Show more. First visit to our site? Try our free app! Android iOS version. Ibu volcano photos Ibu in Halmahera is one of the most active, but rarely visited volcanoes in Indonesia. A spectacular lava dome is growing inside the volcano's breached crater, with frequent explosions. Santorini BC. Redating the Minoan eruption of Santorini A y. Top 20 quakes. Largest 20 earthquakes past 24 hrs Our continuously updated worldwide map and list of the most recent largest earthquakes in the world during the past 24 hours.
A number of theories have been proposed to explain Etna's location and eruptive history, including rifting processes, a hot spot, and intersection of structural breaks in the crust. Scientists are still debating which best fits their data, and are using a variety of methods to build a better image of the Earth's crust below the volcano. Ruins of a small house partially buried by volcanic debris from Mount Etna.
Mount Etna consists of two edifices: an ancient shield volcano at its base, and the younger Mongibello stratovolcano, which was built on top of the shield. The basaltic shield volcano eruptions began about , years ago, while the stratovolcano began forming about 35, years ago from more trachytic lavas. The volcano's slopes currently host several large calderas which formed when the roofs of magma chambers collapsed inward, including the east-facing, horseshoe-shaped Valle de Bove.
Etna's current activity consists of continuous summit degassing, explosive Strombolian eruptions, and frequent basaltic lava flows. Ash clouds from the explosive eruptions are especially hazardous to aircraft, since ash that is pulled into a jet engine can melt, coat moving parts with a layer of glass, and cause the engine to shut down. These dangerous ash clouds are often visible from space.
Etna has also produced pyroclastic flows, ashfalls , and mudflows, but the lava flows are the most immediately hazardous type of activity, especially to the city of Catania. The lava flows themselves usually do not move fast enough to threaten humans, but they can cover large areas and destroy crops and buildings.
In the event of a large flank fissure eruption, evacuating the inhabitants of towns and cities near the volcano would be a huge challenge. Mount Etna ash plume: An oblique photograph of Mount Etna looking to the southeast taken by astronauts onboard the International Space Station on October 30, The dark plume rising from the top of the volcano is an ash cloud. The broad white cloud streaming from areas of lower elevation is smoke produced by forest fires ignited as a hot lava flow moved through a pine forest.
The ash and smoke caused air traffic to be diverted and forced the closing of roads, schools and businesses. Larger Image. This photo is looking to the southeast with the Mediterranean Sea in the background and was taken by astronauts onboard the International Space Station on October 30, The scene shows the ash plume from the eruption being carried by wind across the Mediterranean Sea to Libya, over miles away.
A Sicilian vineyard growing in the shadow of Mount Etna. The inhabitants of Sicily must balance the advantage of rich volcanic soil with the dangers of losing their crops and farms to an eruption from the still-active volcano. Etna's eruptions have been documented since BC, when phreatomagmatic eruptions drove people living in the eastern part of the island to migrate to its western end.
The volcano has experienced more than eruptions since then, although most are moderately small. Etna's most powerful recorded eruption was in , when explosions destroyed part of the summit and lava flows from a fissure on the volcano's flank reached the sea and the town of Catania, more than ten miles away.
This eruption was also notable as one of the first attempts to control the path of flowing lava. The Catanian townspeople dug a channel that drained lava away from their homes, but when the diverted lava threatened the village of Paterno, the inhabitants of that community drove away the Catanians and forced them to abandon their efforts.
An eruption in produced large lahars when hot material melted snow and ice on the summit, and an extremely violent eruption in produced more than 2 billion cubic feet of lava and covered more than three square miles of the volcano's flanks in lava flows. During Phase II, explosion frequencies also decreased with time, from 80 per minute on 11 March to 50 per minute by 13 March, and emission heights decreased from — m initially, to — m by 23 March Bottari et al.
Activity within the summit craters during Phase I was characterized by continuous ash emissions, but during Phase II incandescent scoria was ejected and fractures developed across the summit zone Tanguy and Kieffer A sudden drop of the magma column in the summit conduit was then observed at the end of the eruption Bottari et al.
