Volcanic ash retrieval at Mt. Etna using Avhrr and Modis data

Transcription

1 Volcanic ash retrieval at Mt. Etna using Avhrr and Modis data Claudia Spinetti* a, Stefano Corradini a, Maria F. Buongiorno a a Istituto Nazionale di Geofisica e Vulcanologia, via di Vigna Murata, Roma Italy ABSTRACT The volcanic ash detection procedures are based on Brightness Temperature Difference (BTD) algorithm using the thermal infrared channels centred around 11 and 12 microns of a multispectral satellite sensor. The Mie code has been is included in the retrieval procedure to compute the ash optical properties from the ash microphysical characteristics. The simulations has been carried out using MODTRAN radiative transfer model. The Nasa-Modis and the Noaa-Avhrr measurements of Mt. Etna eruptive plume occurred in November 2006 have been analyzed to retrieve the plume optical thickness, the particle effective radius and the size distribution. Keywords: volcanic ash, MODIS, AVHRR, radiative transfer model, Mt. Etna 1. INTRODUCTION During 2006 Mt. Etna volcano (Sicily-Italy, see Figure 1) was characterize by episodic explosive activity with formation of volcanic ash plumes. These episodes occurred frequently from September to December but showing weak intensity and short duration compared to previous 2001 and 2002 eruptions. The 24 November 2006 at the about 03:00 GMT start the ash emission that ended at about 17 GMT; this episode represent the most voluminous episode occurred at Mt. Etna in term of ash emissions of the entire September-December eruptive period. During this day both NASA-MODIS (Moderate Resolution Imaging Spectroradiometer) and NOAA-AVHRR (Advanced Very High Resolution Radiometer) satellites measurements have been acquired (see Section 2). In the present study the AVHRR and MODIS data have been processed using the BTD algorithm in order to detect the ash presence in the plume; a retrieval procedure was also developed to compute the ash optical thickness, particles effective radius and total ash mass present in air at the time of MODIS and AVHRR acquisition. Fig 1. Mt. Etna location. *spinetti@ingv.it; phone ; fax ;

2 1.1 Mount Etna s 2006 Eruption events The eruption considered in this work is the episode occurred the 24 November 2006 and took place at the SE crater located on Piano del Lago, in the southern flank of Mt. Etna. The explosive activity were characterized by the presence of bombs, scorie, black ash and lapilli column with rising up to 2 km above the vent [1, 2]. During the day the wind was blowing towards the S-SE direction. Due to a wind change, the ash plume moves towards and the city of Catania was affected by ash fallout creating major problems for the Fontanarossa International Airport that was close to air traffic. The finer portion of ash reached above 80 km away from the summit craters over the Mediterranean sea. 2. SATELLITE OBSERVATIONS The eruptive ash plume analyzed has been recorded by NOAA AVHRR and NASA MODIS. The satellite NOAA 18 with on board the AVHRR version 3 acquired TIR image at 11:20 GMT while the Aqua satellite with on board MODIS acquired TIR image 1 hours later at 12:20 GMT. The Figures 2 and 3 shown the images of the 24 November 2006 for AVHRR and MODIS respectively. Fig. 2. NOAA 18 AVHRR image recorded the 24 November 2006 at 11:20 GMT. Fig. 3. NASA Terra MODIS image recorded the 24 November 2006 at 12:20 GMT.

3 3. RETRIEVAL ALGORITHM The discrimination between volcanic and meteorological clouds is based on different spectral absorption between volcanic aerosols and water aerosols in the µm spectral range [3]. The discrimination algorithms are based on the difference between the brightness temperatures computed from two channels around 11 and 12 µm; such difference will be negative for volcanic clouds and positive for meteorological clouds [3,4,5,6]. Using the TIR AVHRR channels 4 and 5 and TIR MODIS channel 31 and 32, respectively, the BTD procedure has been applied to the georeferenced test images. The white pixels of Figures 4 and 5 indicates the volcanic ash (BTD values lesser than -0.2) while the black pixels of MODIS image indicates the meteorological clouds (BTD values greater than 0.2). Fig. 4. NOAA-AVHRR volcanic cloud detection map using BTD technique. The white pixels indicates the volcanic cloud. Fig. 5. NASA-MODIS volcanic cloud detection map using BTD technique. The white pixels indicates the volcanic cloud and the black pixel the meteorological clouds. The Aerosol Optical Thickness (AOT) and effective radius (r eff ) retrieval is based on work on Prata [3] and is realized computing the AOT-r eff curves [4] by means of MODTRAN [7] Radiative Transfer Model (RTM). MODTRAN needs as input the atmospheric profiles (pressure, temperature and humidity) of the day of measurements, the surface characteristics (temperature and emissivity), the plume geometry and the volcanic ash optical properties. The atmospheric profiles considered is the Trapani radiosounding [8] (Trapani is the nearest WMO meteorological station to

