From owner-volcano@ASUVM.INRE.ASU.EDU Thu Dec 28 12:56:12 1995 Received: from mtu.edu (mtu.edu [141.219.70.1]) by alt.geo.mtu.edu (8.6.10/8.6.10) with ESMTP id MAA04386; Thu, 28 Dec 1995 12:56:11 -0500 Received: from wnt.dc.lsoft.com (wnt.dc.lsoft.com [205.186.43.7]) by mtu.edu (8.6.10/8.6.10) with ESMTP id MAA08447; Thu, 28 Dec 1995 12:55:08 -0500 Received: from PEACH.EASE.LSOFT.COM (205.186.43.4) by wnt.dc.lsoft.com (LSMTP for Windows NT v1.0a) with SMTP id 48E32F90 ; Thu, 28 Dec 1995 12:52:03 -0500 Received: from ASUVM.INRE.ASU.EDU by ASUVM.INRE.ASU.EDU (LISTSERV release 1.8b) with NJE id 8366 for VOLCANO@ASUVM.INRE.ASU.EDU; Thu, 28 Dec 1995 10:55:06 -0700 Received: from ASUACAD (NJE origin AIJHF@ASUACAD) by ASUVM.INRE.ASU.EDU (LMail V1.2a/1.8a) with BSMTP id 4235; Thu, 28 Dec 1995 10:55:04 -0700 Message-ID: Date: Thu, 28 Dec 1995 10:50:43 MST Reply-To: VOLCANO Sender: VOLCANO From: Jon Fink Subject: Global Volcanism Network Bull. v. 20, no. 10, October 1995 To: Multiple recipients of list VOLCANO Status: RO Smithsonian Institution Global Volcanism Network Bulletin v. 20, no. 10, October 1995 This information is from a worldwide network of correspondents, who receive a Bulletin that includes illustrations not available in this electronic version of the text. Figures are available via Internet Gopher (nmnhgoph.si.edu). The issue date is the closing date for regular monthly reports, not the publication date; more recent information about ongoing activity is often included. * The Bulletin is available by subscription from the American Geophysical Union, 2000 Florida Ave. NW, Washington, DC 20009 USA, at $20/year for US addresses, $32/year outside the US. * The information in our reports is necessarily preliminary, and subject to change as the reported events are studied in more detail. Please contact the original sources for updates and corrections before using Bulletin information. * The network depends on prompt communication from observers around the world. Please help by sending news of current activity via Internet electronic mail (mnhms017@sivm.si.edu), telephone (202:357-1511), fax (202:357-2476), or airletter (NHB MRC 129, Smithsonian Institution, Washington, DC 20560 USA). Editors: Rick Wunderman, Ed Venzke, and Giuseppina Kysar TABLE OF CONTENTS VOLCANOES Rincon de la Vieja (Costa Rica) New eruption; lahars damage a bridge and lead to evacuations Poas (Costa Rica) High seismicity Soufriere Hills (Montserrat) Small ash explosions continue; three new vents formed; September dome grows Galeras (Colombia) Minor seismicity and fumarolic emissions Llaima (Chile) Minor eruption just after a M 4.0 earthquake 160 km to the east Vulcano (Italy) Fumarolic activity notably diminished from previous years Etna (Italy) Frequent Strombolian explosions and ash emissions from Northeast Crater and Bocca Nuova Ol Doinyo Lengai (Tanzania) New hornitos and lava flows observed in July Merapi (Indonesia) Pyroclastic flows travel down two river drainages Tengger Caldera (Indonesia) Eruption from Bromo sends dark ash plume 700 m above the rim Semeru (Indonesia) Explosions and pyroclastic flows continue Raung (Indonesia) Aviation report of a plume, but not seen on satellite imagery Rinjani (Indonesia) Small ash plume seen on 12 September Dukono (Indonesia) Pilot report of plume on 25 September Manam (Papua New Guinea) Passive degassing Langila (Papua New Guinea) Ash-bearing eruption columns rise hundreds of meters Rabaul (Papua New Guinea) Minor seismicity and vapor emission Ruapehu (New Zealand) Late September-early October eruptions rival those in 1945 Ruby Seamount (Northern Marianas) Submarine eruption Sakura-jima (Japan) Explosive activity continues Aso (Japan) Isolated tremor; ejections of mud and water Kuju (Japan) Additional data on the sudden aseismic eruption of 11 October Kozu-shima (Japan) Earthquake swarm ends in mid-October Oshima (Japan) Minor tremor and 48 earthquakes Izu-Tobu (Japan) Tremor observed again Iwate (Japan) Short tremor episode Adatara (Japan) First tremor since 1965 Akan (Japan) Continued elevated seismicity Medvezhia (Kurile Islands) Measurements of SO2 in Kudriavy plume Yellowstone (USA) New mud volcano, minor mud flow, and associated thermal features Atmospheric Effects Lidar data from Germany and Virginia EARTHQUAKES Rincon de la Vieja Costa Rica 10.83 N, 85.33 W; summit elev. 1,916 m All times are local (= GMT - 6 hours) A new phreatomagmatic eruption followed three months of declining seismicity. During 1995 the number of local earthquakes peaked in July and then progressively decreased (figure 1). Prior to the eruption, during October, OVSCICORI-UNA reported that park rangers who ascended to the main summit saw increased degassing and noted the appearance of fumaroles along cracks at the E and NE crater margins. Rangers described the crater lake~s color as green and the smell as strong and sulfurous. ICE described the eruption as phreatomagmatic, beginning at 1504 on 6 November, and climaxing on 8 November with 25 explosions. They noted the ash-bearing and steam-rich columns rose to 1 and 4 km, respectively, above the crater. Ash blew WSW; medium- to fine-grained ash reached up to 30 km from the volcano (Santa Rosa National Park). According to ICE, on 9 November the eruption entered a steam-rich phase. Columns typically rose 200 m, but sometimes as much as 1.5 km after some steam explosions. During the course of the eruption, ballistic ejecta were thrown over a zone extending to ~1 km N. Ejecta formed lahars that followed two key rivers (Penjamo and Azul rivers) and their tributaries. Heavy rains beginning on 10 and continuing on 11 November triggered secondary lahars and associated floods; a bridge 7 km N of the crater (Penjamo bridge) was damaged but not destroyed, interrupting traffic flow. During this episode, lahars along a tributary of the Penjamo river produced a gully 8-m deep and 25-m wide, isolating some inhabitants. Initial inspections of ash and the lahar matrix indicated that they mainly consisted of hydrothermally altered fragments, lake-sediment mud, and vesiculated glassy andesite fragments. Some residents living near the volcano were evacuated to a safe village 9 km NW of the crater. News reports on 8 November by both Associated Press and Deutsche Presse- Agentur stated that about 100 families were evacuated. Two days later Enrique Coen reported relocation of 300 families. The volcano sits near Costa Rica~s N border, 85 km SE of Concepcion, the volcano forming a prominent island in the W part of Lake Nicaragua. Rincon de la Vieja is a compound volcano that has erupted at least twelve times since 1860. Information Contacts: E. Fernandez, E. Duarte, and V. Barboza, Observatorio Vulcanologico y Sismologico de Costa Rica, Universidad Nacional (OVSICORI-UNA), Apartado 86-3000, Heredia, Costa Rica; G.J. Soto, Oficina de Sismologia y Vulcanologia del Arenal y Miravalles: OSIVAM, Instituto Costarricense de Electricidad (ICE), Apartado 10032-1000, San Jose, Costa Rica; Enrique Coen, Departamento de Fisica, University Nacional, Heredia; Costa Rica; P.O. Box 86, Heredia 3000; Associated Press; Deutsche Presse-Agentur. Poas Costa Rica 10.46N, 84.71W, summit elev. 2,708 m All times are local (= GMT - 6 hours) During October, tremor at Poas reached 101 hours; the last time tremor rose over 12 hours/month was May-September 1994, an interval when tremor ranged between 49 and 307 hours/month. The number of minor earthquakes, which were predominantly of low frequency, continued to climb during the month of October, reaching 9,838 events. This was a value ~8% larger than the total for September, the previous month with the most seismic activity in 1995 (see histogram, Bulletin v. 20, no. 9). The crater lake has risen consistently: by ~5 m during June- October (ICE), and by ~30 cm in the last month (OVSICORI- UNA). During October 1995, the fumarole on the W terrace appeared to have decreased its emissions compared to recent months (< 50-m-high steam plumes), and others on the lake~s NW and SW sides also had diminished output. Fumaroles on the S and SW crater wall produced steam columns reaching 100 m tall. During October, bubbling in the lake still continued. During October OVSICORI-UNA scientists measured the temperatures at several sites: pyroclastic cone, 93 degrees C; fumaroles on the S and SW sides of the crater, 95-97 degrees C; the lake in the inactive crater (Lake Botos), 15 degrees C; and the lake in the active crater, 30 degrees C. Head scarps of landslides that emanate from the dome and flow toward the lake displayed ongoing mass wasting; ICE workers mentioned that this mass wasting may have been triggered by recent heavy rains. In addition, ICE reported that on 17 September (at 0548) a M 3.9 earthquake struck; it had a depth of 5 km and an epicenter 1.6 km SW of the main crater. At the summit, the earthquake~s intensity was MM III-IV. One of the most active volcanoes in Costa Rica, Poas is located 35 km NW of San Jose. The summit area contains several eroded calderas and two crater lakes, one clear and the other, to the N, colored. Information Contacts: OVSICORI-UNA (see Arenal); Mauricio Mora, Escuela Centroamericana de Geologia, Universidad de Costa Rica; G.J. Soto, ICE (see Arenal). "Soufriere Hills Montserrat, West Indies 16.72N, 62.18W; summit elev. 915 m All times are local (= GMT - 4 hours) The observatory was moved on 1 October from the Vue Pointe Hotel to Eifel House on Bishop View Road in Old Towne. A phreatic eruption that day deposited ash across a large area, including the capital city of Plymouth. This eruption was followed by a volcano-tectonic (VT) earthquake swarm, with 70 events located beneath the volcano at depths of 1-6 km. Two of the earthquakes, at 2257 and 2319, had magnitudes of ~2.5 and were felt at the observatory; several were felt in the Long Ground area. After about 0500 on 2 October, the number of located earthquakes dropped to ~5/day. Two episodes of low-amplitude broadband tremor recorded during 1-3 October were related to steam emission. Electronic tiltmeter and EDM observations during that time revealed no significant deformation. EDM measurements at Tar River completed on 3-4 October continued to show a shortening trend, signaling minor inflation. Shallow VT (12 located events) and long-period (2 events) seismicity continued. Moderate levels of seismicity prevailed during 4-8 October, with 30-40 shallow (< 6 km depth) VT earthquakes each day, rare felt events (M 2-2.5), and a few long-period events. No deformation was detected by electronic tiltmeter. An explosion around 2355 on 5 October caused heavy ashfall in Plymouth and in the SW part of the island. On 5 October the government announced that over the next two days they would evacuate Plymouth~s home for elderly people and the hospital, sending residents to the N part of the island. Two eruption signals were recorded at 0235 and 0347 on 8 October, and the EDM line at Tar River continued to show minor inflation. Seismicity began decreasing on 8-9 October, when 24 earthquakes