Fernandina Island is the newest of the Galapagos Islands, closest to the hotspot. It hosts a single massive basalt shield, Fernandia volcano.
Over the years, we’ve done several posts about the Galapagos, neighboring volcanoes on Isla Isabela to Fernandina’s immediate east. I did a 2016 post on the 2015 eruption of Wolf volcano. Granyia did a 2018 post on the Sierra Negra caldera. There was also a 2022 post on Darwin. As we discussed the Galapagos region extensively in the Darwin post, I will skip the normal regional description in this one and spend most of the time on the volcano and its extensive history of eruptions.
Fernandina Island is mostly circular, about 30 km in diameter and tops out at 1,476 m. It has an area of 642 km2. Due to extensive recent volcanic activity, the island has a mostly rocky surface. Most volcanic activity in the Galapagos over the last few hundred years took place on Fernandina and neighboring Isabela islands.
Fernandina Island first appeared on western navigational charts and maps in 1684. The British named it Narborough Island after Sir John Narborough, an English naval commander of the 17th Century. The Spanish named it after King Fernando of Spain, who sponsored Columbus.
Fernandina is the most active and most pristine of the Galapagos volcanoes. There is a single visitor site on the NE edge of the island. Other than that, the island is maintained in its pristine state. It hosts a large land iguana population which nests on the caldera rim and on its floor.
Waters around Fernandina and neighboring Isabela are the richest in the archipelago. Cold upwelling waters provide prime habitat for Flightless Cormorants and Galapagos Penguins.
There are two species of rice rats present on Fernandina. Black / Norway rats introduced by explorers killed off rice rats on the other islands. As with Hawaii (and other islands), the biggest threat to resident wildlife is introduction of aggressive non-resident species by visitors. Generally, this is unintentional. Occasionally, it is intentional. Outcome for either method is usually deadly to indigenous wildlife.
Punta Espinosa is the island visitor site. It offers visitors two hikes, a short one around the small peninsula, and a longer one inland to the edge of a new a’a lava flow. Marine iguanas and Flighless Cormorants are found during the hikes. So is Lava Cactus, one of the first species to grow on new lava. There are two dive sites off Fernandina. One is adjacent to Punta Espinosa. The other S at Punta Mangle. Divers get to observe marine iguanas, Flightless Cormorants, Galapagos Penguins, sea turtles, and various fish species. I did not see any mention of coral reefs surrounding Fernandina, though they do exist around neighboring Isabela Island.
Volcano and seismic observation and reporting in Ecuador is handled by Instituto Geoficico Escuela Politenica Nacional. It is continuously monitored by at least two seismographs on the island.
Volcano
Fernandina is a classic hotspot island basaltic shield, located closest to the Galapagos hotspot plume. It is topped with the 5 x 6 km La Cumbre caldera. The caldera is quite deep, currently around 900 m.
It is the most active volcano in the Galapagos, likely sitting on top of the hotspot plume head. Age of the volcano is at least 32 ka. It has been entirely resurfaced in the last 4.3 ka and has high eruption rates over the last few millennia. The youngest flows cover nearly 360 km2 with a volume of 4.4 km3. Subaerial volume of Fernandina is 140 km3. Aerial volume is 200 km3.
Subsidence of the island is estimated at 7 m over the lifetime of the edifice, returning 2 km3 below sea level. Present day volume of the caldera is 11 km3. Volume of the youngest lavas (over the last 1,000 years) are estimated at 5 km3. That may be as high as 15 km3 if the caldera has filled and reformed at least once during the last millennia.
The summit is nearly 1,500 m above sea level. The elliptical caldera grew in stages over the years, occasionally refilling with lava and collapsing when magma withdrew from the shallow magma chamber beneath. Previous caldera collapses are visible at both ends of the caldera as benches. The most recent of these was in 1968. At that time, the floor was flat, 7 km2, 800 m below the rim.
