Advertisement

26 views

Quantitative scanning-electron microscope analysis of volcanic ash surfaces: Application to the 1982–1983 Galunggung eruption (Indonesia

Quantitative scanning-electron microscope analysis of volcanic ash surfaces: Application to the 1982–1983 Galunggung eruption (Indonesia
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Tags
Transcript
  Geological Society of America Bulletin doi: 10.1130/B26048.1 2007;119, no. 5-6;743-752 Geological Society of America Bulletin   Orkun Ersoy, Alain Gourgaud, Erkan Aydar, Gary Chinga and Jean-Claude Thouret  1983 Galunggung eruption (Indonesia) − Application to the 1982Quantitative scanning-electron microscope analysis of volcanic ash surfaces:   Email alerting services articles cite this article to receive free e-mail alerts when newwww.gsapubs.org/cgi/alertsclick  Subscribe America Bulletin to subscribe to Geological Society ofwww.gsapubs.org/subscriptions/ click  Permission request  to contact GSAhttp://www.geosociety.org/pubs/copyrt.htm#gsaclick  official positions of the Society.citizenship, gender, religion, or political viewpoint. Opinions presented in this publication do not reflectpresentation of diverse opinions and positions by scientists worldwide, regardless of their race, includes a reference to the article's full citation. GSA provides this and other forums for thethe abstracts only of their articles on their own or their organization's Web site providing the posting to further education and science. This file may not be posted to any Web site, but authors may postworks and to make unlimited copies of items in GSA's journals for noncommercial use in classrooms requests to GSA, to use a single figure, a single table, and/or a brief paragraph of text in subsequenttheir employment. Individual scientists are hereby granted permission, without fees or further Copyright not claimed on content prepared wholly by U.S. government employees within scope of Notes GEOLOGICAL SOCIETY OF AMERICA  on June 29, 2012gsabulletin.gsapubs.orgDownloaded from  For permission to copy, contact editing@geosociety.org© 2007 Geological Society of America  743 ABSTRACTQualitative analyses of volcanic ash are time-consuming and subjective, whereas quantitative analyses are methodical and automated. Not only volcanic ash particles, but also many natural particles have been widely described and quantified by their outlines. The qualitative data of volcanic ash surfaces need to be expressed quantitatively, supported by supplementary methods such as statistical analysis and artificial intelli-gence. Well-defined surface descriptors can be applied to volcanic ash particles. In this study, roughness and texture descriptors of pyroclastic material from the 1982–1983 eruption of Galunggung (Java, Indonesia) were used to describe the vesicle surfaces of the particles, alteration intensity, and/or fine particle abundance. These parameters are important for distinguishing the products of magmatic eruptions from those of phre-atomagmatic eruptions. Further application of this method may allow these descriptors to be easily converted to alteration grade, vesicularity index, intensity of the fragmen-tation mechanism, and relative proportions of the pyroclast types. Hence, discrimination between products of different fragmentation mechanisms may permit forecasting of volca-nic hazards.Keywords:  volcanic ash, roughness, gradient analysis, texture descriptors, Galunggung. INTRODUCTIONMicroscopy Analysis of Volcanic Ash Qualitative analyses of volcanic ash are time-consuming and may be affected by the human operator, whereas quantitative analyses are methodical and automated. Multiple samples can be analyzed quantitatively over a short period of time with limited intervention, allowing quan-titative analyses to produce reliable results. Diverse efficient methods have been developed to describe and quantify natural objects in geo-sciences, such as gastropod shells (Dommer-gues et al., 2003) and detrital sediments (Drolon et al., 2003). Qualitative data describing volca-nic ash also need to be expressed quantitatively, which requires supplementary methods includ-ing