Iceland: The Formation and Evolution of a Young, Dynamic, Volcanic Island—A Field Trip Guide
Published:August 12, 2019
- PDF LinkChapter PDF
Brennan T. Jordan, Tamara L. Carley, Tenley J. Banik, 2019. "Iceland: The Formation and Evolution of a Young, Dynamic, Volcanic Island—A Field Trip Guide", Iceland: The Formation and Evolution of a Young, Dynamic, Volcanic Island—A Field Trip Guide, Brennan T. Jordan, Tamara L. Carley, Tenley J. Banik
Download citation file:
Iceland is a young, dynamic, volcanic island with a unique tectonomagmatic setting at the intersection of the Mid-Atlantic Ridge and a mantle hotspot. The combination of this tectonomagmatic setting, a history of glaciation, and the sparse vegetation of a subarctic maritime climate means that Iceland has both intriguing geology and outstanding exposures of this geology. This field trip guide contains an introduction to the geology of Iceland and an itinerary for a 10-day journey around the island. The itinerary consists of 55 stops and 15 optional stops. These stops include exposure to representative examples of most phenomena typical of the island’s geology and all of the major tectonic elements of Iceland. The primary focus of this guide is on volcanic and tectonic features, but topics such as glaciation, geothermal energy, geomorphology, paleontology, soil loss, and geotourism are also addressed.
Iceland (Ísland) is a 103,000 km2 island nation (Fig. 1) situated in the North Atlantic Ocean between Greenland to the west, and Norway and the British Isles to the east (Fig. 2). The main island is between 63.4 and 65.5°N. Numerous islands and skerries surround Iceland, most notably the Vestmannaeyjar (Westman Islands) south of the main island. The northern island Grímsey is the only inhabited part of Iceland that crosses the Arctic Circle (though by 2050 the Arctic Circle will migrate to north of the island due to the gradual change in the Earth’s tilt).
The elevation of Iceland ranges from sea level to 2110 m at the peak Hvannadalshnjúkur on the volcano Öræfajökull (16 km from the sea). The only point in the interior of Iceland where the earth’s surface drops below sea level is the lakebed of Þingvallavatn with a minimum elevation of –13 m.
Eastern and northwestern Iceland—the Eastfjords and Westfjords—are characterized by fjords and U-shaped glacial valleys. The peninsulas of north-central Iceland (Skagi, Tröllaskagi, and Flateyjarskagi), separated by wide fjords, are also deeply incised by glacial valleys. In all three of these regions, the dissected peaks and uplands have locally consistent elevations, with upland plateaus being common in the Westfjords. The interior of Iceland is a relatively little-incised volcanic upland punctuated with ridges and mountains created by subglacial and postglacial volcanic eruptions. Glacial ice covers ~10% of Iceland. Ice is mostly distributed in five ice caps greater than 100 km2: Vatnajökull, Langjökull, Hofsjökull, Mýrdalsjökull, and Drangajökull. Each of the four largest of these ice caps covers one or more volcanic systems. Only Drangajökull—located on a plateau in the northern Westfjords—lies outside of the active volcanic zone.
The climate of Iceland is cool and wet, but moderated by the warm North Atlantic current fed by the Gulf Stream. Three Köppen classes are present in Iceland: subpolar oceanic (Cfc); subarctic (Dfc); and mild tundra (ET). Precipitation varies widely across the country; Vík in south Iceland averages 224.7 cm per year, while Akureyri in north Iceland averages 49.0 cm. Temperature variations are more modest, e.g., average daily high for Akureyri in January is 0.9 °C and in July is 14.5 °C, while the average daily high for Reykjavík (southwest) in January is 1.9 °C and in July is 13.3 °C (all climate data are 1961–1990, from the Icelandic Meteorological Office [Veðurstofa Íslands] data).
Humans have been present on Iceland continuously since Norse settlement sometime between 870 and 874 CE (preceded by sporadic discontinuous habitation by Irish monks, the detailed history of which is not known). (All years are common era, CE, unless otherwise noted.) The timing of settlement is associated with a distinctive tephra, the Landnám or Settlement Layer dated at 871 ± 2 in the Greenland summit ice core (Grönvold et al., 1995). The first permanent Norse settler was Ingólfur Arnarson, who established a homestead at the current location of Reykjavík. The settlement of Iceland was recorded in the Book of the Icelanders (Íslendingabók) and the Book of Settlements (Landnámabók). According to the Book of the Icelanders, the first national assembly (Alþingi) was held in the year 930 at Þingvellir, and this year is commonly given as the end of the “Age of Settlement.” The country existed as an independent commonwealth until 1262 when it, by treaty, became part of the Kingdom of Norway. The incorporation of Norway into a union with Denmark in 1380 began a period of Danish rule over Iceland that lasted until the formation of the Republic of Iceland in 1944. The combination of the long continuous history of Iceland and the strong literary tradition of the country provides a resource for modern scientists in the form of non-scientific descriptions of past geologic events (e.g., Þórarinsson, 1967). A comprehensive academic history of Iceland can be found in Karlsson (2000).
Regional Tectonics of Iceland in the North Atlantic
Iceland is positioned at a unique plate tectonic setting. It is the only place on Earth where a mature mid-ocean ridge system rises above sea level. Of course, it does so due to unique circumstances: the coincidence of the Mid-Atlantic Ridge with a mantle hotspot. This makes Iceland a special place to study both mid-ocean ridge and hotspot processes, as well as the interaction between the two. However, this coupled relationship also means that Iceland can’t be taken as typical of either mid-ocean ridge or hotspot processes.
The Mid-Atlantic Ridge segment south of Iceland is called the Reykjanes Ridge. It extends ~1445 km southwest before reaching a major transform zone, the Charlie-Gibbs Fracture Zone (Fig. 2). The Mid-Atlantic Ridge comes on shore in Iceland at the southwestern tip of the peninsula Reykjanes in southwestern Iceland. The plate spreading rate at Iceland is ~19 mm/yr to ~105° (DeMets et al., 2010). The Mid-Atlantic Ridge segment north of Iceland is the Kolbeinsey Ridge. It was named for a small skerry that represents the northernmost territory of Iceland which was critical for establishing Iceland’s territorial water claims. The Kolbeinsey Ridge is ~580 km in length, with two minor transforms, and ultimately connects to the Jan Mayen Fracture Zone. The Kolbeinsey Ridge connects to Iceland by a transform fault system referred to as the Tjörnes Fracture Zone.
As elsewhere, the Mid-Atlantic Ridge north and south of Iceland has a pattern of linear magnetic anomalies that parallel the ridge reflecting seafloor spreading through a sequence of magnetic polarity reversals. The Reykjanes Ridge was the second place where such magnetic anomalies were studied in detail (e.g., Heirtzler et al., 1966; Vine, 1966). These anomalies supported the seafloor spreading hypothesis and contributed to the development of plate tectonic theory.
Both north and south of Iceland, the Mid-Atlantic Ridge is characterized by features called V-shaped ridges (e.g., Vogt, 1971; Jones et al., 2002). These are topographic ridges with a V-shaped map pattern, with the Vs focused on the axial rift and pointing away from Iceland (Fig. 2). They are particularly well developed on the Reykjanes Ridge. The V-shaped ridges are oblique to seafloor magnetic anomalies. The origin of V-shaped ridges is discussed in the context of hotspots, below.
The bathymetry of the ocean floor north and east of Iceland is complex, reflecting more complicated rifting processes than observed in most ocean basins (Fig. 2). The volcanic island Jan Mayen (Norway) lies ~545 km northeast of Iceland and is east of the intersection of the Kolbeinsey Ridge and the Jan Mayen Fracture Zone. The island is at the northern end of the submarine Jan Mayen Ridge, which is commonly interpreted as a rifted microcontinent (e.g., Breivik et al., 2012). The Ægir Ridge, which is east of Iceland and the Jan Mayan Ridge, is an extinct oceanic rift that was active between ca. 52 and 26 Ma after which rifting shifted to the Kolbeinsey Ridge (e.g., Gernigon et al., 2015).
Iceland constitutes the subaerial portion (about one-third) of a larger topographic/bathymetric feature, the “Iceland basalt plateau.” This plateau includes the Iceland shelf above the shelf-slope break at ~500 m. (Note: confusingly, this is distinct from the Iceland Plateau which is the oceanographic term for the deeper seafloor between Iceland and the Jan Mayen Fracture Zone, a plateau relative to the deeper Norway Basin.) The Iceland basalt plateau is one part of the WNW-trending Greenland-Iceland-Faroe Ridge, which comprises the Greenland-Iceland and Iceland-Faroe Ridges, traversing the North Atlantic at this latitude (Fig. 2). The crust of Iceland varies from 15 to 46 km thick (Allen et al., 2002a), and is thickest under the northwest portion of the Vatnajökull ice cap in the area of the Grímsvötn and Bárðarbunga volcanic systems.
The North Atlantic Ocean Basin between Greenland and Eurasia began to open ca. 55 Ma (e.g., Storey et al., 2007). Igneous rocks of the North Atlantic Igneous Province are found on both sides of the Atlantic (Saunders et al., 1997). The initial phase of the North Atlantic Igneous Province (62–56 Ma) preceded basin opening. The largest subaerial exposures of the North Atlantic Igneous Province are plateau lavas of East Greenland (Fig. 3), but exposures are also found in the northern British Isles, the Faroe Islands, West Greenland, and Baffin Island. The largest volume of North Atlantic Igneous Province volcanic rocks is represented by the seaward-dipping reflector sequences on the volcanic rifted margins of Greenland and Eurasia (including the margins of the Vøring and Rockall submarine plateaus). The volume of the North Atlantic Igneous Province is estimated to be 5–10 × 106 km3 (Storey et al., 2007).
The Iceland hotspot is commonly interpreted as a mantle plume, an interpretation dating back to Morgan’s (1971) proposal of the mantle plume model. Seismic tomography studies have imaged a plume under Iceland (Wolfe et al., 1997; Bijwaard and Spakman, 1999; Allen et al., 2002b; Nolet et al., 2007). However, the plume is not conclusively recognized in the lower mantle (cf. French and Romanowicz, 2015). Iceland is considered the archetype for plume-ridge interaction as a ridge-centered plume. The mid-ocean ridge steps eastward through Iceland to center on the hotspot, and the features that accommodate that step and the history of rift relocation are considered in the next section. The plume-track of the Iceland plume is represented by the Greenland-Iceland-Faroe Ridge (Fig. 2), extending in both directions from Iceland because of the persistence of plume-ridge interaction throughout the history of the system. The V-shaped ridges have been interpreted to reflect variations in plume flux, a pulsing plume (e.g., Ito, 2001; Jones et al., 2002; Poore et al., 2009; Parnell-Turner et al., 2017). In the context of the plume model, the North Atlantic Igneous Province could be viewed as representing the plume head phase (e.g., Spice et al., 2015). Some authors, however, propose a longer history for the Iceland hotspot, including Ellesmere Island ca. 120 Ma (Lawver and Müller, 1994), and even the Siberian Traps (e.g., Lawver et al., 2002).
While most tectonic and petrologic research on Iceland interprets the anomalous character of Iceland in the context of the mantle plume paradigm, the plume model is not universally accepted, in general (e.g., Hamilton, 2003; Anderson, 2007) or as applied to Iceland. Foulger and Anderson (2005) and Foulger et al. (2005a) argue that Iceland fails to meet a number of hotspot criteria and that the seismic velocity anomaly below Iceland does not continue into the lower mantle (Foulger et al., 2001). They propose an alternative to the plume model in which the anomalous magmatic productivity of the Mid-Atlantic Ridge at Iceland does not reflect high temperature. Instead, they attribute Iceland’s magmatic productivity to decompression of mantle that is relatively easily melted (fertile) due to contributions of pyroxenites from Iapetus oceanic crust subducted during the Caledonian orogeny.
Tectonic Features of Iceland
The nomenclature of tectonomagmatic features in Iceland is somewhat inconsistently handled. Varied names are applied to the same feature in the literature. We will generally follow the nomenclature of Thordarson and Höskuldsson (2008) (Table 1), but recognize the variety in nomenclature from other sources (cf. Sigmundsson et al., 2018). We introduce the neovolcanic zones and volcanic systems here to elucidate the tectonic system of Iceland, but these features are discussed in a petrologic and volcanological context in the next section. The neovolcanic zones of Iceland (Fig. 4) are areas of upper Pleistocene (<0.78 Ma, Bruhnes magnetic epoch) and post-glacial volcanism (Sigmundsson et al., 2018). The neovolcanic zones can be separated into the rift zones, which represent the mid-ocean ridge system onshore, and the flank zones where volcanism is not clearly associated with rifting. Following the usage of Thordarson and Höskuldsson (2008) there are 30 active volcanic systems in the neovolcanic zones of Iceland (Fig. 5); the number of volcanic systems reported in the literature varies with the lumping/splitting decisions of various authors.
The Mid-Atlantic Ridge, represented by the Reykjanes Ridge segment south of Iceland, comes on shore at the Reykjanes volcanic system. This is one of four volcanic systems that constitute the Reykjanes Volcanic Belt, also called the Reykjanes Peninsula Oblique Rift (e.g., Sigmundsson et al., 2018). Fissure swarms of the individual volcanic systems trend NE, while the Reykjanes Volcanic Belt as a whole trends east-west. This is oblique to the overall west-northwest (~295°) relative plate motion between the North American and Eurasian plates. Continuing north, two parallel neovolcanic zones accommodate rifting: the Western and Eastern Volcanic Zones. These parallel rifts are linked by the South Iceland Seismic Zone in the south and the Mid-Iceland Belt (also called the Hofsjökull Volcanic Zone and Central Rift Zone) in the north. The Mid-Iceland Belt links up with the Northern Volcanic Zone which continues to the north coast of Iceland. The Northern Volcanic Zone is reconnected to the Mid-Atlantic Ridge, as the Kolbeinsey Ridge segment, by the Tjörnes Fracture Zone. The area of central Iceland bound by the Western Volcanic Zone, Eastern Volcanic Zone, South Iceland Seismic Zone, and Mid-Iceland Belt is sometimes called the Hreppar microplate (Fig. 4).
Earthquakes are common in Iceland but the maximum recognized magnitude is ~M7 (Sigmundsson et al., 2018). The strongest earthquakes (M6–7) occur only in the two transform zones, the South Iceland Seismic Zone and Tjörnes Fracture Zone. The South Iceland Seismic Zone and Tjörnes Fracture Zone are of vastly different tectonic styles (Gudmundsson, 2007). The South Iceland Seismic Zone trends east-west. Left-lateral strike-slip shear across the zone is accommodated by rotation of blocks within the zone. These blocks are separated by north-south–trending right-lateral strike-slip faults, resulting in “bookshelf” type faulting (Einarsson, 2010). Similar geometries of bookshelf faulting are also recognized within the Reykjanes Volcanic Belt (Clifton and Klattenhorn, 2006; Einarsson et al., 2018). The most recent >M6 earthquakes in the South Iceland Seismic Zone were two M6.5 separated by four days in 2000, and an M6.3 in 2008.
