The Atitlan Caldera System, Guatemala

Volcan Toliman (center right), Atitlan (far right) from Volcan San Pedro (foreground). Note cinder cone on northern flank of Toliman. Lake Atitlan in the foreground and to the left of the photo. Heavily vegetated lava flows on the near flanks and north facing flanks of Toliman. Village in the center foreground. North is to the left of the photo. Lake Atitlan fills about half the Atitlan III caldera. Image courtesy DIY Travel HQ

 

Special note to our readers:

Many thanks to all who filled in the survey.  A few great suggestions have come in. Now we have to think about the “logistics” to put some of them into action. As we are allowed to keep the survey app running to a certain limit of entries we have decided to put the post up in our top menu. Perhaps the one or other of you still want to have their say. Be reminded that you may answer even just one question and then click the “Finish” button. That’s okay – we appreciate every single bit of input!

Volcan Atitlan is one of three stratovolcanoes surrounding Lake Atitlan, the most recent caldera created by the system.  The last massive eruption took place some 84,000 years ago.  San Pedro, Toliman and Atitlan were all built in the southern part of the caldera lake since that eruption,

Lake Atitlan is the deepest lake in Central America at 340 m at its deepest point.  It was formed during the most recent caldera forming eruption some 84,000 years ago.  Its name means “between the waters.”  It is described as one of the most beautiful lakes in the world.  It is also a mountainous lake, at some 1,500 m above sea level.

Volcan Toliman to the left, San Pedro to the right. Volcan Atitlan is behind Toliman. Note the twin peak on Toliman. Lake Atitlan in the foreground. View is to the south. Village of Santa Cruz in the foreground. Image courtesy Adventurous Kate

 

Guatemala was in the heart of Mayan empire.  Over half the population call themselves Maya.  Many villages still have Maya culture, indigenous peoples and traditional dress.  Local religion has melded Mayan and Catholicism introduced by the Conquistadors.  While the immediate area is not heavily populated, with just over 70,000 within 10 km of the volcano, there are over 664,000 within 30 km and over 8 million within 100 km.

There is no road that goes completely around the lake nor are there airports in the villages on its shores.  There do not appear to be a lot of float planes in the country.  Most travel between villages is done by boat around and across the lake.  There are mountain roads that do connect the villages, but travel is rough and time consuming.

Statue in San Pedro La Laguna. Brightly painted buildings and statues are common in these heavily Mayan villages. Image courtesy Adventurous Kate

 

Several Mayan archeological sites are found at and around the lake, many of them under water.  The lake has been rising for the last few decades and more recent buildings in the 11 villages surrounding the lake have started getting inundated by lake water.

One town on the lake was overwhelmed by Guatemalan and tourists, becoming quite popular with hippies in the 1960s.  A civil war caused many of them to leave but following the end of hostilities in 1996, visitors started returning.  Today, the main tourist town, Panajachel’s economy is almost entirely dependent on tourism.

The Guatemalan Civil War raged 1960 – 1996.  Things were particularly ugly around the lake as the government decided that the local indigenous peoples were far too supportive of the guerrillas fighting against them and targeted them in response.

Village street in Santa Cruz. Toliman in the background to the south across the lake. Image courtesy Adventurous Kate

 

As with any idyllic location, people always try to improve things.  Usually, these efforts cause more problems than they solve.  One example is the lake itself which was selected as a national park in 1955.  The government was trying to figure out how to increase tourism and boost the local economy.  What better way to do this than to improve fishing?  At the suggestion of Pan American World Airlines, they started stocking the lake with black bass.  The bass liked it a lot.  So did the tourists.  Native species not so much and the bass wiped out two thirds of native fish in the lake and contributed to extinction of the Atitlan Grebe, a bird that only lived in the vicinity of the lake, something the next caldera event is also likely to do.

The lake also started getting thick blooms of toxic algae starting in 2009.  The second occurrence was 2015.  Cause is likely the use of the lake as a wastewater discharge catchment basin.

Oct 2005 landslide location. Caused by Tropical Storm Stan rainfall. Top of image is north. Volcan San Pedro is to the west (left). Landslide on flank of Toliman depicted in red. Closest village Santiago Atitlan is depicted. Lava flows on the north, west and east flanks of Toliman are also clearly visible. Image courtesy Dartmouth EDU DFO 2005 – 139

 

The area also gets hurricanes and tropical storms, which cause local flooding, landslides and lahars.

The primary Guatemala volcano observatories is Insivumeh.  It appears to concentrate its efforts on Pacaya, Fuego and Santiaguito.  http://www.insivumeh.gob.gt/

COORDINADORA NACIONAL PARA LA REDUCCIÓN DE DESASTRES (CONRED) handles disaster response in Guatemala.  https://conred.gob.gt/site/index.php

Annotated satellite photo of Atitlan caldera clearly showing 3 new stratovolcanoes filling the southern portion of the caldera. Major villages are also depicted. Caldera rim is clearly visible. Image courtesy Guatemala Tours and Travel

 

Neighboring Calderas

Atitlan is the westernmost of a chain of at least five caldera lakes stretching from southern Guatemala eastward through El Salvador.  From west to east, these caldera lakes are Atitlan, Amatitlan, Ayarza, Cotapeque and Ilopango.  All caldera centers are silicic in nature.  The relationship between the calderas and the stratovolcanoes of the volcanic front is not yet established, though the total volume of products from the calderas is similar to that of the volcanic front over the last 200,000 years, estimated at 300 – 500 km3 in 1981.  Estimated rate of production of volcanic products is 31 – 62 km3.

The central American volcanic arc stretches 1,500 km from southern Mexico to northern Panama.  It has at least 12 large composite stratovolcanoes, several dacite domes, and five major silicic caldera centers behind the arc.  Caldera activity is generally older than that of the volcanic front, but there is some overlap with the present generation of 50,000-year-old and younger composite volcanoes.

Caldera lakes in Guatemala and El Salvador. Only one not depicted is Ayarza, midway between Amatitlan and Coatepeque. Tectonic setting of these calderas. Image courtesy Saxby, et al

 

The 15 x 25 km Atitlan caldera activity is episodic, with at least three caldera formation events over the last 12 Ma.  The youngest formation event is the massive Los Chocoyos Ash of 85,000 years ago.  This produced 250 – 300 km3 dense rock equivalent, the largest unit known in central Guatemala in the Quaternary.  It is non-welded valley fill deposits and fall deposits in three ocean basins from the Gulf of Mexico near Florida to the Equatorial Pacific near Ecuador.

Amatitlan caldera is 16 km in diameter.  There are at least 9 pyroclastic units identified with a total volume of 60 – 80 km3 DRE.  The largest events produced major pyroclastic flow sheets during the caldera event.  Time between major events is thought to be around 80,000 years, like that of Atitlan.  This may mean that eruptions between the two neighboring caldera systems are controlled by the same process.

Relative ages and volumes of eruptive products of these 5 calderas. The larger the black, the more voluminous the eruption. Largest depicted here is the Chocoyos eruption from Atitlan 84,000 years ago. Size of the unit is depicted by length of black line. Note that this does not capture any event older than 200,000 years. Screen capture from pdf courtesy Rose, et al, 1999

 

Ayarza caldera is a compound caldera with mixed basalt – rhyolite.  The two largest events were a few thousand years apart centered some 25,000 years ago.  The caldera is much smaller at 3 x 5 km and produced around 2 km3 DRE.

Coatepeque caldera is a 6.5 x 11.5 km caldera and the second largest of the young generation of collapse calderas in San Salvador after Ilopango.  Coatepeque produced around 24 km3 DRE during three major and one minor eruptions.  Ignimbrite sheets from these eruptions extend 10 – 15 km from the caldera to the south, east and north.  Sheets to the west are covered by younger deposits from the neighboring Santa Ana volcanic complex.  Ages of these deposits range between 70,000 – 350,000 years ago.

Relative ages and volumes of major eruptions from the 5 recent calderas. Note that the volumes of these eruptions generally increase over time as better data is gathered. Screen capture from pdf, Rose, et al, 1999

 

Ilopango is an 8 x 11 km diameter caldera.  Its major deposits were produced by at least four major eruptions and stretch up to 20 km to the east, north and south, where they changed neighboring topography, and up to 80 km to the west.  The most recent pyroclastic flow deposit is dated at 408 – 550 AD.  Estimated volume from this eruption is greater than 39 km3 DRE.

Depiction of Central American Volcano Arc. Guatemala segment toward the top. Image courtesy Saginor, et al, 2009

 

Volcanoes

Atitlan is part of the Central American Volcanic Arc, though it is located a bit behind the belt.  The immediate neighboring group of volcanoes some 40 km to the west include Volcan Santa Maria, Santiaguito, Cerro Quemado, Siete Orejas and Chicabal.  44 km to its east, we have the next volcanic center which is the Volcan de Fuego – Acatenago complex.  Volcan Agua and Pacaya are a few tens km farther to the east.

Initial volcanic activity at Atitlan took place some 12 – 11 Ma.  Since the first eruptions, there have been three cycles of caldera formation and volcanic growth afterwards.  The most recent caldera event is called the Los Chocoyos Eruption (named after a local bird found in soft ash), produced some 300 km3 of tephras and pyroclastics covering 6 million km2.  Ash can be found from Florida to Ecuador.

Extent of Los Chocoyos Tuff north of Atitlan. Yellow areas flow coverage. Yellow arrows are flow paths. Green numbers are ash flow depths. Image courtesy Brocard, et al, 2014

 

Volcanic activity continued after the most recent eruption, building the three stratocones.  The oldest of these is Volcan San Pedro, which lies in the SW portion of the caldera.  It is likely extinct as activity has moved eastward since activity began.  Volcan San Pedro is around 3,000 high.  It is a forested cone that rises above the SW side of Lake Atitlan.  It has a shallow summit crater breached to the NW, partially filled with a lava dome.  There are two cinder cones on its flanks between the volcano and the SW wall of the caldera.  It does not have any fumarole or hydrothermal activity.  San Pedro is a favorite for visiting hikers.

The next volcano is Volcan Toliman which began growing after activity stopped at San Pedro.  While it has not erupted in historic times, it is still thought to be active.  Toliman is 3,158 m high and is located near the southern wall of the most recent caldera.  It has a parasitic lava dome, Cierro de Oro on its northern flank thought to be the most recent active feature, with lava flows dated a few thousand years based on sediment thickness.  It is a large andesitic stratovolcano.  Unlike neighboring Atitlan, its northern flanks are covered with lava flows.  These mostly erupted from the flanks and many of them made their way to the lake.   It is topped with two cones, both of which have small craters.

Los Chocoyos Tuff exposed south of Guatemala City over 100 km east of Atitlan Caldera. 3 flow units are visible. Top layer is the pinkish layer at the center of the stack. Bottom two layers are the top and bottom halves of the pyroclastic flow deposit. Fall deposits above the pink Los Chocoyos top layer are from Amatitlan and Agua. Image courtesy Smithsonian GVP

 

The most recently active volcano is Volcan Atitlan, at most some 10,000 years old.  Its most recent eruption was in 1853.

There are active hot springs around the lake.

The Newhall thesis identifies remains of ancestral stratovolcanoes around the periphery of the caldera rim destroyed in previous caldera eruptions.  These remains include a hypothetical Volcan San Marcos centered over the NW part of the caldera, Volcan Tecolote which has Cerro Las Minas small basaltic cinder cone on its northern flank, Volcan Paquisis just inside the SW scarp of the current caldera rim, the informally named Volcan Xejolon SE of the caldera, Volcan Tzanjuyub and associated domes, and the more recently emplaced Volcans Zunil, Pecul and Almolonga to the west of the caldera.  The area is complex and difficult to study due to the interplay of stratovolcano deposits and caldera formation deposits.

Sunrise from Volcan Atitlan. Toliman center right and San Pedro Center left. Lake Atitlan in the foreground center and right. Other volcano in the distance to the upper left is Pecul. Image generally looking north. Image courtesy Guatevalley.com

 

Eruptions

Initial activity at Atitlan was basalt / andesite.  This switched over to silicic eruptions at least 14 Ma.  Initial silicic eruptions produced 5 – 10 km3 of rhyodacites as Plinian tephras and pyroclastic flows around 126,000 years ago.  The most recent caldera event 85,000 years ago erupted 270 – 300 km3 of zoned rhyolite, erupted in widely dispersed Plinian tephras and mobile pyroclastic flows.  Pyroclastic flows traveled more than 100 km from the caldera.  Materials erupted late in the event resembled earliest silicic magmas from the system.

There were at least three cycles of stratovolcano growth, major silicic eruptions and caldera formation.  The first cycle took place 14 – 11 Ma and ended with five ash-flow eruptions and formation of the first Atitlan caldera (Atitlan I) centered just north of the present basin.  The system has migrated a few km south over the last 2.5 Ma.

Schematic of multiple Atitlan calderas with possibly associated Ixtahuacan caldera to the NW. Image courtesy braeuNERD

 

Carl Rehnberg at VC has posted the volume of this eruption at 1,800 km, the Maria Tecun Tuff.  I have found nothing in the research for this post to substantiate this claim.  There is a 1980 Dartmouth Thesis by Chris Newhall that describes in detail the history of the Atitlan system.  He finds outcrops of the Maria Tecun Tuff outcrops at least 100 km east and west of Atitlan.  These tuffs range from welded to unwelded.  Thickness of the base tuff is at least 650 m near the caldera.  The second tuff is 150 – 200 m.  The third is substantially thinner and overlain with 3.5 m layer of ash and tephras.  The fourth is 100 – 500 m thick and is the most widespread and voluminous of the five sheets.  Most of the tuff is welded, capped with at least 40 m of unwelded material.  The final tuff sheet is up to 300 m thick, mostly welded, with the top 50 – 100 m unwelded.  In the absence of underlying channelization of these flow sheets, their total volume could be quite massive.

The second cycle 10 – 8 Ma ended with the San Jorge ash flow eruption and caldera collapse (Atitlan II).  This cycle was followed by the growth of three stratovolcanoes 1.8 – 0.4 Ma.

Los Chocoyos Tuff exposed in a quarry near San Juan Ostuncalco. White base layer is a Plinian pumice fall unit. Thicker overlying unit is a pyroclastic flow unit over 200 m thick. The Plinian unit is the largest known fall deposit in Central America. Photo courtesy Smithsonian GVP

 

The Newhall thesis linked in the Additional Information section proposed a fourth caldera cycle a few tens of km to the west of the main Atitlan caldera field.  This produced a set of voluminous tuffs estimated at 25 – 100 km3 from what is named as the Ixtahuacan caldera system some 3.5 Ma.  Unlike Atitlan, resurgent dome activity took place in this system after caldera formation.  Boundary faults for the two systems are separated by only a few km, which means they could have been fed from the same magma body.

The most recent cycle 1 Ma – present has so far grown stratovolcanoes, the Los Chocoyos caldera forming eruption (Atitlan III) and several smaller silicic eruptions.

Three main stratovolcanoes from west looking east. San Pedro La Laguna in the foreground. Image courtesy DiscoverAtitlan.com

 

There were two rhyodacite eruptions at the caldera between 84,000 – 60,000 years.  Another pair of closely spaced eruptions around 40,000 years ago.  These last two tephras overlie San Pedro lavas but predate Toliman and Atitlan lavas.  San Pedro grew to its present size by 40,000 years ago.  Toliman grew since then and may still be active.  Atitlan was built almost entirely within the last 10,000 years.

Rhyodacites erupted in smaller volumes partly through the lake, building three composite volcanoes in the caldera.  The 5 – 13 km3 of post caldera eruptions were bimodal mafic / silicic similar to other volcanism on the Guatemalan Highlands.  Volumes at Atitlan are significantly larger.

Samabaj, Guatemala below the shores of Lake Atitlan. Discovered in 1996. Sunk perhaps 3,500 years ago after a pyroclastic flow dammed the lake outlet. Ruins of the city lay 35 m below the surface of the lake. Image courtesy Orange Smile

 

Atitlan Volcano is the only historically active member of this volcanic complex.  Historic eruptions took place in clusters a couple decades long.  The last of these was 1826 – 1856.  There were five previous clusters since the Spanish arrived in the 15^th Century, 1469, 1505, 1579, 1663 and 1717.  This activity was explosive vulcanian and subplinian in the VEI 2 – 3 range.  They also produced lava flows, mudflows (lahars) and pyroclastic flows during nine different eruptive events.

There several flank lava flows on San Pedro and Toliman volcanoes under the current level of the lake.  Lava flows and more recently pyroclastic flows blocked the original outlet of the lake south of San Lucas Toliman.  There are number of pyroclastic flows on the eastern and western flanks dated within the last 3,000 years.

Panajachel sits on a delta of lahar debris. Heavy tropical rains frequently cause lahar damage in this part of Guatemala/ Image courtesy RocDocTravel

 

Recent earthquakes and heavy rains have triggered lahars from Atitlan.  These were not connected with any recent activity.  A 1991 earthquake triggered two major landslides.  Rain associated with Hurricane Mitch triggered a major landslide in 1998.  A Sept. 2002 heavy rain triggered a late evening lahar that travelled through two small communities, destroying buildings, homes, and killing 37.  The lahar flowed 7 km and came without warning.  An Oct. 2005 lahar destroyed nearby Panabaj, killing at least 200 on the NW flanks of the volcano.  That town was built on pyroclastic flow deposits.  Many of the surrounding towns are built on lahar and pyroclastic deposits.  Historic lahar deposits reach as far as 20 km to the south..

implified regional tectonics of southern Mexico and northern Central America. Traces on the Cocos Plate give approximate age in My of the oceanic plate. Isochrons inland from the trench give approximate depth of the Cocos Plate beneath the Caribbean Plate. Image courtesy Surendra Pal Verma, 2009

 

Tectonics

The major tectonic feature of central Guatemala is formed by the subduction of the Cocos Plate under the Caribbean Plate.  This has created the 1,500 km long Central America Volcanic Arc stretching from southern Mexico to northern Panama.  Volcanoes are closely spaced, but clustered, averaging 27 km between clusters.  Due to active subduction, massive earthquakes are common.  A M 7.5 earthquake in 1976 killed 26,000 and opened fissures under Lake Atitlan dropping its water level by two meters in a month.

The volcanic front along the arc is 165 – 190 km from the offshore trench.  The seismic zone underneath the Arc varies from less than 100 km in Guatemala to 200 km in eastern clusters of small composite cones, shields, domes and cinder cones.  The volcanic front is located with a shallow seismic zone with strike-slip normal faults and grabens.  Much of the terrain dropped by the action of grabens has been filled with volcanic debris.  Basaltic lavas are usually present along the arc, but far more common in areas with thinner crust like Nicaragua.  Lavas range from basalts to rhyolites.  Individual vents within volcanic centers usually have more than one magma batch, indicating zoning of the underlying magma reservoirs.

Lava flows from Volcan Toliman flowing north into Lake Atitlan. Flank cinder cone in the distance. Photo generally looking east. Image courtesy Smithsonian GVP

 

There is a line of calderas dispersed along and slightly behind the main Arc.  These typically erupt silicic tephras and pyroclastic flows.

There is a poorly defined secondary line of volcanoes 20 – 75 km behind (NE) of the volcanic front.  These are typically single composite cones.  They have significantly different chemistry than volcanoes in the main Arc.

Back Arc volcanism occurs in Central America.  It overlaps the second line and extends to the main front.  These are typically clusters of cinder cones, small shields and lava fields in areas of extension.  There is no historic activity in the back arc.

All three Atitlan stratovolcanoes from NW corner of Lake Atitlan. From left to right we see Atitilan in the background, Toliman in front of it, Toliman flank cinder cone barely visible, and San Pedro to the far right. San Pedro is heavily eroded. Image courtesy Just a Pack blog

 

Conclusions

Atitlan is clearly an active, dangerous system.  It is nicely situated so as to receive a prodigious flow of eruptible magma into its underlying magma chamber.  Life cycle of this system seems to be multiple stratovolcano growth, caldera event, followed by a period of quiet and then more stratovolcano growth.  Other than the possibly related Ixtahuacan Caldera, there is no resurgent dome growth following the caldera-forming eruption.  The more this system is studied, the larger its eruptions appear to be, with estimates of the Chocoyo Tuff growing from 200 to greater than 300 km3 over recent decades.  It is likely that the Maria Tecun Tuffs may see similar growth as they are better studied.

Lake Atitlan at sunset. Toliman and Atitlan to the left. San Pedro to the right. Image courtesy Prologo Atitlan

 

Additional Information

http://volcano.oregonstate.edu/atitl%C3%A1n

https://volcano.si.edu/volcano.cfm?vn=342060

https://www.volcanodiscovery.com/atitlan.html

https://pubs.usgs.gov/of/2005/1403/

https://www.sciencedirect.com/science/article/abs/pii/0377027387900540

https://pdfs.semanticscholar.org/7825/59bdc5c152642b05a7f5b1dcc8a362c33208.pdf

http://adsabs.harvard.edu/abs/2003AGUFM.V32D1050V

https://www.sciencedirect.com/science/article/abs/pii/0377027387900539

https://www.adventurouskate.com/the-towns-of-lake-atitlan-guatemala/

https://www.atitlan.com/

http://volcano.oregonstate.edu/tolim%C3%A1n

https://volcano.si.edu/volcano.cfm?vn=355070

http://pages.mtu.edu/~raman/papers2/CentralAmer/CarrRefs/Carretal.pdf

https://www.researchgate.net/figure/Tectonic-setting-of-large-volume-silicic-calderas-in-El-Salvador-and-Guatemala-The_fig7_305627295

https://www.nature.com/articles/ncomms12295

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.725.4757&rep=rep1&type=pdf

https://libarchive.dartmouth.edu/cdm/compoundobject/collection/dcdis/id/332162/rec/7

 

Advertisement