Home Reef (Tonga) Ex ten sive pum ice rafts be tween Tonga and Fiji dur ing Au gust-Oc to ber . . . . . . . . . . 2 Rabaul (Pa pua New Guinea) Strong erup tion at Tavurvur ejected ash and large plumes to the tro po sphere . . 6 Bamus (Pa pua New Guinea) Force ful vapor emis sion seen on 12 July 2006 . . . . . . . . . . . . . . . . . . . 9 Sulu Range (Pa pua New Guinea) Vol cano seis mic ity de clines in Sep tem ber and Oc to ber 2006. . . . . . . . . 9 Bar ren Is land (In dia) On go ing emis sions, in clud ing lava, but late-Sep tem ber news re ports of slow ing pace . 10 Bulusan (Phil ip pines) Ten ex plo sions re corded seis mi cally be tween 21 March and 28 June 2006 . . . . . . . 10 Cleve land (Alaska) Short du ra tion ex plo sions dur ing Au gust-Oc to ber 2006 . . . . . . . . . . . . . . . . . 11 Fourpeaked (Alaska) Erup tion on 17 Sep tem ber, fol lowed by emis sions un til at least early No vem ber . . . . 11 Soufri?re Hills (Montserrat) Ex tru sive dome dy nam ics dur ing May-Sep tem ber 2006 . . . . . . . . . . . . . 14 San Crist?bal (Nic a ra gua) Multi-year up date: 13 June 2004, lo cal ash fall; early 2006, small erup tions . . . 16 Montagu Is land (S Sand wich Is lands) Five years of nearly per sis tent erup tive ac tiv ity . . . . . . . . . . . . 16 Ed i tors: Rick Wunderman, Ed ward Venzke, Sally Kuhn Sennert, and Catherine Galley Vol un teer Staff: Rob ert An drews, Jerome Hudis, Ve ron ica Bemis, Jackie Gluck, Wil liam Henoch, Hugh Replogle, Zahra Hirji, Ste phen Bentley, Paul Berger, Jeremy Book binder, and Antonia Book binder Global Vol can ism Program ? Na tional Mu seum of Natu ral His tory, Room E-421, PO Box 37012 ? Wash ing ton, DC 20013-7012 ? USA Tele phone: (202) 633-1800 ? Fax: (202) 357- 2476 ? Email: gvn@si.edu ? URL: http://www.vol cano.si.edu/ The text of the Bul le tin is also dis trib uted through the Vol cano Listserv (vol cano@asu.edu). Bul le tin of the Global Vol can ism Net work Vol ume 31, Num ber 9, September 2006 Home Reef Tonga Is lands, SW Pa cific 18.992?S, 174.775?W; sum mit elev. -2 m All times are lo cal (= UTC + 13 hours) Pum ice rafts drift ing from Tonga to Fiji oc curred dur ing Au gust-Oc to ber 2006. The source of these pum ice rafts was Home Reef, which was first ob served to be in erup tion on 9 Au gust and was clearly build ing an is land by 12 Au - gust (fig ure 1). A com pi la tion of re port sight ings through mid-Oc to ber 2006, plot ted us ing Google Earth, shows the tim ing and dis tri bu tion of the pum ice rafts that are dis - cussed in this re port (fig ure 2). As is our con ven tion, and as avail able, a list of con tri bu tors (and their ves sels) is noted in last sec tion of this re port. Pum ice trav eled both N and S around Fiji's Lau Group. To the N, pum ice reached Taveuni through the Nanuku pas sage and en tered the Koro sea, wash ing onto south ern Vanua Levu, be fore mov ing into the Bligh Wa ters N of Viti Levu by 20 Sep tem ber. To the S, ex ten sive pum ice was seen N of Vatoa Is land on 16 Sep tem ber, and on Kadavu Is - land by the end of the month. Pum ice was also en coun tered by the En core II W of Viti Levu on 30 Sep tem ber while enroute to New Cal edo nia. Early ob ser va tions of the erup tion. The news ser vice Matangi Tonga On line quoted Allan Bowe, the owner of the Mounu Is land Re sort in south ern Vava?u, re gard ing vol ca nic ac tiv ity in the di rec tion of Home Reef dur ing 9-11 Au gust. Bowe heard ?. . . what sounded like con tin u ous thun der rum bling to the S and there was a huge plume of smoke and cloud ris ing up into the sky.? In an other Matangi news ar ti cle, Siaosi Fenukitau, a cap tain of one of the fish ing boats of the Mar i time Pro jects Co. (Tonga) Ltd., re ported that around mid-Sep tem ber they sighted a new vol ca nic is land near Home Reef that was larger than Fotuha?a, a small is land in Ha?apai with a pop u la tion of about 134 peo ple. The yacht Maiken left Neiafu on 11 Au gust, pass ing the N side of Late Is land. Af ter about 9 km the crew no ticed brown, some what grainy streaks in the wa ter. The streaks be came larger and more fre quent as they con tin ued SW ?un til the whole ho ri zon was a solid line to what looked like a desert.? The brown ish pum ice frag ments the size of a fist were float ing in wa ter that was strangely green. They mo - tored into the vast (many miles wide) belt of densely packed pum ice, and within sec onds Maiken slowed down from seven to one knot. Ini tially the thin layer on the sur - face was pushed away by the bow wave, but when they en - tered the solid field it started to pile up and ?be haved like 2 Home Reef Smithsonian Institution ? Bulletin of the Global Fig ure 1. Pho to graph of the new is land be ing built by the erup tion at Home Reef as seen on 12 Au gust 2006. The is land was ~1.5 km in di am e ter. View is to wards the W from about 2.8 km away. Cour tesy of Fredrik Fransson of the Maiken. Fig ure 2. Map of Tonga (right) and Fiji (up per left) show ing dates and lo ca tions where ob serv ers saw pum ice rafts (placemarks with dots) or where mar i ners cross ing be tween Tonga and Fiji failed to see rafts (placemarks with crosses). Some lo ca tions are ap prox i mate; see text for ad di tional de tails and sources of each ob ser va tion. The base map is from Google Earth with points plot ted by Bul le tin ed i tors. wet con crete? and ?looked like roll ing sand dunes as far as the eye could see.? Af ter re treat ing from the pum ice with only mi nor paint abra sion along the wa ter line, and then clean ing their in take fil ters, they de cided to an chor in Vaiutukakau bay out side Vava?u for the night. The next morn ing, 12 Au gust, they re ceived ra dio con fir ma tion of an erup tion, but the vent and ex tent were un cer tain. They de - cided to go S to avoid the pum ice rafts float ing NW, head - ing SSW un til they en coun tered the pum ice, then sail ing along side un til the rafts were bro ken up enough to safely travel through. As they ap proached Home Reef it be came clear that one of the clouds on the ho ri zon was a vol ca nic plume. Ob ser - va tions from a closer van tage point re vealed that an in ter - mit tent ?mas sive black pil lar shot up wards to ward the sky? and par ti cles were rain ing down. Since the wind was push - ing the plume NW, the Maiken mo tored up to within 2.8 km of the is land (to 18?59.5?S, 174?46.3?W) while the sun was go ing down. Mul ti ple peaks form ing a crater open to the sea on one side were vis i ble, and it looked like it was ?made of black coal.? Not want ing to en coun ter more pum ice rafts af ter dark, they con tin ued SSW to wards the south ern part of the Lau Group. Pum ice sight ings be tween Tonga and Fiji. Boats that later noted see ing pum ice in Fiji did not re port any ac tiv ity or rafts near Tonga dur ing 27-29 Au gust. The Soren Larsen sailed through ?a sea of float ing pum ice? one eve ning that ?sounded like we were sail ing through ice? just be fore reach ing Fiji. This en coun ter was prob a bly on 30 Au gust when their on line tracker lo cated the ship just W of the cen - tral Lau is lands af ter de part ing Neiafu on the 28th. No erup - tive ac tiv ity or pum ice was noted in the on line log of the Soren Larsen for 14-15 and 23-24 Au gust when they transited to north ern Tonga to the E of Home Reef. While the En core II crew was vis it ing the Mounu Is - land Re sort on 2 Sep tem ber there were ?grape fruit-sized? pum ice pieces on the beach. A few days later, while lis ten - ing to the ?Rag of the Air? net broad cast out of Fiji, the En - core II crew learned of pum ice rafts along their ex pected route. The op er a tor of this broad cast, Jim Bandy, pro vides weather re ports for boats go ing be tween Tonga and Fiji. One re port was of a mass of pum ice about 11 km long and at least a me ter (?many feet?) deep. The En core II de parted from Neiafu on 8 Sep tem ber on a course around a set of Fijian is lands and reefs called the Lau Group. The crew be - lieved that this route, go ing NW around the Lau Group, helped them avoid most of the pum ice. As the En core II ap proached their turn ing point about two thirds of the way to Fiji, on 10 Sep tem ber, they en - coun tered ?rivers of pum ice? float ing roughly par al lel to their NW course due to the SE winds (fig ure 3). Some pum - ice frag ments that they col lected were about 5-10 cm in di - am e ter, al though most were about the size of pea gravel. The par al lel streams of pum ice, only a sin gle layer in depth, were some times up to 90 m wide and 400 m long. The crew later heard re ports from sev eral boats that had taken a more west erly route through the Lau Group to Fiji and en coun - tered much larger ar eas of pum ice. The crew on the Nor we - gian sail boat Stormsvalen went through larger and thicker ar eas of pum ice, leav ing a track in the pum ice as they went through (fig ures 4 and 5). They noted that boats trav el ing through the pum ice dur ing higher winds and seas en coun - tered a prob lem of air