The distribution of seismic activity in the Tjörnes Fracture Zone defines three distinct concentrations (Sigmundsson et al., 2018). These are the Húsavík-Flatey fault, which is the main transform; the Grímsey Oblique Rift (which includes four offshore volcanic systems) to the north; and the seismically less active Dalvík lineament to the south (Fig. 4). The most recent M6–7 earthquake in the Tjörnes Fracture Zone was a M6.4 in 1976.
The main representation of the mid-ocean ridge through Iceland is sometimes generalized as the Reykjanes Volcanic Belt and Western Volcanic Zone connecting to the Northern Volcanic Zone through the Mid-Iceland Belt (Fig. 4). However, geodetic measurements demonstrate that far more displacement is partitioned to the Eastern Volcanic Zone (90%–65%, decreasing southward) than the Western Volcanic Zone (Islam et al., 2016).
The flank zones are the Snæfellsnes Volcanic Belt in the west and the Öræfi Volcanic Belt in the southeast (Fig. 4). The Snæfellsnes and Öræfajökull zones are entirely removed from the active rift system. Their tectonic origin is enigmatic. Both have been interpreted to be influenced by the Iceland mantle plume (e.g., Prestvik et al., 2001), with the timing of rejuvenation of the distal Snæfellsnes Volcanic Belt possibly reflecting a pulsing plume (Walters et al., 2013). The east-west trend of the volcanic systems in the Snæfellsnes Volcanic Belt could reflect either the geometry of the earlier Skagi-Snæfellsnes Rift Zone (Walters et al., 2013) or structural control (e.g., Sigurdsson, 1970; Karson, 2017). The Eastern Volcanic Zone south of the South Iceland Seismic Zone is sometimes considered a flank zone, the South Iceland Flank Zone, extending southward to Vestmannaeyjar (e.g., Sæmundsson, 1979; Sigmundsson et al. 2018). It may represent the propagation of a new rift zone south of the Eastern Volcanic Zone, and over time it may be expected to take over rifting from the Reykjanes Volcanic Belt and northern Reykjanes Ridge. Such a postulated future rift relocation would repeat a pattern discerned in the geologic past of Iceland.
The current position of rifting in Iceland is displaced eastward by up to 150 km from the ridge segments immediately to the north and south (Fig. 2). The most widely accepted model for the evolution of rifting in Iceland posits that the current rift configuration is the product of three rift relocation events (Fig. 4). The earliest rift for which there is on-land evidence is the Northwest Rift Zone (Hardarson et al., 1997). It is interpreted to have been active from 24 to 15 Ma, and its axis is postulated to lie offshore of the Westfjords in northwest Iceland. Rifting then shifted to the Skagi-Snæfellsnes Rift Zone, which is also known as the Snæfellsnes Rift Zone, Húnaflói Rift Zone and other permutations of these names. The Skagi-Snæfellsnes Rift Zone was active from 15 to 7 Ma before it was abandoned in favor of a rift system consisting of the Western Volcanic Zone and Northern Volcanic Zone at 6–7 Ma (e.g., Martin et al., 2011). The Eastern Volcanic Zone propagated south of the Northern Volcanic Zone and began to take up significant rifting at 2–3 Ma.
Evidence for the location of abandoned rift zones includes:
radiometric ages, paleomagnetism, and aeromagnetic signatures;
regional synclines similar to those observed for the active rifts; and
angular unconformities between lavas produced in one rift zone and those produced in another, with both subject to regional tilting as per #2.
Regional synclines are produced because of high magmatic productivity relative to the slow spreading rate. Iceland has the spreading rate of a slow-spreading mid-ocean ridge, but the magmatic productivity of a fast-spreading ridge (Karson, 2017). This combination causes loading and subsidence along the rift axis, as modeled by Pálmason (1980, 1986) (Fig. 6). Tilting associated with this process is evident in many parts of the Neogene lava sequence (Bodvarsson and Walker, 1964). Regional anticlines, e.g., the Borgarfjörður anticline, occur between the synclines.
Alternative models for the history of rifting in Iceland have been proposed. For example, Garcia et al. (2003) interpreted the Northern Volcanic Zone to have initiated at 8.1 Ma. They rejected the conventional axis of the Skagi-Snæfellsnes Rift Zone, and suggested instead an alternative paleo-rift south and west of Skagafjörður, which they interpret to have remained active until ca. 3 Ma. Foulger (2006) argues that two parallel rifts have been active in Iceland since at least 26 Ma. Foulger (2006) and Torsvik et al. (2015) also propose that continental crust, an extension of the Jan Mayen micro-continent, may underlie part of east Iceland.
Distribution and Diversity of Volcanic and Intrusive Rocks
Iceland’s unique geologic setting—a coupled spreading center and mantle hotspot—yields a highly active volcanic area with a more varied array of magma types compared to other ocean islands globally. Geochemical diversity in Icelandic rocks is strongly tied to the geographic and geologic locations of volcanism—the volcanic zones. Modern Iceland is nominally divided into six volcanic zones and the Reykjanes Ridge, each of which is magmatically active (Fig. 4; Table 1). These currently active areas are collectively referred to as the neovolcanic zone.
The neovolcanic zone, and its probable precursors, closely follows the location of active rifting areas (Figs. 4 and 5). However, at least two areas—the SVB and ÖVB—comprise active volcanism that lies in flank zones off the main rift axes. Magmatism corresponding to the main rift axes is fairly straightforward to assess, but there is more debate about the mechanism(s) of magma production in the flank zones (see “Origin of Silicic and Intermediate Magmas”).
Icelandic volcanoes have traditionally been considered in the framework of volcanic systems. This is due to a number of factors, including the strong control rift zone geodynamics have on determining locations of volcanism, uncertainty about what constitutes an “active” volcano, and edifice erosion due to glaciation. Volcanic systems are composed of either a fissure swarm or a central volcano edifice, or a combination of both (e.g., Thordarson and Larsen, 2007). While fissures commonly occur without a corresponding central volcano, the vast majority of central volcanoes have an accompanying fissure swarm. These surface features generally reflect the nature of their underlying magmatic systems. Fissures are typically fed by deep (>10 km), elongated, transient magma reservoirs near the base of the crust, while central volcanoes are manifestations of shallower (<10 km), more equant, longer-lived magma reservoirs in the mid-to-upper crust. Central volcanoes are polygenetic—they result from repeated eruptions from the same vents confined to a relatively small geographic area. Central volcano morphologies in Iceland include shields, stratovolcanoes, calderas, and combinations thereof, in addition to subglacially erupted formations (see “Subglacial Volcanism”). Individual fissure eruptions tend to produce monogenetic cones and flows, and rarely occur from the same vents as previous eruptions; however, repeated eruptions from the same fissure swarm are common.
Magmatic systems that feed central volcanoes are longer-lived and therefore capable of producing more compositionally variable magmas compared to those produced by basaltic fissure eruptions (see sections on magma production and composition). Basalt dominates the eruptive volumes of most central volcanoes, but central volcanoes also produce subordinate amounts of higher-Si magmas—primarily rhyolite, but also andesite and dacite (see “Origin of Silicic and Intermediate Magmas”). The ability to produce silicic magmas, which have fundamentally different rheologies and volatile compositions than basalts, leads to a wider array of eruptive products (e.g., ignimbrites, fall deposits, and domes) at central volcanoes than at fissures. These diverse magma compositions and eruptive products contribute to central volcano edifice building.
Historical Icelandic Eruptions
Eruptions of Icelandic volcanoes have persisted throughout human inhabitation of the island. In many cases, eruptions influenced major societal events or provide key benchmarks for such events. For example, the explosive eruption of Vatnaöldur ca. 870 yielded the “Settlement Layer” of tephra that is used to help establish dates of permanent settlement in Iceland (Larsen, 1984). Icelandic lore points to a lava eruption in the Ölfus district, possibly at Hengill, as instrumental in facilitating peaceful conversion to Christianity from paganism ca. 1000 (Thordarson and Höskuldsson, 2014). More recently, the 1973 eruption of Eldfell dramatically reshaped the island of Heimaey, its community, and its harbor. Rates of recent eruptions suggest that, on average, an eruption occurs every 3–5 years (Thordarson and Höskuldsson, 2008), and some volcanic systems are clearly more active than others. A brief overview of notable Icelandic volcanic systems active in historic times (the past ~1100 years) and their major eruptions follows. This list is dominated by systems in the Eastern Volcanic Zone, which account for ~80% of all recorded volcanic events by frequency and volume (Thordarson and Larsen, 2007).
The notable 1875 Plinian eruption of Askja was part of the larger Askja Fires magmatotectonic events that lasted from 1874 to 1929. Volcanic activity in the Askja system initially occurred in response to a crustal rifting episode, resulting in up to 70 km lateral flow of magma within the crust and eventual eruption of ~0.3 km3 of basaltic lavas in the northern part of the system (Sigurdsson and Sparks, 1978, 1981). The March 1875 Plinian eruption in the Askja central volcano lasted for two days and produced a caldera—now filled with the caldera lake Öskjuvatn—and 0.2 km3 (dense-rock equivalent) of rhyolite tephra. Subsequent basalt eruptions occurred intermittently in the vicinity of the newly formed caldera until 1929. Another eruption in 1961 on the caldera rim produced a further 11 km2 basaltic lava field.
The 6-month-long eruption at Holuhraun (August 2014–February 2015) in the Bárðarbunga-Veiðivötn volcanic system was the largest effusive eruption in Iceland since the 1783–1784 Laki eruption (Pedersen et al., 2017). The lava flow field covered ~84 km2 and has an estimated bulk (i.e., including vesicles) volume of ~1.5 km3 (Pedersen et al., 2017; Gudmundsson et al., 2016). The flow resulted from the collapse of Bárðarbunga caldera (Riel et al., 2015; Gudmundsson et al., 2016) and was fed by propagating dikes traveling north along the fissure swarm (Reynolds et al., 2017).
Eyjafjallajökull is a moderately active (~15 eruptions in the past 11 k.y.; Thordarson and Höskuldsson, 2014) volcano that is capped by a glacier of the same name. It became globally known in spring 2010 when new magma injection triggered a series of eruptions, first on the flank of the volcano at Fimmvörðuháls, then at the summit crater (Sigmundsson et al., 2010). The Fimmvörðuháls activity was limited to basaltic fire fountains and flows that persisted from 20 March to 12 April. On 14 April, an eruption of trachyandesite started from the summit crater and quickly melted through the glacier, resulting in an explosive phreatomagmatic eruption and production of very fine ash. This ash was blown south and east by prevailing winds and impacted European air traffic for two weeks cumulatively during the explosive phases of the eruption, which lasted until early June (Gudmundsson et al., 2010b).
Grímsvötn is a sub-Vatnajökull central volcano caldera located directly above the center of the Iceland hotspot. It is Iceland’s most historically active volcano, with more than 70 eruptions in the past 1000 years (Guðmundsson and Björnsson, 1991). Despite producing basaltic magmas, Grímsvötn’s eruptive behavior ranges from entirely effusive and contained under the ice (such as at the Gjálp fissure in 1996) to highly explosive, with an eruption column over 20 km high (as in 2011). This wide range of behaviors results from volcano-ice interaction: addition of meltwater to magma leads to magma fragmentation, either with subsequent effusive behavior (if confining pressure is sufficient) or explosive eruption (if confining pressure is low) (see section on subglacial volcanism). In addition to being highly active, Grímsvötn deserves particular attention because of its intimate association with glacial outburst floods known as jökulhlaups. Jökulhlaups are one of the most severe hazards associated with Icelandic volcanoes. Eruptions of Grímsvötn have both caused jökulhlaups, and occurred as a result of them. For example, the 1996 Gjálp eruption caused a jökulhlaup that had a peak discharge estimated at 45,000 m3/s and destroyed the Route 1 bridges over Gigjukvísl and Skeiðará in Skeiðarársandur (e.g., Björnsson, 2003, 2010). The 2004 eruption in Grímsvötn caldera began when the caldera lake (also called Grímsvötn) drained, which reduced pressure on the crust and facilitated magma movement and eruption (e.g., Björnsson, 2003, 2010).
Hekla is perhaps the most historically famous of all Icelandic volcanoes. Its frequent explosive eruptions prompted medieval Europeans to nickname Hekla the “Gateway to Hell,” and the solitary edifice towers over the relatively monotonous surroundings. Hekla’s unusual morphology as an elongated stratovolcano with a summit fissure results from variable eruptive styles and its location at the junction of transform and rifting tectonic boundaries. Hekla’s ~13 km3 of erupted material from 18 eruptions places it third on the list of “big four” magmatic systems that dominate historic eruptive volumes (the other three are Katla, Grímsvötn, and Veiðivötn). Hekla erupts two dominant magma types (and mixing products thereof): basaltic-andesite (53 wt% SiO2) and rhyolite (74 wt% SiO2), both of which are stored in a >8 km-deep magma chamber (Thordarson and Höskuldsson, 2014). Rhyolite has erupted from Hekla in highly explosive eruptions many times. Hekla eruptions from 3100 (H3), 4200 (H4), and 7000 (H5) years ago, along with the 1104 eruption, dispersed such large volumes of tephra over the majority of Iceland that they form the references used in tephrochronology studies throughout the country (Larsen and Eiríksson, 2008). Hekla eruptions commonly start with a short (<1 h) silicic tephra-producing explosive phase, followed by eruption of andesite along the summit fissure that extends the length of the edifice (Baldridge et al., 1973). Eruptions of Hekla are notable for their short period of eruptive precursors (Einarsson, 2018)—don’t wait for an earthquake before evacuating!
Katla is the better-known name of Mýrdalsjökull central volcano, Iceland’s second-largest volcano and its most productive in historical times (Thordarson and Larsen, 2007). Katla central volcano is entirely covered by Mýrdalsjökull (Iceland’s fourth-largest glacier), but parts of the Katla fissure system extend beyond the ice, including the famous Eldgjá fissures. Katla erupts a variety of magma types from basalts to rhyolites, but almost always erupts explosively due to interaction with glacial meltwater. The resulting phreatomagmatic eruptions produce copious amounts of tephra and high-discharge jökulhlaups. Jökulhlaups during the most recent eruption of Katla, in 1918, added 14 km2 of new material to the sandur south of the volcano and flood waters traveled >36 km/h (Tómasson, 1996). The 934–940 Eldgjá Fires eruption from the fissures northeast of the Katla central volcano is the most voluminous eruption in the past 1100 years. It erupted ~20 km3 of magma, mostly as basaltic lava flows, and released an estimated 220 Mt of SO2 (Thordarson et al., 2001). While information on the effects of this extreme atmospheric pollution are lacking, it is likely that this eruption had environmental effects similar to the Laki 1783 eruption (see below), but spread over a longer time, since the Eldgjá eruption lasted for ~6 years (versus Laki’s ~8 months).