The elliptical summit caldera is 21 km2, with the long axis NE – SW. Its walls slope 30° – 50°. Its current shape is due to multiple cycles of partial filling and collapse. The most recent of these dropped the SE portion by as much as 350 m during the 1968 event. There is no evidence of submarine eruptions during the 1968 event.
Large calderas on the seven western Galapagos volcanoes are taken as evidence that there is shallow magma storage beneath their summits. The remaining portion of magma storage and flow is not as well understood. The typical Galapagos shield volcano has steep upper flanks, large calderas, summit fissures and radial flank fissures. The submerged flanks of Fernandina have three well-developed rifting zones extending SW, W and NW.
Fernandina mainly erupts lavas, with not a large amount of tephras. However, stratigraphy following the 1968 eruption shows that explosive eruptions do happen. Apparently the intracaldera lake is present most of the time and interacts with magma intrusions into the caldera.
The island has been completely resurfaced over the last 4.3 ka. This is comparable with resurfacing on the Big Island of Hawaii, with Mauna Loa surface not older than 30 ka, and Kilauea younger than 3 ka. Fernandina has been extremely active over the last 4 – 5 ka, an eruption rate comparable to Kilauea. The extreme young age of Fernandina is a key factor in species distribution. For example, the giant tortoises are not found on Fernandina, one of the few places in the western Galapagos without them.
Prior to the advent of continuous satellite observations, activity at Fernandina was observed only by passing ships and aircraft. Many of the earlier smaller effusive eruptions were not observed in real time. The island cannot be seen from most other parts of the archipelago.
There is no evidence of large flank intrusions at Fernandina, though a fishing boat tied up off Punta Espinoza in 1927 viewed an uplift of several feet overnight. A similar uplift event took place May 1954 at Urbina (Urvina) Bay on the coastline of neighboring Darwin volcano. This event uplifted the shoreline by as much as 4.6 m in less than an hour, stranding fish in ponds.
InSAR observations of Fernandina 2003 – 2010 suggest a pair of magma reservoirs. The shallow one has a flat top and is 1.1 km below sea level. The deeper one is an oblate spheroid 4.9 km below sea level. They appear to be hydraulically connected which may help explain a rapid uplift episode in 1927 and the withdrawal of 1.5 km3 of magma causing the 1968 caldera collapse without significant eruption.
InSAR data 2003 – 2010 demonstrated multiple active deformation sources active at various times and locations inside the volcano, a complex magmatic network between the upper magma chamber and the surface. The shallow source produces almost continuous displacement beneath the caldera. The deeper chamber is intermittently active and produces edifice-wide displacement. Lateral intrusion of magma from both chambers is common. Two eruptions from intruded dikes took place while the entire summit deflated. Magma seems to easily migrate between the two reservoirs. While most eruptions are tied to intrusions from the shallow magma chamber, there was at least one lateral intrusion observed from the deep chamber in Aug 2007. The deep chamber is periodically recharged with magmas from the hot spot.
Lava from flank eruptions is dangerous to local wildlife in multiple ways. When it enters the sea, it heats up the water, in at least one case as far as 7 km from the shoreline. This kills fish and in turn birds and mammals that eat fish. Lava flowing down the flanks ignites vegetation in the dry climate causing wildfires. Fires ignited during the 2017 eruption burned over 16 km2 on the western flank and lasted for a year. Finally, cold blooded iguanas like the warmth, and spend a lot of time basking in the sunlight to regulate their bodily temperature. Multiple iguanas wandered onto the crust of fresh lava and were promptly cooked before they figured out it was too hot to remain.
Eruptions
Fernandina erupted in 2020, 2018, 2017, 2009, 2005, 1995, 1991, 1988, 1984, 1981, 1978, 1977, 1973, and 1968. There were at least 13 earlier eruptions ranging back to 1813. Most of these were short duration from or near the summit caldera. Lava would flow down the flanks toward the sea or onto the caldera floor. Most recent eruptions were on the flanks. Two of the most notable eruptions were in 1988, which triggered a debris avalanche inside the caldera, and 1968 which featured a caldera collapse increasing the depth of the summit caldera.
