In the Unzendake eruption since May 24th, 1991, frequent ash falls rich in accretionary lapilli were observed to be originated from ash clouds accompanied with running down pyroclastic flows. To clarify the process and condition for generating fine ash falls including accretionary lapilli from pyroclastic flows, we investigated distribution, occurrences, grain-size, and water content of the ash fall deposits accompanied with pyroclastic flows in April 1st, July 14th, and September 27th, 1992. Grain-size distribution of the ash fall in July 14th is similar to each types of occurrences; mud rain, flattenned and spherical accretionary lapilli, and ash aggregates. The ash fall in September 27th shows decrease in the proportion of particles finer than 6φ (＜16μm) because of sorting during their transport. No visible internal structures are found clearly in many specimens of accretionary lapilli, except for outer layer, thickness of about 40 micron, composed of fine ash, but about 10% of them have obscure core and mantle-like structures. On the basis of the measurement of water content in ash fall deposits rich in accretionary lapilli, difference in occurrences of ash fall deposit depends chiefly on the water content; mud rain formed in water content of 20-25%, flattened accretionary lapilli 15-23%, and spherical accretionary lapilli 8-22%, in the case of April 1st. Water content of ash aggregates is generally much lower than that of spherical accretionary lapilli. Using the water content measured and the weight of ash fall deposits estimated, the amount of water included in all ash fall deposits fallen as accretionary lapilli was calculated at 1,300 t for the April 1st and 250 t for the July 14th, respectively. If the water content of dome lava are assumed to be 1.5 wt%, the total amount of water in dome lava blocks collapsed is only 10%-20% of these. On the other hand, using meteorological data, the amount of water included in the atmospheric air of 1 km^3 is estimated about 11,000 t for the April 1st and 15,000 t for the July 14th, respectively. This amount is much more than the total amount of water included in the ash fall deposit. Thus, even if only a part of the actual amount of water are condensed, its amount may be enough to change for all ash fall deposit to spherical accretionary lapilli. In the September 27th, no accretionary lapilli were generated because the ash cloud didn't reach to the dew point height. Genemte condition of accretionary lapilli in the Unzendake 1992 eruption are summarized as follows: 1: The atmospheric air are risen up vertically over dew point height owing to buoyancy, while mixing with hot ash cloud. 2: Fine ash particles are gathered and adhered to dew drops condensed from vapor included in the air, and form accretionary lapilli. 3: Accretionary lapilli generated are transported by wind and fall mainly in and around Shimabara City and Fukae Town.

抄録全体を表示

Hokkaido island is situated at the junction of northeastern Japan and Kuril arcs. Recent K-Ar age dating of Cenozoic volcanics in the island revealed that the distribution of Quaternary volcanoes (＜1.7Ma) is remarkably different from previously accepted one. On the basis of the newly established distribution of the volcanoes, spatial variations in SiO₂-normalized major element values (SiO₂=60%) of Quaternary andesitic rocks (SiO₂=56-64wt%) are investigated. The Na₂O, CaO and P₂O₅ values vary systematically in the across-arc direction irrespective of the tectonic setting and/or the geological structure. On the contrary, the variations in FeO, Al₂O₃ and K₂O values seem to be largely correlated with the tectonic setting and the Bouguer anomaly. The distribution of the volcanoes and the mode of spatial variations in major element chemistry suggest the presence of three Quatemary volcanic fields, the southwestern Hokkaido (SWH), Taisetsu-Tokachi-Shikaribetsu (TTS) and Akan-Shiretoko (AKS) regions. Modes of the spatial variations in normalized major element values in SWH and AKS regions resemble each other and are nearly the same as those of northern Honshu arc; chemical zonation of volcanoes is sub-parallel to the trench axis in both regions. On the contrary, the spatial variations in FeO and K₂O values in TTS region are different from those in SWH and AKS regions. The mode of the variations in FeO and K₂O values is thus discontinuous in Hokkaido island. The values of Bouguer anomaly at TTS region (-40 to 0 mgal) are much lower than those at other two regions (+40 to +100 mgal). Because crustal thickness under the regions is nearly identical (25-34km), the regional variation in the values of Bouguer anomaly suggests that crustal materials of TTS region are less dense than those of SWH and AKS regions. These different crustal materials, which have been produced under the different tectonic settings, are believed to have resulted in mode of spatial variations in major element chemistry of Quaternary andesites.

抄録全体を表示

There is no positive correlation between the long-term eruption rate of large-scale felsic volcanism and its discharge volume of a single eruptive episode. This means that the storage of voluminous felsic magma at high-level in the crust is caused not by high magma production rate but by continuous accumulation of magma during a long repose time, if the long-term eruption rate reflects the averaged magma production rate. If the cruslal defomation is weak, the magma chamber could be stable in the crust; it is favorable for efficient accumulation of voluminous magma. In fact, the large-volume felsic volcanism occurs exclusively in the region of low crustal strain rate. The low crustal strain rate is considered to be essential for the formation of large-scale felsic volcanism. The large-volume felsic volcanic activity is present in the compressional tectonic stress field as well as in the extensional one; the difference in arrangement of principal stress axes is not related to the occurrence of voluminous felsic volcanism.

抄録全体を表示

Unzen volcano located at Shimabara Peninsula, Kyushu, Japan, has been erupted at the summit since November 1990 after 198 years dormant period. Phreatic eruption continued until middle of May 1991, and the appearance of a lava dome was confirmed on May 20, 1991. Since the end of May 1991, the lava dome has been growing, and pyroclastic flows have been frequently generated by collapses of the dome. Total volume of discharged lava as of February 1994 is estimated to be 1.86×10^8m³ by Geographical Survey Institute. GPS surveys have been repeated at several sites around the volcano since January 1991. Until the middle of 1991, the ground around the volcano indicated expansion and then turned into contraction. Applying a point source model to the data on the change in length of GPS baselines, the location of a pressure source is estimated to be approximately 5 km west from the summit and approximately 11 km in depth. Dilatation of the ground in the area which includes the pressure source region closely relates to changes in extrusion rate of lava. While the rate exceeded about 1×10⁵ m³/day, the ground indicated deflation, and in contrast, the contractive movement came to stay when the rate decreased to less than the value. This suggests continuous supply of magma through the lower crust to the zone where the pressure source is located. These results are consistent with those derived from levelling survey. It is concluded that the location of the pressure source estimated here indicates that of the magma reservoir.

抄録全体を表示