Krafla is a unique volcanic system due to its location straddling the Northern Volcanic Zone. The ~9 × 7 km caldera is visibly split by its accompanying 100-km-long fissure swarm that occupies the middle of the Northern Volcanic Zone. The two most notable periods of volcanic activity in the Krafla system, the 1724–1729 Mývatn Fires and the 1975–1984 Krafla Fires, coincided with rifting activity. The Mývatn Fires produced several basaltic lava flows, small explosive cones, and opened 13 km of new rift during intermittent eruptions. The Krafla Fires produced several basaltic flows between 1975 and 1984 associated with numerous rifting episodes. Both Fires resulted in ~9 m of rift extension (Tryggvason, 1984; Hollingsworth et al., 2012). Krafla has the highest surface heat flow in Iceland (300–350 mW/m2) due to its proximity to the rift axis (Hjartarson, 2015), and is the site of one of Iceland’s geothermal power stations.
Lakagígar, or Laki, is a row of cones erupted from a fissure swarm in the Grímsvötn volcanic system beginning in June 1783. The Laki flood basalt eruption lasted 8 months and produced over 14 km3 of new lava and 0.4 km3 (dense rock equivalent) of tephra (Thordarson et al., 1996). The 1783 Laki eruption is well-known for several reasons: (A) its events and effects were well-documented by local reverend Jón Steingrímsson, whose writings provide an unparalleled glimpse into the evolution of the eruption; (B) the Laki eruption had a notorious effect on Icelanders, either directly or indirectly killing more than 20% of the Icelandic population; and (C) the eruption released over 100 Mt of SO2, which turned into more than 200 Mt of H2SO4 once it mixed with atmospheric water. This pollution covered the Northern Hemisphere with an acidic fog for half a year, blocked incoming solar radiation, and ultimately led to a 1.3 °C temperature drop in the Northern Hemisphere that lasted for several years (e.g., Sigurdsson, 1982).
Öræfajökull is Iceland’s tallest, and volumetrically third-largest, volcano. In 1362, it produced Iceland’s largest eruption of explosive rhyolite in historical times. This eruption is estimated to have erupted ~10 km3 of tephra over only 1–2 days, but due to westerly winds, the majority of the tephra accumulation occurred in the sea (e.g., Thórarinsson, 1958; Selbekk and Trønnes, 2007). Parts of eastern Iceland near the volcano were devastated by the tephra accumulation from this eruption and the related jökulhlaups that ensued. What was once a prosperous farming area became a wasteland—öræfi in Icelandic—and the area was abandoned for decades afterward. Öræfajökull erupted again in 1727, but the eruption was confined to the summit caldera and resulted in only minor jökulhlaups (Thórarinsson, 1958).
Plateau Lavas—The Tertiary Basalt Formation
The dominance of basaltic fissure eruptions throughout Iceland’s history is difficult to overstate. Plateau lava eruptions over the past ~25 m.y. constructed the thickened crust that the island of Iceland caps. Tholeiitic basaltic lava flows account for more than 80% of the erupted volume during Iceland’s ~16 m.y. subaerial history. These “plateau building” flows (sometimes called the “volcanic pile”) are repetitive, gently dipping, and laterally extensive. They are easily observed in the Tertiary (a.k.a. Neogene, but Icelandic geology traditionally uses Tertiary) and Plio-Pleistocene bedrock exposed in cross section by glacial erosion in the eastern and western areas of the country. The cumulative thickness of the Tertiary-age basalts is close to 10 km (e.g., Walker, 1964, 1974), and the vertical accumulation of volcanic and sedimentary rocks is coupled with outward spreading of the volcanic pile. Interbedded within the basalts are fossil-containing paleosols that are often red in color and provide information about Iceland’s much warmer past (Fig. 7).
Older Central Volcanoes
As the volcanic plateau of the Tertiary Basalt Formation formed, central volcanoes grew in places with concentrated eruptive vents that allowed repeated eruptions. The majority of central volcanoes older than Holocene have been variably eroded by repeated glaciation, but remnants of their eruptive products and underlying magmatic systems remain accessible (Fig. 8). In addition to providing a glimpse into the petrogenesis of silicic and basaltic magmas throughout the majority of Iceland’s history, these older central volcanoes also provide a template by which to understand Neovolcanic systems. The oldest central volcanic systems are exposed at the western (e.g., Hrafnsfjörður, ca. 14 Ma; Banik, 2015) and easternmost (e.g., Breiðuvík, ca. 13 Ma; Carley et al., 2017) edges of Iceland and generally young toward the Neovolcanic zone, in accordance with Iceland’s crustal evolution. Studies of old central volcanoes confirm that these systems may remain active for upward of 2–3 m.y. (Carley et al., 2017; Banik et al., 2018), and their magmas follow the tholeiitic trend, with one exception. Króksfjörður central volcano in the Westfjords contains copious tholeiitic basalts and rhyolites, but also minor amounts of calc-alkaline dacites thought to form from late, low-temperature crustal melting (e.g., Jónasson et al., 1992). In southeastern Iceland, central volcanic systems experienced such extensive erosion that only the intrusive complexes associated with the volcanic systems remain (e.g., Austurhorn, Furman et al., 1992a; Padilla et al., 2016). These exposed intrusive complexes provide glimpses into the processes operating several kilometers deep in the crust. Pivotal studies regarding the structure of central volcanoes, the importance of basalt and rhyolite dikes in melt transport, and thorough petrogenetic evaluations were undertaken in eastern Iceland by Walker (1958, 1963, 1964, e.g.) and Carmichael (1964, 1967, e.g.) that provide additional background on Tertiary central volcanoes.
Origin of Basaltic Magmas
Basalts are generated via partial melting of the underlying mantle. Due to Iceland’s unique position at the junction of the Mid-Atlantic Ridge and the Iceland hotspot, mantle melting under Iceland is more complicated than at either feature alone. At mid-ocean ridges and other spreading centers, thinning of the crust due to divergence lowers the pressure exerted on the underlying mantle, allowing the upper mantle to partially melt. That melt buoyantly rises and either erupts or infiltrates the crust at the spreading center, thereby creating new crust. Mantle hotspots produce magmas through decompression melting as well, but because of the elevated temperature in the ascending hotspot mantle, melting generally begins at greater depths and yields a greater melt volume than at spreading centers. Radiogenic isotope compositions of Icelandic rocks indicate the presence of at least four mantle components that contribute to producing Icelandic magmas (e.g., Zindler and Hart, 1986; Sigmarsson et al., 1992; Hart et al., 1992; Chauvel and Hémond, 2000; Prestvik et al., 2001; Thirlwall et al., 2004; Kokfelt et al., 2006; Kitagawa et al., 2008; Peate et al., 2010). In Iceland, the combination of hotspot and ridge leads to a melt production rate an order of magnitude higher than that observed at mid-ocean ridges.
The average rate of divergence of the rift system on Iceland is approximately that of the submarine portions of the rift (~2 cm/year), indicating that the extra magma flux attributable to the hotspot influence largely goes to crust-building processes. Icelandic crust varies in thickness from ~15 km at Reykjanes in the southwest to >40 km over the plume head (Darbyshire et al., 2000), which is considerably thicker than Atlantic Ocean crust where it is free of hotspot influence (~7 km) (Wang et al., 2011). The thick Icelandic crust facilitates production of a much wider variety of magma compositions than is typical for ocean islands (see next section on silicic magmas), and the ridge–hotspot combination allows for a greater range of basaltic melt compositions than is typically seen in ocean islands.
In Iceland, three series of magmas are traditionally recognized: tholeiitic, alkalic, and transitional alkalic (Fig. 9) (Jakobsson et al., 2008). Each of these magma series primarily results from different mantle melting depths and materials. Tholeiites result from shallow melting of chemically depleted mantle—mantle that has partially melted previously and is relatively free of incompatible elements that partition into earlier-formed melts. Radiogenic isotope compositions indicate that a crustal component, derived from altered basalts, is evident in evolved tholeiites and indeed in most Icelandic basalts. However, distinguishing between contamination of fresh mantle melts via hydrothermally altered crust versus melting of recycled oceanic crust remains non-trivial (Sigmarsson and Steinthórsson, 2007). Magmas formed in the active, established rift zones—including the vast majority of the Tertiary sequence—are tholeiitic (e.g., Hardarson and Fitton, 1997).
Alkalic and transitional magmas form from lower degrees of melting of deeper, more chemically enriched and heterogeneous mantle (e.g., Peate et al., 2010). Since the source mantle has experienced little, if any, previous melting, incompatible elements are present and partition into the melt. This results in a magma with higher amounts of Na, K, and other incompatible elements than observed in tholeiites (e.g., Sigmarsson and Steinthórsson, 2007). Alkalic magmas may also result from small degrees of partial melting of the lower crust that mixes with more depleted magmas (e.g., Óskarsson et al., 1985; Steinthorsson et al., 1985). However, debate exists as to the importance of this process. For example, the Vestmannaeyjar at the propagating tip of the Eastern Volcanic Zone are characterized by alkaline rocks that are not the result of assimilation of lower crustal melts (e.g., Furman et al., 1991; Mattsson and Óskarsson, 2005). Alkalic rocks are restricted to the Vestmannaeyjar and the Snæfellsnes off-rift area (Jakobsson et al., 2008). Transitional alkalic magmas compositionally lie between tholeiites and alkaline magmas. They occur in volcanic systems at the propagating arm of the Eastern Volcanic Zone (e.g., Tindfjallajökull, Eyjafjallajökull, Hekla, Katla, and Torfajökull) and the off-rift Öræfi Volcanic Belt (Jakobsson et al., 2008).
Origin of Silicic and Intermediate Magmas
Despite its position at the junction of a mid-ocean ridge and an oceanic hotspot (traditionally basaltic settings), Iceland is home to a surprising abundance of silicic igneous rocks. In a survey of Icelandic rocks from the late Pleistocene and Holocene, ~14% of rocks are estimated to be intermediate and 11% silicic, with the remaining 75% basaltic (Jakobsson et al., 2008). The Icelandic crust as a whole is thought to consist of ~10% silicic rocks. This makes Iceland a global anomaly as home to the greatest known concentration of silicic rocks in the modern ocean (e.g., Walker, 1966; Gunnarsson et al., 1998; Jónasson, 2007). Silicic magmas are concentrated at central volcanoes, where the relative abundance of silicic material increases to ~20%–30% of erupted material, and sometimes much more (e.g., ~50% at Breiðuvík-Kækjuskörð: Carley et al., 2017; and >80% at Torfajökull: Walker, 1963; Gunnarsson et al., 1998).
This abundance of silicic rocks in an oceanic environment, which formed in the absence of substantial influence of preexisting continental crust, has motivated many researchers to consider Iceland to be a modern analogue for Early Earth (e.g., Hadean and Archean) proto-continental crust construction (e.g., Martin and Sigmarsson, 2005; Martin et al. 2008; Willbold et al., 2009; Reimink et al., 2014). Further investigations of Iceland’s rare silicic rocks may result in better understanding of the processes that initiated Earth’s earliest continental-like crust. While this is a tantalizing idea, a comparison of Hadean and Icelandic zircons (Carley et al., 2014) revealed more differences than similarities in oxygen isotopes and trace elements. This suggests very different types of petrogenetic processes, source materials, magmatic temperatures, and compositional evolution.
Though the appropriateness of Iceland as a modern analogue for the Early Earth has been called into question, the abundance of silicic magma in Iceland has been, and continues to be, the subject of much debate and study (e.g., Carmichael, 1964; Wood, 1978; Sigurdsson and Sparks, 1981; Macdonald et al., 1987, 1990; Nicholson et al., 1991; Sigmarsson et al., 1991; Furman et al., 1992a, 1992b; Jónasson, 1994; Prestvik et al., 2001; Martin and Sigmarsson, 2007, 2010; Bindeman et al., 2012; Charreteur et al., 2013; and Carley et al., 2014; to name a few). One explanation for the relative abundance of silicic magma is that a sliver of ancient crust might be tectonically pinned beneath Iceland. If present, this pinned crust might influence some of the island’s unusual characteristics, as discussed in the Tectonic Features of Iceland section (e.g., overly thickened crust, abundance of felsic material, heterogeneous isotope compositions: Foulger et al., 2005b; Martin et al., 2011; Torsvik et al., 2015). However, it is possible to explain the abundance of silicic magma in Iceland using the local mantle and young mafic crust, without relying on contributions of ancient crust. Two end-member mechanisms are often called upon to explain the abundance of rhyolite in Iceland from mafic sources. One is extreme fractional crystallization of mafic magma, and the second is partial melting or assimilation of hydrothermally altered crust.
Choosing just one mechanism to explain all of the felsic magma in Iceland is limiting and fraught with contradiction. When considering individual central volcanoes, or even individual eruptions of rhyolite from a single central volcano, there will be instances when one end member process is more compelling than the other. For example, uniquely low oxygen isotope signatures at Torfajökull and Krafla (in volcanic glass, major phenocrysts, and zircon) suggest strong influence of hydrothermally altered crust in the rhyolite generating process (e.g., Bindeman et al., 2012; Carley et al., 2017). Meteoric water interacting with rock at high temperature is the most straightforward way to explain a depletion of oxygen isotope ratios away from expected mantle values. Partially melting (or assimilating) that altered rock effectively imprints the low δ18O signature on rhyolitic magmas and rocks. Alternatively, compositionally homogeneous rhyolites with near-mantle oxygen isotope signatures exist in Iceland, as are observed in the 1362 eruption of rhyolite at Öræfajökull. These data are more easily explained by extreme fractional crystallization from a mafic magma (e.g., Selbekk and Trønnes, 2007; Schattel et al., 2014). An intriguing explanation proposed by Martin and Sigmarsson (2007) is that the local tectonic setting of a central volcano plays a role in the origin of silicic magmas. Fractional crystallization may dominate off-rift where mafic magma is injected into cool crust, whereas partial melting may dominate on-rift where mafic magma is injected into fractured, altered, heated crust. Subsequent studies (e.g., Bindeman et al., 2012; Schattel et al., 2014; Carley et al., 2017; Banik et al., 2018) support the notion that oxygen isotope signatures are more suggestive of assimilation and partial melting on-rift relative to off-rift.