Although the two eruptive phases were separated by 22 days of quiescence Bottari et al. The petrochemical and volatile analysis of Corsaro et al. A petrological study using high resolution chemical maps of clinopyroxene crystals and more specifically the dominance of Cr-rich zones occurring at the rims, revealed that the trigger of this eruption was linked to the new arrival of a mafic magma Ubide et al. Etna cf. Supported by observations of the lava flow emplacement published in Guest et al.
During May , eleven UAV over-flights of the flow field allowed acquisition of overlapping images covering a total area of 1. This allowed generation of a cm spatial resolution DEM and a 3-cm orthomosaic as described in Favalli et al.
Cross-unit profiles were obtained by slicing the DEM down to a datum defined by projecting surrounding surfaces beneath the flow field itself as inset in Figure 3 , and used to define the dimensions, areas and volumes of Table 2 following the methodology of Mazzarini et al.
Since no DEM prior to the eruption was available, the volumes were estimated using the area and the average thickness determined using the base of the flow as baseline. Mapping was aided by comparison with maps and chronologies of Guest et al. Lava flow units are numbered in chronological order see Table 2 for dates of emplacement of each unit. Structural map of the January—March west flank eruption of Mount Etna overlain on the orthophoto of Favalli et al.
Inset represents cross-unit profiles, taken perpendicular to flow direction, to show the overall shape of lava units associated with each phase: Profiles A-B Unit 1. TABLE 2. Dimensions, areas, and volumes of each unit obtained from the DEM.
Dates assigned to emplacement of each unit are following the chronology given in Guest et al. DRE volume is obtained by multiplying the bulk volume by the one minus the vesicularity obtained for each unit. During the field survey, 29 samples were collected for all exposed lava and cone units Supplementary Figure 1 ; Supplementary Table.
Twenty-two lava samples were collected: sixteen from the Phase I flow field, and six from the Phase II flow field. Due to potential textural heterogeneity with distance and depth e. Our sampling strategy was thus designed accordingly, with a focus on sampling of near or at vent lava and lava flow fronts Figure 2 ; Supplementary Figure 1 ; Supplementary Table.
Sampling followed the strategy of Riker et al. This is not too difficult for pahoehoe, but more problematic with 'a'a lava flow type which lacks coherent, glassy surfaces cf.
Robert et al. We attempted to sample all exposed Phase I units, focusing on near-vent accreted levees and squeeze-outs of toothpaste lava; however, units 1, 2, 4 and 6 could not be sampled, either because they were buried, or no suitable sample was available for collection.
Due to problems of oxidation and alteration, only 15 of these 22 lava samples could be used for further analysis Supplementary Table. Of these, two samples were selected one scoria and one bomb sample for each cone, respectively Supplementary Table. Textural analyses were completed for density, vesicularity and crystallinity following the methodologies of Houghton and Wilson and Robert et al. This value was then used to obtain the vesicularity of 19 samples following the Archimedes-based method of Houghton and Wilson Results are given in Table 3.
TABLE 3. Densities, vesicularities and crystallinities by phase and unit. Next, thin sections were prepared for 13 samples eight for Phase I, and five for Phase II; Supplementary Table and gray scale images separating vesicles, crystals and glass were prepared. Crystal size distributions of plagioclases and Fe-Ti oxides were also performed on scoriae from both phases to compare the crystallization kinetics between the two phases e.
Microphenocrysts of plagioclase and Fe-Ti oxides were manually outlined from ten BSE images for each scoria. Subsequently, the crystal measurements including area, 2D aspect ratio, major and minor ellipses were retrieved from the digitized binary images imported into ImageJ. Each population consisted of more than crystals as recommended by Morgan and Jerram Finally, the conversion of the apparent crystal dimensions to 3D crystal shapes was performed using the software CSDCorrections version 1.
The thin sections contained no perceptible fabric and, based on a visual inspection of the samples, a roundness factor of 0. Our results are given, along with those of Corsaro et al. We also report the glass chemistry measured by Corsaro et al. TABLE 4. Bulk rock compositions obtained in this study and by Corsaro et al. TABLE 5. Glass chemistry, eruption temperatures and derivation of viscosity for Phase I and II samples.
To provide order of magnitude estimates on the rheological differences between the two phases, we estimated the viscosity of the lava using a petrographic and empirical model-based approach.
To employ such approach, we considered the petrographic and geochemical characteristics of the pyroclast samples rather than the samples collected for the lava flows, ensuring that the effect of post-emplacement crystallization did not influence the result.