were located beneath the volcano, with a few in the Centre Hills area. A small eruption at 1356 on 9 October generated light ashfall in Amersham and Upper Gages. Vent 2 was emitting a small amount of steam again during 7-9 October. Several episodes of broadband tremor may have been caused by increased steam emission. There were only 6 located earthquakes during 9-10 October, but several episodes of broadband tremor. Another minor eruption around 0012 on 10 October caused light ashfall in Plymouth. Visual helicopter inspection of the crater revealed significant steam emission and an increase in the size of the 25 September dome (Bulletin v. 20, no. 9). Formation of Vent 5 on 11 October. An ash eruption at 0021 on 11 October came from a new vent on the Tar River side of the Castle Peak dome, and damaged the EDM reflector at Tar River. A small earthquake swarm accompanied this vent formation. There were two more small ash eruptions later that day at 1540 and 1700. Although no significant changes to the dome were noted, steaming continued from its top; Vent 1 was also steaming, and appeared to be larger and deeper. Scientists noted that steam emissions from the crater had generally increased. Three more ash eruptions occurred on 12 October, at 0901, 0955, and 1114. Continuous steam emission came from several areas in the crater and Vent 5. Two episodes of broadband tremor during 12-13 October were attributed to increased steam emission. Seismicity was low, with only 22 events during 11-13 October. No deformation was detected following this latest series of explosions. Formation of Vent 6 on 14 October. An eruption at 0708 on 14 October created another vent on the NE flank of Castle Peak dome, generated a significant amount of ash, and ejected blocks as far as the edge of Long Ground, ~1 km E of the vent. A pilot reported that the plume may have reached ~2 km altitude. Another eruption at 1058 caused no reported ashfall. Two gas venting episodes at 2200 and 2345 on the 14th were associated with a small earthquake swarm and broadband tremor episodes. Vent 2 again emitted moderate amounts of steam, accompanied by a loud roaring sound, and Vent 5 continued to emit small amounts of steam. Seismicity decreased from 18 events on 13-14 October to 5 events accompanied by broadband tremor on 15-16 October. Seismicity increased again on 16-17 October with 22 events clustered in two areas: one beneath the volcano and the other just E of Windy Hill. Steam-and-ash eruptions were recorded by the seismic network at 1757 and 2245 on 16 October, and at 1150 and 1522 on the 17th. There were also several episodes of broadband tremor and ~30 minutes of low- frequency harmonic tremor starting around 0414 on 17 October. Later that morning an aerial inspection of the crater showed no significant changes and little steaming. During a second flight at 1145, a large mudflow originating within the crater moat beyond Vent 2 was seen running rapidly down the Hot River and reaching the sea. This was probably the largest mudflow (in terms of volume of material) since the current activity began. During 17-18 October there were 12 scattered earthquakes, several periods of broadband tremor, and some intermediate- frequency tremor. Ash eruptions were recorded at 1739 on the 17th and at 0530 on the 18th. The dome area continued to emit steam, but did not increase in size. Formation of Vent 7 on 18 October. The 31 earthquakes during 18-19 October were clustered beneath the volcano. Several broadband tremor episodes and one period of low-frequency tremor were also detected. An eruption at 1621 on the 18th was associated with the formation of a new vent within the moat area of English~s Crater, just SW of Vent 1. Another eruption was recorded at 2207 on the 18th. An explosive event around 1516 on 19 October generated a mudflow down the Hot River. During 19-20 October there were 28 earthquakes located; the events were scattered throughout S Montserrat, with some clustered beneath Soufriere Hills and St. Georges Hill. There were 15 VT earthquakes on 20-21 October concentrated around the Long Ground/Soufriere Hills area. Several eruption episodes on 21 October resulted in ashfall that affected villages in the E. Ash fell at the airport for the first time, closing it briefly. No deformation was detected at the Tar River EDM or Long Ground tilt stations. Helicopter observations revealed that Vent 1 had extended E and was responsible for the previous ashfall. There was a small mud flow down the Tar River. An average of 35 earthquakes/day occurred during 21-23 October. They were scattered throughout S Montserrat with some concentrations in the Long Ground-Tar River area and beneath the volcano. Some broadband tremor was also recorded. Visual observation of English~s Crater both from helicopter and Tar River on 22 October revealed light steam emission from vents 2 and 5. When observed on the morning of 23 October, the September dome continued to steam, and was covered with sulfur deposits; it may also have grown since last observed on 20 October. Only one other small area SE of the dome was steaming. An eruption at 1337 on 23 October produced ash deposits within the summit crater and at Tar River. Steam emission increased after this eruption. Seismicity decreased following this eruption to 10-14 events/day through 29 October, except for 22 events on the 27th. Locations were mainly beneath the volcano, although some were centered in the Windy Hill area and other parts of S Montserrat. An eruption at 1325 on 25 October caused ashfall in the Tar River area. Eruption signals were again recorded at 2314, 2321, and 2347 on 25 October, and at 0447 on the 26th; no ashfall was reported. Several episodes of low-amplitude broadband tremor were recorded during 25-26 October. EDM measurements at Tar River on 26 October indicated a continuation of the minor inflation observed during the past several weeks. A steam-and-ash eruption at 1317 on 27 October from Vent 1 was followed by more than 30 minutes of low-frequency tremor. Eruption signals were recorded at 0855 and 2018 on 28 October, but no ashfall was reported. Steam emission from Vent 2 was observed that afternoon. Eruptions occurred again at 0326 and 0857 on the 29th, both followed by broadband tremor. An ash-and-steam plume was seen from the observatory following the 0857 event. Steam was seen coming from Vent 1 during a helicopter flight, but no major changes were noted. Seismicity increased on 29-30 October to 55 events; most were clustered in a region just W of Windy Hill, with some scattered in the Centre Hills and Soufriere Hills areas. Eruption signals were recorded at 2110 on the 29th, and at 0244 and 1310 on the 30th. Two small long-period events were recorded after the first eruption. Ash from the first two of these eruptions was observed in English's Crater by helicopter. The third eruption, witnessed by scientists at the Tar River EDM site, produced a high column that caused ashfall over a wide area. This ashfall was the most significant since 21 August, and was accompanied by a density current of ash in the Gages valley. The morning of 31 October visual observations revealed a significant increase in Vent 1~s size, but the 25 September dome appeared unchanged. Seismicity decreased again the next day to 23 events, but they were located in clusters in the Tar River-Long Ground area and W of Windy Hill. There were also 4 long-period events and several episodes of broadband tremor. One eruption at 1118 on 31 October had no reported associated ashfall. EDM measurements at Tar River again showed a slight shortening, associated with continued slow inflation of the upper part of the volcanic edifice. Only 14 seismic events were recorded during 31 October-1 November; most were located beneath the volcano with a few in the Windy Hill and Fox~s Bay area. There were three long- period events and several episodes of broadband tremor. A small eruption at 1129 on 1 November caused ashfall within the summit crater. Information Contacts: Montserrat Volcano Observatory, Eifell House, Olde Towne, Montserrat, West Indies. Galeras SW Colombia 1.22N, 77.37W; summit elev. 4,276 m Activity during August-October remained low. Fumarolic emissions continued from areas near the active cone, with a concentration of fumaroles on the W part of the summit. SO2 concentrations, obtained by the COSPEC method, remained generally low at 53-170 metric tons/day in August and < 100 t/d in September. No deformation was detected by electronic tiltmeters during August-October. Temperature measurements at La Joya and Chavas fumaroles, as well as radon measurements, have begun in order to improve the surveillance. High-frequency seismicity during August was centered NNE of the active crater, and consisted of events of M < 2.2 Seismic activity in September was characterized by volcano- tectonic events, located mainly in three seismogenic regions: W, SW, and NNE of the active crater. Most active was the NNE source, which has shown signs of reactivation since last March. Most earthquakes had magnitudes < 1.5. Four events during September were felt by local residents, on 3, 12, 15, and 16 September, with magnitudes of 2.5, 2.0, 2.7, and 2.7, and depths of 12, 5, 8, and 8 km, respectively. The 16 September earthquake occurred in the SW region and the other three events in the NNE region. The most significant October seismicity consisted of high- frequency events NNE of the active cone at depths of 3-7 km; magnitudes were < 3. The largest earthquake, on the morning of 15 October, was centered ~3 km NNE of the cone at 7 km depth. It had a magnitude of 3 and was felt in Pasto, Jenoy, Narino, and in other local towns. Information Contacts: Pablo Chamorro and Diego Gomez, INGEOMINAS - Observatorio Vulcanologico y Sismologico de Pasto, A.A. 1795, San Juan de Pasto, Narino, Colombia (Email: ovp@mafalda.univalle.edu.co). Llaima Central Chile 38.70S, 71.70W; summit elev. 3,125 m All times are local (= GMT - 4 hours) Beginning on 13 October 1995 Llaima started emitting gases and occasional ash; in addition, during the night the northern principal crater glowed a rose color. Dominant winds dispersed the eruptive columns toward the SE on 13 October. Three days later, Llaima started emitting a continuous, strong blast of steam that occasionally also contained dark-gray scrolls bearing fine-grained ash. The resulting plume blew NE. On the night of 20-21 October, the principal crater discharged a strong explosion. Wind carried ash toward the SW, depositing it on the alpine ice. Some ash fell over the Trufultruful valley and the valley~s most eastern flanking hills, forming a band or stripe up to 12 km in length. On 21 October between 1600 and 1800 the volcano gave off a continuous, intense column of vapor and ash. That night, between 2300 and 0100 in the town of Conguillio, residents heard an explosion accompanied by subterranean noises. The following night, observers saw a ~ring of fire~ over the principal crater, an effect thought to indicate the