Intermediate magmas are less common than silicic magmas in most central volcanoes in Iceland, with ~95% of historically erupted intermediate compositions originating at Hekla (Thordarson and Larsen, 2007). This intermediate magma has primarily been attributed to the compositional mixing of mafic and silicic magmas (e.g., Sigmarsson et al., 1992; Sverrisdottir, 2007), perhaps accompanied by subsequent fractionation. In many central volcanoes, separate trends have been interpreted to reflect the occurrence of both fractionation-dominated and mixing-dominated origins for intermediate magmas in the same system (e.g., Charreteur et al., 2013; Jordan et al., 2013). Many eruptive products, combined with rare intrusive exposures, provide insights into the complicated mafic and felsic interactions that take place in active magma bodies beneath central volcanoes (e.g., mixed and mingled intrusive rocks at Austurhorn, Day 6 of this guide; mixed pumice at Askja, Day 7 of this guide).
Volcanic eruptions under ice are common in Iceland, but they are fundamentally different from their subaerial counterparts due to the influence of large volumes of meltwater and variable confining pressure regimes. In the past ~3 m.y., Iceland has experienced conditions from complete ice cover (Einarsson and Albertsson, 1988) to virtually ice-free, leading to the production of subglacial landforms of Plio-Pleistocene and younger age across Iceland. Approximately 60% of historical eruptions in Iceland are from fully or partially ice-capped systems (Larsen, 2002). Magma, ice, and meltwater combine to generate a variety of eruptive products depending on availability of these components and confining pressure. Pillow lavas form when eruptions initiate under ice 100s of m thick and the confining pressure is sufficient to suppress explosivity. This is also true if eruptions occur beneath substantial meltwater or overlying eruptive products. Hyaloclastite forms in lower pressure environments, when magma quenches and explosively shatters upon encountering significant volumes of meltwater. The decreased pressure environment may be due to eruption under a thin or unconsolidated ice sheet or from meltwater starting to drain from a thicker glacier. Subaerial explosive eruptions result after the confining ice sheet is breached, leading to large volumes of meltwater interacting with shallow magma and tephra-forming phreatomagmatic fragmentation (Gudmundsson et al., 1997).
Subglacial volcanic eruptions produce tuyas and tindars—landforms unique to subglacial formation environments—and commonly lead to jökulhlaups. Tuyas are flat-topped mountains that form in a sequence of pillow lava; tuff cone; slope failure; hyaloclastite delta; and subaerial lava cap (Fig. 10). Depending on the evolution of the edifice and the magmatic system, tuya-building may cease at any time during this sequence. Pillow lavas result from hydrostatic pressure of the overlying ice and water exceeding the gas pressure of the magma. The pillows typically form on the 1 m scale and tend to pile up proximal to the conduit. Hydromagmatic eruptions start when volatiles in the magma exsolve, yielding hyaloclastite shards that form a tuff cone on top of the pillow lava. The slopes of the new tuff cone are commonly unstable and easily fail, especially if earthquakes or ice movement occur. Finally, continued eruption will melt through the glacier, allowing the top of the tuya to be capped by subaerial lava and a lava-fed delta. Although their size is dependent on the volume and frequency of eruptions, modern tuyas are typically less than 10 km in diameter (Allen, 1980). Tuya heights are generally indicative of the size of the glacier they formed under. Subglacial portions of a tuya will never be much more than several 10s of meters higher than the maximum glacial height, with only the lava cap representing subaerial eruption. Basaltic eruptions beneath thin ice (<150 m) produce a discrete set of lithofacies due to the different mechanical properties of thin glaciers. Here, meltwater tends to readily drain away from the eruption site, and pillow lavas are rare (Smellie et al., 1993; Smellie and Skilling, 1994; Smellie, 1999).
Hyaloclastite ridges called tindars are similar to tuyas, but are elongated instead of conical, and have only been described on Earth in Iceland (Chapman et al., 2000). Tindars result from fissure eruptions under ice. They form from the identical eruptive sequence as tuyas and have the same internal morphology. Tindar ridges are approximately the same width as their conical counterparts, but are elongated many 10s of km due to their fissure-fed formation (Chapman et al., 2000).
While the conditions for the lithofacies observed in tuyas and tindars listed above generally apply across all magmatic systems, it is important to note that subglacial eruptions of basalt and rhyolite are capable of producing contrasting styles of deposits (Tuffen et al., 2002; Tuffen, 2011). Subglacial rhyolitic tuyas are widespread in Iceland, occurring at central volcanoes (e.g., Torfajökull, Kerlingarfjöll) and fissure zones (e.g., Hágöngur, Prestahnúkur) (Tuffen et al., 2002; McGarvie et al., 2007; Stevenson et al., 2009), but not as common as basaltic subglacial products. Basaltic eruptions are capable of melting >10 times their eruptive volumes in meltwater due to their high temperature (~1200 °C versus ~850 °C for rhyolite), while cooler rhyolite eruptions melt significantly less ice (Höskuldsson and Sparks, 1997). Rhyolite eruptions also favor draining of meltwater versus retaining it in basaltic subglacial systems (Tuffen et al., 2002). As a result, magma quenching directly against ice is far more common in subglacial rhyolite eruptions than basaltic ones (McGarvie, 2009). Tuffen et al. (2002) suggest that rhyolitic eruptions generally commence with an ice-confined phreatomagmatic phase. Upon transition to a subaerial eruption, little meltwater is likely to be available for interaction with rising magma. Therefore, explosive subaerial activity may only occur if the magma is sufficiently volatile-rich to generate a “dry” magmatic eruption. This is in contrast to basaltic subglacial eruptions, which generate copious amounts of meltwater that do not readily drain, increasing the likelihood of a highly explosive phreatomagmatic eruption (Smellie, 1999).
In addition to magma-induced landforms, glacial-outburst floods or jökulhlaups commonly result from sub-ice volcanism. Jökulhlaups form when the volume of water in en- or sub-glacial lakes overflows the banks, often as a result of eruption-induced glacier melting. These floods initially contain ice, water, and volcanic sediments, but quickly pick up loose debris and topsoil as they move away from their source. Jökulhlaups travel at speeds of up to 10s of km/hr and can carry up to 10 km3 of debris, creating highly distinctive and far-reaching deposits (Thordarson and Höskuldsson, 2014).
Glaciation and Glacial Geology
Iceland is ~10% covered by glacial ice (Fig. 11) distributed amongst 269 named glaciers and a small number of unnamed glaciers (Sigurðsson and Williams, 2008). The five largest ice caps cover 96% of the glaciated area, with 73% in the massive Vatnajökull in southeastern Iceland, the largest glacier in Europe (percentages as of 2000, based on Sigurðsson et al., 2013). Sigurðsson and Williams (2008) classify Iceland’s named glaciers into: ice caps (14), contiguous ice caps (2), outlet glaciers (109), ice-flow basins (8), ice streams (3), cirque glaciers (55), mountain glaciers (73), and valley glaciers (5).
Iceland has both surge-type and non-surge-type glaciers. The patterns of advance and retreat of non-surge-type glaciers are clearly influenced by climate (Sigurðsson et al., 2007). Surge-type glaciers spend most of the time in retreat but periodically have very rapid advances. Iceland’s surge-type glaciers are mostly, but not limited to, large outlet glaciers. Brúarjökull, a large piedmont glacier on the north side of Vatnajökull, surged 8 km at a rate of ~100 m/day in 1963–1964 (Thórarinsson, 1969). The lower portions of many of Iceland’s outlet glaciers are classified as temperate glaciers, meaning that they are at their melting point for at least part of the year, and with water at their bases they often have high glacial flow rates.
The history of glaciation of Iceland has been summarized by Eiríksson (2008) and Geirsdóttir (2011), amongst others. At least 20 glaciations are recognized since 4 Ma. The earliest evidence for glaciation in Iceland dates to ca. 7 Ma. Glacial sedimentary rocks become more common in the Pliocene, but glaciation appears to have remained a localized phenomenon in this time. Pliocene tillites (diamictites) are found intercalated in the volcanic sequences of east and southeast Iceland, and further evidence of glaciation in this time is found in the lower Tjörnes beds in north Iceland. Evidence for widespread glaciation of Iceland first appears in the Pleistocene, in the Hreppar Formation in south Iceland, the volcanic sequences of southwest (Borgarfjörður and Hvalfjörður), east (Fljótsdalur and Jökuldalur), and southeast Iceland (Skaftafell area), and the Tjörnes beds in the north. Evidence to document Pleistocene glaciation between ca. 700 ka and the most recent glaciation is limited, with sedimentary rocks in the Tjörnes beds providing the best record (Geirsdóttir, 2011).
In Iceland, the most recent glaciation is referred to using the name applied to northern Europe, the Weichselian glaciation (equivalent to the Wisconsin glaciation in North America). The Weichselian in Iceland began by 85–90 ka (Norðdahl and Pétursson, 2005) and culminated in the Last Glacial Maximum (LGM), variably given as >25 ka (e.g., Ingólfsson et al., 2010) to ca. 21 ka (e.g., Hubbard et al., 2006). With lower global sea level during the LGM, the Iceland ice sheet extended to near the edge of the Iceland shelf (Fig. 10). Locally, this is more than 100 km beyond the modern shoreline (e.g., Norðdahl and Pétursson, 2005; Geirsdóttir, 2011). The ice sheet is estimated to have had a maximum thickness of 1500 ± 500 m (Norðdahl et al., 2008), tapering at its margins. The elevations of subaerial lavas on tuyas reflect (but are lower than) the maximum heights of the glaciers that they erupted through (e.g., Licciardi et al., 2007).
Retreat of the LGM ice sheet began between 18 and 15 ka. The melting was rapid because rising sea levels floated the lower parts of the ice sheet leading to calving and collapse (Norðdahl et al., 2008; Geirsdóttir, 2011). Marine deposits on land (defining the “marine limit”) at 12–13 ka (Bølling-Allerød interstadial) demonstrate that the shelf had been deglaciated by this time. These deposits are found at 50–225 m elevation around Iceland (Norðdahl et al., 2008; Le Breton et al., 2010) demonstrating that the lower areas of Iceland were below sea level. This relationship reflects the fact that eustatic sea level had risen but little post-glacial rebound had occurred yet in Iceland.
The Iceland ice sheet re-advanced during the Younger Dryas (ca. 13–11.5 ka) and Preboreal (11.5–10.1 ka) climate events (Ingólfsson et al., 2010). The inner and outer Búði moraines mark the extent of the ice sheet in south-central Iceland at these times. The presence of marine fossils in associated deposits demonstrates that the marine limit was >50 km inshore from the modern shoreline in this area (Norðdahl et al., 2008; Geirsdóttir et al., 2009; Ingólfsson et al., 2010). During the subsequent warming of the Holocene Thermal Maximum, Iceland is thought to have been nearly ice-free, as suggested by the possible complete disappearance of the Langjökull ice cap in this time (Geirsdóttir et al., 2009).
A return to more extensive glaciation occurred in Neoglacial time which started around 5 ka and continues (though it is argued that we may be emerging from it now due to anthropogenic climate change). Many outlet glaciers reached their maximum extent during 4–5 advances (Norðdahl et al., 2008) that occurred in the period between 5 and 1.2 ka. The start of the Little Ice Age in Iceland is given a wide range of dates in the literature, ranging from 1200 to 1650, but Geirsdóttir et al. (2009) indicate a return to colder conditions beginning around 1250. The end is generally given as ca. 1900. During this 650-year interval, Iceland’s climate was highly variable, and many glaciers underwent several distinct advances. Since the end of the Little Ice Age, non-surge type glaciers have retreated 100s to 1000s of meters (e.g., Sigurðsson et al., 2007).
Though Iceland is sparsely populated, ~1150 years of human presence has had profound consequences for the vegetation, and consequently the soil, of the island. Dugmore et al. (2009) summarizes research on vegetation changes in Iceland since settlement with estimates of reduction in vegetation cover ranging from a pre-settlement 54%–65% to the present 28%. Woodland diminished from a pre-settlement 15%–39% cover to the present 1%. While wood was used for building materials and fuel, Trbojevic (2016) concludes that the clearance of woodlands for home-fields for animal husbandry was the leading cause of deforestation. Grazing, primarily of sheep, in both cleared lands and upland heath, has had a profound impact on vegetation. However, other stresses such as volcanic eruptions and climate fluctuations also play a role. Overgrazing reduces the resilience of the ecosystem to respond to such natural perturbations, and this combination of factors was responsible for considerable soil loss during the Little Ice Age, from ca. 1250 to 1900 in Iceland (e.g., Gísladóttir et al., 2010; Greipsson, 2012).
Iceland’s soils are dominated by andosols (andisols) (86%, Arnalds, 2004). These soils, which form on volcanic tephra, are highly susceptible to erosion by water and wind, especially when exposed through a reduction in vegetative cover. Thus, Iceland brings together a combination of historic patterns of impactful land use, susceptibility to soil loss, and natural stresses that have contributed to soil loss and desertification. Arnalds et al. (2001a) surveyed soil erosion in Iceland and determined that 39.7% of the land had undergone considerable, severe, or very severe erosion. Overall, the combination of desertification by soil erosion and purely natural processes (e.g., combinations of flooding and volcanic eruptions) has resulted in Iceland being 42.2% desert, using a vegetation-based classification (Arnalds et al., 2016).
Two features provide the clearest field evidence for soil loss in Iceland to the non-specialist. The first is the rofabard (rofabarð) (Fig. 12), an erosional scarp created by removal of weak andosol above a more resistant lava or till substrate and under an often-overhanging coherent vegetative cap (Arnalds, 2000). Another feature reflecting soil loss is “sandy lag gravel” (Fig. 13) which is thought to have been previously covered (Arnalds et al., 2001b) with 0.5–2 m andosol soils and vegetation. This is a particularly common feature of the deserts of the Central Highlands of Iceland.
Geothermal Energy in Iceland
Iceland is a leader in the utilization of geothermal energy. In 2017, geothermal power accounted for 60.7% of primary energy which includes district heating (see below) and 26.9% of electric power generation in Iceland (hydroelectric being the source for 73.1%). Hot water in district heating systems is utilized for space heating, hot tap water, recreational pools, greenhouses, industrial applications, fish farming, and sidewalk/road snow melting in some urban areas.