borne pum ice pelt ing the crews and their boats. One crew re ported pum ice cov er ing their deck. A sail boat blog en try by Sara Berman and Jean Philippe Chabot noted a ?strong sul fur odor? in the di rec tion of the vol cano upon leav ing Ton gan wa ters around 20 Sep tem ber. As they pro gressed SW to wards Fiji they passed through streams of pum ice con tain ing pieces rang ing from very Volcanism Network, Volume 31, Number 9, September 2006 Home Reef 3 Fig ure 3. Pho to graph show ing small ar eas of float ing pum ice just NE of the Lau Group of is lands, Fiji, around 10 Sep tem ber 2006. Cour tesy of the En core II crew. Fig ure 5. View of a large pum ice raft af ter the pas sage of a sail ing ves sel near the Lau Group of is lands, Fiji, on an un known date in early to mid-Sep tem ber 2006. Cour tesy of the Stormsvalen crew via the En core II. Fig ure 4. Pho to graph show ing a large pum ice raft near the Lau Group of is lands, Fiji, on an un known date in early to mid-Sep tem ber 2006. Cour tesy of the Stormsvalen crew via the En core II. small peb bles to larger pieces the size of a base ball. Ev ery time a wave crashed on deck they heard the pum ice mak ing its way onto the boat and into the cock pit. On 30 Sep tem ber the Windbird log noted that ?. . . cruis ers are still hav ing to avoid the huge pum ice field that is float ing about between Tonga and Fiji.? Bob McDavitt?s ?Weathergram? for 15 Oc to ber noted that re ports from yachts sail ing be tween Tonga and Fiji in - di cated an ab sence of pum ice. These ob ser va tions sug gest that the bulk of ma te rial pro duced by the erup tion, or se ries of erup tions, had crossed to Fiji by mid-October. Pum ice rafts in north ern Fiji. The ear li est known di - rect ob ser va tions of the float ing pum ice in Fiji come from a boat with callsign KB1LSY, the crew of which noted that ?thick pum ice? slowed them to 2 knots for 30 min utes dur - ing the early morn ing hours of 28 Au gust. This oc curred as they ap proached the north ern is lands of the Lau Group in Fiji, about 500 km NW of Home Reef. Ac cord ing to Roberta Da vis, the pum ice ar rived at Taveuni, Fiji, on 14 Sep tem ber. There were sev eral rafts ~300 m from shore with other rafts scat tered far ther out. Lo cal mar i ners noted that pieces in the top layer were ap - prox i mately the size of pea gravel. Sus pended be low the sur face were pieces al most as large as foot balls. The beaches on the north ern shores of Taveuni were cov ered in what ap peared to be black pop corn. The pum ice was pres - ent at Taveuni for up to 6 days. On 19 Sep tem ber Da vid Forsythe re ported that large rafts of pum ice were pass ing through the north ern Lau group in Fiji (fig ure 6). He noted gooseneck bar na cles up to 10 mm long on the larg est pieces. Bul le tin ed i tors found com piled growth rates for var i ous stalked bar na cles ( Thiel and Gutow, 2005), which in di cated 17-29 days of growth. The En core II crew ob served pum ice along the S side of Vanua Levu, W of the Lau Group, around 16 Sep tem ber. They noted pum ice at Fawn Har bour that ob scured the chan nel into the har bor and it made a boat at an chor ap pear to be aground on an is land. They also ob served streams of pum ice near the Makogai Chan nel on 20 Sep tem ber. The Fiji Times On line re ported on 20 Sep tem ber that vil lag ers liv ing along the coastal ar eas of Saqani in Cakaudrove (Vanua Levu) were bat tling to clear their pum ice-cov ered sea shores and rivers. Vil lag ers saw the pum ice float ing in the sea near their homes on 18 Sep tem ber, and by the next day the pum ice cov ered the river and vil lag ers could not fish or travel by boats and bamboo rafts to their plantations. While div ing at the ?Bligh Tri an gle? of Fiji at sites NW of Viti Levu, the crew aboard the Nai?a en coun tered float - ing pum ice dur ing 20 Sep tem ber-7 Oc to ber. The pum ice was ?sur round ing the Nai?a and the skiffs with oc ca sional big car pets of float ing rock.? Ro man Leslie, an Aus tra lian vol ca nol o gist who was fish ing in Koro (Lomaiviti Group), also ob served the pum ice in late Sep tem ber. Sci en tists aboard the re search ves sel Yokosuka ob served pum ice set tling to the shore of Viti Levu on 6 Oc to ber. The rafts were in bands up to 70-80 m wide and sev eral hun dred me ters long. The pum ice frag ments were fully abraded, and dom i nantly less than 1 cm in di am e ter with oc ca sional large blocks up to 15-20 cm in di am e ter. The pum ice seemed to be quite phenocryst-rich. The sound of the mov ing, abrad - ing rafts was de scribed as ?siz zling.? Pum ice rafts in south ern Fiji. A bi ol o gist aboard the Na tional Geo graphic mo tor ves sel Endeavour re ported that on the morn ing of 16 Sep tem ber they ob served an ex ten sive re gion of float ing pum ice ?... in long, wind-driven rows, ap prox i mately 1-5 m wide and up to sev eral hun dred me ters long.? Pieces of pum ice av er aged 0.5-8 cm in lon gest di - men sion. The larg est piece ob served was ap prox i mately 15 cm in lon gest di men sion. The ob ser va tions con tin ued over the next 90 km, for 3.5 hours, with lit tle in ter rup tion, un til they made land fall at Vatoa Is land in the Lau Group. Mod - er ate wind rows of pum ice, up to sev eral inches deep, were ob served on the beaches of Vatoa. Roger Matthews ar rived in Kadavu, Fiji, on 30 Sep tem - ber and re ported that pum ice had been com ing ashore for about a week. On the south ern coast of the is land near the air port, the layer of pum ice on 30 Sep tem ber was 10-15 cm thick float ing on top of ~1 m of wa ter (fig ure 7). Far ther NE, pum ice that be gan com ing ashore at the Matava Re sort on 3 Oc to ber car ried goose bar na cle shells that mea sured about 2-3 mm on the big ger clasts. By 7 Oc to ber bar na cle size on ar riv ing pum ice had in creased to around 4-6 mm. While scuba div ing, Matthews noted neu trally buoy ant bits of pum ice, gen er ally in the 3-10 mm size range, down to at least 40 m wa ter depth. The pum ice did not ap pear to have an even size dis tri bu tion (fig ure 8). There were a num ber of big clasts, 2-3 cm, with a large amount of ma te rial in the 8-15 mm range. In the shore de pos its there ap peared to be a 4 Home Reef Smithsonian Institution ? Bulletin of the Global Fig ure 6. Pan oramic view of In digo Swan Beach filled with pumice, Naitauba Is land, Fiji, as seen in Sep tem ber 2006. Cour tesy of Da vid Forsythe. large vol ume of fines in the sub-2 mm size. The ma te rial was clean with no al gae, just the oc ca sional bar na cles. The clasts con tained pheno crysts up to 2 mm long. The raft drifted in and out de pend ing on wind con di tions, at times ex tend ing 75-100 m from shore, and in vaded streams at high tide. On shore there were 20-cm-thick de pos its, some of which was used as fill be hind the sea walls (fig ure 9). A 31 Oc to ber story in the Fiji Times de scribed trans por - ta tion dif fi cul ties be tween Daviqele Vil lage, on the W end of Kadavu, and other parts of the is land due to pum ice that a res i dent said had ?cov ered [Naluvea Bay] for over two months now.? Sim i lar prob lems were re ported by Adrian Watt at Matava Re sort on the S shore of Kadavu. In an email re layed by Roberta Da vis, Watt noted that by 2 No - vem ber the pum ice had mostly stopped com ing in, with ?... just a few strands of small pieces be ing blown along wind lines here and there.? The pieces were gen er ally 5-10 mm in di am e ter, but sev eral were big ger, and one was larger than 30 cm across. Large bays on Kadavu's SE side were pum ice choked, ham per ing boat travel, and clogged cool ing sys tems dam aged or de stroyed many out board en gines. Geo logic Sum mary. Home Reef, a sub ma rine vol cano mid way be tween Me tis Shoal and Late Is land in the cen tral Tonga is lands, was first re ported ac tive in the mid-19th cen tury, when an ephem eral is land formed. An erup tion in 1984 pro duced a 12-km-high erup tion plume, co pi ous amounts of float ing pum ice, and an ephem eral is land 500 x 1,500 m wide, with cliffs 30-50 m high that en closed a water-filled crater. Ref er ence: Thiel, M., and Gutow, L., 2005, The ecol - ogy of raft ing in the ma rine en vi ron ment. II. The raft ing or - gan isms and com mu nity: Ocean og ra phy and Ma rine Bi ol - ogy: An An nual Re view, 2005, v. 43, p. 279-418. In for ma tion Con tacts: Fredrik Fransson and H?kan Larsson, Yacht Maiken, 32 Mac ros san St., Unit 70, Bris - bane 4000, Aus tra lia (URL: http://yacht-maiken.blogspot. com/, Email: fredrikfransson@ya hoo.com); Paul and Nancy Horst, En core II (URL: http://www.encorevoyages. com/; Email: en core.crew@sbcglobal.net); KB1LSY Crew (URL: http://www.pangolin.co.nz/yotreps/tracker. php?ident=KB1LSY); Matangi Tonga On line, Vava?u Press Ltd., PO Box 958, Nuku?alofa, Tonga (URL: http:// www.matangitonga.to/, Email: mfonua@matangitonga.to); Roger Matthews, Pri vate Bag 93500, Takapuna, North Shore Ci ty 1332, New Zea land (Emai l: roger . matthews@northshorecity.govt.nz); Ken Tani, R/V Yokosuka (Email: kentani@jamstec.go.jp); Da vid Forsythe , Nai tauba Is land , Fi j i (Email : Da-vid_Forsythe@adidam.org); Da vid Cothran, 1211 Colestin Rd., Ashland, OR 97520, USA (Email: da - vid@gmail.com); Bob McDavitt?s Weathergram (URL: http://www.pangolin.co.nz/yotreps/list_man ager.php#Bob McDavitt?s Pa cific Weathergrams); Nick Sambrook, Tall Ship Soren Larsen, P.O.Box 60-660 Titirangi Auckland 0642, New Zea land (URL: http://www.sorenlarsen.co.nz/ 2006/V237_Tonga-Fiji/V237_Tonga-Fiji_Nick.htm, http:// www.sorenlarsen.co.nz/Voylog_Track.htm, Email: es - cape@sorenlarsen.co.nz); Windbird Crew (URL: http:// handleysai l .com/logs/?cat=1&paged=2); NAI?A Liveaboard Scuba Div ing, Lautoka, Fiji (URL: http://www. naia.com.fj/, Email: ex plore@naia.com.fj); Roberta Da vis, Makaira by the Sea, Taveuni, Fiji (URL: http://www. fijibeachfrontatmakaira.com, Email: makaira@con nect. com.fj); Adrian Watt, Matava Re sort, Kadavu, Fiji (URL: http://www.matava.com/, Email: matava@con nect.com.fj); Sara Berman and Jean Philippe Chabot (URL: http:// zayasail.blogspot.com/2006/09/east.html). Volcanism Network, Volume 31, Number 9, September 2006 Home Reef 5 Fig ure 8. A close up view of pum ice seen near Matava Re sort on the S shore of Kadavu, Fiji, 3 Oc to ber 2006. Cour tesy of Roger Matthews. Fig ure 7. Pum ice found float ing in North Bay along the south ern coast of Kadavu, Fiji, on 30 Sep tem ber 2006. Cour tesy of Roger Matthews. Fig ure 9. Pum ice de pos its seen at ebb tide near Matava Re sort on the S shore of Kadavu, Fiji, 8 Oc to ber 2006. Some of the pum ice has been used as fill be hind the sea wall. De pos its can be seen on the steps into the wa ter, and waves prop a gat ing through the pum ice could still break. Cour tesy of Roger Matthews. Rabaul New Brit ain, SW Pa cific 4.271?S, 152.203?E; sum mit elev. 688 m All times are lo cal (= UTC + 10 hours) A 7 Oc to ber Rabaul erup tion ob scured vis i bil ity in and around the cal dera, which sits at the NE end of New Brit ain Is land (fig ure 10). The erup tion took place at the intra-cal - dera cone Tavurvur, and emis sions in cluded lava flows. In - ter mit tent erup tions had oc curred at Tavurvur since 1994, the last of which took place on 15 Jan u ary 2006 (BGVN 31:02). Pho tos by pi lots shortly af ter the erup tion doc u - mented a dra matic um brella-shaped plume, which rose to the tropo pause and cre ated an SO2 cloud that later di vided into two parts, one mov ing NW, the other SE. Rabaul Vol cano Ob ser va tory (RVO) ob ser va tions. The RVO an nounced that a sus tained erup tion from Tavurvur did not ap pear to have been any im me di ate pre cur sors apart from a small de fla tion. The sub-Plinian erup tion be gan at about 0845 on 7 Oc to ber 2006 and con tin ued into the early af ter noon. Semi-con tin u ous to rhyth mic air blasts were ob - vi ous in Rabaul town, with doors slam ming and win dows rat tling. Rabaul re ceived mod er ately heavy ashfall; heavy lapilli of ~ 1 mm di am e ter fell, and a few lithics up to 3 cm across fell around the S and SW parts of the cal dera. Ac - cord ing to Herman Patia at RVO, a small pum ice raft ac cu - mu lated in Greet Har bor and pum ice was still drift ing about sev eral weeks later. Ashfall af fected the whole of the Ga zelle pen in sula (the name given to the bul bous, 50-km-di am e ter NE end of New Brit ain is land). About 1 cm of ash was de pos ited on the SW side of the cal dera in the Blue La goon-Vul can sec tor. Ashfall oc curred ~ 7 km SE of Rabaul cal dera?s cen ter point in Kokopo and -20 km S of the cen ter point in Warangoi. The den sity of ashfall was such that Tavurvur was ob scured from all di rec tions. In the town of Rabaul the ex pe ri ence was very sim i lar to the Oc to ber 1996 and Jan u - ary 1997 Strombolian eruptions. At 1200 on 7 Oc to ber 2006 the RSAM was about 1900 units and its rate ap peared to be de creas ing. (The Real-time Seismic Amplitude is an of ten-used tool to sum ma rize seis - mic ac tiv ity dur ing vol ca nic cri ses by pre sent ing a mea sure of the av er age am pli tude of ground shak ing over suc ces sive 10-min intervals.) Thick ash clouds rose to a height of about 18 km. The cloud sub se quently dis persed over a broad west ern swath (N to W to S). The na ture of the erup tion changed to Strombolian at 1415 hours, with ac tiv ity char ac ter ized by fre quent ex plo - sions ac com pa nied by shock waves. At 1730 hours, the Strombolian ac tiv ity be gan to sub side. A mod er ate to bright glow was vis i ble dur ing the eve ning of 7 Oc to ber on Tavurvur?s N rim, ac com pa nied by oc ca sional ex plo sions and loud roar ing noises throughout the night. In the morn ing of 8 Oc to ber, thick white and blue va por ac com pa nied oc ca sional ash ex plo sions drift ed N and NW of Tavurvur. In spec tion from Rapindik (2 km NNW from Tavurvur) re vealed lava flows emplaced down the cone?s W and N flanks. The W flank flow went into the har bor and caused small sec ond ary ex plo sions; vis i bil ity of the N flank was poor due to the white va por emis sion. The RSAM level de creased to the back ground value of ~ 70 units. Herman Patia re ported that by 28 Oc to ber 2006 the erup tion had qui eted down with only oc ca sional ash emis - sion ac com pa nied by rare ex plo sions. Seis mic ac tiv ity was at a low level and ground de for ma tion was at a low rate. On 30 Oc to ber mild erup tive ac tiv ity con tin ued at Tavurvur. The ac tiv ity con sisted of con tin u ous emis sion of thick pale to dark gray ash clouds that drifted N to NW of the vol cano. Fine ash fall oc curred in the NE cal dera at Namanula, and also in sur round ing ar eas down wind and on the E side of Rabaul Town. There were no au di ble noises and no glow vis i ble. The low-level erup tive ac tiv ity con sisted of oc ca - 6 Rabaul Smithsonian Institution ? Bulletin of the Global Fig ure 10. (Top) In dex maps in di cat ing the lo ca tion and ge og ra phy around Rabaul cal dera. (Bot tom) A map of Rabaul de rived from work by Al mond and McKee and pre pared by Lyn Topinka (US Geo log i cal Sur vey). For other maps see pre vi ous Bul le tin re ports on Rabaul (most re cently, BGVN 28:01). sional ash emis sions sim i lar to those that have occurred regularly since 1994. Pi lot ob ser va tions. Fig ures 11 and 12 are pi lot?s pho to - graphs pro vided by Tony Gridley, Air Niugini, in di cat ing the well-de vel oped ash clouds vis i ble 1-2 hours af ter the erup tion. The pho tos are rem i nis cent of the 20 Sep tem ber 1994 photo of the erup tion cloud taken from the or bit ing Space Shut tle, an oblique, down ward-look ing per spec tive from the NE about 24 hours af ter the start of that erup tion (BGVN 19:08). Sat el lite ob ser va tions. Ac cord ing to An drew Tupper, the 7 Oc to ber erup tion was clearly vis i ble on in fra red and vis i ble im ag ery (to around tropo pause al ti tudes). Fig ure 13 shows the ash cloud im aged from the MODIS sat el lite on 7 Oc to ber 2006. Fig ure 14 de picts the sul fur di ox ide (SO2) in Dobson Units (DU) from the Ozone Mon i tor ing In stru ment (OMI) for 7-9 Oc to ber 2006. Fur ther de tails ap pear in the fig ure cap tion. The SO2 con cen tra tion-pathlengths on the fig ure are shown us ing the log a rith mic scale of Dobson Units. (As one ex pla na tion of this unit, if all SO2 in the air col umn the sat el lite ob served was flat tened into a thin layer at the sur face of the Earth at a tem per a ture of 0? C, then 1 Dobson Unit would make a layer of pure SO2 0.01 mm thick.) Based on in for ma tion from the RVO, the Dar win VAAC re ported that a brief erup tion of Rabaul on 11 Oc to - ber pro duced a plume that reached an al ti tude of 7.6 km al - ti tude and dis si pated NW. Con tin u ous low-level emis sions and vulcanian erup tions pro duced plumes to 1 km al ti tude during 12-17 October. Mod er ate Res o lu tion Infrared Spectroradiometry (MODIS) ther mal anom a lies. Ta ble 1 shows the ther mal anom a lies as mea sured from the MODIS sat el lite dur ing the erup tion pe riod. Note that there were no anom a lies for sev - eral months be fore this pe riod. The anom a lies are in har - mony with the observed lava flows. News re leases. Ac cord ing to Reuters news ser vice the 7 Oc to ber blast shat tered win dows up to 12 km from the cal - dera. In 1994, a large erup tion at Tavurvur and the nearby Vul can peak de stroyed much of Rabaul, cov er ing the air - port and much of the town with ash, and forc ing the con - struc tion of a new cap i tal, Kokopo, 20 km away. Ash was fall ing on Kokopo, caus ing power and phone cuts. There were no re ports of death or in ju ries. In ad di tion Reuters Volcanism Network, Volume 31, Number 9, September 2006 Rabaul 7 Fig ure 11. Ae rial photo taken 1 or 2 hours af ter the erup tion of 7 Oc to ber 2006 at ~ 3.7 km (~ 12,000 ft) and ~ 90 km (~ 50 nau ti cal miles) from Tokua air port (Rabaul?s new air port, on the S side of the cal dera) while fly ing at a head ing of about 