Gunung Semeru, East Java, Indonesia

Morning view across Bromo – Tengger and Sandsea caldera toward Semeru in eruption.  Image courtesy Tan Yilmaz, flickr, Aug 2008

Mt Semeru (Mahameru – Great Mountain) is a 3,657 m tall andesitic volcano on Java.  It is one of the most active Java volcanoes, the tallest of all Java’s volcanoes, and erupts variations of andesites and basalts.  Semeru is surrounded by a ring plain with over a million people, several hundred thousands of whom are in range of regular lahars and debris flows.  As a steep, active volcano in a wet climate, lahars and debris flows are a very real threat to people, farms and buildings up to at least 35 km from the crater.  Semeru is one of the most prolific lahar producer on the planet. 

Semeru is located at the southern end of a volcanic massif with Bromo – Tengger on the N end.  There are multiple faults in the region which seem to bound the location of volcanic activity. 

Cemoro Lawang in Bromo Tengger Semeru National Park.  Note village built on volcanic scarp with massive volcano in the immediate background.  Image courtesy Axel Drainville via ThousandWonders.net

This portion of Java is home to the Bromo Tengger Semeru National Park, known locally as Taman Nasional Bromo Tengger Semeru (TNBTS).  It is located generally to the E of Malang, S of Pasuruan and Probolinggo, SE of Surabaya, capital of East Java.  The original portion of this park was the Sand Sea, the floor of the Tengger caldera protected in 1919.  The greater park was declared a national park in 1982.  The Sand Sea is a plain of volcanic sand surrounding five volcanic structures on the floor of the Tengger caldera. 

Climate in the park is generally cold, mainly due to the altitude, during the winter from May – Sept.  Heavy rainfall takes place during the summer with little precipitation in winter. Nighttime temperatures in the winter can fall below 0° C, with frost and light snow.  It is classified as a subtropical highland.  Average highs are 16 – 17° C.  Average lows are 8 – 9° C. 

Combined map of Bromo-Tengger-Semeru-Arjuno Biosphere Reserve.  Oceans are off the map to the bottom and upper right.  Image courtesy Hakim & Soemarno, 2017 via bromotourtravel.net

Lowlands in the park are tropical rainforests.  As you get higher on the mountains, plant species are less prolific until you end up in a subalpine zone above 2,400 m.  The Sand Sea is a special ecosystem, covered in volcanic sand.  It is believed to be the only known desert-like area in Indonesia.  There is a wide variety of plants and animals in the park and neighboring R Sojero Grand Forest area.  Both have been combined into the Bromo Tengger Semeru Arjuno Biosphere Reserve, approved by UNESCO in 2015.

The area in and around the park is home to the Tengger people, one of the few significant Hindu communities remaining in Java.  Their population is around 600,000 in at least 30 villages in the mountains in and around the park.  A million people live within 30 km of Semeru.  Over 20 million live within 100 km.

Semeru is monitored by Magma Indonesia.  There are multiple webcams available for Semeru, Bromo – Tengger.  Volcano Discovery is probably as good a place to start as any though as of this writing, neither of their webcams resolve a picture.  There are also webcams available looking at Bromo.  Semeru should be visible in the background of both.  The Bromo webcams are not currently resolving images either. 

Topographic map of general region surrounding Bromo – Tengger – Semeru volcanic massif.  Image courtesy Lupi, et al, 2018

Region

This blog and our previous home covered multiple volcanoes on this end of Java.  These include Arjuno – Welirang and the Lusi Mud Volcano in 2021, eruptive activity at Bromo in 2015, and five pieces on the 2014 eruption of Kelud.  All are good places to start for a review of events in the region.

Volcanoes are concentrated in this part of Java.  Arjuno – Welirang is 57 km NW Semeru.  Kelud is 70km WNW from Semeru.  And the Bromo – Tengger caldera is 20 km NNE.  Semeru is considered to be part of the Bromo – Tengger volcanic massif. 