Iceland has the highest per capita electricity use in the world because of its intensive demands in aluminum, ferroalloy, and silicon smelting (75.5% of Iceland’s electricity use). These industries utilize smelting processes that are electricity intensive, rather than heat intensive, and are thus developed in places where electricity is cheap and abundant (aluminum smelting is by electrolysis and ferrosilicon and silicon by electric arc furnace). While the emphasis on renewable energy in Iceland is viewed as an environmental positive, the construction of new hydroelectric and geothermal power facilities to produce more electricity for heavy industry is controversial (e.g., Newson, 2010). The two most recent projects at this time, Fljótsdalur Power Station (hydroelectric, commissioned 2009) and Þeistareykir Power Station (geothermal, commissioned 2017) were constructed primarily to serve individual industrial facilities, the Alcoa Fjarðaál aluminum smelter and PCC Bakki silicon plant, respectively. Iceland also continues to explore the financial feasibility of direct export of electricity to Europe via underwater cable.
Geothermal resources are commonly classified as either high-temperature (reservoir temperature >175 °C at ~1 km depth) or low-temperature (temperature <150 °C). The former is suitable for electricity generation and the latter is generally suitable only for district heating (Arnórsson et al., 2008). High-temperature fields occur only in the volcanic systems of the Neovolcanic zone. Low-temperature areas are found throughout Iceland. High-temperature geothermal areas are typically marked by natural geothermal features including hot springs, geysers, mud pots, and fumaroles (all of these features, except hot springs, are uncommon in low-temperature fields).
There are currently eight geothermal electric power plants in Iceland (Table 2). All but one of the plants are in high-temperature geothermal fields; the small Húsavík Power Plant in the Hveravellir geothermal field is the exception (Georgsson et al., 2005), though it meets some definitions of a high-temperature field (Arnórsson et al., 2008).
Geoparks in Iceland
Geotourism is a relatively new concept. Dowling and Newsome (2010) define the term as follows: “Geotourism is a form of natural area tourism that specifically focuses on geology and landscape. It promotes tourism to geosites and the conservation of geo-diversity and an understanding of earth sciences through appreciation and learning.” “Geopark” is a designation developed by United Nations Educational, Scientific and Cultural Organization (UNESCO). The UNESCO Global Geopark Network was founded in 2004, building upon the European Geopark Network, which was initiated in 2000 (McKeever and Zouros, 2005). The concept of geoparks is intended to serve three objectives: (1) to conserve significant geological features; (2) to provide an educational framework for geoscience and environmental education in park areas (Fig. 14); and (3) to stimulate sustainable economic development via geotourism with an emphasis on local impacts (UNESCO, 2006). There is a formal application process for the creation of a new geopark.
As of 2019, the UNESCO Global Geopark Network consists of 140 geoparks in 38 countries. Iceland has two geoparks in the UNESCO Global Geopark Network at the time of this writing. These are the Reykjanes Geopark, which includes sites visited on Day 1, and Katla Geopark, which we will visit on Day 4. A third “aspiring” geopark is the Saga Geopark Project in Borgarbyggð (the area around Borgarfjörður) in southwest Iceland.
FIELD TRIP ITINERARY
The itinerary that follows is a ten-day excursion, counter-clockwise around Iceland, beginning and ending in Reykjavík. The focus of the trip is on igneous and tectonic features, but geomorphological and sedimentary features are also covered where compelling sites are encountered along the way. Ten days is still a quick trip around Iceland, and many interesting features are bypassed. Some of these additional points of interest are noted in “On Route” sections, and several are described in more depth as “Optional Stops.” Known dates and ages of features formed during historic time are given as calendar years of the “common era” (CE = AD). All location coordinates are given in latitude and longitude decimal-degrees in the WGS 84 datum.
DAY 1. REYKJANES VOLCANIC BELT
The first day of this excursion is focused on Reykjanes (Fig. 15). The peninsula has a mountain spine of eroded subglacial volcanic edifices, with an apron of post-glacial lava flows that extend to the sea. Four volcanic systems (or six, depending upon the author) make up the Reykjanes Volcanic Belt (or Reykjanes Peninsula Oblique Rift). From west to east these are the Reykjanes, Krýsuvík, Brennisteinsfjöll, and Hengill volcanic systems. These systems are arrayed en echelon along the EW-trending peninsula with NE-trending fissure swarms. The route takes us from Reykjavík out to the southwestern end of the peninsula where the mid-ocean ridge comes on shore, and then makes stops coming back east along the peninsula with the final stop at the Hellisheiði geothermal power plant in the Hengill volcanic system.
Departing from Reykjavík, follow signs to Keflavík and the international airport via Route 41. You will pass through the western suburb Hafnarfjörður. Upon dropping out of the hills at the west end of Hafnarfjörður, a large industrial facility appears. This is the Straumsvík aluminum smelter, operated by Rio Tinto at the time of this writing. Opened in 1969, this was the first aluminum smelter built in Iceland. The smelter is built on the lava flow Kapelluhraun, which erupted from 1151 to 1188 and is the youngest lava to have flowed into what is now part of greater Reykjavík (Sæmundsson et al., 2016).
Approximately 5 km west of Straumsvík is an exit for Hvassahraun, and on the south side of the road at the exit are spectacular ~5-m-high tumuli in the ca. 7 ka Hrútagjárdyngja lava flow (Sæmundsson et al., 2016). A parking lot is available by taking this exit and passing south under Route 41. Farther west, around and west of the Vogar exit, the view to the south includes several low scarps of north-dipping normal faults cutting through early post-glacial (ca. 14 ka) Þráinsskjaldarhraun lavas (Sæmundsson et al., 2016).
After the traffic circle at Njarðvík, turn left (south) onto Route 44 for Hafnir and Reykjanesvirkjun. After Hafnir this becomes Route 425, and ~7.7 km south of Hafnir, look for the sign to turn for the parking lot for the “Bridge between Continents” (Brú Milli Heimsjálfa), Stop 1.1.
Stop 1.1. Bridge between Continents
The “Bridge between Continents” is an ~15 m footbridge over a well-defined ~5-m-deep narrow graben, “open fissure,” or “dilatational crack” (Einarsson et al., 2008). The fissure trends 050°, and widens to 25 m southwest of the bridge (Fig. 16). The fissure floor is covered by windblown sand from the sand beach and dunes at Stóra-Sandvík. About 60 m north of the bridge is a prominent south-facing normal fault scarp. The scarp trends 074° here, oblique to the fissure.
The walls of the fissure and the scarp expose basalt lavas erupted from the early post-glacial (13.6 ka) Sandfellshæð shield volcano (Sæmundsson et al., 2016). Early post-glacial lavas are voluminous in the neovolcanic zone (e.g., Helgason et al., 1978; Sinton et al., 2005). This is thought to reflect enhanced melt production during isostatic rebound associated with deglaciation. These are excellent examples of compound lavas. While it is evident that there are multiple flows, they do not necessarily reflect multiple eruptive events. Instead, during the course of a continuous eruption event, individual flow lobes override one another. The result is a sequence of lavas in which individual flows cannot be traced laterally very far in stratigraphic exposure (as observed at this location). Compound lavas are distinguished by the absence of soil or tephra between flows and petrographic and compositional similarities between individual flow units. Compound lavas are favored over sheet flows when effusion rates are low.
The educational point of this site is evident in its name; the idea is that we are at the boundary between the North American and Eurasian plates. The bridge is sometimes used in mass media to demonstrate the plate boundary. While it has educational value in this perspective, it is worth recognizing that the bridge does not literally cross from one plate to the other. We are indeed where the Mid-Atlantic Ridge comes on shore, but the axial rift is 5–6 km wide here. The fault scarp at this site is the northernmost of several southeast-facing normal fault scarps on the northwest side of the rift zone. Looking southeast one can see several northwest-facing fault scarps that represent the other edge of the rift zone; some will be seen at Stop 1.3.
Do fissures like this form in singular events or over a period of years? A similar fissure will be visited as Stop 8.5 at Hlíðargerði in the north of Iceland. Based on relationships we describe at that stop, we understand that features like this can form in a matter of hours and are likely to form over dikes propagating at depth. The oblique intersection of the fissure and the fault here has been interpreted to reflect differences in the orientation of the stress field during a magmatic event versus magmatic quiescence (Clifton and Klattenhorn, 2006).
Return to Route 425 and continue south for ~2.2 km to a pullout on the west side of the road. Note the metal sphere on a columnar basalt pedestal here, similar to others you may have seen on Route 425 after Hafnir. These are part of a Solar System display with the distances between planets and the spheres representing the planets to scale, giving one a real sense of the vast distances between planets. The planet here is Saturn. Several more planets will be passed on the way to Stop 1.3, and the Sun is in the parking lot of the Reykjanes Power Station.
Stop 1.2. Stampar Cone Row
A short trail leads from the parking area to a small spatter cone ~180 m to the west. Along the trail, appreciate the outstanding basaltic lava flow features including ropey pāhoehoe, shelly pāhoehoe, entrail pāhoehoe, and pāhoehoe toes (helluhraun is Icelandic for pāhoehoe). The cone is constructed of agglutinated spatter, well-exposed in the interior crater walls.
From the rim of the crater, the broader context of the cone can be appreciated (Fig. 17). This cone is the northernmost of a series of cones aligned along ~035°—the Stampar cone row. The cone row extends southwestward to the shore. The nearest vents to the northeast are older. The lava that erupted from the Stampar cones is called Stampahraun. The alignment of cones reflects the geometry of the dike system that fed their eruption. The eruption likely began as a fissure eruption before becoming localized to the cones. A dike can be seen feeding the flows in the sea cliff at Stop 1.3.
This cone and most of the distinct cones to the southwest are called the younger Stampar (yngri Stampar). They formed during a historic eruptive event called the Reykjanes Fires which occurred between 1211 and 1240. More features formed in this eruption will be seen at Stop 1.3. Farther east and northeast (~6 km) is another, larger volume, lava field that erupted during the Reykjanes Fires event. Individual flows that comprise the lava field include Eldvarpahraun, Arnarseturhraun, Illahraun, and Sundvörðuhraun. (We will drive through these lavas between Stops 1.3 and 1.4.) These flows erupted from a fissure system marked for some of its length by Eldvörp, an even longer cone row.
Continue south on Route 425 a short distance (~1.7 km) and turn west at signs for Reykjanesviti. The route passes the Reykjanes Power Station then takes several sharp bends; continue to follow signs for Reykjanesviti. At one point there is a T-intersection with the route to Reykjanesviti to the right and Gunnuhver to the left. Gunnuhver, an area of fumaroles and mudpots, is not an official stop in this field guide, but is worth a short side-trip. At the time of this writing, the main parking area is at the shore with a secondary parking area at the base of the lighthouse Reykjanesviti. At times the coastal parking area is closed, requiring a short walk (500 m) to reach the shore.
Stop 1.3. Reykjanesviti Area
We are at the southwest tip of the peninsula Reykjanes, and, confusingly, the smaller promontory at this location is also called Reykjanes. This stop includes walking to several locations from the parking lot (Fig. 18):
A. the hills at the coastal parking area, including Valahnúkar;
B. a walk to the southeast side of Valahnúkar;
C. the sea cliffs in the low hill Vatnsfell, ~400 m west of the parking area; and
D. a walk farther west to the west side of the bay Kerlingarbás.
The parking area is on younger Stampar lavas from the 1211–1240 Reykjanes Fires eruptive event. Next to the east side of the coastal parking area are two hills separated by a small bay, the larger hill on the east called Valahnúkar. The view from the small bay is informative. Examine the cliff and slopes of the lower west hill. They are composed of pillow lavas. These pillow lavas can be examined on the gentler slopes of the hill, and the sea cliff reached from the west side. Underlying the pillow lavas is palagonite tuff with interbedded breccia. The cliffs of Valahnúkar to the east expose mostly breccias. The small bay between the hills is interpreted as a graben. The pillows, tuff, and breccia here, and in the lighthouse hill Bæjarfell, have been variably interpreted to have formed in a subglacial (e.g., Etienne and Paris, 2010) or submarine environment (Thordarson and Höskuldsson, 2014) during the late Weichselian glaciation. The subglacial interpretation may seem obvious as the Iceland ice cap extended past Reykjanes and global sea level was lower at this time, but the crust was also depressed by the presence of the ice cap, complicating the interpretation.
At the time of this writing, the trail to the top of Valahnúkar is closed; if available, it provides a commanding view of the area. Proceed east and southeast on a trail that skirts the base of Valahnúkar. Stop at the wood overlook platform (63.8116°N, 22.7095°W). The younger Stampar lava, which is slabby pāhoehoe at the level of the platform, drops over a short step and becomes an ‘a‘ā (apalhraun in Icelandic) flow below. This is a common relationship. Two things can cause the transition from pāhoehoe to ‘a‘ā: increasing viscosity or increasing strain rate. When a lava flow goes over a steep slope the velocity increases and strain rate increases correspondingly, frequently inducing transition from pāhoehoe to ‘a‘ā.
Looking beyond the younger Stampar lavas there are several other features to appreciate from this viewpoint. The shield volcano to the east and northeast is Skálafell. The upper vent of Skálafell is a ca. 3.2 ka spatter cone, and most of the upper slopes of the volcano are lavas of this age (Sæmundsson et al., 2016). It is built upon an older edifice that dates to >8 ka, and older lavas of this age are exposed on the more distal flanks of the volcano and in fault scarps.
Also prominent from this viewpoint is the northwest-facing Valbjargagjá fault scarp. This is one of two prominent scarps that mark the southeast side of the rift zone (the other being on the east side of Skálafell). Where these faults cut the older Skálafell lavas, displacements can exceed 15 m. The Valbjargagjá scarp exposes several compound lavas from Skálafell. Note that the scarp becomes less prominent to the north where younger Skálafell lavas flowed over it.
Continue down the stairs and farther east and southeast around Valahnúkar. A prominent boulder beach (63.8104°N, 22.7109°W) connects Valahnúkar to the Valbjargagjá scarp (Fig. 19). The sea cliff exposes palagonite intruded by irregular dikes. The ridge of the beach reaches a maximum elevation of 9.8 m, and a boulder field extends ~140 m landward into the younger Stampar lava and tidal lagoon (Etienne and Paris, 2010). This is one of many boulder beaches and boulder ridges along the shore of Reykjanes, representing transportation and deposition during powerful storm events (centennial wave heights are estimated at 17–18 m).