060? (i.e. look ing ENE). The flight was ?on the Hoskins-Tokua track.? Cour tesy of Tony Gridley, Air Niugini. Fig ure 12. Ae rial photo taken 1 or 2 hours af ter the erup tion of 7 Oc to ber 2006 at ~ 3.7 km (~ 12,000 ft) and ~ 90 km from Tokua air port, head ing about 060?. Cour tesy of Tony Gridley, Air Niugini. Fig ure 13. True-color (above) and false-color (be low) im ages of a Rabaul erup tion cloud cre ated by the Mod er ate Res o lu tion Im ag ing Spectroradiometer (MODIS) on NASA?s Aqua sat el lite, 7 Oc to ber 2006. Vol ca nic emis sions block the view of most of the is land but Rabaul?s ap prox i mate lo ca tion is at the solid tri an gle. The brown or tan plume in the E clearly bears vol ca nic ash. The bright ?cloud? to the im me di ate left of the brown ash rep re sents a por tion of the vol ca nic ash plume that reached a high enough al ti tude for the wa ter con tent of that plume to turn to ice crys tals that ?white out? the ash con tent that would oth er wise ap pear tan or brown. Cour tesy of the NASA Earth Ob ser va tory web site. noted that ?Rabaul Cham ber of Com merce Pres i dent and ho te lier Bruce Al ex an der told Aus tra lian As so ci ated Press that around 2,000 peo ple?or 90 per cent of the lo cal pop u - la tion?had fled the town as Mt. Tavurvur erupted. All flights into Tokua air port across the har bor from Rabaul had been can celed due to ash falls.? Ac cord ing to The Syd ney Morn ing Her ald, with 90% of the res i dents ab sent and only es sen tial per son nel in Rabaul, lo cal of fi cials feared loot ers. Ac cord ingly, ex tra po lice were called in, and armed po lice pa trols were stepped up. Geo logic Sum mary. The low-ly ing Rabaul cal dera on the tip of the Ga zelle Pen in sula at the NE end of New Brit - ain forms a broad shel tered har bor uti lized by what was the is land?s larg est city prior to a ma jor erup tion in 1994. The outer flanks of the 688-m-high asym met ri cal pyroclastic shield vol cano are formed by thick pyroclastic-flow de pos - its. The 8 x 14 km cal dera is widely breached on the east, where its floor is flooded by Blanche Bay and was formed about 1,400 years ago. An ear lier cal dera-form ing erup tion about 7,100 years ago is now con sid ered to have orig i nated from Tavui cal dera, off shore to the north. Three small stratovolcanoes lie out side the north ern and NE cal dera r ims of Rabaul . Post-cal dera erup t ions bui lt ba - saltic-to-dacitic pyroclastic cones on the cal dera floor near the NE and west ern cal dera walls. Sev eral of these, in clud - ing Vul can cone, which was formed dur ing a large erup tion in 1878, have pro duced ma jor ex plo sive ac tiv ity dur ing his - tor i cal time. A pow er ful ex plo sive erup tion in 1994 oc - curred si mul ta neously from Vul can and Tavurvur volcanoes and forced the temporary abandonment of Rabaul city. In for ma tion Con tacts: Steve Saunders and Herman Patia, Rabaul Volcanological Ob ser va tory (RVO), De part - ment of Min ing, Pri vate Mail Bag, Port Moresby Post Of - fice, Na tional Capitol Dis trict, Pa pua, New Guinea (Email: hguria@global.net.pg); An drews Tupper, Dar win Vol ca nic Ash Ad vi sory Cen tre (VAAC), Bu reau of Me te o rol ogy, Dar win, Aus tra lia (Email: A.Tupper@bom.gov.au); Pe ter Webley, ARSC/UAF, 909 Koyukuk Drive, Fair banks, Alaska (Email: pwebley@gi.alaska.edu); Si mon Carn, Joint Cen ter for Earth Sys tems Tech nol ogy (JCET), Uni ver sity of Mary land Bal ti more County (UMBC), 1000 Hill top Cir - cle, Bal ti more, MD 21250, USA (Email: scarn@umbc. edu); Na tional Aero nau tics and Space Ad min is tra tion Earth Ob ser va tory (URL: http://earthobservatory.nasa.gov/ NaturalHazards); RSAM def i ni tion (URL: http://vul can.wr. u s g s . g o v /M o n i t o r in g / D e s c r ip t io n s / d e s c r ip t io n _RSAM_SSAM.html); HIGP MODIS Ther mal Alert Sys - tem, Hawai'i In sti tute of Geo phys ics and Plan e tol ogy (HIGP), Uni ver sity of Ha waii at Manoa, 168 East-West Road, Post 602, Ho no lulu, HI 96822, USA (URL: http:// modis.higp.ha waii.edu/). 8 Rabaul Smithsonian Institution ? Bulletin of the Global Fig ure 14. The Rabaul erup tion in jected SO2 into the at mo sphere and mea sure ments from sat el lite spec trom e ters led to cre ation of this se ries of im ages map ping the SO2 con cen tra tions over the re gion dur ing 7-9 Oc to ber 2006. Data are from the Ozone Mon i tor ing In stru ment on NASA?s Aura sat el lite. On 7 Oc to ber, high SO2 con cen tra tions lin gered over New Brit ain. By 8 Oc to ber, the orig i nal plume had split into two clouds, one spread ing NW, the other, SE. On 9 Oc to ber, the SO2 had dif fused more, but a core of el e vated con cen tra tion-pathlength val ues re mained in the north ern plume. Cour tesy of NASA Earth Ob ser va tory and Si mon Carn, Uni ver sity of Mary land Bal ti more County. Date Time (UTC) Num ber of Pix els Sat el lite (A=aqua, T=Terra) 07 Oct 2006 1140 4 T 08 Oct 2006 0000 2 T 08 Oct 2006 1220 6 T 08 Oct 2006 1520 4 A 10 Oct 2006 1210 2 T 11 Oct 2006 0035 1 T 11 Oct 2006 1250 1 T 15 Oct 2006 1230 1 T 15 Oct 2006 1525 3 A 17 Oct 2006 1215 1 T 22 Oct 2006 0015 2 T 22 Oct 2006 1535 1 A 24 Oct 2006 1220 2 T Ta ble 1. MODIS ther mal anom a lies for Rabaul vol cano for 7-17 Oc to ber 2006. Cour tesy of Hawai?i In sti tute of Geo phys ics and Plan e tol ogy. Bamus New Brit ain, SW Pa cific 5.20?S, 151.23?E; sum mit elev. 2,248 m All times are lo cal (= UTC +10 hours) Ac cord ing to the Pa pua New Guinea De part ment of Min ing (DOM), re ports com ing from Bialla Lo cal Level Gov ern ment (LLG) in di cated that Bamus showed signs of un usual ac tiv ity. At 1010 on 12 July 2006 ob serv ers saw white va por com ing out at the sum mit. The emis sion was force ful at about 1110 that day, with a tint of gray color in the emis sion. The va por-rich plume blew in land to the SSE. No ashfall was re ported. Of fi cials from Bialla LLG to gether with a DOM ob - server wit nessed the ac tiv ity, as did Max Benjamin from Walindi Re sort (~ 40-50 km away). Benjamin called the Rabaul Vol cano Ob ser va tory to re port the ac tiv ity. No sat - el lite-de tected ther mal anom a lies at the vol cano were re - ported by the MODIS website for this time frame. Geo logic Sum mary. Sym met ri cal 2,248-m-high Bamus vol cano, also re ferred to lo cally as South Son, is lo cated SW of Ulawun vol cano, known as the North Son. These two vol ca noes are the high est in the 1,000-km-long Bis - marck vol ca nic arc. The andesitic Bamus stratovolcano is draped by rain forest and con tains a breached sum mit crater filled with a lava dome. A satellitic cone is lo cated on the south ern flank, and a prom i nent 1.5-km-wide crater with two small ad ja cent cones is sit u ated half way up the SE flank. Young pyroclastic-flow de pos its are found on the vol cano?s flanks, and vil lag ers de scribe an erup tion that took place during the late-19th century. In for ma tion Con tacts: Rabaul Vol cano Ob ser va tory (see Rabaul). Sulu Range New Brit ain, SW Pa cific 5.50?S, 150.942?E; sum mit elev. 610 m On 31 Oc to ber 2006 the Rabaul Volcanological Ob ser - va tory (RVO) is sued a followup re port to the erup tive ac - tiv ity in the Sulu Range through much of Oc to ber. Sulu Range was pre vi ously dis cussed in BGVN 31:07, but that re port was am big u ous on the na ture of the ac tiv ity that had taken place dur - ing July 2006. This re port and per sonal com mu ni ca tions es tab - lishes that RVO staff are doubt ful that the most en er getic events were mag matic in char ac ter. Fur - ther more, RVO re ported that in the weeks that fol lowed, seis mic - ity continued to decline. The seis mic un rest that be gan on 6 July de clined from over 2,000 daily vol cano-tec tonic (VT) events to be low 50 dai ly VT events dur ing Oc to ber (fig ure 15). The num ber fluc tu ated be tween 35 and 50 from late Sep - tem ber to early Oc to ber and be tween 5 and 25 dur ing the third week of October. RVO noted that about two to three felt earth quakes with in ten sity 2 con tin ued to be felt daily at ir reg u lar in ter vals within the Bialla area and that white steam emis sions from the Silanga Hot Springs were still vis i ble from Bialla. In ad - di tion, a mod er ately strong sul fur smell from the Silanga and Talopu hot springs con tin ued to be reported. An anal y sis by RVO sci en tists con cluded that at no point did magma reach the sur face. The de clin ing trend in seis mic ac tiv ity from early to late Oc to ber may in di cate that the new magma that ap par ently in truded to shal low levels in July is be gin ning to stall. A per ma nent seis mic sta tion will be in stalled at Kaiamu in De cem ber 2006 to pro vide con tin u ous mon i tor ing of ac - tiv ity from the Sulu Range and sur round ing ar eas. In an ex ten sion of el e vated re gional tec tonic seis mic ity, a strong earth quake, M ~ 6.5, struck the S side of cen tral New Brit ain on 17 Oc to ber. The USGS com puted the fo cal depth as ~ 60 km, with epi cen ter ~ 50 km S of the Sulu