presence of lava within the crater. The Servicio Sismologico de la Universidad de Chile reported that seismic activity one day before the eruption, on 12 October, included a M 4.0 earthquake that struck the region; its depth was 70 km; its epicenter fell at the extreme S end of Lake Lieulleu in the Cordillera de Nahuelbuta (38.28S, 73.408W), a spot about 160 km E of Llaima. During 20 and 22 October, portable seismometers picked up 1.0-1.5 Hz tremor; on 20 October the tremor appeared about 15-20 seconds before the above-mentioned explosion. It should be noted that such sub-continuous episodes of 1.0-1.5 Hz tremor are relatively rare at Llaima. The 13-22 October eruptions followed fumarolic activity (Bulletin v. 20, no. 2) and, before that, an outbreak of ash-bearing eruptions in late August 1994 (Bulletin v. 19, no. 8). On the basis of the above behavior, the 24 October SERNAGEOMIN report stated that the volcano had been assigned an alert status of yellow. Llaima, an ice- and snow-covered stratovolcano, is one of the largest and most acive in Chile; it erupted in 1990, 1992, and 1994. Information Contacts: Hugo Moreno^1, Gustavo Fuentealba, and Paola Pena, Observatorio Volcanologico de los Andes del Sur, SERNAGEOMIN, Temuco, Chile. ^1Also at Universidad de Chile, Departamento de Geoligia y Geofisica, Casilla 13518 Correo 21, Santiago, Chile. Vulcano Aeolian Islands, Italy 38.40N, 14.96E; summit elev. 500 m Fumarolic activity, vigorous in the late 1980s and through 1994, notably diminished in 1995 (Bulletin v. 20, nos. 4 and 6). During observations in September, the steam and gas output of the most conspicuous fumaroles, at the N rim of the Fossa Grande crater, was back to pre-1985 levels, and no longer formed sizeable gas plumes. Some of the formerly most vigorous fumaroles and steaming cracks were no longer active. Strong gas emission still occured from fumaroles in the oversteepened and unstable Forgia Vecchia area, below the N rim of the Fossa Grande, and hydrothermal alteration continued to weaken the rock. Several blocks of strongly altered rock with volumes of ~100-500 m^3 each had already detached and subsided by 10-20 cm, and may fall. However, it was uncertain whether they would reach the S margin of the village below the Fossa cone. Fumarolic activity also continued from numerous places on the beach N of the ~Faraglione~ and on the low isthmus connecting Vulcanello to the main body of Vulcano island. During a visit to the western-most (and most recent) crater of Vulcanello on 13 September, no evidence of recent fumarolic activity was found in its NE part where intense fumarolic activity took place until the mid-19th century. Vulcano last erupted 1888-90 when numerous meter-sized bombs and blocks fell in the area now occupied by the village of Vulcano Porto, which hosts thousands of tourists daily during the summer season. Vulcanello, the youngest part of Vulcano Island, began to form only ~2,100 years ago as an isolated island that later became connected with the main island. The latest activity at Vulcanello occurred in the 16th century when lava flows now covered by large hotel complexes were extruded. Information Contacts: Boris Behncke and Giada Giuntoli, Department of Volcanology and Petrology, GEOMAR, Wischhofstr. 1-3, 24148 Kiel, Germany (Email: bbehncke@geomar.de; URL: http://www.geomar.de/personal/bbehncke/STROMBOLI.html). Etna Sicily, Italy 37.73N, 15.00E; summit elev. 3,350 m All times are local (= GMT + 1 hour) The Istituto Internazionale di Vulcanologia (IIV) report below provides an overview of activity during October. IIV reports generally summarize the temporal evolution of volcanic phenomena during the whole month, skipping some trivial details, and frame the ongoing activity in the context of phenomena over a period of years. Reports detailing activity during short visits made by visiting volcanologists provide a different perspective on the volcanism. One such report for some days in October was provided by a team led by Open University (OU) volcanologists conducting routine deformation measurements during 9 September-14 October. Short visits to the summit craters on 7, 12, and 14 October were also made by Boris Behncke, with additional observations from Carmelo Monaco and Marcello Bianca (University of Catania), Maria Felicia Monaco (Bari University), and others. Review of July-September 1995 activity. Strombolian activity resumed at Bocca Nuova on 30 July and in Northeast Crater on 2 August (Bulletin v. 20, v. 8). On 30 July spatter was observed inside Bocca Nuova from a new pit crater on the N part of the crater floor. The activity climaxed on 2 and 3 August, when lava jets rose above the crater rim, then stopped on the night of 4 August. Strombolian explosions during 2-3 August issued from a small vent in the lowest part of the crater. Two more Strombolian episodes occurred on 18 and 29 August. A strong explosion from Northeast Crater on 13 September sent an ash plume 100 m above the rim. Ash emissions from Bocca Nuova and Northeast Crater continued until about 20 September, but explosions were heard throughout the month (Bulletin v. 20, no. 9). The OU team noted light ashfall 2-3 km away in the third week of September, and heavier ashfall 50 m from the Bocca Nuova rim on 27 September. Overview of October 1995 activity from IIV. After a short period of Strombolian activity at Bocca Nuova and Northeast Crater at the beginning of October, alternating mild Strombolian activity and ash emission characterized their activity for the rest of the month. On 8 October almost continuous rumbling noises (like roaring jets) were heard from both craters. On the morning of 12 October intense ash emissions took place from both craters. Bocca Nuova displayed small short-lived ash puffs (5-7/hour), while from the Northeast Crater a dense ash column rising as high as 900 m developed repeatedly (2/hour). IIV field parties working in the summit area reported that the ash emission were accompanied by falling rock noises. However, successive surveys observed neither juvenile nor lithic blocks on the crater rims. After 12 October Strombolian activity progressively resumed at Northeast Crater and continued with variable intensity until the end of the month. On 19 October Strombolian activity was relatively vigorous and the scoria ejections, up to few tens of meters from the crater rim, were almost continuous. A survey on 25 October revealed an appreciable decrease of the explosion frequency. Bocca Nuova exhibited intermittent ash emissions after 12 October. As during previous activity, they originated in a depressed area of the NW crater floor. Explosions observed on 19 October were accompanied by ejection of a black (lithic?) block to a few tens of meters above the crater floor, but neither glowing at the vent or ejection of incandescent bombs were observed. After 19 October intermittent ash emission progressively decreased, and in the last week of the month weak Strombolian activity resumed at Bocca Nuova. Significant eruptions on 9 and 14 November will be reported in the next Bulletin. Deformation measurements. Preliminary results from the OU team indicate little ground deformation since October 1994 over most of the network. Summit levelling showed insignificant movement (-5 mm near the summit, +7 mm on the N flank) apart from the area above the 1991-93 dike, which between the W side of Cisternazza and Belvedere showed a fairly consistent subsidence of 17-24 mm. Preliminary GPS computations suggested a radial expansion about the summit of ~15 mm. Dry-tilt stations showed no large tilts. Details of 1-7 October activity. Observations from the Northeast Crater rim on the afternoon of 1 October by the OU team revealed two faintly glowing vents, ~3-5 m across, on the crater floor. The following night, bright summit glow was seen from Nicolosi (15 km S), and on the morning of 3 October loud explosions from Northeast Crater were heard from the trail 800 m W, which had been covered with a thin layer of red ash overnight. Explosions were again heard late in the afternoon from ~7 km away, and light ash fell near Monte Corbara (5 km NW). While approaching the crater at 1815 on 3 October, two guides and an Italian TV camera crew returning from the rim warned of bombs falling outside the crater. As the OU team moved towards the high ground behind the crater, a large explosion sent brightly-glowing juvenile bombs just over the rim, rolling toward them. A few seconds later a single bomb ~20 cm across landed 10 m away, 100-200 m from the rim. Similar bomb ejections to smaller distances occurred about every 2 minutes until the team descended at 1845. On 7 October, Behncke noted a dense steam-and-gas plume from Northeast Crater. Most of the plume and occasionally some ash rose from the SSE part of the crater floor; falling stones were frequently heard. Detonations from within Bocca Nuova heard by the OU team on 1 October were only audible from the rim. One vent on 4 October was explosively exhaling gas, and the other was collapsing, producing brownish ash clouds. Behncke observed small Strombolian explosions from Bocca Nuova on 6 October, but only ash emissions the next day. On the 7 October visit, Behncke observed frequent ash plumes from Bocca Nuova accompanied by rumbling noises and the sound of falling stones; Strombolian explosions were frequent. The Chasm (La Voragine) quietly emitted fumes on 1 October. On 4 October the OU team climbed into Southeast Crater to the edge of the vents, which emitted gas quietly and not under pressure, apart from one area just below the S rim. On 7 October, Behncke heard small explosions, but no ejections or incandescence were seen after sunset. Details of 12-14 October activity. Between 0800 and 0900 on 12 October a series of collapses within Northeast Crater generated a thick ash cloud. Pulses of rapidly rising ash plumes resulted in a vertical column 800-1,000 m above the summit. After 0900, a dilute gas plume rose from Northeast Crater while Bocca Nuova sent frequent ash emissions 200-300 m above the summit. When Behncke reached the crater rim shortly after 1230, there were vigorous steam emission and explosions from Northeast Crater. Behncke saw incandescent spots in the central Northeast Crater floor that gradually increased in number and intensity. Pyroclastic ejections became more frequent and vigorous, and soon the incandescent areas were hidden by gas and dilute ash plumes. The ash plumes first rose slowly to ~100 m above the crater floor, but gradually rose higher and became more heavily ash-laden. About 5 minutes after the onset of ash venting, dense convoluting ash clouds began to rise above the rim. Bomb and ash emission steadily increased. The high-pressure gas emission noise at the beginning of this activity changed to a dull rumbling connected with the ash emission. Short pulses of bomb emissions every 5-10 seconds were followed by a dark ash puff. After ~10 minutes, the ash puffs merged into a continuous column that rose hundreds of meters