Walk back through the coastal parking area and continue westward. Descend from the younger Stampar level, around another boulder field, and onto the coastal basalt platform at the base of the low hill Vatnsfell, ~400 m west of the parking area (63.8150°N, 22.7237°W). The cliff is composed of fresh basaltic hyaloclastite tuff and lapilli tuff. The tuff is well bedded and contains numerous sedimentary and volcanological structures including cross bedding, dunes, and bomb sags (Fig. 20). This is the remnant of a tuff cone deposit. Tuff cones are formed by phreatomagmatic eruptions—products of explosive interaction of lava/magma with water. This type of volcanism is called Surtseyan volcanism, named for the 1964–1967 eruption of the island Surtsey in the Vestmannaeyjar. The absence of palagonitization indicates that this site is on a distal part of the cone. The deposits here and in the cliffs to the west represent two distinct tuff cones (Sigurgeirsson, 1995); the distinction between the two tuff cones will be most evident at the last site at this stop. The tuff cones were formed early in the younger Stampar eruptive event. The small sea stack Karl, ~300 m offshore, is a remnant of the palagonitzed core of the younger tuff cone. Fourteen kilometers out to sea is the island Eldey which is also a remnant of an emergent volcano erupted during the younger Stampar event (Thordarson and Höskuldsson, 2014). The straight cliff to the west, above the bay Kerlingarbás, exposes thinly bedded tuff of the younger cone. It is cut by several younger Stampar dikes (Fig. 21) that can be reached with some difficulty by navigating the angular talus under the cliff. The eastern dike can be seen to cut through not only the tuff but also the lowermost flow of the two younger Stampar lavas above.
To reach the last site at this stop, ascend the slopes east of the cliff above Kerlingarbás and proceed northwest along a rough road. Leave the road it where it turns northward and descend past a small spatter cone. Go around to the northwest side of Kerlingarbás (63.8182°N, 22.7327°W). The pāhoehoe lava seen here underlies scattered remnants of younger Stampar tuff cone deposits, and thus represents an earlier event. These are called the older Stampar (eldri Stampar) lavas because they are also associated with cones in the Stampar cone row, and are estimated to be ~1900 years old (Sæmundsson and Einarsson, 2014). One of the older Stampar cones is the small spatter cone passed on the way to this site. The sea cliff due east exposes a dissected older Stampar cone (Fig. 22). The core of the cone includes a lens of lava that ponded in its crater. The eastern flank of the dissected cone is mantled with the two distinct tuff cone deposits, which are in turn overlain by the younger Stampar lava.
Return to Route 425 and head east toward Grindavík. After passing the northern route to Gunnuhver, the small shield volcano to the southeast is Háleyjabunga, which erupted picrite lava at ca. 14 ka. Beginning ~200 m after the turnoff for Brimketill, the route passes through the Sundvörðuhraun at the southern end of the larger Reykjanes Fires for ~1 km. As the route approaches Grindavík, the prominent hill to the north is Þorbjörn. It is composed of subglacial palagonite tuff and breccia deposits, and is conspicuously cut by northeast-trending normal faults that define a graben on its upper slopes. Pass the junction leading to the Blue Lagoon (Bláa Lónið), a spa resort utilizing the eerie blue silica rich effluent of the Svartsengi geothermal power plant. At the T-intersection, turn right (south) on Route 43 toward Grindavík. In Grindavík, follow signs for Route 427 to Þorlákshöfn through Grindavík. After Grindavík, the route skirts steep slopes of subglacial (Húsafell and Fiskidalsfjall) and submarine (Festarfjall) volcanic edifices and then drops on to post-glacial lavas again. Look for the sign for Selatangar on the right (south) and take this road to a parking area, Stop 1.4.
Stop 1.4. Selatangar
The parking area is at a beach between two lava flows, Katlahraun on the west and Ögmundarhraun on the east (Fig. 23). Both flows are products of the Krýsuvík volcanic system. Katlahraun erupted ca. 2 ka from vents northwest of Selsvallaháls ridge, and Ögmundarhraun erupted between 1151 and 1188 from vents southeast of Selsvallaháls (Sæmundsson et al., 2016). The Ögmundarhraun eruption is part of the event called the Krýsuvík Fires. Another lava erupted in this event, Kapelluhraun, was passed on the way from Reykjavík to Stop 1.1. Thus, flows from this event reached both the north and south shores of Reykjanes.
Ögmundarhraun is an ‘a‘ā flow at this location; Katlahraun is a pāhoehoe flow. A short walk into Katlahraun west of the parking area leads to a large (~150 m) oval pit. The pit is ringed by an inflationary margin (outward tilted pāhoehoe slaps), indicating that the feature represents first pooling and inflation within the flow, and then subsequent draining and deflation, leaving the disrupted crust in the floor of the pit (Sæmundsson et al., 2016).
As the site of a major historic eruption, and with proximity to greater Reykjavík and roads on both sides of the peninsula, the Krýsuvík volcanic system is monitored for indications of possible volcanic activity. GPS monitoring and InSAR (interferometric synthetic aperture radar) data documented a period of uplift beginning in late 2009 and continuing through 2011 (Gudjónsdóttir et al., 2018). Uplift was focused on the Krýsuvík high-temperature geothermal system. Though there is concern about a possible magma chamber at depth driving inflation, Hersir et al. (2018) conclude that the source of the uplift was a gas flux. The Krýsuvík geothermal system is the site of several proposed geothermal power production facilities. Such development is controversial because the area is highly valued for recreation by the nearby greater Reykjavík population.
Looking back northwest, the slope that rises above the two young lavas is the southwestern end of a long ridge, known as Núpshlíðarháls, Selsvallaháls, and other names along its length. This ridge is made primarily of subglacial hyaloclastite dating to the Weichselian and earlier glaciations. In Iceland these are called móberg ridges, and they represent subglacial fissure eruptions. The NE-SW trend of the ridge is consistent with the geometry of post-glacial fissures and cone rows. Normal faulting, parallel to the fissures, may locally account for some of the topographic relief.
Return to Route 427 and continue east. Driving through the Ögmundarhraun, you will observe several kipuka (“islands” of older rock within a lava flow) to the south. These kipuka include portions of the former Krýsuvík settlement, which was abandoned in the wake of the Krýsuvík Fires event (Thordarson and Höskuldsson, 2014). Turn left (north) on Route 42. At ~4 km you pass the first of several maar craters on both sides of the road, and at 4.4 km turn off for the parking lot for Grænavatn, Stop 1.5.
Stop 1.5. Grænavatn
The parking area is at the southeastern corner of the lake Grænavatn (Fig. 24). Grænavatn occupies the largest of 15 maars in this area (Sæmundsson et al., 2016). Maars are formed by phreatomagmatic or phreatic eruptions involving groundwater that have excavated craters below the preexisting surface. These maars formed between 11 and 7 ka, with the earlier eruptions being purely phreatic (no juvenile magmatic component in the tephra). The later eruptions began as phreatic, transitioned to phreatomagmatic, and, in the case of Grænavatn, finished with a small effusive eruption.
The Grænavatn maar has a maximum diameter of ~370 m. Near the parking area, lithic clasts can be seen weathering out of the crater walls. At the east and northeast edges of the crater is a thin effusive basaltic lava flow erupted late in the eruptive sequence. Two smaller prominent maars occur east of the parking area.
Leave the parking lot turning right (north) on Route 42. At 1.5 km is a parking area on the left (west) for Seltún. This location is potentially worth a stop, though it is not an official stop in this itinerary. Seltún is a fumarole field featuring mudpots among the geothermal features. The original tourist attraction here was a geothermal exploratory borehole drilled in 1949 that erupted in geyser-like fashion until it plugged in 1999 and then exploded (Clifton, 2006). There is a boardwalk through the geothermal features. Continue along Route 42, which provides several passes between the lake Kleifarvatn and the móberg ridge Sveifluháls. The route provides several viewpoints out to the lake as well as opportunities to look at the subglacial palagonite tuff and breccias—many with interesting volcanic-sedimentary structures—that form Sveifluháls. Gudmundsson (2017) provides an outstanding structural overview of Sveifluháls and the Kleifarvatn area.
Kleifarvatn is a 90-m-deep lake with a base elevation of 45 m and no external drainage (Friðriksson, 2014). The lake is notable for a sudden partial draining event in 2000. A M6.5 earthquake in the South Iceland Seismic Zone in June 2000 triggered three M4–5 earthquakes in central Reykjanes (Clifton et al., 2003). Surface fractures were observed in association with these quakes. Fractures near the north end of Kleifarvatn led to the draining of the lake, which dropped 4 m in ~15 months, losing ~12% of its volume (Clifton et al., 2003).
Leaving Kleifarvatn, the route climbs over the Sveifluháls móberg ridge at Vatnskarð and descends onto Kapelluhraun (the previously mentioned north-flowing Krýsuvik Fires lava). Turn right (east) onto Route 417; the next stop will be ~6.1 km down this road. The route climbs past quarries in the móberg ridge and then onto post-glacial lavas, first Skúlatúnshraun then the younger Tvíbollahraun, which will be visited in the next stop. Turn into a parking area on the left for Leiðarendi.
Stop 1.6. Leiðarendi Lava Tube
From the parking area, you will follow a footpath through a moss-covered lava field for ~250 m. The path terminates at a collapsed skylight that serves as an entrance into the Leiðarendi Lava Tube (Fig. 25A). As you transition from the subaerial to the subterranean world (Fig. 25B) you will scramble over blocks that have detached from the ceiling of the lava tube. Take this as a warming that lava tubes are unstable features that should be approached with appropriate caution—particularly in Iceland where fractured rocks are made even more vulnerable by freeze-thaw erosion and frequent seismic activity. We encourage all those exploring the lava tube to bring a headlamp or flashlight (with extra batteries) and head protection. It is also a good precaution to check recent earthquake activity reported by the Icelandic Meteorological Survey (https://en.vedur.is/earthquakes-and-volcanism/earthquakes/) to see if any events may have recently destabilized the lava tube ceiling.
Once you have descended into the cave you will see—with the help of your headlamp—several classic lava tube features. Looking up, you will see several varieties of “lava-cicles” (like icicles) protruding from the ceiling, often generated by partial melting of the lava tube ceiling as hot lava flows below (Fig. 25C). These take the form of elongate stalactites and smaller drip and flow features. The walls of the lava tube record a history of multiple episodes and pulses of flow with ridges and textural changes serving as “high lava marks” similar to high water marks after a flood (Fig. 26). Venturing further back into the cave, you will encounter delicate spires, or stalagmites, rising from the floor of the cave (Fig. 25D). Leiðarendi has a mapped length of ~800 m (Reich, 1993). From our entrance point you will be able to traverse ~200 m into the cave with (relative) ease. The ceiling height is mostly ≥2 m, down to 1.5 m for a short stretch (Reich, 1993).
In addition to classic lava tube features, Leiðarendi is home to an ovine surprise: the skeleton of a sheep who met its untimely demise in this dark subterranean environment (Middleton, 2003). This sheep, who “came to the end of the road” in the most fateful sense of the expression, is the namesake for Leiðarendi, which loosely translates to “road’s end.”
Proximity to Reykjavík (only ~30 km when traveling directly) has made Leiðarendi an increasingly popular tourist destination. There are several reports of vandalism, littering, and general wear-and-tear due to the increased foot traffic. In response, there are increasing numbers of chains and ropes in place to cordon off fragile features (e.g., stalagmites, stalactites) and protect them from prying fingers. Please respect these conservation efforts and do your part to help preserve this lava tube for posterity.
Get back on Route 417 headed east. At ~6.5 km the road is on the Mið-Húsafellsbruni lava, a tenth century historic lava (Sæmundsson et al., 2016). At ~8 km the road passes below Þríhnúkagígar (1.3 km south), the cone in which magma drainage left an ~120 m deep vertical cave that is the focus of the “Inside the Volcano” tour. (Note: the attraction is visited via guide from an access point to the southeast.) Continue on Route 417 until it reaches Route 1, and turn right (east). At ~4.5 km on Route 1 the road is on Kristnitökuhraun, a historic lava erupted in 1000, its name reflecting its eruption in the year in which the Icelanders decided to formally adopt Christianity via deliberation at Alþingi (national assembly). After ~9 km on Route 1 turn left (north) on the road signed for Hellisheiðarvirkjun and proceed to the parking area in front of the large modern building that houses the Hellisheiði Geothermal Power Plant turbines and visitor displays, Stop 1.7.
Stop 1.7. Hellisheiði Geothermal Power Plant
The Hellisheiði geothermal power plant has visitor’s facilities that make it an excellent place to both learn about geothermal power in Iceland (exhibit) and see the workings of the plant (windows into turbine rooms). The Hellisheiði plant is one of two geothermal power plants utilizing geothermal fields in the Hengill volcanic system, the other being Nesjavellir to the north. The Hengill volcanic system is the first system, working northeast from where the mid-ocean ridge comes on shore, to have a central volcano, the Hengill central volcano. The exposed geology is dominated by subglacial hyaloclastites, though subaerial lavas are encountered at depth (Franzson et al., 2010). The drilling targets were young volcanic fissures (2–5 ka) and fault structures on the western edge of the Hengill graben (Franzson et al., 2010).
The Hellisheiði plant was commissioned in 2006, and was expanded in 2007, 2008, and 2011. Its installed capacity at the time of writing is 303 MW electric power and 133 MW of thermal energy for district heating, making it the largest geothermal power plant in Iceland (Table 2). The plant is operated by ON Power, a subsidiary of Reykjavík Energy, the former operator.
At Hellisheiði, steam and hot water are collected at 30 production wells (one typically on standby) over an 8 km2 area mostly above the plant (Hallgrímsdóttir et al., 2012). After separation of water and steam, electricity is generated by six 45 MW high-pressure turbines and one 33 MW low-pressure turbine. All are condensing turbines where low pressure is produced by condensation of steam on the outlet side in condensers cooled by freshwater. For the high-pressure turbines the inlet pressure is 6.5–9.5 bar and the condensing pressure is 0.1–0.22 bar. For the low-pressure turbine the design inlet pressure is 2 bar and the condensing pressure is 0.068 bar (Hallgrímsdóttir et al., 2012). Fresh water is initially heated in the condensers and then further heated by heat exchangers interacting with separated water (Hallgrímsdóttir et al., 2012).
There have been some challenges with the Hellisheiði plant. Due to high production density (production rate per unit area) the system has experienced pressure draw-down and reduced enthalpy of the produced fluid (Gunnarsson and Mortensen, 2016). As a result production has been decreasing since 2013. Reinjection wells have not, as of this point, produced desired rejuvenation of the system. Reinjection, started in 2011, has apparently produced increased seismicity and geodetically measurable surface deformation (Juncu et al., 2018).
Return to Route 1 and resume travel east (left). The route climbs over the pass Hellisheiði on ~1900-year-old Nesjahraun lavas then descends to the town Hveragerði, a town notable for its geothermally warmed greenhouses. Turn left (north) on Route 35, signed for Laugarvatn. The first few kilometers on Route 35 are at the base of Ingólfsfjall to the west, notable for large boulders of hyaloclastite shed during South Iceland Seismic Zone earthquakes (Gudmundsson, 2017). After 4.5 km there are views of the confluence and mixing of the clear river Sog, which originates in Þingvallavatn, and the turbid river Hvítá, which originates at the glacially fed Hvítavatn (both lakes are on the itinerary for Day 2); below the confluence the river is called the Ölfusá.