The most recent eruption from Fernandina was a 2-day long fissure eruption Jan 2020. The eruption was preceded by an increase in seismicity and deformation weeks before the eruption began. The eruption started 12 Jan, with 11 seismic events. The largest of these was M 4.7 at 5 km deep. Around 1800 L, a circumferential fissure formed below the E rim of the La Cumbre crater. This produced lava flows down the flank. A rapidly increasing in magnitude seismic swarm started a couple hours before the eruption began, with a maximum around 1650. 70 min later, a second increase in seismicity marked the start of the eruption. A hotspot visible from satellite was visible just after 1800, along with a visible gas plume to the W and NW. The eruption lasted 9 hours.
Slow deformation took place Jun 2020 – Nov 2021 visible from satellite InSAR data. It was centered on the summit caldera and NE flank. Deflation was observed on the upper W and SW flanks. Oct 2021, the deflation changed to inflation and the rate of inflation increased. Fumarole activity was visible in the summit crater before 17 Nov. This may have been related to a small episode of tremor 16 Nov. Periodic fumarole activity and small episodes of tremor at Fernandina without subsequent eruptions are common.
The previous eruption was Jun 2018. It began with a seismic swarm on 16 Jun. The largest event was M 4.1 at a depth of 4 km NE of the island. The eruption began just after 1100, visible from a passing boat and satellite thermal imagery. The eruption took place from a radial fissure on the NNE flank, producing gas plumes with low ash content. These plumes were 2-3 km high, drifting WNW. Lava flows reached the sea within a few hours. After 2 days of vigorous eruption, tremor and thermal anomalies decreased, along with a significant drop in SO2 emissions.
A short eruption took place Sept 2017. It began on 4 Sept with earthquakes and long period events. Tremor announced the beginning of the eruption. Lava was erupted from a circumferential fissure near the SSW rim of the caldera. It flowed down the S and SW flanks but did not reach the sea. A low ash content plume rose 4 km above the crater, drifting W. Flows continued on 5 Sept. Eruption intensity decreased on 5-6 Sept.
The April 2009 eruption began 10-11 April. It was first observed via satellite thermal imagery of active lava flows. Tourists observed the eruption on 11 April. Galapagos National Park personnel overflew the eruption and saw a fissure on the SW flank, 500 m from the summit caldera rim. The fissure was 200 m long and 10 m wide. Fountains were estimated at 15 m high. The gas and ash plume drifted up to 370 km W, SW and S. This eruption took place near the site of the previous 2005 eruption. There was a large thermal anomaly with SO2 detected. There was also smoke from burning vegetation.
This eruption lasted for 20 days, ending the end of April. Lava entering the sea killed fish which were seen floating. The hot water also killed sea lions and birds looking for an easy meal of the floating fish. Magma erupted from radial and circumferential fissures on the flank below the caldera rim. Satellite observation hours before the eruption began captured sill propagation toward the surface.
A May 2005 eruption was first observed by satellite photos of an eruption plume 13 May. A flyby noted a large convection cloud that obscured the caldera rim. Active lava flows were observed on the S and SW slopes. Source was in the same area as the 1995 eruption from a circumferential fissure near the rim. The fissure was not observed, but the 1995 eruption fissure was 4.5 km long around and just below the caldera rim. The plume was observed up to 5 km above the volcano. Satellite thermal imagery noted hot spots on 14-15 May, none on 16 May, and down the S flank on 17 May. A short, follow-on plume of ash and gas was visible 29 May. The plume was gone the next day, though a thermal anomaly was observed by satellite on 30 May.
This was the first circumferential dike intrusion ever observed by InSAR and GPS measurements. Such observations allow the examination of subsurface structures like dikes, sills and magma reservoirs. At least two magma reservoirs existed at the time of the 2005 eruption.