statistical analysis and artificial intelligence. Scanning electron microscopy (SEM) provides classification of volcanic ash based on surface morphology and texture (Wohletz and Krinsley, 1982; Büttner et al., 1999; Ersoy et al., 2006), and the most extensive SEM studies of pyroclast shapes to date have been presented by Heiken (1972, 1974), Wohletz (1983), and Heiken and Wohletz (1985). A study of explosive fragmen-tation dynamics by examination of the mor-phological features of natural and experimental ash particles has demonstrated the significance of morphological microfeatures on ash grains (Wohletz, 1983). The lack of analogue particles produced by scaled experiments at that time, however, allowed only a qualitative assess-ment. Although a study of natural pyroclastic sequences combined with scaled laboratory experiments identified the different fragmenta-tion mechanisms related to the water/magma mass ratios during their interaction (Büttner et al., 1999), the surface features of the ash were not quantified. Bayhurst et al. (1994) developed an automated program for the characterization of volcanic ash particles of Redoubt volcano (Alaska) based on size, density, shape, and min-eralogy using a SEM with an energy-dispersive X-ray detector (EDS); however, classification of volcanic ash surfaces is still limited to descrip-tive terms such as stepped, planar, crack pat-terns, and hydration skin. Surface Measurements Several texture descriptors have been devel-oped to characterize the detailed surface structure of aluminum (Lee et al., 1998), aggregate (Rao  et al., 2003), wear particles (Stachowiak, 1998), and paper surfaces (Chinga et al., 2003; Chinga, 2004). Fractal dimension, autocorrelation, gra-dient analysis, band-pass filtering, wavelet anal-ysis, roughness statistics, and quadtree decom-position have been applied to assess complex surface structures (Panozzo, 1992; Costa, 2000; Chinga et al., 2003; Chinga, 2006). Analysis Quantitative scanning-electron microscope analysis of volcanic ash surfaces: Application to the 1982–1983 Galunggung eruption (Indonesia) Orkun Ersoy †  Department of Geological Engineering, Hacettepe University, 06532, Beytepe-Ankara, Turkey, and Université Blaise Pascal, 5 rue Kessler, 63038 Clermont-Ferrand, France Alain Gourgaud Observatoire de Physique du Globe de Clermont-Ferrand and Institut de Recherche pour le Développement, Université Blaise Pascal, UMR-CNRS 6524, 5 rue Kessler, 63038 Clermont-Ferrand, France Erkan Aydar  Department of Geological Engineering, Hacettepe University, 06532, Beytepe-Ankara, Turkey Gary Chinga GCSCA, Trondheim, Norway Jean-Claude Thouret OPGC and IRD, Université Blaise Pascal, UMR-CNRS 6524, 5 rue Kessler, 63038 Clermont-Ferrand, France † E-mail: oersoy@hacettepe.edu.tr. GSA Bulletin ; May/June 2007; v. 119; no. 5/6; p. 743–752; doi: 10.1130/B26048.1; 7 figures; 4 tables.  on June 29, 2012gsabulletin.gsapubs.orgDownloaded from   Ersoy et al. 744 Geological Society of America Bulletin, May/June 2007of surfaces of volcanic ash particles has been commonly limited to their outlines (Dellino and LaVolpe, 1996; Dellino and Liotino, 2002, Riley et al., 2003). Durant et al. (2004) and Horwell et al. (2003) measured the surface area of volcanic ash particles to investigate the ice-nucleating ability and to measure the production of radicals on the surfaces, respectively. Ersoy et al. (2006) calculated quadtree decomposition parameters and the surface descriptors derived from gradient analysis in order to quantify the structural changes of the ash surfaces due to variable explosion conditions in the phreatoplin-ian eruption of Nemrut volcano. The quadtree decomposition was performed by assessing the local gray-level standard deviation. The com-putation of the gradient of the presumed height data in the presumed topographical image was based on the partial derivatives at every pixel in the image. The 1982–1983 Eruption of Galunggung During the last eruption of Galunggung, Java, in 1982–1983, the composition of the erupted magma evolved from andesite (58% SiO 2 ) to Mg-rich basalt (47% SiO 2 ) (Gerbe et al., 1992), while the style of