At ~12 km from Route 1 the road begins passing the Tjarnarhólar crater row, a NNE-trending group of 4 cinder cones in the Grímsnes volcanic system (Jakobsson, 1966). The northernmost cone, Kerið, is deep with a small lake. Jakobsson (1966) interpreted Kerið as a maar, but Gudmundsson (2017) interprets it a scoria/spatter cone whose great depth reflects drainage of the conduit during the eruption. A hiking trail around Kerið requires a fee. Continue on Route 35 to Route 37, signed for Laugarvatn, turning left (north). Take Route 37 to Laugarvatn. End of Day 1.
DAY 2. GOLDEN CIRCLE AREA
This day starts and ends at Laugarvatn (Fig. 27) and includes stops at the three sites commonly called the “Golden Circle” because they are the stops on the classic tourist day trip from Reykjavík: Þingvellir, Geysir, and Gullfoss. The Golden Circle sites all have a variety of well-exposed geology of Plio-Pleistocene to modern age. The day includes a stop to view products of subglacial volcanism and finishes with a short trip into the central highlands of Iceland.
From Laugarvatn, head south out of town on Route 37/Laugarvatnsvegur/Dalbraut. At the roundabout, take the first exit onto Route 365 toward Þingvellir. Most of the journey traverses basalt flows. The first of these were sourced from the shield Lyngdalsheiði to the south (left). Those flows are overlain by younger Þingvallahraun lava erupted from the fissures east of Hrafnabjörg to the north (Fig. 28); we are now in the eastern side of the Þingvellir graben. Approximately 14 km from Laugarvatn, the road veers to the right and changes to Route 36/Þingvallavegur. Continue another 12 km through the graben past the Service Centre on the right side of the road and continue up the hill (which is a fault scarp). Follow Route 36 another ~3 km and park at the Visitor Centre. The landscape highpoints are subglacially erupted tuyas (e.g., Hrafnabjörg, Ármansfell) and tindars (e.g., Tindaskagi, Kálfstindar), as well as the shield Skjaldbreiður (Figs. 28 and 29).
Stop 2.1. Þingvellir
Þingvellir is renowned for its important role in Iceland’s cultural and political history, the beautiful lake Þingvallavatn, and the well-exposed landforms that illustrate many fundamental aspects of Iceland’s geology in the Western Volcanic Zone. Weeks-long summer meetings of the Alþingi (parliament) began at Þingvellir in 930 and were a high point of the year. Law and justice, marriage arrangements, trade and sale of goods, and socializing were all important matters. Encampments were built where the river Öxará flows into the north end of Þingvallavatn.
At this location, Þingvallahraun, an early post-glacial lava, flowed into the northern end of the lake 10,200 years ago (Sinton et al., 2005; Stevenson et al., 2012). Þingvallahraun dammed the outflow from the lake, causing the surface to rise to 11 m above the current level; post-emplacement subsidence of the graben by up to 20 m has since submerged parts of the flow (Sæmundsson, 1992). Younger, but mostly early post-glacial, lavas overlie Þingvallahraun at the north end of the lake (Sinton et al., 2005). The many lobes of Þingvallahraun—likely erupted in a single, long-lived eruption—are also exposed in a compound sequence in Almannagjá, the obvious fault scarp that borders the western edge of the graben. Almannagjá is ~7.7 km long and has a maximum throw of 30–40 m (Thordarson and Höskuldsson, 2014). Almannagjá is another feature that is often used in popular media to represent the plate boundary. However, the fissure here is a tensional fracture created primarily by tilting of the east-dipping block caught between the upthrown and down-dropped blocks of the Almannagjá fault (Figs. 30 and 31). The axial rift is actually much wider. The fault marking the other side of the graben, Hrafnagjá, is ~6.6 km southeast and is 11 km long and has a maximum throw of 30 m. Another manifestation of extension is the presence of open fissures in lavas within the graben, several of which can be observed between the Service Centre and the lake (e.g., Sinton et al., 2005). In 1789, the most recent rifting episode in the graben resulted in 1–3 m of subsidence that forced Alþingi’s permanent relocation to Reykjavík (Thordarson and Höskuldsson, 2014). Overall, the rift valley at Þingvellir is widening ~7 mm/yr and subsiding ~4 mm/yr (LaFemina et al., 2005; Islam et al., 2016).
From the parking area, we will retrace the journey toward Laugarvatn by returning to Route 36 for ~12 km, then veering left onto Route 365. Approximately 8 km east on Route 365, turn left (north) onto Road 367, signed for Laugarvatnshellir. After ~3 km, turn right and proceed a short distance to park in the Laugarvatnshellir parking area.
Stop 2.2. Laugarvatnshellir
Signage at the parking lot describes the history of habitation of the small cave (Laugarvatnshellir) cut into palagonite just northwest of the parking area. The site is at the base of the mountain Kálfstindar, the southern end of a tindar (móberg ridge) that extends ~25 km (Jones, 1970). Erosion into the base of this móberg ridge provides an opportunity to see the lower facies of subglacial basaltic volcanism. The first ravine past (northeast of) the cave leads to terrific exposures of pillow lavas. At the farthest point that one can walk up the ravine there is a wall of pillows (Fig. 32). These pillows represent the initial phase of subglacial eruption, while the pressure of meltwater and ice was great enough to suppress explosive expansion of heated meltwater. The pillows exhibit classic characteristics including radial columnar joints, concentric vesiculation bands, glassy rinds, and hyaloclastite/palagonite accumulations between pillows. In the ravine, some hyaloclastite has been weathered out providing an unusually good three-dimensional exposure of pillows.
Walking back out the ravine, one encounters the hyaloclastite and breccia layers that overlie the pillows. These pyroclastic layers are quite varied, reflecting variation in eruption and sedimentation processes. One of the first layers is an ~3-m-thick reversely graded bed. This reflects a high-energy emplacement in a subaqueous pyroclastic density current derived from slope failure higher on the edifice. Breccia units are interbedded with massive and thinly bedded hyaloclastite layers that reflect steady or pulsed (respectively) explosive eruptions under lower pressure as the subglacial volcano built up and the glacial surface melted down.
Hike up on the surface above the cave, southwest of the ravine, to gain an overview and further examine the higher facies of the system. Soft-sediment deformation features can be observed in the thinly bedded hyaloclastite tuffs (Fig. 33). Bedding in the tuffs and breccias flattens out southeastward (Fig. 34). This is interpreted to reflect a delta-like wedge mantling the underlying pillow mound.
The eruption of Kálfstindar constructed a volcanic edifice up to the height necessary to erupt subaerially. A sequence of approximately ten subaerial ‘a‘ā lava flows is exposed beginning at ~715 m on the north peak of Kálfstindar (Jones, 1970), out of reach of field trip visitors. The ~525 m difference between the base of Kálfstindar and the lowest subaerial lavas represents a minimum thickness for the glacier through which it erupted.
Retrace the route back to the paved road, Route 365, and head east toward Laugarvatn. At the traffic circle south of Laugarvatn continue around it and make a left (north) onto Route 37 traveling through Laugarvatn. After Laugarvatn the route passes several tuyas—Fagradalsfjall, Efstadalsfjall, and Bjarnarfell—before turning into Route 35 ~23.5 km from Laugarvatn. Geysir is another 5 km on Route 35. Pull into a parking lot at Geysir, Stop 2.3.
Stop 2.3. Geysir
The Geysir geothermal field in Haukadalur comprises several geysers, fumaroles, hot springs, and nearby mudpots (Fig. 35). The Geysir area trends NNE-SSW and lies between Laugarfjall, a rhyolite dome on its western edge, and the river Beiná to the east. Geothermal activity here is driven by volcanic intrusions in the roots of a now extinct central volcano. Residual heat from the Laugarfjall system forms a reservoir in the basalt bedrock with temperature of 200–250 °C (Pasvanoglu et al., 2000). Heat is transferred as groundwater at 1–2 km depth that moves through mainly NE-trending fractures. In high-temperature areas, rapid heat transfer leads to geothermal waters rising to the surface. The namesake geyser, Geysir, is rarely active, but its neighbor, Strokkur (Fig. 36), erupts a 20–50-m-high (rarely up to 100 m) jet of boiling water every 5–10 minutes or so.
Geothermal waters in the Geysir area are alkaline. Geysir, Strokkur, and Smiður hot springs plot in the bicarbonate water range, but are close to the chloride and sulfate water fields on a Cl–SO4-HCO3 diagram (Pasvanoglu et al., 2000). Geothermal features at Geysir commonly produce silica deposits called sinter around their openings, and the clear waters at Geysir provide excellent views into the basins of many of the springs.
The geothermal area experiences heavy use from tourists; please make every effort to tread lightly and always stay on marked paths.
Resume travel northward on Route 35 toward Gullfoss for another ~9.5 km. Follow signs for the main parking area at Gullfoss, Stop 2.4.
Stop 2.4. Gullfoss
At the head of Hvítágljúfur, the Hvítá river gorge, meltwater from Langjökull plunges over Gullfoss (golden falls). Gullfoss is a two-step waterfall formed by variable erosion of the underlying bedrock succession by the river. Hvítá’s average discharge varies seasonally between 80 and 140 m3/s. The stratigraphy in the Gullfoss area—visible in the canyon walls—consists of thin sedimentary units capped by basalt lava flows (Thordarson and Höskuldsson, 2014). The lavas are more erosion-resistant and the river erodes the underlying sedimentary units at a greater rate. Once the lavas are undercut to the point where they are no longer supported, the lava collapses and the step retreats upstream. For those who brave the mist and walk out to the edge of the falls, sedimentary interbeds are well exposed on several levels below the trail (Fig. 37A).
The formation of Gullfoss has been active since the retreat of the last glaciers ~10,700 years ago (Thordarson and Höskuldsson, 2014), although uncertainty exists surrounding the origin of Gullfoss and 70-m-deep, 2500-m-long Hvítágljúfur. An average erosion rate of 25 cm/yr could account for the gorge’s dimensions (Thordarson and Höskuldsson, 2014), but other researchers suggest that the erosion happened through a series of jökulhlaups during glacial retreat (e.g., Tómasson, 1993).
Around the upper rim of Gullfoss, glacial striations are observable in the bedrock (Fig. 37B).
Continue on Route 35 north. On leaving the parking lot at Gullfoss there are good views to be had of Langjökull, the ice cap to the north. Langjökull is the second largest ice cap in Iceland. The ice cap partially covers two volcanic systems: Prestahnúkar in the south and Hveravellir in the north, neither of which is known to have produced a historic eruption. Route 35 is called the Kjölur route through the interior of Iceland. It turns into a gravel road, and it used to be classified as a fjallvegur (mountain road), but that distinction was recently removed. The gravel road is good, but rough in places; though no longer a fjallvegur road, be aware that driving on it may void the insurance on two-wheel-drive rental cars.
The route drops through several stream valleys carved into intraglacial volcanic rocks before ascending to a pass between Bláfell (table mountain to east) and Skálpanes (shield to west). At ~27 km a road takes off to the left (west) that leads to Skálpanes where a close-up view of the Langjökull ice cap can be had. At 33.7 km, pull over by the hut before the bridge crossing the river Hvítá, Stop 2.5.
Stop 2.5. Central Highlands
This stop is at a bridge over the river Hvítá, which we saw downstream at Gullfoss, immediately north of the mountain Bláfell (a tuya/table mountain). The purpose of this stop is twofold: (1) to appreciate the diversity of subglacial and interglacial volcanoes of the central highlands of Iceland (e.g., Jakobsson and Johnson, 2012); and (2) to examine soil loss and desertification.
A better view can be had of the Langjökull outlet glaciers by proceeding across the bridge and ~300–500 m farther up the road. Two outlet glaciers of Langjökull are visible here, Norðurjökull (to the north) and Suðurjökull (to the south). Norðurjökull calves directly into the lake Hvítárvatn, the source of the Hvítá. The 1948 U.S. Army Map Service 1:50,000 Hvítárvatn map of the area, compiled in 1948 from aerial photos and road reconnaissance during World War II (1943–1945) and older maps, shows Suðurjökull also reaching Hvítárvatn. Between the outlet glaciers is Skriðufell, a tuya (Fig. 38). To the north of Norðurjökull and south of Suðurjökull Langjökull banks up against shield volcanoes, the post-glacial Sólkatla to the north and the interglacial Skálpanes to the south. Northeast of Sólkatla is another tuya, Hrútfell. To the ENE are the rugged peaks of Kerlingarfjöll, a central volcano with peaks of subglacial intermediate to rhyolitic volcanic rocks (e.g., Stevenson et al., 2009).
The area around the bridge is representative of the Central Highlands “desert.” It is not a desert in a climatological sense, but is a vegetative desert, nearly devoid of soil and vegetation. This is not the natural state of this land, but reflects soil loss. Charcoal pits have been found in this area indicating that the area was vegetated with birch forest until the sixteenth to seventeenth centuries (Arnalds, 2015; Ó. Arnalds, 2019, personal commun.). In addition to stream erosion of the grazing-impacted land, strong katabatic winds between the glaciers in the Central Highlands lead to particularly effective wind erosion and transportation. This effectively removed the andosol soils and winnowed the underlying tills leaving behind “sand lag gravel,” sand, gravel, and erratic cobbles and boulders (Fig. 38). Southwest toward Bláfell and southeast toward the lower hill Lambafell rofabards (e.g., Fig. 12) occur at the interface between areas of remnant vegetation and eroded areas.
Retrace the route to Laugarvatn.
DAY 3. HEKLA AREA
This day starts at Laugarvatn, where a fine view of Hekla can be had on a clear day. Today we will examine numerous facets of the geology of the Hekla area (Fig. 39) including: (1) Hekla tephras and lavas; (2) a rootless cone field; (3) the most extensive Holocene lava flow on Earth; and (4) features created by an early twentieth century earthquake which caused displacement in a Holocene lava flow. The day ends with a ferry ride to the Vestmannaeyjar (Westman Islands) positioning us to examine the geology there in the morning of Day 4.
Depart Laugarvatn on Route 37 south, taking it until it terminates. Turn left onto Route 35. Take a right on Route 31, signed for Skálholt, then a left on Route 30 for only 1.4 km, and then a right on Route 32. The road approaches the river Þjórsá for the first time ~3 km after turning on to Route 32. At 230 km long, the Þjórsá is the longest river in Iceland. The Þjórsá is intensively utilized for hydroelectric power. There are six hydroelectric power plants on the Þjórsá system with a total installed capacity of 1040 MW. Three more power plants on the lower Þjórsá are planned, two along the river on this trip route (and one farther downstream).