The interstitial melt viscosity was calculated following Giordano et al. The temperature was obtained following Corsaro et al. In this approach maximum packing is calculated for rough particles following Mader et al. The relative viscosity is then calculated via the Maron—Pierce equation:.
Finally, the relative effect of bubbles was calculated following Llewellin and Manga by considering bubbles as rigid spheres high capillary number :. The average viscosity and yield strength values were then used to extract the mean velocity v m e a n and effusion rate for lava entering the master channels.
For this, we used the velocity equation of Jeffreys , as modified by Moore for a Bingham fluid:. Finally, we used v mean w, and h , to estimate instantaneous effusion rate E r in Harris et al.
Our geological map Figure 2 , coupled with the detailed description of lava flow emplacement during the first phase of the eruption by Guest et al. Guest et al. Our field survey and sampling then allowed us to characterize the two phases in terms of dimensions, morphology, texture and chemistry. The eruption began during the afternoon of January 30, with nearly continuous Strombolian activity being observed at MDF I Guest et al.
The first three lava units of Phase I units 1. These flows initially moved north, before turning west around the base of the Monte Nuovo cinder cone Figure 1 to attain lengths of — m Table 2.
These three lava flows were active for two-to-three days 30 January—1 February; Guest et al. These three lava flows were the first, though, to move into the pine forest that surrounds MDF I to the north.
As a result, tree molds and the remains of burnt tree trunks and branches can be found here Figure 4A ; Supplementary Photos A—C. A Tree mould m in unit 1. Pits and toothpaste lava associated with late stage extrusion from the vents feeding 1. D View looking north down unit 1. Ox-s show locations where oxidized scoriae were sampled Supplementary Table. Toothpaste lava associated with late stage extrusion into the head of the 2. F Levee of the 2.
H Flow front of 2. Our field observations and mapping confirm that unit 1. Unit 1. These fed units 1. The medial and distal segments of Unit 1. Instead, units 1. They respectively extended m and m Table 2. By 17 February, all Phase I activity had stopped Guest et al. Field-based descriptions for Phase II are lacking, but from our mapping we can say that four main east-west orientated channel-fed 'a'a lava flow units were emplaced Figure 2.
A first basal lava unit unit 2. This second flow, unit 2. This was, in turn, covered by unit 2. Unit 2. Lava flow unit thicknesses for Phase I are typically 2—3 m, except for the two units 1. Phase I flow lengths are m—1, m average of m , for a total area coverage of 0. Instead, Phase II lava flow unit thicknesses are typically 5 m, with lengths of m—1, m average of m , for a total area of 0. For Phase I, the average width is 70 m and the aspect ratio is 10; for Phase II, average width is 90 m and aspect ratio is 12 Table 2.
Thus, our results show that Phase I lava flow units are generally thinner, shorter and narrower than Phase II units. In both cases, this is about half of that estimated by Corsaro et al. Units 1. Lava flow units comprising the Phase I lava flow field are generally channel-fed 'a'a with a maximum clast size for clinker comprising the surface crust being 10—30 cm Figure 4B ; Supplementary Photo D.
Each unit is characterized by a single master channel Figure 3 issuing from one of ten source vents distributed around the northern, western and southern sides of the MDF I cone e. Levees to channels in Phase I units are incipient, discontinuous and low Figure 4C ; Supplementary Photo D , typically being 1—3 m high and 4—10 m wide, with flow in the channel being more-or-less bank-full see profiles as inset in Figure 3. Most Phase I flows also lack cross-flow ridges Figure 3 and, where they are apparent, have a wavelength of 4—6 m, heights of around 0.
The exceptions are units 1. There is also a preponderance of cone-cored lava balls on the flow surface and cone material of oxidized scoria in the levees and at the flow front, i. Lava flow units comprising the Phase II lava flow field are also channel-fed 'a'a, but have a maximum clast size of 30—60 cm Figure 4E—F and well-formed cross-flow ridges Figure 3. Folds are well-formed, and the associated cross-channel ridges have wavelengths of 14—20 m, heights of around 1. Proximally, the channel system is bifurcated with the master channel having four branches Figure 2.