The Jan – Apr 1995 eruption was by far the largest of the historic eruptions, putting lava flows into the ocean down the SW flank of the volcano. Lava increased water temperatures and changed the color of the water up to 2 km from the shore. The high water temperature also killed fish. Hydrophones recorded explosions at least 7 km from the point the lava flow entered the water, likely the end of a lava tube down the submerged flank. Initial activity took place high on the volcano flank, near fissures around the rim but migrated down the slope, building a pair of new cones, one low on the flank and one near sea level, both of which erupted lavas along the new flow path.
Lava from a 1961 eruption flowed to the sea from a vent 400 m above sea level on the SE flank. The cloud from this eruption was photographed by an early TIROS weather satellite pass overhead, marking the beginning of satellite monitoring (inadvertent in this case) of volcanic activity in remote parts of the globe.
Lava flows from a 1958 eruption covered the caldera floor and evaporated the intracaldera lake. Total coverage of this lava was 15 km2.
1988 Debris Avalanche
The 14-16 Sept 1988 eruption produced basaltic tephras, lavas, and the largest known debris avalanche from the caldera wall. It appears that a feeder dike emplaced before the eruption destabilized a block in the E caldera wall. This triggered the debris avalanche that immediately exposed the newly emplaced dike, triggering the eruption. Dike propagation may have caused pre-eruption earthquakes in Feb, Apr, May, and finally in Sept. Largest of these was M 5.5.
The avalanche took place around 1.6 hours after a M 4.6 earthquake on 14 Sept. The block separated from the E caldera wall. It was 2 km long by 300 m wide, with a total volume of 0.9 km3. Its deposit on the caldera floor was 250 m thick. The avalanche buried a 110 m high tuff cone on the caldera floor and displaced an intracaldera lake, whose waters disappeared for a time. The eruption began within minutes of the avalanche, producing lava flows and a tephra plume. The plume dissipated within 12 hours. Lava flowed for the next two days.
1968 Caldera Collapse
The 1968 eruption was the largest recorded on Fernandina, a VEI 4. It was the largest known caldera collapse since the 1912 Katmai – Novarupta eruption. The eruption proceeded in two phases, starting with withdrawal of magma below the caldera. This started dropping the floor of the caldera vertically, which immediately exposed open dikes and sills in the caldera rim to water from the intracaldera lake, quickly starting an explosive phreatic eruption. The explosive eruption lasted perhaps 4 days. Meanwhile, magma withdrawal continued for a total of 12 days, as did the collapse of the caldera floor. There was no effusive phase of this eruption. The explosion took place from a small alcove on the western wall of the caldera. Solfataras on the floor of the alcove increased in activity at least a year before the 1968 eruption began.
A precursor eruption took place 21 May around 600 m up the E flank of the volcano, starting with a small vapor cloud. This was 5 km NNW of the 1961 eruption, 6 km E of the caldera. The cloud grew to 4 km, and lava fountained to 70 m. Lava flows covered 10 km2. There may have been earlier activity, as a vapor cloud was observed in the same location a few days earlier. Activity finished by 25 May. There were a few more earthquakes recorded before the beginning of the main eruption on 11 Jun.
The main eruption began with a small quake on 1 Jun quickly followed by a small plume. 4 hours later, the eruption produced a major ash plume at least 22 km high. Observers reported a red glow, flashes, and lightning in the plume. Ash from this plume fell at least 150 km from the volcano. An hour after the ash plume began, infrasonic stations recorded a major explosion. Seismic activity peaked on 19 Jun and mostly ended by 23 Jun. The caldera was 1 -2 km3 larger (deeper) following this eruption and collapse.
The 4-day long explosive part of the eruption is described as hyrdomagmatic (or phreatic) as basaltic magmas interacted with water ponded in the caldera. Over the course of the eruption and for the following week, magma withdrew below the caldera, dropping the caldera floor up to 350 m in places. The caldera floor collapse triggered the explosive eruption, as lake water interacted with exposed magma. The caldera did not appreciably enlarge in diameter. Rather, it got deeper. The caldera floor tilted SW along its ring fault during the collapse. Another oddity is that the shallow lake that covered the caldera floor enlarged as it subsided.