Range. Ac cord ing to a USGS ma chine-gen er ated shak ing and in ten sity map, the Sulu Range lies within the zone of high est com puted in ten sity (VI). Geo logic Sum mary. The Sulu Range con sists of a group of par tially over lap ping small stratovolcanoes in north-cen tral New Brit ain off Bangula Bay. The 610-m Mount Malopu forms the high point of the ba - saltic-to-rhyolitic com plex at its SW end. Lava Point (also known as Lara Point) forms a pen in sula of volcaniclastic-cov ered lava flows with a small lake ex tend - ing about 1 km into Bangula Bay at the NW side of the Sulu Range. The Walo hy dro ther mal area, con sist ing of sol fa ta - ras and mud pots, lies on the coastal plain west of the SW base of the Sulu Range. Prior to 2006, no his tor i cal erup - tions had oc curred from the Sulu Range, al though some of the cones dis play a rel a tively undissected morphology. In for ma tion Con tacts: Rabaul Vol cano Ob ser va tory (see Rabaul); USGS Earth quakes Haz ard Pro gram (URL: http://earth quakes.usgs.gov/) Volcanism Network, Volume 31, Number 9, September 2006 Sulu Range 9 Fig ure 15. Sulu Range seis mic ity plot of daily VT earth quakes from 22 July 2006 to 24 Oc to ber 2006 at the Kaiamu Seis mic Sta tion. The sta tion did not op er ate on the days that lack earth quakes. Cour tesy of RVO. Bar ren Is land Andaman Is lands, In dian Ocean 12.278?N, 93.858?E; sum mit elev. 354 m Our last re port on Bar ren Is land dis cussed events through much of Jan u ary 2006 (BGVN 31:01); since that time we have only found spo radic re ports of ac tiv ity. Ac cord ing to a news ar ti cle by The Indo-Asian News Ser vice, a team of sci en tists that vis ited Bar ren Is land around 12 March 2006 found that the vol cano was still very ac tive. The height of the vol ca nic cone had in creased by 50 m since erup tive ac tiv ity be gan in May 2005. In ad di tion, lava flows cov ered the NW side of the island. Since March 2006 there have been only a few sat el lite im ages and pi lot re ports of con tin ued ac tiv ity. Based on a pi lot re port and sat el lite im ag ery, the Dar win VAAC re - ported that an ash plume was emit ted dur ing 5-6 April that did not rise higher than 4.6 km al ti tude. On 19 April a low-level plume ex tend ing W was visible on satellite imagery. On 2 May sat el lite im ag ery de tected a plume from Bar - ren Is land near 3.7 km al ti tude. The fol low ing day low-level ash plumes ex tended N. Based on a pi lot re port, the Dar win VAAC re ported an ash plume at 1230 on 26 May that re mained be low 3 km al ti tude and drifted N. On 23 Sep tem ber a news re port in The Hindu stated that In dian Coast Guard of fi cials in di cated that the con tin u ing erup tion at Bar ren Is land was de creas ing in in ten sity. The news piece cited a sur veil lance re port state ment that there was less lava but more ?smoke? from the volcano. Geo logic Sum mary. Bar ren Is land, a pos ses sion of In - dia in the Andaman Sea about 135 km NE of Port Blair in the Andaman Is lands, is the only his tor i cally ac tive vol cano along the N-S-trending vol ca nic arc ex tend ing be tween Su - ma tra and Burma (Myanmar). The 354-m-high is land is the emer gent sum mit of a vol cano that rises from a depth of about 2,250 m. The small, un in hab ited 3-km-wide is land con tains a roughly 2-km-wide cal dera with walls 250-350 m high. The cal dera, which is open to the sea on the W, was cre ated dur ing a ma jor ex plo sive erup tion in the late Pleis - to cene that pro duced pyroclastic-flow and -surge de pos its. The mor phol ogy of a fresh pyroclastic cone that was con - structed in the cen ter of the cal dera has var ied dur ing the course of his tor i cal erup tions. Lava flows fill much of the cal dera floor and have reached the sea along the west ern coast dur ing his tor i cal erup tions. In for ma tion Con tacts: The Hindu (URL: http://www. hinduonline.com); Indo-Asian News Ser vice (IANS) (URL: http://www.eians.com/); Geo log i cal Sur vey of In dia, 27 Jawaharlal Nehru road, Kolkata 700 016, In dia (URL: http:/ /www.gsi.gov.in/bar ren.htm); In dian Coast Guard, Na - tional Sta dium Com plex, New Delhi 110 001, In dia (URL: http://indiancoastguard.nic.in/indiancoastguard/); Dar win Vol ca nic Ash Ad vi sory Cen ter, Bu reau of Me te o rol ogy, North ern Ter ri tory Re gional Of fice, PO Box 40050, Ca sua - rina, North ern Ter ri tory 0811, Aus tra lia (URL: http://www. bom.gov.au/info/vaac/). Bulusan Luzon, Phil ip pines 12.770?N, 124.05?E; sum mit elev. 1,565 m All times are lo cal (= UTC + 8 hours) On 19 March 2006, the Phil ip pine In sti tute of Vol ca nol - ogy and Seis mol ogy (PHIVOLCS) raised the sta tus of Bulusan from Zero Alert (no alert) to Alert Level 1 to re - flect el e vated seis mic, fumarolic, and other un rest (BGVN 31:05). From that date un til an ash ex plo sion on 28 June 2006, 10 ex plo sions were re corded (see table 2). Af ter the ash ex plo sion of 28 June 2006, Bulusan?s mon i tored pa ram e ters grad u ally de creased to near base line lev els. The daily count of vol ca nic earth quakes was very low, and SO2 emis sion rates and ground-de for ma tion data 10 Bar ren Is land Smithsonian Institution ? Bulletin of the Global Date (2006) Lo cal time Plume Al ti tude (km) Drift Di rec tion Com ments 19 Mar ? ? ? Seis mic swarm which lasted un til 21 Mar; Alert Level raised to 1 21 Mar 2258 1.5 N, W, SW 1st ex plo sion-type (E-type) earth quake lasted 20 min; to tal of 4 E-type earth quakes re corded 08 Apr 2000 ? ? Lahar at Cogon spill way 09 Apr 1036-1058 ? ? Lahar at Cogon spill way 29 Apr 1044 1.5 WSW. NW 2nd E-type earth quake; to tal of 3 E-type earth quakes re corded 25 May 2117-2130 cloud-cov ered sum mit ? 3rd E-type earth quake; ash de pos its, trace to 2 mm thick in Juban, Irosin 31 May 1617 1.5 W, WNW 4th E-type earth quake 07 Jun 2017-2030 2.0 N, W, SW 5th E-type earth quake; smaller E-type earth quake at 0225 on 8 Jun; Alert Level raised to 2 10 Jun 1218 1.0 NE, E 6th E-type earth quake, last ing 25 min 13 Jun 1904 1.5 NW 7th E-type earth quake, last ing 13 min 18 Jun 1556 1.5 W 8th E-type earth quake 20 Jun 2013 cloud-cov ered sum mit ? 9th E-type earth quake ? mild; event not ob served; seis mic sig nal re corded for 17 min; rains gen er ated some lahars 24 Jun 2300 ? ? Lahar at Cogon spill way 28 Jun 0206 cloud-cov ered sum mit ? 10th E-type earth quake; the as so ci ated vol ca nic event was not ob served but seis mic sig nal re corded as E-type earth quake lasted 4 min 29 Jun 0800 ? ? Con tin u ous de cline in Bulusan ac tiv ity; Alert Level low ered to 1 Ta ble 2. Sum mary of sig nif i cant events through late July 2006 at Bulusan . Num ber ing of ex plo sion-type (E-type) quakes be gan 21 March 2006. Cour tesy of Phil ip pine In sti tute of Vol ca nol ogy and Seis mol ogy (PHIVOLCS). re vealed the vol cano?s de flated con di tion, in di cat ing the ab sence of ac tive magma as cent. Ash emis sion stopped and steam ing from the ac tive vents and fis sures grad u ally re - turned to nor mal lev els. Due to the de cline in ac tiv ity, on 29 July PHIVOLCS low ered the status of Bulusan from Alert Level 2 to 1. On 10 Oc to ber 2006 at 1256 UTC, the To kyo Vol ca nic Ash Ad vi sory Cen ter an nounced that an erup tion plume from Bulusan was vis i ble on sat el lite im ag ery reach ing al ti - tudes of 3 km and drift ing SW and SSE. Un like nearby Mayon vol cano (~ 70 km NW) (see BGVN 31:08), no ther mal anom a lies were de tected at Bulusan by sat el lite or re corded by the Hawai?i In sti tute of Geo phys ics and Plan e tol ogy (HIGP) MODIS/ MODVOLC web site from the be gin ning of 2006 to 10 October 2006. Geo logic Sum mary. Luzon?s south ern most vol cano, Bulusan, was con structed along the rim of the 11-km-di am - e ter dacitic-to-rhyolitic Irosin cal dera, which was formed about 35,000-40,000 years ago. Bulusan lies at the SE end of the Bicol vol ca nic arc oc cu py ing the pen in sula of the same name that forms the elon gated SE tip of Luzon. A broad, flat moat is lo cated be low the top o graph i cally prom - i nent SW rim of Irosin cal dera; the NE rim is bur ied by the andesitic Bulusan com plex. Bulusan is flanked by sev eral other large intracaldera lava domes and cones, in clud ing the prom i nent Mount Jormajan lava dome on the SW flank and Sharp Peak to the NE. The sum mit of 1,565-m-high Bulusan vol cano is unvegeta ted and con ta ins a 300-m-wide, 50-m-deep crater. Three small crat ers are lo - cated on the SE flank. Many mod er ate ex plo sive erup tions have been re corded at Bulusan since the mid-19th century. In for ma tion Con tacts: Phil ip pine In sti tute of Vol ca nol - ogy and Seis mol ogy (PHIVOLCS), Uni ver sity of the Phil ip - pines Cam pus, Diliman, Quezon City, Phil ip pines (URL: http://www.phivolcs.dost.gov.ph); To kyo Vol ca nic Ash Ad - vi sory Cen ter (VAAC) (URL: http://www.jma.go.jp/jma/ jma-eng/jma-cen ter/vaac/in dex/html); HIGP MODIS Ther - mal Alert Sys tem, Hawai'i In sti tute of Geo phys ics and Plan - e tol ogy (HIGP), Uni