Annotated photo of volcanic structures of the Tengger – Semeru massif (left).  Ajek-Ajek and Jambangan calderas tagged.  Screen capture courtesy Solikhin, et al, 2012

The Semeru – Tengger volcanic massif covers around 900 km2 with a cluster of calderas and strato- cones.  Bromo – Tengger is the farthest north.  The Jambangan and Ajek-Ajek calderas are in the middle of the string.  Mount Kepolo and the composite cone of Mahameru – Semeru overlap.  The volcanic centers align generally in a N-S direction.  The northernmost part of this group is Lake Grati, a maar at the northern foot of Tengger.

Bromo – Tengger is 16 km in diameter.  It consists of five overlapping composite volcanoes.  Each of these are topped with a caldera.  The most recent of these is the 9 x 10 km Sandsea (or Sand Sea) caldera, which formed in multiple eruptions 45 – 33 ka.  It is the most recent caldera in the region.  There is an overlapping cluster of cones on the floor of the Sandsea caldera built within the last several thousand years.  The youngest of these is Bromo tuff cone, one of Java’s most popular and active volcanoes.  Over 50 mild to moderate explosive eruptions from Bromo have taken place since 1804.

Schematic of central and southern end of Tengger – Semeru massif annotating the Jambangan and Ajek-ajek calderas (center top).  Mt Kepolo is on the southern flank of Ajek-ajek.  Mahameru is the next volcano built followed by Semeru.  Note general movement of volcanic activity to the south over time, with the newer systems moving south.  Note also satellite vents on the eastern flank of Semeru.  Screen capture courtesy Solikhin, et al, 2012

The nested Jambangan and Ajek-Ajek calderas are located between Tengger and Semeru.  They are open to the E, which may be connected to debris avalanches.  It is unknown if these were associated with volcanic eruption.  There are few hummocky landforms associated with these deposits.  There are at least two parasitic cones in the area.  There are several intra-caldera vents including the deep Rana Kumbolo crater lake.  This lake may indicate a third eruption vent from Ajek-Ajek. 

The Holocene Mt Kepolo is a stratocone near the southern rim of the Jambangan caldera.  It is aligned N-S with a recent lava cone and lava flow at the foot of Semeru’s northern flank.  There are two sets of faults that cut through the two groups of nested calderas. 

Photo of Semeru in eruption in 2016.  Note the scar downslope from the active crater, lack of greenery around the upper cone, and heavily eroded flanks.  Image courtesy NASA Digital Globe via Smithsonian GVP

Volcano

Semeru is one of the few persistently active volcanoes on Earth.  The composite andesitic cone is constructed from two edifices, the Mahameru ‘old’ Semeru, and the Seloko ‘young’ Semeru.  The SE flank of the summit cone has a scar that branches on the active Jonggring – Seloko vent.  This is the current pathway for rockslides and pyroclastic flows.  The summit of Semeru is covered with tephras. 

The flanks are covered by dense evergreen forest up to 2,800 m in elevation.  Seasonal monsoons produce 200 – 370 cm of rain every year.  It is not uncommon for 20 cm of rain to fall in a 24-hour period.  The ring plain below the volcano covers nearly 1,800 km2 and supports more than a million people.

Schematic of new Semeru and old Semeru including structural boundary between the two cones.  Image courtesy Thouret, et al, 2007

The bulk of the young Semeru cone is about 60 km3 of material consisting of lava flows, pyroclastic flows, tephras, debris avalanche and lahar deposits.  It grew on the older Mahameru edifice.  The N to SW part of the Mahameru cone and the NW slope of the Semeru cone are covered by thick tephras.  The Mahameru summit shows three scarps, likely vent rims.  Its vents appear to have migrated NW – SE over time and are located at the intersection of three fractures.

The upper NW flank of the summit cone is more weathered, with unconsolidated deposits.  To the NE of Jonggring – Seloko are landslide scars from 1909 and 1981 landslides and gullies.  These are still active today.  The unconsolidated deposits are related to a previously active hydrothermal system beneath this area.

Annotated schematic of major flows from Semeru since 1884.  Screen capture courtesy Solikhin, et al, 2012

The SW to E part of the young Semeru cone is fractured and deeply cut by erosion.  The summit area has thick tephra layers between lava flows.  The Jonggring – Seloko crater is 500 m SE from the summit and has a pile of lava flows layered with pyroclastic fall and flow deposits.  There is a lava tongue from a lateral flank eruption to the W of the crater.  A lava flow S of the vent came from the crater itself. 