Total volume of the magma withdrawal was in the neighborhood of 1.5 km3, far more than was erupted during the precursor eruption in May. The collapsing wet plume created multiple pyroclastic flows, surges across the western summit dune field. Violence of the eruption also ejected blocks large enough to create impact craters up to 1 km from the vent.
The eruption in late May and the Jun 11 – 14 explosive eruptions were accompanied by minor seismicity. Major seismicity did not start until the caldera collapse (overlapping with the explosive portion of the eruption for a few days). The collapse was the only volcanic event that included major ground deformation.
The caldera collapse in 1968 triggered at least 638 earthquakes over a 9-day period. The quakes are consistent with withdrawal of magma below the caldera. The quakes in the M 5 – 5.4 range were large enough to be sensed worldwide. Larger shocks were concentrated in the early part of the swarm, the first 3 days of the eruption. Seismic data is consistent with a cylindrical block dropping some 300 m in 75 dislocation stages. There were 75 earthquakes larger than M 4.5,
Initial earthquakes were periodic, with large earthquakes occurring in 6-hour intervals, coinciding with tidal extremes. The interval between larger earthquakes decreased in subsequent days, 60 – 90 minutes by the height of activity on 19 Jun. Another oddity was that the larger quakes were preceded by a few minutes by one of more foreshocks. There were no aftershocks. This oddity is thought to be due to frictional resistance by the collapsing blocks. In other words, the blocks break loose as a foreshock, and reach their destination as the main shock. Once they arrive in place, there is no aftershock.
An inspection team reached Fernandina on 19 Jun at the height of the swarm and climbed to the N rim of the caldera. Dust clouds obscured the caldera, earthquakes were nearly continuous as was rockfall noise, which seemed to be concentrated at the S end of the caldera. The team did not linger. They experienced 56 felt tremors during a 6-hour period.
Tectonics
Our 2016 post on the most recent eruption of Wolf and our 2022 post on Darwin are good starting places tectonics associated with the Galapagos. This section will take a look at more recent tectonics over the last 5 Ma. It is a shortened version of what we posted for Darwin.
The Galapagos are tied to the Nazca Plate traveling over the head of a mantle plume located some 1,000 km off the west coast of Ecuador. Rate of travel is around 5 cm/yr. The Galapagos Spreading Center, the boundary between the Nazca Plate to the S and the Cocos Plate to the N, lies some 250 km N of the Galapagos. Most eruptible magma is available at the plume head underneath the spreading center. The farther you get from that ridge, the less eruptible magma there is.
The Nazca Plate is currently moving some 5 cm/year relative to the plume head / hot spot. The Galapagos Spreading Center (GSC) is located some 250 km N of Fernandia Island. There is a neighboring transform fault to the S. The GSC is currently migrating from the Galapagos toward the NE at 4.7 cm/year. The GSC jumped southward at least twice 3.5 – 2.5 Ma and around 1 Ma.
Conclusions
Fernandina is a young, large, actively growing hot spot basaltic shield. It has been very active over the last 2,000 years, completely resurfacing itself, which is hard on local wildlife. While most of its activity is effusive, it does erupt explosively from time to time when magma intrusions interact with the intracaldera lake. Given that the island is all but uninhabited, satellite and emplaced instrumentation will provide the best warning of future eruptions. And there will be future eruptions.
Additional information
Fernandina Island, Galapagos Conservancy
Fernandina Volcano, Galapagos, NASA Earth Observatory
Fernandina, Galapagos Islands, Ecuador, Oregon State University
Fernandina Volcano, Galapagos Islands, P Mouginis-Mark, University of Hawaii
Volcanic eruption in Galapagos Islands, D Geist, Charles Darwin Foundation, Sept 2017
Seismicity of a caldera collapse: Galapagos islands, 1968, Filson, et al, Dec 1973