ver sity of Ha waii at Manoa, 168 East-West Road, Post 602, Ho no lulu, HI 96822, USA (URL: http://modis.higp.ha waii.edu/). Cleve land Aleu tian Is lands, USA 52.825?N, 169.944?W; sum mit elev. 1,730 m All times are lo cal (= UTC - 10 hours) Cleve land?s com monly ob served ac tiv ity con sist ing of short du ra tion ex plo sions, such as those seen ear lier in the year on 6 Feb ru ary 2006 (BGVN 31:01) and on 23 May 2006 (BGVN 31:07), con tin ued dur ing Au gust and Oc to ber 2006. This re port will cover the 24 Au gust and 28 October eruptions. At 1955 on 24 Au gust a brief erup tion was seen by mar - i ners on a pass ing ship. The erup tion was un con firmed by sat el lite data. Video foot age sent to the Alaska Vol cano Ob - ser va tory (AVO) on 28 Au gust showed that an ash cloud rose to an ap prox i mate al ti tude of 3 km and pro duced mi nor ashfall. Shortly af ter the erup tion, mi nor steam ing was ob - served from the vent on ad di tional foot age. In re sponse to the erup tion, the AVO raised the level of Con cern Color Code from ?un as signed? to ?Yel low? on 7 Sep tem ber. A weak ther mal anom aly in the sum mit crater was present in subsequent satellite images. Clouds ob structed vis i bil ity through most of Sep tem ber and Oc to ber. A pi lot re ported that a mi nor erup tion started at 1345 on 28 Oc to ber. Sat el lite data con firmed the pres ence of an ash cloud drift ing ENE of the vol cano. The height of the cloud was es ti mated at an al ti tude of 6 km us ing the sat el lite im - ag ery. One pi lot re ported the plume top at an al ti tude of 9 km. The AVO raised the alert level to ?Or ange? dur ing 28-29 Oc to ber. On 30 Oc to ber the AVO low ered the level to ?Yel low? be cause of no further evidence of activity. Geo logic Sum mary. Beau ti fully sym met ri cal Mount Cleve land stratovolcano is sit u ated at the west ern end of the un in hab ited, dumb bell-shaped Chuginadak Is land. It lies SE across Carlisle Pass strait from Carlisle vol cano and NE across Chuginadak Pass strait from Her bert vol cano. Cleve - land is joined to the rest of Chuginadak Is land by a low isth mus. The 1730-m-high Mount Cleve land is the high est of the Is lands of the Four Moun tains group and is one of the most ac tive of the Aleu tian Is lands. The na tive name for Mount Cleve land, Chuginadak, re fers to the Aleut god dess of fire, who was thought to re side on the vol cano. Nu mer - ous large lava flows de scend the steep-sided flanks of the vol cano. It is pos si ble that some 18th-to-19th cen tury erup - tions at trib uted to Carlisle should be as cribed to Cleve land (Miller et al., 1998). In 1944 Cleve land pro duced the only known fa tal ity from an Aleu tian erup tion. Re cent erup tions from Mount Cleve land have been char ac ter ized by short-lived ex plo sive ash emis sions, at times ac com pa nied by lava fountaining and lava flows down the flanks. In for ma tion Con tacts: Alaska Vol cano Ob ser va tory (AVO), a co op er a tive pro gram of the U.S. Geo log i cal Sur - vey, 4200 Uni ver sity Drive, An chor age, AK 99508-4667, USA; Geo phys i cal In sti tute, Uni ver sity of Alaska, P.O. Box 757320, Fair banks, AK 99775-7320, USA; and Alaska Di vi sion of Geo log i cal & Geo phys i cal Sur veys, 794 Uni - ver sity Ave., Suite 200, Fair banks, AK 99709, USA (URL: http://www.avo.alaska.edu/). Fourpeaked Alaska Pen in sula, USA 58.770?N, 153.672?W; sum mit elev. 2,105 m All times are lo cal (= UTC - 9 hours [or 8 hours early April-late Oc to ber]) Un til the erup tion of Fourpeaked on 17 Sep tem ber, ev i - dence for erup tive ac tiv ity in the past 10,000 years was uncertain. The vol cano is largely gla cier cov ered with only iso lated out crops (fig ure 16). This re port dis cusses the ini - tial ob ser va tion of plumes and sub se quent ac tiv ity un til the end of Oc to ber 2006. Fourpeaked is in S Alaska ~ 320 km SW of An chor age. It is SW of the mouth of Cook In let and within NE Katmai National Park (figure 17). On the eve ning of 17 Sep tem ber, AVO re ceived sev eral re ports of two dis crete plumes ris ing from the Cape Douglas area. The plumes were pho to graphed at an un - stated time on 17 Sep tem ber from the town of Homer (fig - ure 18). At this stage, nei ther Douglas nor Fourpeaked had de voted seis mic instruments. Volcanism Network, Volume 31, Number 9, September 2006 Fourpeaked 11 Ret ro spec tive anal y sis of data from the NEXRAD Dopp ler ra dar in King Salmon showed an un usual cloud start ing at 1200 on 17 Sep tem ber. The max i mum cloud height de ter mined by ra dar dur ing the first hour of the event was 6 km al ti tude. The ra dar re turn from the cloud con tin - ued un til at least 2145 (fig ure 19). A cloud of sul fur di ox ide gas was ob served by col - leagues at the Vol ca nic Emis sions Group at the Uni ver sity of Mary land Bal ti more. They used data col lected at 1500 by the Ozone Mon i tor ing In stru ment (OMI) on NASA?s Aura sat el lite (figure 20). On the ba sis of the suite of vi - sual, ra dar, and sat el lite ob ser va - tions, all the 17 Sep tem ber clouds were in ferred vol ca nic in or i gin. Al though sat el lite data did not de - tect ash dur ing this event, AVO re ceived re ports of a trace of ashfall at Nonvianuk Lake out let (110 km WNW) and near Homer (150 km NE). Field ob serv ers saw deep scour ing of a gla cier flow ing W from the sum mit, in di cat ing flood ing, prob a bly from the 17 September event. In the cap tion to a 20 Sep tem - ber AVO photo by K.L. Wallace there was noted a "con tin u ous layer of dis col ored snow and ice above [~1 km el e va tion,]~3,000 feet as l on the NE f lank of Fourpeaked vol cano (S of Douglas vol cano). Could pos si bly be ash from the 9/17/06 event." Both fixed-wing and he li cop - ter over f l igh ts in the Cape Douglas area on 20 Sep tem ber con firmed the source of vol ca nic ac tiv ity to be Fourpeaked vol - cano. AVO raised the Level of Con cern Color Code from ?Not As signed? to YEL LOW on 20 September. A 23 Sep tem ber ob ser va tion flight con ducted in rel a - tively good weather per mit ted the first look at the sum mit since the event of 17 Sep tem ber. Ob serv ers saw a lin ear se - ries of vents run ning N from the sum mit for about 1 km. Most of these vents vig or ously emit ted steam and other vol - ca nic gases. Gas mea sure ments in di cated abun dant quan ti - ties of sul fur di ox ide, hy dro gen sul fide, and car bon di ox ide. Ther mal mea sure ments of up to 75?C were re corded at the vents, al though steam was likely ob scur ing hot ter ar eas. Ad ja cent gla cial ice had been dis rupted and showed signs of sub si dence. Air borne gas mea sure ments taken on 23, 24, and 30 Sep tem ber again doc u mented high emis sion rates of 12 Fourpeaked Smithsonian Institution ? Bulletin of the Global Fig ure 16. Fourpeaked vol cano, the gla cier-cov ered peak at the up per left is one of a group of poorly known vol ca noes NE of Katmai Na tional Park. In the fore ground of this photo is Kaguyak cal dera, which hosts a 2.5-km- wide lake. Pre-erup tion photo at un cer tain date taken by Chris Nye (Alaska Di vi sion of Geo log i cal and Geo phys i cal Sur veys, Alaska Vol cano Ob ser va tory. Fig ure 17. A map show ing the lo ca tion of Fourpeaked and Douglas vol ca noes, Cook In let, and ad ja cent set tle ments in clud ing the city of Homer on the SW Kenai Peninsula. Cre ated by Seth Snedigar and Janet Schaafer, AVO-ADGGS. Fig ure 18. A pho to graph of the erup tion of Fourpeaked on 17 Sep tem ber 2006. The photo was taken from Main Street in Homer at an un stated time. Copy righted pho to graph by Lanny Simpson, Alaska High Moun tain Im ages (shown on AVO?s website). sul fur di ox ide, hy dro gen sul fide, and car bon di ox ide, and a dis tinct sul fur smell was ev i dent up to 50 km from the sum - mit. An AVO sta tus re port on 3 Oc to ber noted that cloudy con di tions had pre vented vi sual or sat el li te ob ser va tions, but l imited seismic data being received did not indicate significant volcanic activity. The AVO re ported that vol ca - nic un res t con t in ued a t Fourpeaked dur ing 30 Sep tem - ber-24 Oc to ber. A seis mom e ter in stalled on 25 Sep tem ber in di - cated on go ing low-level seis mic - ity. Due to the lim ited num ber of seis mom e ters, earth quake epi cen - ters were not lo cated. Emis sion rates of sul fur di ox ide were high dur ing 4-10 Oc to ber and on 27 Oc to ber. Ob ser va tions were hin - dered due to cloud cover, but on 12 Oc to ber AVO staff re ported that two prom i nent vents were emit ting steam and gas. Fig ure 21 shows sev eral shots il lus trat ing the enlarged opening in the ice on 15 October. On 20 Oc to ber, field crews in - stalled a web cam era lo cated 16 km (10 miles) N of Fourpeaked. Volcanism Network, Volume 31, Number 9, September 2006 Fourpeaked 13 Fig ure 19. Im age from the King Salmon NEXRAD weather ra dar show ing the vol ca nic cloud at Fourpeaked on 17 Sep tem ber 2006 at 1240 (2040 UTC). In color the ra dar re flec