The horseshoe-shaped scar in the crater measures 1.8 x 0.5 km.  It channels rockfalls and pyroclastic flows.  It is oriented along a N 160° fault.  The top of the volcano is quite steep at just over 30°.  There appears to be some structural deformation of the new Semeru cone with the slip plane along the Mahameru edifice.  While there is no geophysical data to support this, there is a bulge and faults along the SE cone flank which could be due to slippage along a known weak basal layer. 

Lava flow tongue down crater scar of Semeru jan 2021.  Image courtesy Agus Harianto / AFP via New Straits Times, Jan 2021

The summit crater has physically changed since 1923.  The SE-trending scar did not exist then, though a small channel may have been concealed between a pair of lava flows.  In 1963, dome collapse sent pyroclastic flows along this forming scar.  After 1977, the summit had been eroded and a horseshoe shaped amphitheater formed which today channels eruption related flows.  By 1994, the structure extended downward into the head of three eroded flank river headwaters.  This scar channels pyroclastic flows up to 11 km to the S of the volcano. 

There are three styles of eruptive activity from Semeru since 1818.  First is persistent vulcanian and phreatomagmatic activity marked by short steam and ash eruption columns several times a day.  Second is an increase in activity every 5 – 7 years that produces multi-kilometer high eruption columns, ballistic bombs, and thick tephra falls around the vent.  Ashfall up to 40 km from the vent due to these is not uncommon.  Dome extrusion as part of these eruptions in the vent and subsequent collapses produce block and ash flows generally toward the SE that have traveled as far as 11 km.  The final eruption type are flank lava flows erupted on the lower SE and E flanks in 1895 and 1941 – 1942.

The following video of a lahar from Semeru took place following heavy rains on Feb 8, 2021.  The driver did escape.  The vehicle did not.  Image courtesy Pas Mateng via The Watchers blog, Feb 9, 2021. 

Lahars are considered to be the primary hazard from Semeru.  There have been five large scale lahars since 1884.  Lumajang, with 85,000 people was devastated by lahars in 1909.  Analysis of runoff from the cone into a river on the ESE flank in 2000 reported a rate of sedimentation similar to other active composite cones in wet environments.  Unlike these other cones, the rate of erosion of cone material does not decline drastically in the years following a major eruption.  This is because the material is continually resupplied by persistent eruptive activity.  That material is then remobilized down the flanks by rain. 

There are three primary hazard prone areas below Semeru.  The first is a triangular shaped area from the crater, open to the SE frequently swept by dome collapse avalanches and pyroclastic flows.  The second are valleys to the S and SE that funnel tens of rain-triggered lahars each year up to 25 km across the ring plain.  Valleys crossing the ring plain to the E have carried debris flows and lahars up to 35 km.

Close up of active lava dome in Semeru late Nov 2014.  Image courtesy Andi / Volcano Discovery Nov 2014

Eruptions

Semeru has been in almost continuous activity since 1967.  During this time, eruptions are typically Strombolian eruptions with lava flows.  The lava flows exit the crater via the Kobokan breach on the SE slope of the volcano.  The lava tongue is unstable, collapses creating pyroclastic flows after a sufficient volume is extruded.  The pyroclastic flows travel up to 11 km from the crater.  Sometimes the collapse is triggered by an eruption.  Access to the summit is permitted, though ballistics may reach the crater rim and have caused injuries or fatalities in the past.

As previously stated, the main danger from the volcano are not eruption and collapse of lava tongues or dome destruction.  Rather it is the secondary process of lahar and debris flows which can extend tens of kilometers from the crater.  Heavy local rainfall causes landslides and lahars of unconsolidated eruptive products.  The flow of mixed water and landslide materials has devastated large areas.  These flows are unpredictable with more watery viscosity than the lahars.  They are distributed over wide areas and threaten cultivated lands, densely populated and developing urban areas. 