the eruption changed markedly through time (Katili and Sudradjat, 1984; Sudradjat and Tilling, 1984; Gourgaud et al., 1989). Observers from the Volcanological Survey of Indonesia (Katili and Sudradjat, 1984) recognized three distinct phases with three different eruptive styles: an initial Vulcanian phase 1, a phreatomagmatic phase 2, and a Strombolian phase 3. Gourgaud et al. (1989, 2000) emphasized the increase in explosivity (expressed as volcanic explosivity index; Newhall and Self, 1982) from the Vul-canian phase 1 to the phreatomagmatic phase 2, including higher plume height, larger volume of deposits, and great changes in crater mor-phology. In particular, Gourgaud et al. (2000) identified the transition between the phases with systematic variations of clast vesicularity and xenolith proportions and aimed to issue a volcanic-hazard forecast. Phase 1 was Vulca-nian, but according to the limited involvement of groundwater in Vulcanian eruptions and Schmincke’s proposal (Schmincke, 1977), the dynamism of phase 1 was already phreatomag-matic, and phase 2 represented an increase in phreatomagmatic activity (Gourgaud et al., 2000). The grain-size analysis did not reflect the transition from phase 1 to phase 2 (Gourgaud et al., 2000). The slight but continuous decrease in the vesicularity index of juvenile clasts and progressive increase in the ratio of xenolith ver-sus juvenile lava clasts showed evidence of the increasing efficiency of groundwater-magma interaction during eruption (Gourgaud et al., 2000). Irrespective of the grade of phreatomag-matic activity, we use the term “Vulcanian” herein for phase 1 to suggest that less water was interacting with magma than during phase 2. Types of Pyroclasts in 1982–1983 Tephra Three main sequences of tephra properties during the 1982–1983 eruption are related to the three distinct eruptive phases. The varia-tions in xenolith proportions and vesicularity in pyroclastic-flow deposits record an increase in magma-water interaction through the transition from Vulcanian phase 1 to phreatomagmatic phase 2 (Gourgaud et al., 2000). Here, we have available only xenolith and vesicularity data for pyroclasts from the pyroclastic flows of Vul-canian phase 1 and phreatomagmatic phase 2 (Gourgaud et al., 2000) to make a comparison with our quantitative results. Thus, we deal with pyroclastic-flow deposits of phase 1 and phase 2 from the Hot River and Cibanjaran valley sec-tions (Fig. 1). The detailed stratigraphy includes four sections (Hot River, Cikunir, Cipanas, Cibanjaran), and all units can be found in Gour-gaud et al. (2000). The grain-size interval of 250–500 µm was selected to enable comparison of the textures with those recorded in other stud-ies (Wohletz, 1983; Heiken and Wohletz, 1985). Furthermore, the fine ash fractions often show shapes or textures that are distinct from coarser fractions, and they may be more definitive of explosive fragmentation (e.g., Zimanowski et al., 2003). Micrographs of whole grains and detailed surface configurations were obtained with JEOL JSM-5910 and CAMECA SU-30 operating at secondary electron modes with 5–15 keV at Laboratoire Magmas et Volcans (Clermont-Ferrand, France) and Hacettepe Uni-versity (Ankara, Turkey), respectively.In this study, two dominant juvenile pyroclast types were distinguished on 220 SEM micro-graphs. Their shapes correspond to type 1 and type 2 recognized by Wohletz (1983). The sur-faces of particles show the influence of water-magma interaction, a form of “fuel-coolant interaction (FCI)” (e.g., Sheridan and Wohletz, 1983). FCI involves the contact of two fluids, where the fuel has a temperature above the boil-ing point of the coolant (Board et al., 1974; Buchanan, 1974; Board and Hall, 1975; Frölich et al., 1976; Drumheller, 1979; Corradini, 1981). The interaction generally results in vaporization of the coolant and chilling or quenching of the fuel. The fragmentation-vaporization process can be a cyclic process of vapor film genera-tion and collapse. The energy of this collapse is partially cycled back into the system, generat-ing new contact surfaces so that the system is self-sustaining. The