above the rim. Around 1345 vigorous emissions ejected black ash plumes ~1 km above the summit. Periodic ash emissions from Northeast Crater gradually became less vigorous before ceasing that evening. On 12 October (0800-0900), the OU team heard detonations from Bocca Nuova, mainly from a vent on the E side of the floor, but the larger vent on the NW side occasionally threw 20-cm-diameter lithic blocks 30-50 m high. Ash emissions seen by Behncke after 1230 occurred every 2-5 minutes from the pit on the NW crater floor. Each emission began with block and/or bomb ejections followed by a dense ash plume. The bombs and blocks rose out of the ~50-m-deep pit but remained ~100 m below the rim, whereas the ash plumes rose 100-500 m above the summit. An open vent in the SE crater floor displayed continuous gas emission with occasional explosions that ejected dense gas clouds. Shortly after 1700 on 14 October Behncke saw a central glowing vent in Northeast Crater. Vigorous high-pressure gas emission produced a roaring noise, and the plume was almost vapor-free. During the first 30 minutes of the visit, glowing spatter was occasionally ejected from the vent. As degassing increased, numerous incandescent spots became visible, aligned more or less concentrically around the vent. After the first half hour, Strombolian bursts became more vigorous, ejecting bombs ~50 m above the pit. About 10 minutes later, the explosions again intensified, and the crater floor around the vent, which appeared more funnel- shaped, was covered with incandescent bombs. Ejections rose ~100 m above the vent but remained far below the crater rim. Information Contacts: Massimo Pompilio, CNR Istituto Internazionale di Vulcanologia, Piazza Roma 2, 95123 Catania, Italy (Email: max@iiv.ct.cnr.it); John B. Murray and Fiona McGibbon, Dept. of Earth Sciences, The Open University, Milton Keynes MK7 6AA, United Kingdom; Nicki Stevens, NUTIS, Reading University, Whiteknights, P.O. Box 227, Reading RG6 2AB, United Kingdom; Phil. Sargent, Sue Elwell, and Sarah Cooper, Civil Engineering Dept., Nottingham Trent University, Burton Street, Nottingham NG1 4BU, United Kingdom; Boris Behncke, Dept. of Volcanology and Petrology, GEOMAR, Wischhofstr. 1-3, 24148 Kiel, Germany. Ol Doinyo Lengai Northern Tanzania 2.75S, 35.90E; summit elev. 2,890 m All times are local (= GMT + 3 hours) Intermittent explosive activity and extrusion of carbonititic lava on the crater floor began in January 1983 and continued for over ten years. Vigorous effusive and explosive activity in June 1993, perhaps the strongest of that eruptive episode, covered most of the crater floor and upper W flank with fresh lava flows and deposited ash on the flanks (Bulletin v. 18, nos. 7-10). In September 1994 a deep central depression contained a hornito from which highly vesicular brown lava was erupting (Bulletin v. 19, no. 9). Activity observed in mid-July 1995 was the first reported since September 1994, although the appearance of the recent flows indicated that they were a few months old. Members of the Societe de Volcanologie Geneve (SVG) visited the summit on 15 July 1995. A visit to the summit crater was also made by Celia Nyamweru on 19 July. Activity on 15 July 1995. SVG observers reported a new active hornito (T36), ~4 m high, close to the S foot of T20 (figure 2). Fluid carbonatitic lava flows were emitted from its base through a channel in the direction of a rounded collapsed new opening ~15 m in diamteer, close to T5/T9. The lava in the channel was pale brown and frothy, with a velocity estimated at 1.5 m/second; temperature was ~550 degrees C. At the end of the channel, the flow moved N through different tubes. Lava breakouts from some downstream openings were still very fluid and completely black. Both small pahoehoe and aa lava fronts were observed. Ejecta were rare from the summit vent of T36. The new lava field was mainly directed N, with one branch passing W of T20 and the other going through and filling the oval-shaped depression first noted in October 1993 (see Bulletin v. 19, no. 4). Activity on 19 July. At 1000 the crater was full of cloud, hiding features on the crater floor, but frequent sharp cracks, bangs, and thumps were heard, as well as bubbling noises. Conditions improved so that activity could be observed after 1115. White to brown steam was escaping continuously from the top of T20, and a little from T5T9. Sulfurous fumes were emitted from cracks on the E crater rim and wall. The lower slopes of T23 were made up of many small parallel pahoehoe flows, now soft and pale brown; T23 was not emitting steam. The new cones, T34 to T37, lay W of the depression that had been virtually filled by lava flows from these centers. T34 was a double cone, pale gray, with an open vent on its upper slope from which no steam or heat was being emitted. T35 was light brown to white, with no sign of fresh lava. T37 was a shallow circular crater W of and close to the base of T5/T9; it appeared fresh but showed no activity on 19 July. T36 was a compound cone of which T36A was the largest component; it was composed of cascades of pahoehoe lava, some whitened and others black and very fresh. T36B was a rounded dome with a small vent at its base from which lava was emitted. T36C appeared to have a crack along its crest that emitted gas-rich lava. T36B and T36C were ~5 m apart and very close in elevation. Activity from hornito cluster T36 (figure 3) consisted of clots of lava thrown ~1 m above T36B, gas-rich lava escaping from the top of hornito T36C and flowing down its N slope, and very fluid, black shiny lava escaping from a small crack (T36E) on the lower slopes of this feature and flowing N across very recent pahoehoe. At 1137 a small spray of gas-rich lava escaped from hornito T36D, on the W side of T36. Warm pahoehoe flows on the W slope of T36, < 100-m long, 2-m wide, and 4-cm thick, had probably formed that morning. Crater morphology. Features from June 1993 and earlier (see map in Bulletin v. 19, no. 4) were still visible, but major new cones had formed in the area between T5/T9, T20, and T23 (figure 2). T5/T9 remained a very prominent feature, and the tops of the T8, T14, and T15 cones remained visible, although all were surrounded by many younger lava flows. T24, T26, and T30 were not inspected closely, but there seemed to be no change in these large features in the S part of the crater; they were gray and white, with no sign of recent activity. West of T36 were two low lava domes with pale brown open craters, now inactive. To the W of them, on the edge of F34, was a low wide feature, possibly a collapsed cone, probably the features identified as T22, T31, and T32 in September 1993 (Bulletin v. 18, no. 9). There was also a rather new hornito in this area. Recent pahoehoe flows ~10 cm thick had reached the base of the E, N, and NW walls. Crater walls appeared lowest to the NW. The rugged F34 and F35 lava flows of June 1993 were heavily weathered and beginning to soften and crumble. They were quite dark gray; a great contrast to the flows that had formed over the last several months (thin pahoehoe flows that whiten within a few weeks of eruption). No recent ash was observed on the outer slopes of the cone, the crater rim, or the inner walls; the vegetation was green and healthly. Brown vegetation was observed in a few areas near the base of the inner wall, probably due to contact with hot lava reaching the wall, and on part of the S wall below the summit. This symmetrical stratovolcano in the African Rift Valley rises abruptly above the plain S of Lake Natron. It is the only volcano known to have erupted carbonatite tephra and lavas in historical time. The cone-building stage of Ol Doinyo Lengai ended about 15,000 years ago and was followed by periodic Holocene ejections. Historical eruptions have consisted of smaller tephra ejections and emission of numerous natrocarbonatite lava flows on the floor of the summit crater. Information Contacts: Celia Nyamweru, Department of Anthropology, St. Lawrence University, Canton NY 13617, USA; M. Vigny and P. Vetsch, Societe de Volcanologie Geneve, B.P. 298, CH-1225 Ch nebourg, Switzerland. Merapi Java, Indonesia 7.54S, 110.44E; summit elev. 2,911 m During August-October 1995 pyroclastic flows (~glowing avalanches~) continued flowing down the Boyong River; others entered the Krasak River and reached ~1-1.5 km from the source. Seismic activity was dominated by multiphase and lava-avalanche (rockfall) earthquakes. The number of multiphase earthquakes increased in October to 793 events, compared to 186 in September. Earthquakes associated with lava avalanches or rock falls gradually decreased from 1,195 events in August to 806 in September and 605 in October (figure 4). Shallow volcanic (B-type) earthquakes (~1 km depth) were recorded on 25 October and a deep volcanic (A- type) earthquake (2.7 km depth) was detected on 30 October. Observations in October indicated an inflation associated with 40 ~rad of tilt. Measurement of SO2 by COSPEC indicated that the emission rate during October fluctuated between 18 and 112 tons/day (average 63). Merapi is one of the most active volcanoes in Indonesia. The stratovolcano has an exposed, summit lava dome, the source of abundant glowing avalanches directed down its S to W slopes toward a region settled by at least 50,000 people. In historical time, instability of the growing dome has led to pyroclastic flows that have caused many fatalities. Collapse of the growing lava dome in November 1994 triggered pyroclastic flows that killed 64 people (Bulletin v. 19, nos. 10 and 12). Additional collapses continued into January 1995 (Bulletin v. 20, no. 2). Information Contact: Wimpy S. Tjetjep (Director), Volcanological Survey of Indonesia (VSI), Jalan Diponegoro 57, Bandung, Indonesia. Tengger Caldera Java, Indonesia 7.94S, 112.95E; summit elev. 2,329 m All times are local (= GMT + 7 hours) On 9 September, dark gray emissions were observed reaching a height of 70 m above the rim of Bromo Crater. Volcanic tremor associated with the emission events (maximum amplitude of 1-3 mm) was recorded continuously beginning on 8 September, using a PS-2 seismograph installed 750 m from the active crater. After 10 September the plume was denser than during the March-May 1995 activity (Bulletin v. 20, no. 3). An international Notice to Airmen (NOTAM) on the morning of 22 September reported an ash cloud with a top at ~3 km altitude and a SW drift. The height of the ash column gradually increased, peaking at 700 m (~3 km altitude) on 25 September (figure 5); during the emission, maximum tremor amplitude was 49 mm. A thick dark gray ash cloud caused ashfall in nearby villages, reported as far away as ~20 km E (around the area of Sukapura). The eruption vent, with a diameter of ~25 m, was located on the N part of the crater floor, similar to the last eruption. Ash eruptions were continuing at the end of October, but the activity was gradually decreasing. In October the maximum plume height was 200-450 m above the crater rim; the