Webcam capture of Jan 2021 pyroclastic flow and ash from Semeru drifting 4 km SE of the crater.  Image courtesy MAGMA via Smithsonian GVP

The current persistent eruptive activity from Semeru in 1967 began with continuous emission of gas from a vent in the southern part of the Jengging – Seloko crater.  This was followed by lava extrusion that created a flow down the south slope.  Eruptive activity continued until 1974.  Lava extrusion created a lava dome that grew in the W and S areas of the summit.  Growth later shifted to the SE but remained in the summit area.  The extrusion also had ashy explosions, lava avalanches, and block and ash flows.  This widened the SE-trending channel.  Since 1979, the rate of dome growth decreased, but activity continued with lava avalanches from the dome, dome collapse block and ash flows and explosions.  After 1979, the scar widened rapidly.  The 1994 activity extended the scar down slope.  Its width in 2009 is almost twice that of 1981. 

Annotated photos and schematic of active crater and scar used in 2002 – 2003 eruption cycle.  Image courtesy Solikhin, et al, 2012

The 2002 – 2003 VEI 2 – 3 activity announced itself by an increase in seismic activity followed by emission of a short ash plume.  Lava was extruded and several lava avalanches and pyroclastic flows traveled down slope.  Eruptive activity increased until Aug 2002 followed by a slight decrease until late Nov.  During Dec, several block and ash flows erupted, traveling up to 5 km from the dome.  Voluminous block and ash flows and lahars took place on Dec 29 and 30 on the SSE flank.  One of these made it 11 km to the suburb of Pronojiwo.  The Dec 29 block and ash flow contained blocks of dome and lava larger than 12 m in diameter.  Eruptive activity continued after these dates and declined until the end of Jan 2003 when the episode ended.  Block and ash flows and lahars from these eruptions were visible via satellite remote sensing.  The 2014 eruption sequence was also classified as a VEI3 by Smithsonian GVP. 

Over 500 killed by Semeru over the last 30 years, mostly via lahars, avalanches and debris flows.  Nearly 600 were killed during the 1909 and 1981 eruptions.  Mudflows and secondary mudflows have been the culprits. 

Semeru in eruption with pyroclastic flow down the SE flank Dec 2020.  Image courtesy Antara Foto / Seno via Reuters via Aljazeera Dec 2020

The first historical report of fatalities from Semeru lahars took place in Apr 1885 following a partial collapse of a lava tongue from the crater followed by a powerful eruption.  It was immediately followed by a lahar that inundated two villages, killing 70.  A similar number died in 1895 following multiple lahars down multiple drainages.  Heavy rainfall in Mar 1909 triggered a massive landslide / debris flow that traveled as far as the city of Lumajang, 35 km E of the volcano.  38 settlements were inundated in 110 km2.  1,450 buildings were destroyed and at least 220 killed.  In 1976 multiple lahars in multiple drainages claimed 133.  The flow direction of some of these unexpectedly changed, they jumped the banks of the drainages and hit villages not normally in the path.  A lahar outbreak two years later in 1978 claimed another 12. 

The first fatality from the current (post-1967) eruptive cycle took place in 1981 when pyroclastic flows reached 7 km downslope after destruction of a lava dome.  The event was poorly documented, though a crater opened at the former location of the dome.  In May 1981 over 30 cm of rain in 3 hours triggered a landslide down the E flank.  This took a similar course to the 1909 event and took 275 lives.  There are at least 350 km2 of river valleys and low-lying areas on the ring plain at risk from Semeru lahars. 

Progression of Jan 19, 2002 debris flow front in the Curah Lengkong river.  Channel is 25 m across.  Three photos on the left show the approaching flow frong.  Three photos on the right show closeup of the front.  Note height measurement at the lower right of 3.5 m.  Screen capture from Lavigne, et al, 2003

In Feb 1994, increasing activity triggered a powerful eruption that put pyroclastic flows 11.5 km down a river valley on the SE flank.  Six fatalities were reported from this eruption.  Pyroclastic flows were initially channeled down the SE crater scar.  In Sept 1997, a pair of climbers were killed.  July 2000 a group of vulcanologists were surprised by an unusually violent eruption, killing two of them.  Several other were injured.  Pyroclastic flow activity from the Dec 2002 eruptions extended 9 km from the summit. 