collapse of a superheated vapor film or the explosive expansion of the film produces stress waves in the melt. If these exceed the bulk modulus of the melt and if it fractures in a brittle fashion, blocky type 1 or type 2 pyroclasts may form (see Figure 19 in Wohletz, 1983). Type 1 pyroclasts are found in compositions ranging from basaltic to rhyolitic. Type 2 pyroclasts were only found in basaltic compositions, especially in Surtsey tephra, by Wohletz (1983).Pyroclasts of different types have unique shapes and textures and can therefore be dis-tinguished visually on 220 SEM micrographs by an operator. Type 1 pyroclasts have blocky, equant morphology. Typically, vesicular sur-faces are rare and cut by curviplanar fracture surfaces. The irregularities on surfaces of type 1 pyroclasts belong to alteration products and adhering dust or vesicle embayment. Type 2 pyroclasts have surfaces controlled by vesicle walls. Vesicle edges are rounded and smoothed, and overall grain shape is irregular. The smooth curved surfaces between vesicles are lumpy and appear to be fused and fluid-formed. Quenching and solidification during and after brittle frac-ture probably preserved the blocky shapes with curviplanar surfaces (type 1). Solidification and formation of a quenched crust was probably incomplete after fracture, and subsequent move-ment of fragments out of the zone of interaction formed smooth, fluid-like surfaces on fragments (type 2) (Wohletz, 1983). Definitive Characteristics of 1982–1983 Tephra The first sequence of the 1982–1983 erupted tephra, including initial, high-energy pyroclas-tic surges at the onset of eruption, lithic-rich pyroclastic flows, scoriaceous block-and-ash flows and, brief pulses of surges interspersed with brief and small-volume tephra fallouts, is related to the Vulcanian phase 1 of 5 April to 13 May 1982. Samples from pyroclastic flows of the Vulcanian phase have free crystals or crystals with vesicular glass. Some crystals have stepped fractures on their surfaces, char-acteristic of brittle breakage. Fine adhering dust (maximum 10 µm) on the surfaces of crystals is common. Several juvenile glass fragments are nonvesicular and have equant, blocky sur-faces (type 1) (Fig. 2). In particular, surfaces have irregularities caused by poorly developed vesicles. Many of these vesicles are cut by cur-viplanar fractures, possibly as a result of hydro-volcanic fragmentation of the melt (Wohletz, 1983; Heiken and Wohletz, 1985). The surfaces are covered by aggregates of fine ash and altera-tion products, probably clay minerals. Vesicular on June 29, 2012gsabulletin.gsapubs.orgDownloaded from  Quantitative SEM analysis of volcanic ash surfaces  Geological Society of America Bulletin, May/June 2007 745glass fragments have frosted, smooth, fluidal surfaces (type 2) (Fig. 2). The vesicles are filled with fine ash and alteration products. Vesicle edges are rounded and chipped by transport abrasion or the fragmentation mechanism. Glass fragments from the Vulcanian phase have elon-gated particles with a width:length ratio of up to 0.25. Some show large crystal molds. Hydration cracks are absent in samples from the first phase of the eruption. Nonjuvenile particles of the Vul-canian phase are more altered and rounded than the juvenile ones.Following a major break in the eruptive activity, several pulses of energetic base surges deposited dune-like, cross-bedded deposits on the crater slopes. This suggests a sudden increase in magma-water interaction in the phreatomag-matic phase (Gourgaud et al., 2000). Repeated tephra-fall deposits of small volume and a few lithic-rich pyroclastic flows were emplaced in the first part of the phreatomagmatic phase. Moreover, a new crater that largely intersected the 1918 dome was opened on 17–19 May and 13–17 July. It has been interpreted as a maar crater, and it removed 80% of the 1918 dome (Gourgaud et al., 2000). The crystals of the phre-atomagmatic phase have the same characteris-tics as those of the Vulcanian phase. Juvenile blocky pyroclasts (type 1) are more common than smooth fluidal ones (type 2). The vesicles are filled with adhering dust, and vesicle edges