Phase II levees are well-formed, continuous and high Figure 4F—G , typically being 10 m high and 20 m wide with lava in the channel being 2—5 m below-bank see profiles as inset in Figure 3. Like the lava flow units associated with the two phases, the cone morphology associated with each of the two phases is also quite different Figure 3 , see also Supplementary Figure 2. The MDF I cone has an extremely complex form, with the summit being cut by coalesced pit craters and the flanks are modified by faulting and collapse.
Collapse of the east flank prior to extrusion of lava units 1. These were constructed during lava fountaining activity, before being covered by cinder from subsequent violent Strombolian activity. Three vents opened around the southern base of the cone feeding lava flow units 1. B View of the 1. C View SE down the exit channel from the 1. Proximal portion of lava unit 1. D View of the 1. F Toothpaste lava in the 1. The largest depression on the flank of MDF I is that associated with the vent that fed unit 1.
This is drained by a 2—3 m wide exit channel which runs into a 10 m wide master channel to feed unit 1. The vent area associated with unit 1. The W flank is cut by a graben, with its head in the summit pit and the vent structure for flow 1. The vent of unit 1.
Likewise, a graben on the NE flank within which four pit craters formed, Figure 3 has the unit 1. Again, this is filled by a late-stage squeeze-out of toothpaste lava. The MDF II cone, instead, has a simple form, being a symmetric, horseshoe-shaped and open to the west, with no collapses, peripheral vents or fault structures Figure 3. As with the MDF I vents, late stage extrusion of toothpaste lava into the head of the channel draining the vent zone is apparent Figure 4E.
Following Favalli et al. For lava flow units 1. For these lava flow units near-vent vs. A Density and B crystallinity and geothermometer-derived eruption temperature for samples analyzed during each phase.
Each lava flow unit sample is sequenced in chronological order. Scatter plots of DRE volume vs. Empty diamonds represent lava front samples in A and vesicle-free crystallinity in B. Clinopyroxene microphenocrysts from both phases are often associated with hourglass sector-zoning Figures 7B,D ; Downes, Phase II lavas and pyroclasts have more ferric clinopyroxenes, more Ti-magnetite compared to that of Phase I, and are also distinctive due to the occurrence of Fe-rich rims in olivines.
These convert to vesicle-free crystal fraction of 0. Following the same pattern, the crystal content of the scoria samples from Phase II is higher than for Phase I. We measure a vesicle-free crystal fraction of 0. The mean aspect ratio of crystals was 3. Back-scattered electron images and element maps of two eruptive products from the flank eruptions of Mt. Table 3. Note that hourglass sector zoning occurs in clinopyroxenes cpx.
Also, note that Phase II lavas contain more Ti-magnetites than those of Phase I, as well as Fe-rich rimmed olivine crystals red arrows , which are inexistent in Phase I lavas. Plg, ox, cpx and ol respectively stand for plagioclase, oxide, clinopyroxene and olivine.
Comparisons of the size distributions of plagioclase E and Fe-Ti oxides F for both phases. Note that CSDs were generated from the same number of images from each phase in which plagioclases and oxides were considered for Phase I and 1, plagioclases and oxides were considered for Phase II. The crystal size distributions CSD for plagioclases and oxides between the two phases Figures 7E,F result in linear ln n -L relationships that are similar for both phases.
Here, n is the number of crystals and L is their dimension. For the four CSD plots, we obtain an R 2 of 0. Bulk rock compositions show, in agreement with Corsaro et al. Analysis of matrix glass by Corsaro et al. Our analyses are again in agreement with those of Corsaro et al. While Guest et al. Following Harris and Allen , Etnean lavas have water contents of 0. With the glass chemistry and temperatures of Table 4 , the melt viscosity for the lava on eruption is about 5.
The effect of adding crystals and vesicles to the mixture increases the viscosity of Phase I lavas to 0. For Phase II, the higher crystal fraction probably induced higher crystal-crystal interaction and therefore the lava could have developed a yield strength that can be modeled by a Bingham-like behavior.
The lava yield strength calculated using Eq. The velocity of the lava, calculated using Eq. Phase I produced ten relatively thin 2—3 m, Table 2 lava flows units 1. That is, they were near bank-full with poorly defined levees Figure 3 , inset. Phase I also built a complex cone system with coalesced summit vents, flank vents containing lava lakes, grabens, pits, and collapse deposits Figure 3. Instead, Phase II produced four relatively thick 5 m, Table 2 units 2.