Pre-eruption inflation has been measured at Semeru with tilt meters installed close to the active crater.  Vulcanian eruptions, which release large amounts of ash are preceded by accelerating inflation 200 – 300 sec before the eruption.  This is interpreted as volume expansion of liquid into gas in the shallow portion of the conduit.  Non-accelerating changes in inflation start 20 sec or so before smaller gas bursts, which explosively emit water steam.  Increasing tilt is measured for both eruption styles.  This means the volume or pressure of magma or gas stored in the conduit before the eruption drives the size of the eruption.  This also means there is a possibility of predicting eruption size from measurements of volcano inflation for Semeru.

Tectonic elements of East Java.  Image courtesy S Husein, Dec 2015

Tectonics

As Semeru is located only 20 km S of Bromo – Tengger, continental tectonics are similar.  This system is still a subduction volcano driven by the subduction of the Australian Plate under the Eurasian Plate.  Crust thickness remains some 25 km.  Location of volcanic activity and structures appears to be controlled by multiple clusters of faults in the area.  Our Tectonics section in the 2015 Bromo / Tengger post is a good place to start with regional tectonics.  A more extended discussion can be found in the 2015 Rinjani post. 

Conclusions

Semeru is an example of a huge stratovolcano with a history of deadly debris flows, lahars, and over 50 years of persistent volcanic activity.  The combination of volcanic activity and massive rainfall during the wet season powers this threat.  Its cone appears to be moving, shearing at a weakened base layer.  In recent years, most of the lahar, debris flow, avalanche and pyroclastic flow has taken place on the SW – E portion of the cone and ring plain.  The system appears to have a vigorous magma source and is immediately adjacent to a pair of undated calderas to the N. 

Two things bother me about this volcano.  First is the presence of multiple calderas within 20 km of the volcano.  Second is the apparent mobility of its flank.  We have written about slumping cones before, Conception which was built on a bed of clay deposits in Nicaragua, and Anak Krakatau, which lost most of its cone via flank collapse into the 1883 Krakatau caldera Dec 2018.  Structurally weak strato cones can and will move, sometimes with devastating effect.  The two closest calderas to Semeru, Ajek-Ajek and Jambangan are both associated with what appears to be debris flows and openings AT one end of each caldera.  However, neither of the stratocones built since caldera formation (as yet undated), Mount Kepolo or Mahameru appear to have gone caldera.  On the other hand, part of Mahameru is missing, likely due to a flank collapse. 

Semeru in eruption Jun 2020.  Image courtesy Cap Allon, Electroverse, Jun 2020

Additional information

Smithsonian GVP – Semeru

The threat of hazards in the Semeru volcano region in East Java, Indonesia

Spectral analysis of harmonic tremor signals at Mt Semeru volcano, Indonesia

Contrasts between debris flows, hyperconcentrated flows and stream flows at a channel of Mount Semeru, East Java, Indonesia

The coalescence and organization of lahars at Semeru volcano, Indonesia

Biodiversity conservation, community development and geotourism development in Bromo Tengger Semeru Arjuno biosphere reserve, East Java

Volcanic hazards at Mount Semeru, East Java (Indonesia), with emphasis on lahars

Inflations prior to vulcanian eruptions and gas bursts detected by tilt observations at Semeru volcano, Indonesia

Lahars deposits architecture and volume in the C Lengkong valley at Semeru volcano, Indonesia

Transverse architecture of lahar terraces, inferred from radargrams: preliminary results from Semeru volcano, Indonesia

Geology, tectonics and the 2002 – 2003 eruption of the Semeru volcano, Indonesia: Interpreted from high spatial resolution satellite imagery

A real-time assessment of lahar dynamics and sediment load based on video-camera recording at Semeru volcano, Indonesia

The persistently active Semeru volcano and Merapi composite dome (Indonesia):  Mapped and interpreted from high-spatial resolution imagery

Semeru Volcano – Photovolcanica