maximum tremor amplitude was 8-40 mm. The most recent previous eruption began on 3 March 1995 (Bulletin v. 20, no.3) but ceased by the end of May. Bromo cone (700 m in diameter) is the only historically active center in the 7-km-diameter Tengger Caldera. Information Contact: Wimpy S. Tjetjep, VSI (see Merapi); Bureau of Meteorology, Northern Territory Regional Office, P.O. Box 735, Darwin, NT 0801, Australia. Semeru Java, Indonesia 8.11S, 112.92E; summit elev. 3,676 m All times are local (= GMT + 7 hours) Pyroclastic-flow activity increased during June and July, with significant events on 20-21 July (Bulletin v. 20, no. 7). On 3 August, Qantas airlines reported an ash cloud over Semeru with a plume to 4.5 km altitude drifting W (Bulletin v. 20, no. 7). The Volcanological Survey of Indonesia (VSI) reported that by 3 August a tongue of glowing lava had reached 300 m long; at 1932 that evening the lava collapsed to feed lava avalanches. Qantas airlines reported additional activity at 1510 on 8 August, describing volcanic ~smoke~ near Semeru to above 4 km. Two days later, around 1530 on 10 August, a Qantas flight reported an ash cloud to 9 km altitude with a SW drift. VSI noted that during August-October small-to-moderate explosions and avalanches continued from the Jonggring Seloko summit crater. Plumes rose to a maximum of 600 m above the summit; the average plume height was 300-500 m. In August and September, pyroclastic flows often traveled down the Kember River, then descended the Kobokan River, reaching a distance of 1-3 km. The frequency of lava avalanches increased in September, extending along the Kember River for up to 500 m from the summit. Earthquakes associated with the pyroclastic flows were variable, with 1-16 events/day through early October; after that the frequency of earthquakes decreased. Increasing numbers of volcanic earthquakes (both A-and B-type) started on 11 October and continued until the end of the month, fluctuating at 1-14 events/day (figure 6). The number of explosion earthquakes was typically 45-109/day (figure 6), except on 26 and 27 September, when there were only 33 and 24 events, respectively. Information Contacts: Same as for Tengger Caldera. Raung Java, Indonesia 8.125S, 114.04E; summit elev. 3,332 m All times are local (= GMT + 8 hours) An aviation report stated that at 1705 on 15 August ~smoke~ from Raung at an altitude of 6 km was drifting W. Following this report, aviation notices were posted in Indonesia, New Zealand, and Australia for the next 24 hours. No plume was observed by Australian meteorologists on satellite imagery from 1800 on 15 August through 2050 the next day. Raung is a massive stratovolcano in eastern Java with a circular summit caldera that has been the site of frequent historical eruptions. A prehistoric collapse on the W flank produced a large debris avalanche that traveled 60 km. The last reported eruption, which occurred sometime between January and June 1993, generated an ash column 600 m above the rim and caused ashfall in the surrounding area. Information Contact: Bureau of Meteorology (see Tengger Caldera). Rinjani Lombok Island, Indonesia 8.42S, 116.47E; summit elev. 3,726 m All times are local (= GMT + 8 h) A Notice to Airmen (NOTAM) about volcanic activity from Rinjani was issued by the Bali Flight Information Region on the morning of 12 September. An ash cloud was reportedly drifting SW with the cloud top around 4 km altitude. As of 1200 that day, Australian meteorologists had not observed a significant plume on satellite imagery. Synoptic Analysis Branch analysts detected no ash cloud on either visible or infrared GMS imagery. However, at 1600 the Bureau of Meteorology in Darwin advised aviators that a weak low-level plume was intermittently evident on satellite imagery as far as 28 km SW of the volcano. The first reported historical eruption at Rinjani occurred in September 1847. Fairly regular activity continued from then until 1966 when a small new cone formed and lava flows entered the caldera lake. An eruption during June-October 1994 (Bulletin v. 19, nos. 5-6) sent ash plumes 2,000 m above the summit, resulting in ashfall on nearby villages. A cold lahar from the summit area on 3 November killed 30 people (Bulletin v. 19, no. 10). The last reported explosion occurred on 21 November 1994 (Bulletin v. 20, no. 5). Information Contact: Bureau of Meteorology (see Tengger Caldera); NOAA/NESDIS Synoptic Analysis Branch, Rm. 401, 5200 Auth Road, Camp Springs, MD 20746 USA. Dukono Halmahera, Indonesia 1.70N, 127.87E; summit elev. 1,087 m A pilot report from a Qantas flight on the morning of 25 September described a plume to 6 km altitude that was drifting ESE. Visible satellite imagery failed to detect volcanic ash, but weather clouds in the SE sector were identified with infrared imagery. Ash clouds were last reported by aviation sources in December 1994-January 1995 (Bulletin v. 19, no. 12, and v. 20, no. 2). Halmahera lies E of the Sangihe Islands across the Molucca Sea. Dukono is the northernmost volcano in the group, 200 km ESE of Karangetang. The first of its five historical eruptions was in 1550, and it has been erupting almost continuously since 1933, although reports are infrequent due to its remote location. Information Contact: Bureau of Meteorology (see Tengger Caldera). Manam off the N coast of New Guinea, Papua New Guinea 4.10S, 145.06E; summit elev. 1,807 m Activity was low during October. During the month, both summit craters released only white vapors at low to moderate rates and both audible sounds and summit-crater night glow were absent. During the first three weeks of October, the daily totals of low-frequency earthquakes were at 200-500, but by month~s end they increased to 800-1,300. Coincident with the increase, earthquake amplitudes also rose by ~50%. No visual changes accompanied the increase in seismicity. However, data from tiltmeters (4 km SW of the summit) showed a deflation of approximately 1.5 m microradians beginning around the second half of the month. Information Contact: Ben Talai, Rabaul Volcano Observatory, P.O. Box 386, Rabaul, Papua New Guinea. Langila New Britain Island, Papua New Guinea 5.53S, 148.42E; summit elev. 1,330 m The increased eruptive activity at Crater 2 that began during late September continued throughout October. The activity was marked by intermittent audible explosions. The bigger explosions developed plumes that rose several hundred meters above the summit crater, resulting in ashfalls on the volcano~s N-NW side. Langila produced steady but weak crater glow on most nights during October; it threw incandescent lava fragments on 23-24, 26, and 31 October. Crater 3 was quiet, only giving off weak white emissions towards late October. Seismic recording restarted on 5 October after both seismographs had been inoperative since January 1995. October seismic activity was moderate. Information Contact: same as for Manam. Rabaul New Britain, Papua New Guinea 4.27S, 152.20E; summit elev. 688 m All times are local (= GMT + 10 hours) The volcanoes at Rabaul Caldera continued to remain quiet in October. Tavurvur~s summit area released bluish white vapors at very low rates; however, the emission rates rose during rainy days at the end of the month. No emissions came from Vulcan. Only 19 earthquakes were recorded in October. Two of the 13 low-frequency earthquakes originated from Tavurvur while the rest came from either within or just outside the caldera~s N sector. The six high-frequency earthquakes took place on the 20th (2 earthquakes), 23rd (2), 26th (1), and 29th (1). Most of these high-frequency earthquakes occurred in the caldera~s NE sector (Namanula area). One high-frequency earthquake (ML 1.9, on the 23rd) originated near Tavurvur at about 1 km depth. October ground deformation remained very low. Information Contact: same as for Manam. Ruapehu North Island, New Zealand 39.28S, 175.57E; summit elev. 2,797 m All times are local (= GMT + 12 hours) Ruapehu~s current eruptive period began with a vent-clearing blast on 29 June 1995 and a series of larger eruptions began on 23 September (Bulletin v. 20, no. 9). More recently available information (in Immediate Report RUA 95/06) highlighted 18 and 20 September observations summarized below. These are followed by brief comments on eruptions during October. Activity during 18-20 September. An eruption at 0805 on 18 September was accompanied by a ML 3.6 earthquake; the eruption produced the largest lahar down the ESE flank since 1975. The ESE drainage is called the Whangaehu River. Two days later, at 0122 on 20 September, another eruption associated with a smaller earthquake (ML 3.2) also sent a smaller lahar down the Whangaehu River. At roughly 0800 on 18 September the ski field manager heard what he initially thought was wind noise while he was inside a ski lodge building on Ruapehu~s flanks, a spot 400 m N of the Whangaehu channel (Aorangi lodge at Tukino). He went closer to the river and saw a 12-18 m deep lahar in the narrow channel. Later that day, a flood warning gauge 27 km downstream was triggered at 1123, suggesting the lahar moved at an average speed of roughly 2.3 m/sec (8.3 km/hr). By around noon at Tukino the lahar was 40-m wide and had covered the snow up to 20-30 m above the Whangaehu valley floor. The lahar~s surface rose about 11 m on the outside of one turn. A preliminary estimate of peak flow was >1,000 m^3/second; the local velocity, 15 m/second. An early phase of the lahar had cut out 2-3 m of ice and snow formerly filling the valley. The 18 September lahar arrived at a point 57 km downstream from Crater Lake (Karioi) at 1515, 7 hours after the eruption. Volume of the lahar at this point was estimated (by groups identified as NUWA Wanganui and ECNZ) at ~2 x 10^5 m^3; the peak flow, at ~34 m^3/second. The lahar destroyed a hiking bridge, leaving only its 0.2-m-high concrete abutments on either side of the river. The smaller 20 September lahar arrived at 57 km downstream (Karioi) 8 hours after the eruption; its size there was estimated at ~0.9 x 10^5 m^3; its peak flow, at ~21 m^3/second. In an area above ~2,000 m elevation, the 18 and 20 September lahar deposits were separated by an intervening snow layer. Still higher, above ~2,400 m elevation, both lahars had emerged from the upper Whangaehu valley~s snow and ice tunnel system. Lahars passing through and over the uppermost part of this system had produced considerable new crevasses and collapse features in the snow and ice. On 20 September, collapsed holes downstream of the large ice cave (located below the crater lake~s drainage point at Outlet, figure 7) were filled with non-steaming water that had apparently cooled. The ice cave itself appeared largely intact. A helicopter was used to visit the crater on 20 September. A large column of steam rose from the waterfall immediately below Outlet. A large volume of lake water continued to spill over the waterfall even though recent eruptions through the lake had expelled substantial lahar-forming discharges. Ash