tivity ranges from light blue (low) to dark green (mod er ate), which cor re sponds to greater num bers and/or sizes of par ti cles. It can not be de ter mined whether the sig nal is due to large wa ter drop lets, ice par ti cles, coarse-grained ash, or a mix ture. Im age cre ated by Dave Schnei der, AVO/USGS, us ing data and soft ware from the NOAA Na tional Cli ma tic Data Cen ter. Fig ure 20. Im age show ing the to tal amount of sul fur di ox ide over Fourpeaked on 17 Sep tem ber 2006 as mea sured by the Ozone Mon i tor ing In stru ment on NASA?s Aura sat el lite. Sul fur di ox ide is dis played in Dobson Units (DU, a mea sure of the num ber of mol e cules in a unit area of the at mo spheric col umn). Im age cre ated by the Vol ca nic Emis sions Group at the Uni ver sity of Mary land Bal ti more County. Fig ure 21. Pho to graphs of the steam ing vent area at Fourpeaked volcano on 15 Oc to ber 2006. Cour tesy of Kate Bull (AVO-ADGGS). Fig ure 22. A 4 No vem ber 2006 pho to graph doc u ment ing steam ing on the up per most sec tion of the north ern flank of Fourpeaked vol cano. Cour tesy of Jennifer Adleman (AVO/USGS). Steam plumes orig i nat ing from vents along the sum mit were vis i ble via the web cam era on 27 and 30 Oc to ber. Steam ing con tin ued through at least 4 No vem ber (figure 22). Geo logic Sum mary. Poorly known Fourpeaked vol cano in NE Katmai Na tional Park con sists of iso lated out crops sur rounded by the Fourpeaked gla cier, which de scends east ward al most to the Shelikof strait. The ori en ta tion of lava flows and ex ten sive hy dro ther mal al ter ation of rocks near the pres ent sum mit sug gest that it prob a bly marks the vent of Fourpeaked vol cano (Swanson, in Wood and Kienle 1990). In for ma tion Con tacts: Alaska Vol cano Ob ser va tory (see Cleve land); S.A. Carn, N.A. Krotkov, A.J. Krueger, and K. Yang, Joint Cen ter for Earth Sys tems Tech nol ogy (JCET), Uni ver sity of Mary land Bal ti more County (UMBC), 1000 Hill top Cir cle, Bal ti more, MD 21250, USA (Email: scarn@umbc.edu). Soufri?re Hills West In dies 16.72?N, 62.18?W; sum mit elev. 915 m All times are lo cal (= UTC - 4 hours) Since the 20 May 2006 dome col lapse, the lava dome at Soufri?re Hills has con tin ued to grow. Only weeks af ter the col lapse, the alert level was raised to 4 as a re sult of in - creased seis mic ac tiv ity. At ap prox i mately 1300 on 30 June, the lava dome par tially col lapsed again, pro duc ing pyroclastic flows that trav eled E. Ac cord ing to the Wash - ing ton VAAC, a pi lot re ported an ash plume that reached ~ 3 km al ti tude and drifted NW. At 1830 on 30 June, Montserrat Vol cano Ob ser va tory (MVO) in di cated a sec - ond dome col lapse that also gen er ated ash plumes to an al ti - tude of 3.0-3.5 km (fig ure 23). Ac cord ing to MVO, on 27 June (prior to the col lapse on 30 June) the lava dome had an es ti mated volume of 27 million cubic meters. On 7 July, the alert level was low ered from 4 to 3. In - creased rockfall ac tiv ity and dome growth to the NE were ob served on 21 July, and the post-col lapse dome de vel oped an asym met ric pro file ow ing to a blocky spine on the NE. On 18 July the spine?s sum mit stood at ~ 895 m elevation. As the dome con tin ued to grow dur ing July (fig ure 24), vi - sual ob ser va tions re vealed that the still in tact blocky spine began leaning E. Dur ing Au gust the dome lost spines from its crest, giv - ing it a more sym met ri cal pro file as it con tin ued to grow E. Height ened ac tiv ity dur ing the last week of Au gust in - cluded an in crease in seis mic ity and pyroclastic flows. On 29 Au gust, pyroclastic flows reached the Tar River val ley and gen er ated a steam-and-ash cloud that reached an al ti - tude of ~ 9 km. Heavy rain fall pro duced mudflows around the base of the volcano. At 0300 on 31 Au gust, two vig or ous ash-and-steam vents opened on the W and N flanks of the dome (fig ure 25). The vent ing ep i sode was au di ble at times from the town of Sa lem and the sur round ing ar eas. MVO noted the con tin ued dome growth and the open ing of these vents when on 31 Au gust they raised the alert level to 4. Height ened ac tiv ity con tin ued in Sep tem ber. The dome con tin ued to de velop sub stan tially with a ma jor ity of growth on the W side. The vents that opened on 31 Au gust re mained ac tive, with the vent above Gage?s wall emit ting a plume of hot gases and the N vent on the dome pro duc ing mainly ash-and-steam (fig ure 26). The open ing of these vents co in cided with high lava ex tru sion rates and consequent dome growth. At 0100 on 10 Sep tem ber, the vent above Gage?s wall be came more vig or ous through out the day, broad en ing the vent and gen er at ing a wide ver ti cal ash col umn. By 1300 the vent ing there be came vi o lent and ex plo sive with black jets of ash ris ing ~ 100 m. Pyroclastic flows trav eled down the Gages val ley for ~ 1 km (fig ure 27). The vent formed a crater in the Gages wall, re duc ing its height com pared to that of Chances Peak by 30-50 m. By 11 Sep tem ber, pyroclastic flows from vent emis sions had ceased, but vig - or ous ash vent ing con tin ued. At 0830 an over hang ing lava lobe that de vel oped on the NE col lapsed send ing a pyroclastic flow al most to the sea at the end of the Tar River valley. 14 Soufri?re Hills Smithsonian Institution ? Bulletin of the Global Fig ure 23. A photo taken on 30 June 2006 of Soufri?re Hills as viewed from the Montserrat Vol cano Ob ser va tory show ing the first par tial dome col lapse of the day. The par tial col lapse be gan just be fore 1300 lo cal time and lasted ~ 20 min utes, gen er at ing ash clouds to an al ti tude of ~ 3.5 km that drifted WNW. Pyroclastic flows (left side of pic ture) were con fined to the Tar River val ley and ul ti mately reached the sea. Most of the lava dome re mained in tact. Photo cour tesy of MVO. Fig ure 24. A photo of Soufri?re Hills taken on 25 July show ing spines at the sum mit of the lava dome as viewed from the NE. Photo cour tesy Greg Scott of Ca rib bean He li cop ters. Al though vol ca nic tremor ended early on 16 Sep tem ber, an in tense ep i sode of vol ca nic tremor last ing just half an hour started at 1400 on 19 Sep tem ber. It was ac com pa nied by in tense rockfall ac tiv ity giv ing rise to mi nor pyroclastic flows down the N and NE flanks of the lava dome. On 21 Sep tem ber the alert level was reduced to 3. Geo logic Sum mary. The com plex, dom i nantly andesitic Soufri?re Hills vol cano oc cu pies the south ern half of the is land of Montserrat. The sum mit area con sists pri - mar ily of a se ries of lava domes emplaced along an ESE-trending zone. Eng lish?s Crater, a 1-km-wide crater breached widely to the E, was formed dur ing an erup tion about 4,000 years ago in which the sum mit col lapsed, pro - duc ing a large sub ma rine de bris av a lanche. Block-and-ash flow and surge de pos its as so ci ated with dome growth pre - dom i nate in flank de pos its at Soufri?re Hills. Non-erup tive seis mic swarms oc curred at 30-year in ter vals in the 20th cen tury, but with the ex cep tion of a 17th-cen tury erup tion that pro duced the Cas tle Peak lava dome, no his tor i cal erup tions were re corded on Montserrat un til 1995. Long-term small-to-mod er ate ash erup tions be gin ning in that year were later ac com pa nied by lava-dome growth and pyroclastic flows that forced evac u a tion of the south ern half of the is land and ul ti mately de stroyed the cap i tal city of Plym outh, caus ing ma jor so cial and eco nomic dis rup tion. In for ma tion Con tacts: Montserrat Vol cano Ob ser va - tory (MVO), Flem ing, Montserrat, West In dies (URL: http:/ /www.mvo.ms/). Volcanism Network, Volume 31, Number 9, September 2006 Soufri?re Hills 15 Fig ure 26. A photo show ing lava-dome glow viewed from the S at MVO at 2200 on 7 Sep tem ber 2006. In can des cent rocks can be seen tum bling down all flanks of the lava dome on this clear night. A faint glow is vis i ble from the very hot and ac tive gas vent just in side the Gages wall (just right of the dome in the pic ture). Photo cour tesy of MVO. Fig ure 25. Pho tos show ing ac tiv ity at Soufri?re Hills on 31 Au gust 2006. (top) Emis sions from the vig or ous new vent in side Gages wall (Gages Moun tain to the left of the vent and Chances Peak to the right). (bot tom) N-look ing photo show ing the N crater wall, lava dome, and the new vig or ous ash vent on the N side of the lava dome. Cour tesy of MVO. Fig ure 27. A photo show ing ex plo sive ash vent ing from a spot above Gages val ley at 1530 on 10 Sep tem ber. Pyroclastic flows can be seen ad vanc ing into Gages val ley in the fore ground. Photo cour tesy of MVO. San Crist?bal Nic a ra gua 12.702?N, 87.004?W; sum mit elev. 1,745 m All times are lo cal (= UTC - 6 hours) San Crist?bal was last re ported on in BGVN 28:10, cov - er ing in ter mit tent gas and ash emis sions be tween Au gust 2002 and Sep tem ber 2003. The Instituto Nicarag?ense de Estudios Territoriales (INETER) noted that low seis