R., Allan, J. D., Robock, A.: Volcanic eruptions and climate, Rev. Geophys., 38, Coe, H., Ulbrich, I., Alfarra, M. R., Takami, A., Middlebrook, 191–219, 2000. A. M., Sun, Y. L., Dzepina, K., Dunlea, E., Docherty, K., De- Rose, W. I., Bluth, G., and Watson, I.: Ice in volcanic clouds: When Carlo, P. F., Salcedo, D., Onasch, T., Jayne, J. T., Miyoshi, and where?, in: Proc. of the 2nd Int. Conf. on Volcanic Ash and T., Shimono, A., Hatakeyama, S., Takegawa, N., Kondo, Y., Aviation Safety, OFCM, Washington, DC, Session 3, 61, Ex- Schneider, J., Drewnick, F., Borrmann, S., Weimer, S., Demer- tended observations of volcanic SO and sulfate aerosol in the jian, K., Williams, P., Bower, K., Bahreini, R., Cottrell, L., stratosphere, 2004. Griffin, R. J., Rautiainen, J., Sun, J. Y., Zhang, Y. M., and Rothman, L., Gordon, I., Barbe, A., Benner, D., Bernath, P., Birk, Worsnop, D. R.: Ubiquity and dominance of oxygenated species M., Boudon, V., Brown, L., Campargue, A., Champion, J.-P., in organic aerosols in anthropogenically-influenced Northern Chance, K., Coudert, L., Dana, V., Devi, V., Fally, S., Flaud, J.- Hemisphere midlatitudes, Geophys. Res. Lett., 34, L13801, M., Gamache, R., Goldman, A., Jacquemart, D., Kleiner, I., La- doi:10.1029/2007GL029979, 2007. Atmos. Meas. Tech., 5, 581–594, 2012 www.atmos-meas-tech.net/5/581/2012/

Journal

Atmospheric measurement techniques – Unpaywall

Published: Mar 13, 2012

There are no references for this article.