from the 18 September eruption was plastered on some steep slopes. Ash from the 20 September eruption was plastered on the new snow around the lake margins. On the E side of the lake there was a N-trending, 100-m-long lobe of ash on the glacier surface. Scoria clasts found near Outlet (the largest, 20-50 cm across) formed a continuous layer trapped behind a low lava ridge. Their distribution suggested they were deposited by a passing surge rather than as impacting ballistics. Absence of snow on the surface of the scoria indicated they had probably arrived during the 20 September eruption and some clasts still had warm interiors. Sampled clasts were black in color, and consisted of an unaltered plagioclase-, augite-, orthopyroxene-bearing andesite. The lack of Fe-Ti oxides makes them similar to 1966 ejecta; in contrast, ejecta from 1971 and 1975 did contain minor amounts of Fe-Ti oxides. Three ash samples collected from within the crater contained lapilli up to 25 mm in diameter and composed of angular lithic material. Ash finer than 2-mm diameter was dominated by gray shiny spheroids and globules of sulfur with lesser amounts of gray comminuted lake bed material. In the interval 15 August-20 September the deformation of the area about Crater lake was significant and indicated moderate inflation (figures 7 and 8). The deformation survey was hampered by snow and ice, which deeply buried most survey stations. Survey mark D had been bent 70 mm out of position immediately prior to the August survey, but eccentricity corrections enable a valid comparison with all former observations at D. Maximum changes took place in the E-W direction. These changes were similar to those computed by comparing the mean of the five surveys made earlier this year to the September survey (first column, bottom of figure 7). Non-elastic inflation of the style seen was previously noted as much as 2 weeks prior to eruptions on 8 May 1971 and 24 April 1975. This short-term inflation (lasting weeks) was also seen on 12 occasions during 1980-91; these occasions were tentatively correlated with intense heating and minor eruptions. Still, the relation between inflation magnitude and the corresponding eruption remains uncertain. The 20 September crater visit yielded the following lake observations. The lake's temperature was 48.5 degrees C (on 15 August it had been roughly 20 degrees C cooler, figure 8). There was a strong smell of SO2. The volume of water escaping at Outlet was estimated visually at 1 m^3/second (on 15 August it was only ~50 liters/second). This exceptional output was the largest seen in 24 years. Lake water sampled on 20 September showed clear increases in the concentrations of Mg, Cl, and SO4 ions, and in the ratio of Mg/Cl (figure 8). The observed concentrations for 15 August and 20 September, respectively, were as follows: Mg, 584 and 713 ppm; Cl, 8,154 and 8,619 ppm; and SO4, 26,600 and 30,600 ppm. Increases in Mg began in May and pointed to dissolution of fresh andesitic material into the hydrothermal system. Although previously it was not clear if the source of Mg was juvenile or older andesites, the increased amounts of Cl and SO4 firmly established the input of fresh magmatic material. SO4 concentrations stand at the highest levels ever recorded at Ruapehu. In the absence of synchronous increases in K, and noting that Ca continues to be controlled by gypsum solubility, it is clear that the increases in SO4 were not attributable to dissolution of secondary hydrothermal minerals. Instead the SO4 increases indicated greater SO2 flux into the lake. Assuming a lake of 9 x 10^6 m^3, the increase in SO4 from 15 August to 20 September equates to a minimum input of ~700 metric tons/day of SO2 into the lake. This behavior differs from that observed prior to the 1971 eruptions: The indication is that the quantity of magma involved in the current activity is larger than in the 1971. Taken with the rather moderate degree of cross-crater deformation seen, the quantity of SO2 discharged into the lake indicates connection to larger volumes of degassing magma at depth. Volcanic tremor remained at background from early July until early September; its amplitude was ~1 micrometers/sec for signals centered around 7 Hz, and at this value or slightly lower for signals centered around 2 Hz. During a five day interval starting on 6 September, the amplitude of 2-Hz tremor increased. During the 24 hours prior to the 18 September eruption and earthquake (Bulletin v. 20, no. 9), predominantly 7-Hz tremor occurred, at one point doubling in amplitude. Later, ~80 minutes prior to the eruption and earthquake, tremor again increased by a factor of 2-3, with 2-Hz tremor becoming dominant. Although dramatic, Ruapehu often displays wide-ranging shifts in tremor amplitude and, in retrospect, the increased amplitudes seen would not have been a useful way to predict the eruption. The 18 September earthquake took place at 0805, continuing for 6 minutes. Analog seismograms from the three local stations (Dome, Chateau, and Ngauruhoe) were pegged, and the M 3.6 estimate was made based on amplitude recorded by the tremor-monitoring system. After the earthquake, predominantly 2-Hz tremor prevailed, remaining at or above the pre-earthquake amplitude. Later the same day (18 September), strong 1-Hz tremor occurred--for the first time at Ruapehu since the early 1970s. Further minor earthquakes were recorded during the next few days. On 19 September seismometers registered a ML 2.2 earthquakes as well as four other discrete earthquakes; on 20 September there were ML 3.1 and 3.2 earthquakes followed by another interval of strong 1-Hz tremor until 0900. October eruptions. At the time of this writing, IGNS reports for October are incomplete, but a brief survey of available ~Science Alert Bulletins~ and aviation reports suggested that minor eruptions continued and in mid-October moderate ash-rich eruptions took place. On 11 October a plume was seen in satellite imagery; on 12 and 14 October, pilot and associated aviation reports indicated ash to at least ~10 km altitude. The 11 October eruption was described as near-continuous moderate eruptive activity that included hot ballistic blocks and lightning. Subsequent lower intensity eruptions presumably fed the plume so that its proximal end remained attached to the volcano. The eruption deposited ash in a blanket with a tentative volume between 0.01 and 0.05 km^3. Thus, the steam-rich plumes seen in the 3 weeks prior to 11 October gave way to more ash-rich plumes during this eruption. A thin blanket of ash was also deposited during the 14 October eruption. The absence of a crater lake was confirmed on 14 October. By 17 October, partly impeded views into the crater revealed steam and ash emitted from at least 3 vents, and a still-dry crater floor. COSPEC measurements around this time suggested the SO2 flux was over 10,000 metric tons/day. A COSPEC flight on 21 October gave viewers their first look at a possible new lava dome, however, there were no subsequent confirmations of the dome in available reports. Information Contacts: C.J.N. Wilson, B.J. Scott, P.M. Otway, and I.A. Nairn, Institute of Geological & Nuclear Sciences (IGNS), Private Bag 2000, Wairakei, New Zealand (Email: bscott@gns.cri.nz); Bureau of Meteorology (see Tengger Caldera). Correction: The most recent analysis indicates that there were 18 hydrothermal eruptions recorded between 0600 and 1640 on 20 September. Table 2 in Bulletin v. 20, no. 9 indicated ~15 small phreatic eruptions witnessed.~ Ruby Seamount Mariana Islands, W Pacific 15.62N, 145.57E; summit elev. -230 m (submarine) Ruby is a prominent, active submarine volcano in the Mariana Arc (2,300 km S of Tokyo) located NW of the Island of Saipan (figure 9). Although signs of an eruption were first noted by fishermen about 11 October, initial attempts to confirm their early observations failed. On 23 October fishermen reported that they could hear submarine explosions in that vicinity. A vessel from the Wildlife and Emergency Management Office of the Commonwealth of the Northern Marianas Islands confirmed these reports. An Associated Press news report stated that early on 25 October observers had seen dead fish and bubbles, and had smelled a sulfurous odor. On 27 October the Pacific Daily News reported the eruption site as 15 deg 36 min 22"N, 145 deg 34 min 33"E (15.6061N, 145.5758E). This spot clearly lies on the edifice identified by Bloomer and others (1985, p. 215) as Ruby (only ~1.7 km from the point specified in this report's heading). Prior to the eruption, published estimates of the summit elevation suggested a 230-m depth, a refinement an earlier estimate of 549 m (Bloomer and others, 1985, p. 215). On 6 October 1995, the Pacific Daily News report stated the summit was measured at 185-m depth. This newly reported depth remains unconfirmed. According to Mike Blackford, on 23 October a marine depth finder reportedly measured a depth of ~60 m. Although this could be a reflection off the eruptive plume, in the absence of any discussion of instrument type and calibration, this depth remains equivocal. According to Koyanagi and others (1993), the two seismic stations nearest the eruption were on Saipan (~50 km SE of Ruby) and Pagan islands (~130 km N of Saipan), both too distant to detect subtle seismic effects. Despite the lack of a nearby seismic station, tremor appeared on seismic records at the time of the eruption and the next day. Given the temporal coincidence between the eruption and the tremor, the two were probably associated. A fish recovered at the eruption site was found to have small particles of ash in its gills and HVO researchers planned to analyze this ash. News of the eruption caused concern about a possible local tsunami and on 25 October, the Commonwealth of the Northern Marianas Islands issued an alert. Evidence for Ruby~s active status came from 1966 hydrophone data, followed later by dredging of extremely fresh volcanic rocks bearing plagioclase, clinopyroxene, and olivine (Bloomer and others, 1985). The larger and much shallower Esmeralda Bank (figure 9) has been the site of 6 possible eruptions in a record starting in 1944. References: Koyanagi, R., Kojima, G., Chong, F., and Chong, R., 1993, Seismic monitoring of earthquakes and volcanoes in the Northern Mariana Islands: 1993 summary report: Prepared for the Office of the Governor, Commonwealth of the Northern Mariana Islands, Capitol Hill, Saipan MP 96950 (revised 21 February 1993), 34 p. Reference: Bloomer, S.H., Stern, R.J., and Smoot, N.C., 1989, Physical volcanology of the submarine Mariana and Volcano arcs: Bull. Volcanol., no. 51, p. 210-234. Information Contacts: Robert J. Stern, Center for Lithospheric Studies, University of Texas at Dallas, Box 830688, Dallas, TX 75083-0688 USA (Email: rjstern@utdallas.edu); Robert Koyanagi, USGS Hawaiian Volcano Observatory, Hawaii Volcanoes National Park, HI 96718 USA; Ramon C. Chong, Commonwealth of the Northern Mariana Islands (CNMI), Disaster Control Office, Capitol Hill, Saipan, MP 96950 USA; Mike Blackford, Pacific Tsunami Warning Center, 91-270 Fort Weaver Road, Ewa Beach HI 96706, USA; Associated Press; Pacific Daily News. Sakura-jima Kyushu, Japan 31.58N, 130.67E; summit elev. 1,117 m Activity at Minami-dake crater became high during both early and late October. On 28 October, 9 explosive eruptions occurred and significant volcanic ash fell in Kagoshima City. During October, seismic station B (2.3 km NE of Minami-dake crater) recorded 720 earthquakes and 1,206 tremors. On 27-28 October there were seismic swarms. During October the volcano produced 31 eruptions, 23 of them explosive; the highest ash plume, on 28 October, rose 3 km above the summit crater. October ashfall (measured 10 km W at the Kagoshima Meteorological Observatory) was 117 g/m^2. Sakura-jima, situated within the Aira Caldera, began to grow about 13,000 years ago. Whereas the volcano had formed an island in the N half of Kagoshima Bay, lava flows of the 1914 eruption established a subaerial connection to the adjacent peninsula. Frequent historical eruptions have deposited ash on Kagoshima, a city located 7 km W of the summit. Information Contact: Volcanological Division, Seismological and Volcanological Department, Japan Meteorological Agency (JMA), 1-3-4 Ote-machi, Chiyoda-ku, Tokyo 100 Japan. Aso Kyushu, Japan 32.88N, 131.10E; summit elev. 1,592 m All times are local (= GMT + 9 hours) During October the floor of Aso~s active crater (Naka-dake Crater 1) remained covered by a pond of hot water. The pond~s surface was disrupted by occasional fountaining up to 5-m high. Elevated tremor continued since last month, and some October days had over 200 earthquakes; the daily mean amplitude of continuous tremors sometimes reached over 0.5 ~m. Personnel 800 m W of the crater (at Aso Weather Station) felt earthquakes at 1829 and 1909 on 11 and 22 October, respectively. The 24-km-wide Aso Caldera contains 15 central cones. One of these cones, Naka-dake, has erupted more than 165 times since 553 AD. Information Contact: JMA (see Sakura-jima). Kuju Kyushu, Japan 33.08N, 131.25E; summit elev. 1,788 m All times are local (= GMT + 9 hours) On 11 October, aseismic phreatic eruptions started within the Kuju volcanic group, on Hosho (Hosyo) dome~s E side (Bulletin v. 20, no. 9). On 12 October observers found an E- W trending line of vents ~300-m long; also, at that time an ash-bearing plume rose to ~1 km above the crater. The eruption deposited a 100 m^2 blanket of fist-sized volcanic clasts; it also emitted mud that flowed down an adjacent valley. After that, the volume and height of the plume gradually decreased until finally ash-bearing eruptions ceased at the month~s end. Seismicity stayed low during October. The Kuju group (or volcanic field) resides SW of Oita Prefecture, Kyushu, and is located NE of Aso Caldera. The group consists of about 20 volcanoes, including stratovolcanoes and many lava domes. Information Contact: JMA (see Sakura-jima). Kozu-shima Izu Islands, Japan 34.22N, 139.15E; summit elev. 574 m All times are local (= GMT + 9 hours) As reported in Bulletin v. 20, no. 9, on 6 October a M 5.6 earthquake occurred adjacent to Kozu-shima and a seismic swarm followed for the next few days. After that, seismic events continued but decreased toward the end of October; in total, during October there were 246 felt earthquakes. Information Contact: JMA (see Sakura-jima). Oshima Izu Islands, Japan 34.73N, 139.38E; summit elev. 758 m All times are local (= GMT + 9 hours) On 4 October, local instruments recorded volcanic tremor of short duration and small amplitude. Throughout the month a significant but undisclosed number of earthquakes occurred in the adjacent N and W coastal areas. During October there were 48 earthquakes beneath the cone. Oshima stratovolcano lies in Sagami bay, forming the northernmost of the Izu Islands. Morphologically broad and low, Oshima is capped by a 4-km-wide caldera that holds the active cone called Mihara. Mihara has erupted dozens of times since the late 1700~s. Information Contact: JMA (see Sakura-jima). Izu-Tobu Honshu, Japan 34.92N, 139.12E; summit elev. 1,406 m All times are local (= GMT + 9 hours) Mid- and late-September micro-earthquake swarms occurred offshore near Capes Kawana-zaki and Shiofuki-zaki (Bulletin v. 20, no. 9), an area adjacent Ito City on the E coast of the Izu Peninsula. In late September and early October pulses of seismicity continued off these Capes, trailing off toward mid-October (figure 10). Located ~5 km SW of the epicenters, Kamala Seismic Station recorded 5,881 October earthquakes. The largest earthquake struck at 1142 on 1 October with M 4.8; nearby Into City sustained a JMA-scale intensity of IV. Small-amplitude tremors occurred on both 4 October (4 times), and 12 October (1 time); low-frequency earthquakes took place on 4 October (4 times) and 6 October (1 time). Volumetric strain at Higashi-Izu and Ajiro acted in the sense of compression. The Izo-Tobu volcanic group lies scattered over a 400 km^2 area of the eastern Izu Peninsula. About 70 subaerial monogenic volcanoes formed during the last 40,000 years, and chemically similar ones lie offshore. Thirteen eruptive episodes have been documented in the past 32,000 years. Information Contact: JMA (see Sakura-jima). Iwate Honshu, Japan 39.85N, 141.00E; summit elev. 2,041 m All times are local (= GMT + 9 hours) Tohoku University seismometers near Iwate volcano continued to register tremor (Bulletin v. 20, no. 9). Beginning at 0009 on 20th October, the tremor lasted ~25 minutes. Located on northern Honshu NW of Morioka City, Iwate sits adjacent to an older cone. Several distinct lava flows drape Iwate~s flanks including a fresh lava flow erupted in 1719. Subsequent eruptions did not produce lavas; the latest eruptions took place in 1919. Information Contact: JMA (see Sakura-jima). Adatara Honshu, Japan 37.62N, 140.28E; summit elev. 1,718 m During 27 October, volcanic tremor of about 3-minutes duration was recorded at a site 4.8 km NE of Adatara~s summit (station A). This was the first case of tremor since the local observatory began observations in 1965. Adatara, which sits E of Bandai volcano, consists of numerous coalescing stratovolcanoes and lava domes. The volcano erupted on 24 August 1899 and 17 July 1900. Information Contact: JMA (see Sakura-jima). Akan Hokkaido, Japan 43.38N, 144.02E; summit elev. 1,499 m Seismicity during October, and thus far in 1995, remained slightly higher than was typical for the past several years (figure 11). The highest daily number of earthquakes during the month took place on 2 October and consisted of 33 events (recorded at Station A, 2.3 km from Ponmachineshiri Crater). The monthly total for October consisted of 395 events. The 13 x 24 km Akan Caldera contains a lake, post-caldera stratovolcanoes, and various cones including the historically active Me-Akan. Information Contact: JMA (see Sakura-jima). Medvezhia Iturup Island, Kurile Islands, Russia 45.38N, 140.80E; summit elev. 1,124 m The Kudriavy cone at Medvezhia has been the subject of great interest since Korzhinsky and others (1994) discovered a pure rhenium sulfide mineral in its high-temperature (535 degrees C) fumaroles. Given the concentration of Re found in gas samples (2-10 ppb), the occurence of ReS2 or Re2S3 (exact form still uncertain) requires enrichment of Re by eight orders of magnitude. During a 21 August-5 September visit, Stan Williams, Tobias Fisher, and Russian colleagues made COSPEC measurements of SO2 flux. Gas samples were also collected from crater fumaroles. The COSPEC was operated from the base camp (150 m elevation) 2.7 km SE of the elongate ENE-WSW summit (990 m elevation), while the wind velocity was measured at the summit using a hand-held anemometer. The first measurements were made on 28 August in vertically oriented stationary mode with the wind blowing the wide gas plume directly over the camp. Wind velocity was measured at 3 m/second and the plume was estimated to be rising only 50 m above the summit before being blown downwind. The flux was found to be 100 +- 20 metric tons/day (t/d) for the three measurements possible before the wind shifted to the SW, making any measurements essentially oblique to the plume axis, and therefore of great uncertainty. An occasional strong odor of H2S was detected at the camp during measurements. During a crater visits on 29 and 31 August, and during a helicopter flight downwind in the plume, there was always a strong odor of H2S. Under clear skies on 31 August the wind carried the plume SW, allowing stationary mode vertical and horizontal measurements. One specific goal of the research was to quantify the output of the isolated fumarole fields with different temperatures, which was possible for part of the day. Wind velocity was measured at 3 m/second and the SO2 flux was calculated to be only 30 +- 10 t/d. Kudriavy has been consistently degassing passively at high temperatures since at least 1961, when it was first visited by Russian scientists; annual fieldwork began in 1989. No change in activity was noted during this visit. At the low levels at which Kudriavy was found to be degassing, a realistic SO2 norm may be 75 +- 50 t/d, with the oscillations potentially reflecting meteoric conditions. These low-pressure fumaroles, some with temperatures up to 950 degrees C, produced hissing to roaring to deafening levels of noise. There was more noise and higher flux than at Momotombo (Nicaragua), where 950-degree-C fumaroles were studied through the early to mid-1980s by U.S. and Russian collaboration. Medvezhia is an eroded 8-km-wide caldera at the NE tip of Iturup Island containing several cones and domes and a small caldera lake. Three large volcanic cones lie on an E-W line. The easternmost, Medvezhii cone, lies outside the caldera; the central cone, Sredni, has produced lava flows that reached the coast. The westernmost, Kudriavy, has been historically active with known summit crater eruptions in 1778 and 1883, a possible phreatic eruption in 1946, and a mild explosive eruption in 1958. Menshoi Brat, a lava dome on the W side of Kudriavy, has produced a series of young lava flows. Reference: Korzhinsky, M.A., Tkachenko, S.I., Shmulovich, K.I., Taran, Y.A., and Steinberg, G.S., 1994, Discovery of a pure rhenium mineral at Kudriavy volcano: Nature, v. 369, p. 51-52. Information Contacts: Stanley N. Williams and Tobias P. Fischer, Geology Dept., Arizona State University, Tempe AZ 85287, USA; Kirill I. Shmulovich and Mikhail A. Korzhinsky, Inst. of Experimental Mineralogy, Russian Academy of Sciences, 142432 Chernogolovka, Moscow District, Russia; Genrich S. Steinberg, Inst. of Volcanology & Geodynamic ANSRF, 693008 Yuzhno-Sakhalinsk, Box 18, Russia. Yellowstone Wyoming, USA 44.43N, 110.67W; summit elev. 2,805 m On the SW flank of Sour Creek resurgent dome W of Astringent Creek in the 0.6 Ma Yellowstone caldera, is an extensive, unnamed acid sulfate hydrothermal system (figures 12 and 13). Surface expression of the ~3 km^2 