mic ity and mi nor gas and ash emis sions char ac ter ized the pe riod from Oc to ber 2003 to June 2004. On 7 June 2004 a lahar flowed more than 600 m. On 13 June 2004, an erup tion caused ash to fall in the com mu ni - ties of Las Rojas, El Chonco, and El Viejo. On 20 July 2004 at 1430, an M 4.3 earth quake oc curred to the N of the vol cano at a depth of less than four km. The earth quake was felt in the re gions of Carlos Fonseca, Villa 15 de Julio, La Suiza, Las Rojas, Mocor?n, San Jose del Obraje, Santa Carlota, San An to nio, Rancher?a, and bor der - ing re gions. Some houses were dam aged and the pop u la tion was alarmed. The earth quake was felt in Matagalpa and Ocotal, and San Crist?bal emit ted abun dant gases for the fol low ing two days. Dur ing the rest of July, 95 after shocks were reg is tered; res i dents felt two more earth quakes, which oc curred on 23 and 30 July. Dur ing Au gust to early De cem ber 2004, mi nor seis mic - ity and ash and gas emis sions were the norm. Ash ex plo - sions oc curred on 3, 4, and 7 De cem ber. Ac cord ing to lo cal peo ple, ash fell in Chinandega and El Viejo. The next avail able re port dis cussed 16-22 No vem ber 2005. INETER de tected an in crease in seis mic ity be gin ning on 19 No vem ber. In creased tremor was in ter preted as be ing re lated to gas and ash emis sions. Ash fell W of the vol cano and near the town of Chinandega, ~ 15 km SW of the vol - cano. The amount of tremor decreased later. Ac cord ing to an As so ci ated Press news re port, ex plo - sions on 6 March 2006 pro duced col umns of ash and gas that rose above the vol cano. The ac tiv ity ceased by 8 March and there were no evacuations. INETER noted that phreatomagmatic erup tions be gan at San Crist?bal on 21 April 2006. Seis mic tremor in creased the same day around 1300. Small ex plo sions pro duced gas-and-ash plumes dur ing 21-23 April that de pos ited small amounts of ash in nearby towns. Geo logic Sum mary. The San Crist?bal vol ca nic com - plex, con sist ing of five prin ci pal vol ca nic ed i fices, forms the NW end of the Marrabios Range. The sym met ri cal 1,745-m-high youn gest cone, named San Crist?bal (also known as El Viejo), is Nic a ra gua?s high est vol cano and is capped by a 500 x 600 m wide crater. El Chonco, with sev - eral flank lava domes, is lo cated 4 km to the west of San Crist?bal; it and the eroded Moyotepe vol cano, 4 km to the NE of San Crist?bal, are of Pleis to cene age. Volc?n Casita, con tain ing an elon gated sum mit crater, lies im me di ately east of San Crist?bal and was the site of a cat a strophic land - slide and lahar in 1998. The Plio-Pleis to cene La Pelona cal - dera is lo cated at the east ern end of the San Crist?bal com - plex. His tor i cal erup tions from San Crist?bal, con sist ing of small-to-mod er ate ex plo sive ac tiv ity, have been re ported since the 16th cen tury. Some other 16th-cen tury erup tions at trib uted to Casita vol cano are un cer tain and may pertain to other Marrabios Range volcanoes. In for ma tion Con tact: Vir ginia Tenorio , Emilio Talavera, and Mar tha Navarro, Instituto Nicarag?ense de Estudios Territoriales (INETER), Apartado Postal 2110, Managua, Nic a ra gua (Email: ineter@ibw.com.ni; URL: http://www.ineter.gob.ni/geofisica/); As so ci ated Press, (URL: http://www.ap.org/). Montagu Is land South Sand wich Is lands 58.42?S, 26.33?W; sum mit elev. 1,370 m All times are lo cal (= UTC - 2 hours) Mat thew Pat rick re ported that the month of Oc to ber rep re sents the 5-year an ni ver sary of the start of the still-on - go ing erup tion at Mount Belinda on Montagu Is land. The first sat el lite ther mal alert for the vol cano oc curred on 20 Oc to ber 2001, and was the first de fin i tive re cord of his tor i - cal vol ca nic ac tiv ity on the is land (BGVN 28:02) (Pat rick and oth ers, 2005). The MODVOLC mon i tor ing sys tem uses MODIS (Mod er a te Res o lu t ion Im ag ing Spectroradiometer) sat el lite data pro cessed at the Uni ver - sity of Hawai?i-Manoa. Cur rent MODVOLC re sults, shown in fig ure 28A, in di cate more-or-less per sis tent ac tiv ity through out the 5-year pe riod, with ra di ant heat flux apparently peaking in late 2005 and early 2006. 16 San Crist?bal Smithsonian Institution ? Bulletin of the Global Fig ure 28. Plots of MODVOLC data at Belinda vol cano on Montagu Is land from 2001 to Oc to ber 2006. (A) Chro no log i cal graph of ra di ant heat out put from Mount Belinda mea sured from sat el lite sen sors. (B) Chro no log i cal plot show ing the dis tance of sat el lite-mea sured ther mal anom aly pix els from the Mount Belinda vent. Cour tesy of HIGP Ther mal Alerts Team. Land sat and ASTER (Ad vanced Spaceborne Ther mal Emis sion and Re flec tion Ra di om e ter) im ag ery has shown that the erup tion con sisted of cen tral vent ac tiv ity pro duc - ing lava flows. Small-scale ex plo sive ac tiv ity has also com - monly blan keted the E side of the is land. Three ef fu sive events have been ob served in AS TER/Land sat im ag ery, with the most re cent (Sep tem ber-Oc to ber 2005) pro duc ing a lava flow that trav eled 3.5 km and reached the sea to build a 500-m-wide delta of lava (BGVN 30:09 and 30:11). Fig ure 28B shows rel a tive lo ca tion (dis tance from the vent) com par ing Mount Belinda?s vent with the lo ca tions of MODVOLC alert pix els. This plot clearly shows lon ger flows dur ing the Sep tem ber 2005 ef fu sive event. Fol low ing this pe riod, there were sev eral other long-dis tance events. It is un clear if these re flect ad di tional effusive events. In ad di tion, the first two ef fu sive events ob served in the AS TER/Land sat im ages do not ap pear on the MODVOLC plot (fig ure 28B), due ei ther to cloud cover or their short flow lengths. Since the be gin ning of 2006, no cloud-free AS TER im ages have been available. Geo graphic ter mi nol ogy. The no men cla ture of vol ca nic fea tures on Montagu Is land, par tic u larly in re gard to Mount Belinda, has been quite vari able. Al though the name Montagu has been ap plied to the ma jor vol ca nic ed i fice form ing the is land (LeMasurier and Thomson, 1990), the name Mount Belinda has been var i ously ap plied to the en - tire vol cano, the cur rently ac tive young cone on the north - ern side of the is land, the 6-km-wide sum mit cal dera, and a peak on the south ern cal dera rim that is the is land?s high point. In con sul ta tion with John Smellie of the Brit ish Ant - arc tic Sur vey, we have used Montagu to re fer to the vol - cano form ing the is land and Mount Belinda for the currently active cone. Ref er ences. LeMasurier, W.E., and Thomson, J.W. (eds.), 1990, Vol ca noes of the Ant arc tic Plate and South ern Oceans: Wash ing ton, D C: Amer i can Geo phys i cal Un ion, 487 p. Pat rick, M.R., Smellie, J.L., Har ris, A.J.L., Wright, R., Dean, K., Izbekov, P., Garbeil, H., and Pilger, E., 2005, First re corded erup tion of Mount Belinda vol cano (Montagu Is land), South Sand wich Is lands, Bul le tin of Vol ca nol ogy, v. 67, no. 5, p. 415-422. Geo logic Sum mary. The larg est of the South Sand wich Is lands, Montagu con sists of a mas sive shield vol cano cut by a 6-km-wide ice-filled sum mit cal dera. The sum mit of the 10 x 12 km wide is land rises about 3000 m from the sea floor be tween Bris tol and Saunders Is lands. Around 90% of the is land is ice-cov ered; gla ciers ex tend ing to the sea typ i - cally form ver ti cal ice cliffs. The name Mount Belinda has been ap plied both to the high point at the south ern end of the sum mit cal dera and to the young cen tral cone. Mount Oceanite, an iso lated 900-m-high peak with a 270-m-wide sum mit crater, lies at the SE tip of the is land and was the source of lava flows ex posed at Mathias Point and Allen Point. There was no re cord of Ho lo cene or his tor i cal erup - tive ac tiv ity at Montagu un til MODIS sat el lite data, be gin - ning in late 2001, re vealed ther mal anom a lies con sis tent with lava lake ac tiv ity that has been per sis tent since then. Ap par ent plumes and sin gle anom a lous pix els were ob - served in ter mit tently on AVHRR im ages dur ing the pe riod March 1995 to Feb ru ary 1998, pos si bly indicating earlier unconfirmed and more sporadic volcanic activity. In for ma tion Con tact: Mat thew Pat rick, Dept. of Geo - log i cal and Min ing En gi neer ing and Sci ences, Mich i gan Tech no log i cal Uni ver sity, 1400 Townsend Drive, Hought - on, MI 49931, USA (Email: mpatrick@mtu.edu; URL: http://www.geo.mtu.edu/~mpatrick); HIGP MODIS Ther - mal Alert Sys tem, Hawai'i In sti tute of Geo phys ics and Plan - e tol ogy (HIGP), Uni ver sity of Ha waii at Manoa, 168 East-West Road, Post 602, Ho no lulu, HI 96822, USA (URL: http://modis.higp.ha waii.edu/); John Smellie, Brit ish Ant arc tic Sur vey, Nat u ral En vi ron ment Re search Coun cil, High Cross, Madingly Road, Cam bridge CB3 0ET, United King dom (URL: http://www.anarctica.ac.uk/, Email: jtsm@bas.ac.uk). Volcanism Network, Volume 31, Number 9, September 2006 Montagu Is land 17