have been smoothed by chipping (Fig. 3). The  juvenile particles are rounded and altered. The dusty appearance is attributable to clay-like debris in hollows. Hydration cracks and pitting are present on the surfaces of the juvenile par-ticles (Fig. 3). The nonjuvenile particles show extreme alteration with rounded outlines. The adhering dust completely covers the surface of particles. Uncharred wood fragments are pre-sent within samples from the phreatomagmatic phase. The aggregation of small particles, strong surface alteration, and uncharred wood frag-ments indicate low-temperature emplacement during the phreatomagmatic phase. Gourgaud et al. (2000) also proposed these characteristics as evidence for the phreatomagmatic phase.The products of eruptions on 17–19 May and 13 July, when new maars formed (Gour-gaud et al., 2000), show the highest proportions of blocky type 1 pyroclasts (Fig. 4; Table 1). Despite the washing procedure, small par-ticles still adhered to the analyzed surfaces. To understand the sensitivity of our quantita-tive descriptors to altered surfaces, an operator labeled all particles as altered or fresh (Fig. 4; Table 1). Samples from the first maar forma-tion were completely altered. At the end of the phreatomagmatic phase, the phreatomagmatic fall deposits preceded the transition to the late Strombolian activity. The activity clearly shifted to the long-lasting Strombolian phase (3) of November 1982–January 1983, which emplaced a thick pile of lapilli and ash-fall deposits (Gour-gaud et al., 2000). METHODOLOGYImage Processing and Analysis The roughness of a surface can be measured in different ways. Figure 5 depicts a schematic representation of a surface and the applied terms used to describe it. Roughness amplitude surge deposit July 13May 17-19(1-2) lahar PHASE2A May 6 (2)May 6 (1) PHASE1 April 25April 8 2     m     CibanjaranSection PHASE2B&3 A i    r  -f    al    l     s  reworked PHASE2A  3  0  c m August surge depositlaharair-fall July 28 air-fall/surge depositair-fall July 13 Hot RiverSection ? V  UL  C ANI   ANP HRE AT  OMA GMAT I    C       P     H     R     E     A     T     O      M     A     G      M     A     T     I     C      P     H     R     E     A      T     O      M     A     G      M     A     T     I     C     +     S      T     R     O      M     B     O      L     I     A     N : Samples analyzed in this study Figure 1. Stratigraphic sections of the 1982 erupted deposits from Hot River area and Cibanjaran valley after Gourgaud et al. (2000). The samples analyzed in this study are labeled with a hammer symbol.  on June 29, 2012gsabulletin.gsapubs.orgDownloaded from   Ersoy et al. 746 Geological Society of America Bulletin, May/June 2007       A      B      C      D      E      F      G      H      A      B      C      D      E      F      G      H    F   i  g  u  r  e   2 .   T  y  p  e   1   (   A ,   B ,   C ,  a  n   d   D   )  a  n   d   t  y  p  e   2   (   E ,   F ,   G ,  a  n   d   H   )  p  y  r  o  c   l  a  s   t  s   f  r  o  m    V  u   l  c  a  n   i  a  n  p   h  a  s  e   (   1   ) .   T  y  p  e   1  p  y  r  o  c   l  a  s   t  s  a  r  e  n  o  n  v  e  s   i  c  u   l  a  r  a  n   d   h  a  v  e  e  q  u  a  n   t ,   b   l  o  c   k  y  s   h  a  p  e  s .   T  y  p  e   2  p  y  r  o  c   l  a  s   t  s  a  r  e  v  e  s   i  c  u   l  a  r  a  n   d   h  a  v  e   f  r  o  s   t  e   d ,  s  m  o  o   t   h ,   fl  u  -   i   d  a   l  -  s  u  r   f  a  c  e  s .   F   i  g  u  r  e   3 .   T  y  p  e   1   (   A ,   B ,   C ,  a  n   d   D   )  a  n   d   t  y  p  e   2   (   E ,   F ,  a  n   d   G   )  p  y  r  o  c   l  a  s   t  s   f  r  o  m   p   h  r  e  a   t  o  m  a  g  m  a   t   i  c  p   h  a  s  e   (   2   ) .   H  y   d  r  a   t   i  o  n  c  r  a  c   k  s  a  n   d  p   i   t   t   i  n  g  a  r  e  p  r  e  s  e  n   t  o  n  s  u  r  -   f  a  c  e  s  o   f  s  o  m  e  p  y  r  o  c   l  a  s   t  s   f  r  o  m   p   h  a  s  e   2   (   H   ) .  on June 29, 2012gsabulletin.gsapubs.orgDownloaded from
Advertisement
Related Documents
View more
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks