Pohnpei Island is the eroded remnant of a basaltic shield volcano active 9 – 1 Ma.  Its tallest point is Mount Nanlaud (Nahnalaud, Big Mountain), at 783 m.  The island measures 334 km2 and hosts a current population of nearly 37,000.  The island is most notable for basalt megaliths and the ruined city of Nan Madol, built on artificial islands off the island’s eastern shore. 

Why would a volcano inactive for the last 5 Ma be of interest?  It is because of what the residents did with the volcanic products.  In the case of Pohnpei, they quarried lavas and used the blocks to build structures and artificial islands in a lagoon on the E edge of the island, Nan Madol.  While a bit older than Easter Island, the residents did much the same thing as they did on Easter Island, used what the volcanoes gave them to build their civilization.  We took a look at Easter Island from a volcanic perspective in 2022.  Today’s post will do the same thing with Pohnpei. 

Micronesia

Pohnpei (Ponape, Ascension) Island is part of the larger Caroline Islands group.  It is politically part of Pohnpei State in the Federated States of Micronesia.  The capital of both the state and the nation are located on Pohnpei.  It is the largest island of the Federation, the highest point, the most developed island, and the most populous. 

There is a lot of Pacific Ocean between Pohnpei and the rest of the world.  It is 5,200 km SW from Honolulu, 1,700 km SE of Guam, and 1,400 km NE from Rabaul, PNG.  Undeveloped land on the island is mountainous, with a thick jungle.  Climate is a tropical rainforest, and one of the wettest places on earth with an average yearly rainfall nearly 480 cm/yr.  This spikes to nearly 900 cm/yr in some of the mountainous regions inland.  Average temperatures run between 22° – 30° C daily throughout the year. 

While most of the inhabitants are Polynesian ancestry, considering themselves Pohnpeians, the population is considered to be the most diverse of any in the Federation.  This is due to over a century of foreign colonia occupation by Spanish, German, Japanese, Chamoro, Filipino, US, Australian and other western Europeans.  Pohnpei shows up in fiction over the last century including several horror stories by HP Lovecraft a century ago. 

Federation economic activity is centered around subsistence agriculture and fishing.  While the potential for a tourism industry exists, the islands are remote, making travel to and from difficult and expensive. 

The first settlers likely traveled by boat from either the neighboring Solomon Islands or Vanuatau somewhere around the beginning of the modern era, 0 AD.  The first rulers are called Saudeleur (Period of the Lord of Deleur) and ruled until 1100 AD.  They also built the first megaliths on the island and their capitol city of Nan Madol.  The Spanish showed up in 1529, though they made no attempt to land.  They did land in 1595 followed by an Australian in 1825.  That visit did not go well with the locals attacking the visitors. 

Spanish retained control until 1898 when Germany purchased the island from Spain.  It was occupied by the Japanese during WWI, transferred to Japan by the League of Nations.  Japan sovereignty ended in 1945 after WWII when it was transferred to the US.  The Federated States of Micronesia became independent in 1985, though it is still under indirect US control. 

Geography

The island is roughly square in shape, well-watered and hilly.  It is surrounded by a barrier reef with many small islets.  It has fertile volcanic soil that supports luxuriant tropical foliage.  There are mangrove swamps along the coasts and rainforests in the central hilly region.  The barrier reef is divided into multiple low coral atolls.  Most of these are wooded and support coconut palm trees. 

One of the things we’ve described over the years is the lifecycle of intraplate volcanic islands in the Pacific.  Hawaii is the best example.  Magma breaks through the Pacific Plate, builds an edifice that may or may not breach the surface of the ocean.  It goes through a shield-building stage, volcanism wanes, erosion takes place, there is usually some resurgent magmatism at the end.  A combination of erosion and subsidence eventually removes all the aerial portions of the cone leaving an atoll lagoon.  Eventually, that also disappears, and we are left with a seamount some distance below the surface.  Note that subsidence takes place from the first day magma erupts above the surface of the Pacific Plate. 

Pohnpei is a very long-lived island that does not have broad plains near sea level.  These have mostly returned to the sea, today populated by mangroves.  Subsidence was taking place during the construction of Nan Madol 1,200 years ago and the builders clearly had to deal with encroaching sea water.  There is an ongoing argument in the popular press whether this was caused by subsidence or rising sea levels. 

While most of the scientific literature mentions changing sea levels, these are specifically assigned to the low sea levels during the height of the last glaciation and the abrupt worldwide sea level increase around 130 m some 7.6 ka.   On the other hand, subsidence is usually a slow, continuous process. 

Nan Madol

Nan Madol is one of more than 100 sites of historic significance constructed on the island over the last 2,000 years.  These sites include both pre-contact and post-contact structures.  It is a city constructed in a lagoon on a series of small, artificial islands linked by a network of canals.  The core of the city measures 1.5 x 0.5 km and connects 92 artificial islets, sone and coral fill platforms, linked by tidal canals.  The name Nan Madol means “within the intervals” and is thought to refer to the canals.  The original name was Soun Nan-leng (Reef of Heaven).  Stone was quarried from a volcanic plug on the opposite side of Pohnpei.

Human activity here can be traced back to the First or Second Century AD, almost immediately after humans arrived on the island.  Construction of islets started by the 8^th or 9^th Century.  Construction of the megalith structures started 1180 – 1200 AD.  The city was the ceremonial and political center of the Saudeleur Dynasty until 1628 AD. 

While little is known about the actual construction, local tradition claims the builders migrated from Kosrae and brought construction skills used to build huge stone buildings with them and used those skills to build Nan Madol.  However, the Kosrae structures are younger than the Nan Madol structures, so migration must have worked in the other direction.  Saudeleur Dynasty founders ruled with a light hand, something their successors forgot over time. 

The center of the city was a residence for nobility and mortuary activities presided over by priests.  Its population was somewhere between 500 – 1,000.  In some ways, Nan Madol was a way to control potential rivals by requiring the nobility and religious leadership to live in the city rather than in their home districts.  There is no fresh water or food in the city.  It was collected and / or grown inland. 

Invading natives overthrew the Saudeleurs in 1628.  They had to grow their own food and gather their own water, leading them to abandoning Nan Madol.  Today, Nan Madol is an architectural district covering over 18 km2.  Nan Madol has been interpreted by some as the remains of fabled lost continents, showing up in popular culture as settings for lost races and various ancient apocalypses. 

There were reports in 1978 – 1979 of submerged basalt columns just offshore leading to searches for a pair of legendary submerged cities.  This was further investigated 1988 – 1989 in an underwater archaeological survey.  While they found a number of prone basalt columns seaward of Nan Madol, the team concluded they were lost, discarded or fallen building materials.  They took a look at two supposed underwater clusters of columns.  One was raised and studied, finding only coral.  A 2013 underwater investigation into a blue hole near the city found it to be a sinkhole that developed in reef limestones during the height of a glaciation when sea level was very low. 

Preservation of these structures is an issue over time.  The primary problem is modern residents recycling basalt from ancient structures for use in new ones and the modification of ancient ones for gardening / farming purposes.  There is also a continuing threat from tidal weathering and flooding which causes washouts and degradation of coral rubble fill over time.  Vines and creepers may or may not assist in preservation by preventing stone and rubble from shifting.  There is an ongoing argument whether vegetation assists preservation or accelerates it.  Basalt used in these structures is very dense and hard.  It is also somewhat brittle.  Building stones were selected as elongated columns, 5-6 sides.  They are prone to transverse breakage. 

Volcano

The island of Pohnpei was built of volcanic materials erupted over the last 9 Ma.  Pohnpei has two satellite atolls, one of which (Ant) appears to be part of the larger edifice.  Ant Atoll is 11.5 km WSW.  Its saddle is 750 m deep.  Pakin Atoll is 30 km WNW.  Its saddle is 1,000 m deep.  They were likely active the same time Pohnpei was.  Neither has been closely studied.

The action of subsidence and erosion shaped today’s island.  There was also what appears to be one to several flank collapses from the NW side of the island that removed significant portions of it.  To date, there has not been sufficient bathymetric surveys to characterize the debris field on the ocean floor, though it may show up in Google Maps satellite view. 

As with any intraplate volcano, subsidence and sea level rise of 5.5 m over the last 6 ka drowned any previous coastal plains of sedimentary material around the island.  The coastline today consists mostly of volcanic rock and fringing mangrove swamps. 

The visible shape of the island with a drowned coast and no coastal plains is due to the combined effects of subsidence and erosion.  Estimates of subsidence range between 2,000 – 1,200 m over the lifetime of the shield.  Evidence for this comes from channels up to 91 m deep carved into the surrounding reef and lagoon in five locations.  They suggest rapid local subsidence since the low sea level of the most recent glacial.  This sort of rapid regional subsidence took place at several other islands in the region, likely due to thermal rejuvenation of the crust, increasing subsidence for a time.  Most currently eroded materials are trapped and deposited in the fringing mangrove swamps building new land.

Rainfall also recharges an island-wide groundwater aquifer.  There have been attempts to exploit the groundwater for human use via wells on the N half of the island.  There may be more favorable conditions in the S half of the island, though this work has not yet begun.  There are no reported hot springs or thermal features, though much of the volcanic rock has been altered over time. 

The central part of the island either subsided or eroded to a greater depth than elsewhere.  A caldera is not visibly apparent but is suggested by gravity measurements.  The center of volcanic activity during the shield-building stage was considerably N of the current geographic center. 

The current barrier reef outline suggests much of the N half of the island has disappeared.  The NE corner of the barrier reef extends 7 k from the main island, while the NW portion of the reef is much closer to the rest of the island.  This bay on the NW portion of the island was created by either a large landslide or a series of smaller slides early in the island’s history.  Note that on Pohnpei, this is only circumstantial evidence of a flank collapse, as there have not yet been sufficient bathymetric surveys done to date.  The few that have been done are as yet insufficient. 

Long duration volcanism on Pohnpei and Chuuk (Truk), 8 Ma and 10 Ma respectively are more typical of hot spot volcanoes on slow-moving tectonic plates.  Other islands and atolls among the Carolines had a much shorter active lifetime than these two, closer to the typical 1.0 Ma seen elsewhere in the Pacific.  This short active lifetime has no post-shield eruptions, quickly creating atolls.  

General Life Cycle

At the top level, Pohnpei had only two main cycles of activity, shield building and post shield building eruptions.  These have been divided into 3-6 stages by various writers based on the desired level of detail.  I will present both of these. 

Pohnpei was built in three main stages of construction.  The first built a shield with frequent eruption of basalts.  Activity stopped for 1.5 Ma and was followed by sporadic volcanic activity that erupted basanites and hawaiites (Awak volcanic stage).  A large flank collapse took place shortly after the end of the shield building stage, removing much of the NW portion of the original shield.  The final stage (Kupwuriso volcanic stage) erupted voluminous lavas, blanketing much of the S half of the island.  The thick, massive post-shield lavas of the Awak and Kupwuriso stages dominate present day geology and hydrology of the island.  These volcanics are separated based on age, flow and chemistry.  Contact between the two units is only visible in a few isolated locales on the island. 

Major construction of the shield took around 1 Ma 9.0 – 8.0 Ma.  This is a similar timespan to the growth of other intraplate oceanic shields.  Activity became more explosive toward the end of shield building, blanketing the S half of the island with extensive ash and the N half with volcanic breccias.  These are best exposed on the Param and Mwahd Peidi barrier island on the NE end of Pohnpei.  Ash deposits (Rohi Ash) are best exposed on the S half of the island along the main road.  Significant erosion took place during a 1.5 – 1.0 Ma period of quiet. 

Awak volcanics erupted on top of the shield building and Parem Breccias in the N half of Pohnpei.  Individual flows are rather thick, with the thicker flow units jointed.  Many flows of this stage entered rivers or standing water producing highly fragmented hyaloclastites.  There were relatively long periods between individual eruptions of this stage demonstrated by relatively thick erosion deposits (up to 15 m) found between individual flow units.  This stage produced several trachyte domes on the S half of the island.  Erosion removed softer shield building lavas beneath ponded lavas from Awak, creating vertical scarps.  Awak eruptions lasted for a long time, 7.1 – 3.1 Ma. 

Kupwuriso volcanics are the most recent stage on the island.  These erupted relatively thick, columnar jointed lava flows with thin layers of sediment between them.  These flows are up to 60 m thick around the perimeter of the island as lava flows ponded in valleys.  Temwen Island is probably the youngest volcanic feature on Pohnpei.  It is a tuff cone created by explosive interaction between magma and either groundwater or seawater.  There is a small lava flow from near the center of the cone that flowed W.  Tuff cones are easily eroded.  This one is mostly intact, indicating it is very young.  The line of contact between Kupwuriso and Awak volcanics is poorly exposed but seems to be around 450 m elevation.  Kupwursio lava flows tend to overlie those of Awak.  Kupwuriso volcanics date 2.1 – 1.0 Ma, with the tuff cone likely being younger.  How much younger is currently unknown.

Detailed Life Cycle

Stage 1.  The original shield volcanoes of Pohnpei, Ant and Pakin were likely active at the same time 10.0 – 8.5 Ma.  Activity built two separate platforms for the three systems.  Activity on Pohnpei took place from a radially oriented rift system.  Volcanism at Ant likely took place from a similar system.  Activity at Pakin took place from an elongated NW trending rift system.  Estimated maximum heights of the shields above sea level are estimated at 2,200 m for Pohnpei, 950 m for Ant, and 450 m for Pakin. 

Stage 2.  Near the end of the shield building stage, volcanism became more explosive, erupting extensive pyroclastics.  Distribution of these was controlled by prevailing NE winds.  Coarse breccias were deposited near the center of activity.  These are best exposed today on Parm and Mwahd Peidi.  Deep depressions under these deposits suggest pre-existing depressions such as pit craters, subsidiary calderas or other collapse features.  The deposits might have been products from erosion of a steep-sided collapse scar. 

Stage 3.  During the break in eruptive activity between Stages 1 and 4, a single or series of large catastrophic flank collapse landslides removed much of the NW corner of the island.  Subsidence of the volcanic platforms led to growth of fringing and barrier reef systems around all three islands.  Pakin probably eroded more quickly than Ant, as Ant is located in the Pohnpei rain shadow. 

Stage 4.  Volcanic activity resumed after a quiet period of 1.5 Ma, some 7.0 Ma.  Pakin was probably close to becoming an atoll due to subsidence and erosion.  Ant was likely at a near-atoll stage similar to Chuuk (Truk).  The new lavas covered the rapidly eroding shield lavas.  Magmas became more evolved partway through this stage, erupting trachyte domes.  Activity was infrequent with thick sediment layers between individual flow units.  Some flows ponded in steep, V-shaped valleys.  Erosion afterwards created prominent elongated ridges on the island’s N end.

Stage 5.  Individual eruptive events became larger 3.0 – 2.0 Ma, creating broad plateaus around the island’s S half.  Somewhat smaller basalt flows flooded valleys of the N half around 1.0 Ma.  These flows today form the floors of the valleys.  The youngest volcanic event on Pohnpei was probably the phreatomagmatic event that created Temwen Island.

Stage 6.  Erosion and continued subsidence over the last 1.0 Ma created the island’s current shape. 

Tectonics

The Caroline Islands present a bit of a geologic oddity.  They are situated on two plates, the almost completely subducted Caroline Plate and the Pacific Plate.  Pacific Plate movement in this part of the world is over 8 cm/yr E–W.  The four main islands of the 607-island group from W–E are Yap, Chuuk (Truk), Pohnpei (Pohape), and Kosrae (Kusaie), with over 1,700 km between Yap and Kusaie.

The island arc system is located along the Caroline Ridge on the E convergent margin of the Philippine Sea Plate and are bounded by the Palau Island Arc to the S and the Izu – Mariana arc system to the N.  While Yap is described as being located on the remaining margin of the Caroline Plate by Rehman, et al, Mar 2013, it is on the W side of the trench, which would put it on the Philippine Sea Plate.

Hot spot island chains on the Pacific Plate usually present themselves as a narrow line of volcanic islands that get younger the closer to the hot spot they are.  Active volcanism usually takes place on top of the hot spot as the plate slides over it.  The four main Carolines are indeed volcanic, with ages generally younger as you move from Yap to Kosrae.  But they are not aligned in a narrow line.  Rather, they and their satellite islands are spread out in a broad area either side of the line between them.  Worse, there is an overlap in dates of activity, with widely separated islands being active simultaneously.

Yap is considered to be a subduction-driven, volcanic arc system.  It was active 11 – 8 Ma.  While the overall subduction of the Caroline Plate to the W seems to be continuous, various spreading center and basin formation changes took place over the last 11 Ma.  The edges of the basin near the Yap and Ayu Trenches was formed 12 – 10 Ma.  Spreading activity along the Ayu Trough stopped 6 – 5 Ma, after which the boundary between the Caroline and Pacific Plates became extensional.  There are numerous normal faults up to 3 km along that boundary.

Volcanic rocks on Chuuk date 15 – 4.3 Ma.  There is a 1.5 Ma period of quiet 7.0 – 5.5 Ma.  Volcanic shield building activity on Pohnpei took place 8.7 – 0.9 Ma.  Post erosional lavas erupted 1.5 – 0.9 Ma.  Kosrae main activity took place 2 – 1 Ma.  Activity at Chuuk and Pohnpei overlapped in time despite being separated by 720 km.  Activity at Pohnpei and Kosrae took place at the same time despite being separated by 550 km.  Clearly there is another explanation.

The Ur, et al paper suggests that volcanism is related to fractures in the Pacific Plate allowing magma to reach the surface as an alternative to hot spot powered volcanism.  It proposes additional work on erupted rocks to solve the mystery, which sounds like a very reasonable thing to do. 

Conclusions

Pohnpei is an example of an elderly intraplate volcanic island that has been returning to the ocean since activity began there 9.0 Ma.  It does not show signs of any continuing or residual volcanic activity, nor does its neighbors.  Residents made extensive use of jointed columnar basalt quarried from lava flows to build a capitol city a thousand years ago.  The building material has been reused over the centuries for more modern needs.  Unlike what we saw on Easter Island, the population of the island never outstripped the ability of the island and surrounding ocean to support them. 

Additional information

Geology and hydrogeology of the island of Pohnpei, Federated States of Micronesia, Spengler, et al, Mar 1991

Geology and hydrogeology of the island of Pohnpei, Federated States of Micronesia, SR Spengler, 1990

Pohnpei, Wiki 

Pohnpei island, Micronesia, Britannica

Geology and geochemical evolution of lavas on the island of Pohnpei, Federated States of Micronesia, Spengler, et al

Pohnpei State, Wiki

The archaeology of Nan Madol and Temwen Island, Pohnpei, site distribution, architecture, and early agricultural features, Pohnpei, Federated States of Micronesia, Levin, et al, Aug 2015

Recently vanished islands of Micronesia, P Nunn, research originally published in 2017

Study of the volcanic rocks from the Pohnpei Island, Western Pacific, Itoh, et al, May 2011

Geological origin of the volcanic islands of the Caroline group in the Federated States of Micronesia, Western Pacific, Ur, et al, Feb 2013

Source and magmatic evolution of ocean island basalts from the Pohnpei Island, northwest Pacific Ocean:  insights from olivine geochemistry, Zong, et al, Jan 2022

Pohnpei eco-adventure guide, Nan Madol Ruins, Jun 2013

Nan Madol:  “In the space between things”, B Lythberg, Google Classrooms, Kahn Academy