That is, they had well-defined levees with below-bank flow Figure 3 , inset. These differences imply a change in rheology, and hence flow dynamics, between Phases I and II; as well as a change in eruption style. To gain an order of magnitude assessment of the viscosity difference between the two phases, we take the typical chemical, temperature and textural values of each phase. In agreement with Corsaro et al. Considering only the melt phase, we find a two-order of magnitude increase from 10 2 to 10 4 Pa s between the two phases.
This is due to the change in glass chemistry, decrease in water content and decrease in temperature between the two phases Table 5. If we also consider the effect of differing crystal and bubble contents on the viscosity of the erupted lava, there is a difference of two order of magnitudes 10 3 vs. This increase in viscosity between the two phases is consistent with the observations of Favalli et al.
They are also consistent with previous viscosity estimates made for Etnean lavas. Measurements of the lava made by Pinkerton and Sparks using a viscometer gave a viscosity of 9. Calvari et al. Thus, while viscosities during Phase I of the eruption are typical for Etna, those of Phase II were relatively high and consistent with a cooled and more crystalline magma. To assess potential peak effusion rates, we use our measurements for our best constrained unit, this being unit 1.
Our velocity for vent-leaving lava of 1. Based on measurements of 17 lava flow units emplaced on Etna between and , Calvari and Pinkerton provided the empirical relation between flow length L and effusion rate E r :. However, Phase I lava flows flow units only attained lengths of 0. Hence, we infer that the episodic nature of Phase I meant that units were volume limited.
That is, during each episode supply was cut before the flow could attain its maximum extent Guest et al. Higher viscosities, though, meant that velocity for vent-leaving lava were much lower 0. Using Equation 10 , we expect flow units during Phase II to have extended to 1. Judging from the channel morphology, the shorter lengths of the other units can be explained by levee failure high in the channel system, within m of the vent Figure 2.
As a result, volume limits to flow extension were the result of beheading of any given unit and creation of a new channel, rather than termination in supply from the vent.
Thus, we envisage eruption of lava during Phase II being much more sustained, with one flow unit 2. Units were active for long enough to reach a degree of maturity, showing all four parts of the Lipman and Banks channel system Figure 3 : 1 well-formed master channels over the proximal-medial sections, 2 a short zone of transitional channel feeding, 3 a zone of channel-free dispersed flow behind, 4 a well-developed flow front Figure 4H.
The lower yield strength and viscosity of Phase I lavas resulted in flows of markedly different lava flow unit dimensions and morphologies than those of Phase II, as is immediately apparent from the cross-flow profiles Figure 3. The Phase I units are all thinner with poorly developed channels. Poor channel development, though, is also likely due to the episodic, volume-limited nature of the emplacement of the Phase I units. Instead, levees remained incipient and following Sparks et al.
Naranjo et al. Harris et al. However, two trends and groupings are present representing 1 the volume-limited control on Phase I units, and 2 the higher viscosity and yield strength of Phase II units Figure 6C—E. For the relationship of volume vs.
Hulme, This also means that, the same volume spreading with the Phase I rheology will attain greater area and length than when spreading under Phase II rheological conditions. This is apparent in the different relations in the linear increase in area with volume Figure 6D and length with volume Figure 6E between the Phase I and II units. The difference in style of eruption between the two phases episodic vs.
In this regard, the Phase I cone represents the result of explosive eruptions from the summit craters and opening of ten different events around its base Figure 2. The form of MDF I was further modified by collapse events, which ranged from large scale flank failure i.
Wilson and Head, ; Figure 3. Instead, the Phase II cone was breached at a single point to the west where a continuous discharge of lava from a stable point meant that the cone was not able to build in this sector, but was able to build symmetrically in all other sectors Figure 3 , Figure 4E ; Supplementary Figure 2.
Although there was no change in bulk rock geochemistry between the two phases of the eruption, in terms of vesicularity, crystallinity, glass chemistry temperature and, therefore, also lava rheology, flow dynamics and system morphology, the two phases were completely different cf.
Tables 2 — 6.
0コメント