thermal area consists of discontinuous high temperature altered ground, turbid springs, pools, seeps, fumaroles, mud pots, a large gas- and sulfur-rich acid lake, and numerous sublimated sulfur mound deposits interspersed among low-temperature forest-covered ground. During early 1990, a significant rise in temperature in the upper NW end of the hydrothermal system began killing old- growth pine trees. Within a year, a new super-heated fumarole emerged, blanketing the downed trees and roots with a layer of hydrothermally altered coarse sand from a directed blast to the N. The temperature and volume of dry steam venting from the deep ~shaft-like~ vent steadily increased over the next three years, with the temperature reaching a maximum of 104.3 degrees C on 8 October 1994, ~11 degrees C higher than the local boiling point. The dynamic activity of the fumarole and surrounding hot ground was only monitored about twice a year over the three years following its 1990 inception due to its remote location and restricted access. A similar progression was previously seen during 1985 in an area ~4.5 km to the E. This area, the upper E margin of the Mushpots thermal area, sits on the W flanks of Pelican Cone (Bulletin v. 17, no. 3). The progression went from new hot ground and dying mature forests, to the vigorous breakout of a dry, super-heated fumarole with progressively hotter temperatures over time, followed by sudden emergence of a large and violent mud volcano. Both the 1985 and recent thermal features had similar fluid compositions. During 1992-94 the unnamed thermal area W of Astringent Creek developed a series of seven large craters that evolved as the Mushpots thermal area did in 1985. The craters were progressively younger towards the SW, ending at the site of the current new hot ground and fumarole (figure 14). In December 1993, National Park Service research geologist R. Hutchinson predicted that the newest superheated fumarole would soon evolve into a large mud volcano. As a part of routine monitoring, the thermal area W of Astringent Creek was inspected on 7 June 1995. The former 104.3 degrees C fumarole was replaced by a large vigorous mud pot with ejecta extensively scattered around it. In addition, two new smaller roaring fumaroles at or slightly above boiling point, three new moderate-sized churning caldrons (pits containing hot, agitatated aqueous fluids), numerous smaller muddy pools, collapse pits, and frying-pan springs (audibly degassing springs) were apparent then. Extensive areas of unstable quicksand-like saturated ground made up of scalding mud were found under the fallen trees. Some regions were heavily encrusted with sulfate minerals or sulfur crystals; others were covered by baked organic matter on the pine forest's floor. Extending NW from the largest parasitic churning caldron, below the new mud volcano crater, was a spectacular white kaoline clay mud flow (figure 14, dark shading and arrow showing flow direction). It spread rapidly to reach an average width of 13.8 m in the first 55 meters of its length in dead forest grove and eventually terminated 114 m from its source on the open, acid thermal-basin floor. The relative freshness of the ejected mud and incorporated semi-coarse sandy material indicated that the super-heated fumarole transformed into the powerful mud volcano between mid-April and mid-May. The distribution of large mud bombs suggested that their trajectories reached 20-30 m above the crater rim. Ejecta were seen along the following compass bearings with the stated maximum distances from the crater: N, 13.6 m; E, 30.2 m; S, 25.4 m; and W, 12.1 m. When visited on both 7 June and 9 September, the mud volcano still continued to throw mud 0.5-1.5 m high from dozens of points around the crater floor. The mud volcano crater was 13.5-m long, 11.3-m wide, and 3.9-4.9 m deep. A conservative estimate of the crater volume was 315 m^3. The total area covered by the ejecta and crater was ~2,100 m^2. In the SW quarter of the crater a large, slightly elevated projection was visible with an arcuate line of dry, white, probably super-heated fumarole vents. The largest parasitic caldron had numerous points of ebullition in its irregularly shaped pool (maximum dimensions of 10.8 x 7.9 m), with a water level 0.7-1.4 m below the former forest floor. The churning water was near boiling, opaque, light tan in color, and partially covered with brown organic-rich foam derived from cooked plant material. Each of the caldrons were interpreted as being parasitic to the mud volcano crater because they appeared to have evolved shortly after the initial fumarole collapse and then subsequently drained much of its fluids. This relationship seems to have rapidly lowered the crater floor, preventing the accumulation of a thick ejecta cone on the crater rim. The mud volcano crater, parasitic features, vents, and the associated hot ground remain extremely dangerous and unstable. Additional alterations in the creation of new or enlarged springs, and perhaps even another mud volcano crater are anticipated. With respect to geologic hazards, the acid sulfate thermal area should be checked again in the near future. Photographs were taken on 7 June. The Yellowstone Plateau volcanic field developed through three volcanic cycles spanning two million years and included some of the world~s largest known eruptions. Eruption of the > 2,500 km^3 Huckleberry Ridge Tuff ~2.1 million years ago (Ma) created a caldera more than 75 km long. The Mesa Falls Tuff erupted around 1.3 Ma, forming the 25-km-wide Island Park Caldera at the first caldera~s W end. A 0.6 Ma eruption deposited the 1,000 km^3 Lava Creek Tuff and associated caldera collapse created the rest of the present 45 x 75 km caldera (figure 13). Resurgent doming then occurred; voluminous (1,000 km^3) intercaldera rhyolitic lava flows were erupted between 150,000 and 70,000 years ago. Phreatic eruptions produced local tephra layers during the early Holocene. Distinctive geysers, mud pots, hot springs, and other hydrothermal features within Yellowstone caldera helped lead to the establishment of the National Park in 1872. Information Contact: Roderick A. Hutchinson, National Park Service, P.O. Box 168, Yellowstone National Park, Wyoming 82190, USA. Atmospheric Effects Lidar data from Germany for July and August (table 1) again revealed the presence of a volcanic aerosol layer centered at 17-19 km altitude. Backscattering ratios have decreased since the last reports (Bulletin v. 20, nos. 2 and 7). October lidar data from Hampton, Virginia, showed an aerosol layer at 18-19 km altitude; these values are similar to the previous report (Bulletin v. 19, no. 11). Backscatter data declined to the range of 1.22-1.25 from 1.38-1.50. Table 1. Lidar data from Germany and Virginia, USA, showing altitudes of aerosol layers. Backscattering ratios are for the ruby wavelength of 0.69 microns. The integrated value shows total backscatter, expressed in steradians^-1, integrated over 300-m intervals from the tropopause to 30 km. LOCATION DATE LAYER ALTITUDE (KM) BACKSCATTERING (peak) RATIO INTEGRATED Garmish- 07 Jul 11-27 (19.7) 1.12 (1.3) ** Partenkirchen, 19 Jul 12-26 (19.8) 1.13 (1.3) ** Germany 21 Jul 13-29 (18.0) 1.12 (1.3) ** (47.5N, 11.0E) 26 Jul 11-28 (19.1) 1.13 (1.3) ** 31 Jul 13-24 (18.8) 1.09 (1.2) ** 03 Aug 12-27 (17.5) 1.12 (1.3) ** Hampton, 23 Mar 12-25 (17.8) 1.36 0.135 ~ 10^-3 Virginia 04 May 12-25 (18.7) 1.3 0.104 ~ 10^-3 (37.1N, 76.3W) 19 Oct 15-30 (18.1) 1.22 0.059 ~ 10^-3 23 Oct 15-30 (18.8) 1.25 0.065 ~ 10^-3 Information Contacts: Horst Jager, Fraunhofer -- Institut fur Atmospharische Umweltforschung, Kreuzeckbahnstrasse 19, D-8100 Garmisch-Partenkirchen, Germany; Mary Osborn, NASA Langley Research Center (LaRC), Hampton VA 23665, USA (Email: osborn@arbs5.larc.nasa.gov). EARTHQUAKES Large and damaging earthquakes during October are listed on table 2. On 1 October, an earthquake occurred in Turkey, near the town of Dinar, 250 km SW of Ankara and 155 km SE of the Kula cinder cone field. The main shock, was variably estimated between M 5.9 and 6.3. The earthquake was strongly felt in much of western Turkey; it was preceded by a series of foreshocks during the previous week and followed by several aftershocks in the magnitude range 3.5-4.7. In Dinar and vicinity, reports noted 101 casualties and 348 injuries; 40% of the buildings were destroyed, 30% were heavily damaged. A M 6.8 earthquake struck near the Peru-Ecuador border on 3 October: It was felt throughout Ecuador, in northern Peru, and as far away as Bogota, Colombia. A large number of aftershocks followed within the next few days, five above M 5.0. The earthquake killed two people, injured a few people, and damaged both local infrastructure and private homes. The epicenter was located 90 km SE of Sangay volcano, Ecuador. On 5 October, a M 6.8 earthquake took place in Western Sumatra, Indonesia. The epicenter, ~725 km NW of Jakarta, struck a highly populated area. The death toll was 84, ~2,000 people were injured, and 64,000 people were left homeless. This earthquake was located in the highly active volcanic front associated with the Sumatra-Java trench. On 9 October the largest event of the month took place offshore of the State of Colima, on the W coast of Mexico, 530 km W of Mexico City. The shock, M 7.6, was felt as far away as Houston and Oklahoma City. The State of Colima reported the heaviest losses, including 48 people killed in the collapse of a hotel. The resulting tsunami reached 200 cm at nearby Manzanillo (peak to trough), and 12 cm at Kahului (Hawaii). Many aftershocks were recorded; the strongest, M 5.6 on 12 October, caused further damage in the epicentral area. Volcan de Colima, Mexico~s most active volcano, lies 87 km NE of the epicenter. Two M 6.9 earthquakes were recorded on 18 and 19 October in the Northern Ryukyu Islands. Both epicenters were offshore of Kikai-shima Island. Within the next three days these two shocks were accompanied by at least 12 aftershocks with M > 5. The shocks were strongly felt in the Amami Islands, but no report of casualties or damages was received. Kikai-shim Island is located 180 km ESE of Tori-shima volcano and 340 km S of Sakura-jima volcano. Yunnan province of S China was hit by a M 6.4 earthquake on 24 October. The shock was felt also in Sichuan Province and Northern Vietnam; the epicenter plotted 100 km NNW of Kunming, Yunnan's provincial capital, at a spot 400 km W of Tengchong volcano. Reports noted 36 deaths, 200 injuries, and several collapsed houses. Table 2. Summary of large and damaging earthquakes, October 1995. Courtesy of the National Earthquake Information Center. DATE GMT MAG LAT LONG DEPTH REGION 1 Oct 1557 Mw 6.0 38.04N 30.15E shallow Turkey 3 Oct 0151 Mw 6.8 02.70S 77.86W shallow Peru-Ecuador 5 Oct 1809 Mw 6.8 01.95S 101.53E shallow Indonesia 9 Oct 1536 Mw 7.6 18.87N 104.15W shallow Mexico 18 Oct 1037 Ms 6.9 28.10N 130.40E shallow Japan 19 Oct 0241 Ms 6.9 28.40N 130.60E shallow Japan 24 Oct 2246 Mw 6.1 25.86N 102.22E shallow China