4 the Etnean area) at 12 GMT, the surface emissivity has been set constant and equal to 1 (the retrieval has been made over the sea) and the sea surface temperature, computed using the Split Window (SW) algorithm [9,10] applied to MODIS channels 31 and 32 and AVHRR channels 4 and 5, set at 292 K. The plume altitude (5000 m) is computed comparing the ash plume brightness temperature of channel around 11 µm and the radiosounding temperature profile and the thickness has been set to 1000 m. The ash optical properties (extinction coefficient, single scattering albedo and asymmetry parameter) has been computed using the Oxford University Atmospheric Oceanic and Planetary Physics (AOPP) Mie code [11] and considering as input the spectral refractive index tabulated by Volz et al. [12], a log-normal distribution and varying the effective radius from 0.7 to 10 µm. The simulations has been fulfilled varying the AOT from 0 to 5. Figure 6 shown the AOT-r eff curves computed from the MODTRAN simulations and convoluted with MODIS slit functions; the gray cross are the MODIS measurements. Fig. 6. AOT-r eff curves considering T s =292 K, a plume top altitude at 5000 m and thickness of 1000 m. The AOT vary along the solid lines (from 0 to 5) and the r eff vary along the dash lines (from 1.13 to 6.95µm). The gray cross are the MODIS measurements. 4. RESULTS AND CONCLUSIONS The volcanic ash pixel AOT and effective radius derive from the comparison between the simulated AOT-r eff curves and the AVHRR and MODIS computed BTD (see Figure 6). The mean AOT and effective radius result 0.78 and 2.41µm from AVHRR image and 0.50 and 2.37µm from MODIS image. The map of ash mass has been computed from the equation suggested by Wen and Rose [4]: ( n,m ) ( n,m ) ( n,m ) M ( n, m) = S4 / 3ρreff τ / Qe ( reff ) (1) where M (n,m) is the pixel ash mass, S is the pixel surface, ρ is the volcanic particles density, r eff (n,m) is the retrieved (n,m) pixel effective radius, τ (n,m) is the retreved (n,m) pixel AOT and Q e (r eff (n,m) ) is the extinction efficiency factor at effective radius r eff (n,m). In Figures 7 and 8 has been shown the ash mass maps for AVHRR and MODIS obtained considering ρ=1.3 g/cm 3.

5 Fig. 6. Map of volcanic ash mass derived from NOAA-AVHRR 24 November 2006 at 11:20 GMT. Fig. 7. Map of volcanic ash mass derived from NASA-MODIS 24 November 2006 at 12:20 GMT. ( ( ) The estimated total mass of ash M = Σ n Σ m M ( n,m) results and tons respectively for AVHRR and MODIS. The retrieved mean ash plume particles effective radius for the two sensors result very similar, while the AOT retrieved by AVHRR is meaningfully greater than the MODIS retrieval. This lead up to a greater ash mass plume estimation using AVHRR than MODIS. The difference between the two retrievals can be due mainly to the difference in time acquisition: the AVHRR data is acquired about 1 hour before the MODIS data. This imply that the eruptive event was not constant and the ash was decreasing after the 11:20 GMT. ACKNOWLEDGEMENTS The work is supported by MIUR - FIRB B5 Project and ASI SRV Project. We thank Dr. E. Carboni for the AOPP database; Dr. C. Tirelli and Dr. S. Pughaghi for the theoretical study. We thank the LABTEL-CNT group to satellite system maintenance. MODIS data are provided under Telepazio and ESA accordance.

6 REFERENCES 1. E. De Beni, G. Norini, M. Polacci, 2. C. Spinetti, M.F. Buongiorno, F. Doumaz, M. Musacchio, V. Lombardo, A. Harris, A. Steffke and S. Amici, 3. A. J. Prata and I. J. Barton, Detection and discrimination of volcanic clouds by infrared radiometry I:theory, Proc. of the first international symposium on Volcanic ash and aviation safety, (1989). 4. S. Wen, W. I. Rose, Retrieval of sizes and total masses of particles in volcanic clouds using AVHRR bands 4 and 5, J. of Geoph. Res., 99, D3, (1994). 5. G. P. Ellrod, B. H. Connel, D. W. Hillger, Improved detection of airborne volcanic ash using multispectral infrared satellite data, J. of Geoph. Res., 108, D12, (2003). 6. T. Yu, W. I. Rose, Atmospheric correction for sateliite-based volcanic ash mapping and retrievals using ``split window'' IR data from GOES and AVHRR, J. of Geoph. Res., 107, D16 (2002). 7. G. P. Anderson, F. X. Kneizys, J. H. Chetwynd, J. Wang, M. L. Hoke, L. S. Rithman, L. M. Kimball, R. A. McClatchey, E. P. Shettle., S. A. Clought, W. O. Gallery, L. W. Abreu, J. E. A. Selby, FASCODE/MODTRAN/LOWTRAN: Past/Present/Future, 18 th Annual Review Conference on Atmospheric Transmission Models, 6-8 June (1995). 8. University of Wyoming Department of Atmospheric Science: 9. S. Corradini, S. Pugnaghi, S. Teggi, M. F. Buongiorno, M. P. Bogliolo, Will ASTER see the Etna SO2 plume?, Int. J. of Remote Sens., 24 (6), (2003). 10. S. Pugnaghi, G. Gangale, S. Corradini, M. F. Buongiorno, Mt. Etna sulfur dioxide flux monitoring using ASTER- TIR data and atmospheric observations, J. of Volc. and Geoth. Res, 152, (2006). 11. Oxford University AOPP department: F. E. Volz, Infrared optical constants of ammonium sulfate, Sahara dust, volcanic pumice and fly ash, App. Optics, 12, (1973).