After the village of Árnes the road passes below low hills and cliffs of the Plio-Pleistocene Hreppar Formation, which consists of a sequence of sedimentary and volcanic rocks recording the early glacial history of Iceland (e.g., Geirsdóttir, 2011). Cliffs in the hill Bringa expose cross-bedded fluvial conglomerates (64.0907°N, 20.0028°W) within the sequence (Fig. 40).
At ~31.8 km on Route 32, soon after crossing a bridge, take the dirt road on the left (north) signed for Stöng (Route 327). Proceed ~1.5 km on the route to a rough pullout, Stop 3.1.
Stop 3.1. Þjórsárdalur
This point provides an excellent overview of the valley Þjórsárdalur. Below is an extensive rootless cone field with numerous small volcanic cones (Fig. 41A). Rootless cones (also called pseudocraters) are formed when lava flows over wetlands or shallow lake beds. The hot lava produces steam explosions in the underlying saturated material which erupt through the flow creating small cones whose tephra is a mixture of scoria from the lava and material from the substrate. The river Fossá has eroded through several cones below, facilitating a distant view of a cross section through the cones (Fig. 41B).
Rootless cone fields will be encountered at several other locations in this itinerary, including Álftavershólar in the south and Mývatn in the north. Similar features have been observed on Mars, and Iceland’s rootless cones have been studied as terrestrial analogs to the Martian features (e.g., Fagents and Thordarson, 2007).
The lava flow in which the rootless cone field is developed is Þjórsárdalshraun. It is 3200 years old and belongs to a group of flows called the Tungnaá lavas (Kaldal et al., 2018). These flows erupted from the Veiðivötn fissure swarm >40 km to the ENE, part of the Bárðarbunga-Veiðivötn volcanic system, which was also the source of the 2014–2015 Holuhraun lava in north Iceland. The Tungnaá lavas are notably plagioclase-phyric. Þjórsárdalshraun is isolated from most of the Tungnaá lavas, having flowed through a narrow ravine in an area to be visited in the next stop, Stop 3.2.
Continue north on Route 327. The road drops down to the level of Þjórsárdalshraun. Sporadic concentrations of white tephra occur along the base of the mountain to east; this is the tephra from the last major eruption of Hekla in 1104. Pull into the parking area at Stöng.
Stop 3.2. Stöng and Gjáin
Two sites will be visited at this stop: (1) Stöng, an archaeological site; and (2) Gjáin, a scenic area of complex lava flow-topography interaction. Unfortunately, the best signage for the Stöng archaeological site, including a building plan, is at the parking lot. Remember to review the signs before proceeding!
Stöng is a homestead (Fig. 42), excavated in 1939 (Thordarson and Höskuldsson, 2014). A modern structure covers the excavated ruins. One can enter the structure and tour the excavated remains of the dwelling. The presentation of the site is simple with rock walls covered by several layers of turf and limited other structural material. To get a better sense of life in this dwelling, visit the re-creation of the house and its outbuildings Þjóðveldisbærinn, discussed in the On Route section following this stop. Outside, on the north side of the structure, stones represent the positions of the smithy and the chapel, with abundant 1104 Hekla tephra present on the site. This homestead and others in Þjórsárdalur were abandoned after the devastation of the 1104 eruption of Hekla which deposited 10–30 cm of pumice at this location (Thordarson and Höskuldsson, 2014).
Continue hiking up the valley, staying at about the level of the Stöng site. The trail ends at a waterfall overlook. The cliffs surrounding the falls expose a series of lavas with irregular contacts and varied geometries of columnar jointing, reflecting the paleotopography through which these lavas flowed (Fig. 43). The uppermost flow here, between the waterfalls, is Þjórsárdalshraun. Numerous side-trails to the east lead to other overlooks and provide access to water level. Hike to an overlook with a view to the east to see springs emerging from different horizons in the lava sequence, demonstrating the significance of interflow-breccias as aquifer units (e.g., the Snake River Plain of Idaho).
Retrace the route south on Route 327, back to the paved road (Route 32).
Two Optional Stops are available near the intersection of Routes 32 and 327:
Option 1: West of the intersection (right), backtrack across the river Fossá and take a left (south) at the road on the other side signed for Hjálparfoss. The waterfall Hjálparfoss can be viewed from a parking area ~1 km down this road (64.1140°N, 19.8548°W). Hjálparfoss provides another look at the rootless cone process by exposing the interior of the Þjórsárdalshraun lava. Irregular geometries of columnar joints reflect conduits feeding through the flow where rootless cones formed above.
Option 2: Back on the east side of the Fossá, <200 m east of the Routes 32 and 327 intersection, a road heads south (right). This road leads to the Búrfell Power Station (270 MW installed capacity), but there is no public access. However, a short distance down the road, a left turn leads to the attraction Þjóðveldisbærinn (64.1177°N, 19.8230°W), a reconstruction of the Stöng homestead (the archaeological site at Stop 3.2).
Whether or not you made the Optional Stops, head east on Route 32, which ascends to a higher surface on Búrfellshraun (one of the Tungnaá lavas, which may correlate with Þjórsárdalshraun; Pinton et al., 2018; Kaldal et al., 2018). Hekla is prominent to the south. The road passes works associated with the Búrfell Power Station, including the diversion controls that either send flow to the power station or allow it to flow in its natural channel east of the prominent flat-topped mountain to the south, Búrfell, and over the waterfall Þjófafoss, Stop 3.4. Farther along Route 32, a bridge crosses a straight man-made ravine through which water flows. This water has been diverted 3.5 km through a tunnel from a higher dam and through Sultartangi Power Station (120 MW installed capacity) just upstream of the bridge. Many of Iceland’s hydroelectric facilities are like this, with a dam, then diversion of water through an underground tunnel and through a turbine some distance from the dam.
At the intersection of Routes 32 and 26, we turn south on Route 26. The junction offers the option to take a side-trip to Landmannalaugar, a scenically spectacular area with easy access to historic rhyolite lava flows and subglacial rhyolite deposits. If you wish to travel to Landmannalaugar without stream crossings, the only option is to stay on the paved road here which becomes Route 26 north and ultimately leads over the central highlands but also provides access to Route 208, a jarring route to Landmannalaugar. Heading south on Route 26 the road passes Route F225—the Landmannaleið—which provides a shorter route to Landmannalaugar with stream crossings. We continue south on Route 26 getting ever closer to Hekla, with Búrfell looming nearer on the west with the Þjórsá in between. A series of pumice quarries is on the east side of the road, many of which have been refilled. At the time of this publication, active quarrying of pumice is occurring in the southernmost of the pits. Pull over near the active quarry pits, but take care not to block access routes to the quarry floor, Stop 3.3.
Stop 3.3. Hafið Quarry
The Hafið quarry is an active quarry so please take care to not disturb work at the site. The quarry generally exposes 3–4 m of Hekla tephra. The summit of Hekla is 9.5 km southeast of the quarry. Before discussing the site we will provide a general introduction to Hekla volcano.
Hekla (1491 m) is a central volcano, the focus of the Hekla volcanic system. It is at the western edge of the Eastern Volcanic Zone, adjacent to the South Iceland Seismic Zone. Hekla is an intermediate to silicic stratovolcano (Fig. 44). From the northwest, as we view it here, it presents a relatively gentle profile because it is elongate NE-SW, a geometry controlled by the trend of the 5.5-km-long fissure system that feeds Hekla’s effusive eruptions. Hekla has erupted 18 times since the settlement of Iceland, making it the third most active Icelandic volcano in historic time (behind Grímsvötn and Katla). The most significant historic eruption took place in 1104, producing ~2 km3 of tephra, with a dense rock equivalent of 0.61 km3 (Thordarson and Larsen, 2007). The last eruption of Hekla prior to this publication was in February and March of 2000. This eruption was one of a series of four eruptions with an unusually short and regular recurrence interval of ~10 years, 1970, 1980, 1991, and 2000.
The 1104 Hekla tephra is also called the H1, with older prehistoric Plinian silicic tephras from Hekla being progressively abbreviated H3, H4, and H5 (what was originally designated H2 was later recognized to underlie H3 and be part of the Selsund pumice, HS). Other prehistoric tephras are noted by a H-letter combination, and historic tephras are represented by the year of eruption.
Most Hekla eruptions begin with a subplinian-Plinian explosive eruption followed by eruption of lava flows. Tephras produced by Hekla eruptions are commonly compositionally zoned; individual tephras can vary from 57 to 74 wt% SiO2 (Larsen et al., 1999). The composition of the initial explosive phase of the eruption correlates with the repose time between eruptions, becoming more silicic and more explosive with longer repose times (Þórarinsson, 1967). The lavas erupted from Hekla are basaltic andesite to andesite (~53–61 wt% SiO2; Chekol et al., 2011). These lavas erupted from the NE-trending fissure which is responsible for the elongate nature of Hekla (Thordarson and Larsen, 2007). There are no known basaltic eruptions of Hekla volcano, but basaltic volcanism has occurred in the Hekla volcanic system on fissures east and northeast of the main Hekla summit ridge fissure. The most recent basaltic eruption was in 1913 with lava erupted at two widely spaced locations: Mundafell and Lambafit (Jakobsson, 1979).
Returning to the site, the bulk of the quarried face is generally in the H3 tephra erupted 3100 years ago (Fig. 45A). A small conspicuously zoned tephra is deposited on top of the H3 and makes up the upper rim of the pit, transitioning to the surface above (Fig. 45B). This tephra is likely the HN or HM (J. Stevenson, 2013, personal commun.). The H3 here consists of lapilli to blocks up to ~30 cm. The H3 has an estimated dense rock equivalent of 2.2 km3, equal to the sum of all historic Hekla tephras combined (Thordarson and Larsen, 2007). The H3 phenocryst assemblage is plagioclase > fayalite + clinopyroxene > ilmenite + magnetite > apatite + zircon (Weber and Castro, 2017). The zonation of H3 is more silicic at the base and mafic at the top. The H3 tephra whole-rock composition varies from 56 to 69 wt% SiO2, with glass being up to 72 wt% SiO2 (Sverrisdottir, 2007)
The origin of Hekla magmas has been the subject of considerable study, and there remains some controversy about the relative importance of fractional crystallization versus crustal anatexis in the origin of the silicic magmas. The basaltic andesites that dominate the effusive products originate by ~62%–83% fractional crystallization of a parental basaltic magma (Sigmarsson et al., 1992; Moune et al., 2007). Andesites which are the most frequent explosive products are produced by magma mixing between basaltic andesite and dacite magmas (e.g., Sigmarsson et al., 1992; Sverrisdottir, 2007). The largest Plinian eruptions at Hekla are dominated by dacitic to rhyolitic compositions (the latter derived by fractional crystallization of the former; Schuessler et al., 2009), and are commonly zoned as the H3 is here. The vertical zoning is interpreted to be an inversion of the zoning of the source magma chamber, produced by mixing between the silicic end-member and a basaltic andesite whose injection triggered the eruption (e.g., Sverrisdottir, 2007). Most workers have interpreted the silicic magmas as originating primarily by crustal melting (e.g., Sigmarsson et al., 1992; Sverrisdottir, 2007; Chekol et al., 2011; Bindeman et al., 2012), but Portnyagin et al. (2012) argue that they are produced primarily by fractional crystallization from mafic parent magmas.
Continue south on Route 26, and take a right (west) at a dirt road signed for Þjófafoss, ~0.75 km south of the southernmost pumice pit (mileage at time of this writing). Take the road ~3.7 km to a parking area at an overlook at the waterfalls Þjófafoss on the river Þjórsá, Stop 3.4.
Stop 3.4. Þjófafoss
Þjófafoss is an 11-m-high waterfall on the river Þjórsá. The discharge of the river here varies considerably depending on how much water is diverted through the Búrfell Power Station, discussed between Stops 3.2 and 3.3. The prominent mountain across the river is Búrfell, which consists of rocks equivalent to the Hreppar Formation: Plio-Pleistocene glacial and interglacial basalt lavas, palagonite, and sediments.
The basalt lavas of the canyon walls and adjacent plateau are part of the Tungnaá lava sequence that originated in the Veiðivötn fissure swarm >45 km to the ENE, part of the Bárðarbunga-Veiðivötn volcanic system (as discussed at Stop 3.1 and in the section between Stops 3.2 and 3.3). The Tungnaá lavas are commonly abbreviated THx with letters increasing from the oldest (THa, the “Great Þjórsá lava,” 8600 years old) to the youngest (THj 1850 years old). Three lava flows are exposed below the level of the overlook and one above (Fig. 46). The lowest flow exposed in the area of Þjófafoss, visible in the lower canyon walls as viewed ~200 m downstream is THa, the Great Þjórsá lava (Kaldal et al., 2018). This lava flow is at least 140 km in length (extending beyond the modern shoreline) and is thought to be the largest Holocene lava on Earth, with an estimated volume of 25 km3 (Halldorsson et al., 2008). The higher flows are THd, THf, and THi, the 3200-year-old Búrfell lava (Kaldal et al., 2018).
Although the Tungnaá lavas span >6700 years, they share in common a plagioclase-dominated porphyritic texture. The abundance of plagioclase phenocrysts declines over time in the sequence (Vilmundardóttir, 1977). Such plagioclase-ultraphyric basalts are common in Iceland (Hansen and Grönvold, 2000). Halldorsson et al. (2008) studied the Great Þjórsá lava and demonstrated that the plagioclase phenocrysts were not in chemical or isotopic equilibrium with the host magma. Thus, they follow Hansen and Grönvold (2000) in interpreting such plagioclase as xenocrystic, reflecting mixing of the host magma and plagioclase-dominated crystal mush.
Turn left (east) on Route 268 signed for Næfurholt. Continue on this road for 3.1 km to the edge of a prominent lava less than 30 m off of the east side of the road, Stop 3.5.
Stop 3.5. 1845–1846 Hekla Lava
The volcanic edifice of Hekla is constructed primarily of lava flows from its northeast-trending fissure. At this location the terminus of the 1845–1846 lava flow Næfurholtshraun (Figs. 47 and 48) reaches the level of the surrounding plains at an elevation of 130 m, 1360 m below and 11.6 km due west of the summit of Hekla. Just up slope the lava flowed through a valley, confining its width to as little as 67 m. At this location the ~7 m thick flow has an ‘a‘ā to block lava surface morphology and a steep margin typical of intermediate lavas. The lava is aphyric and basaltic andesite to andesite in composition (54.5–57.0 wt% SiO2; Chekol et al., 2011). The 1845–1846 lava has an estimated volume of 0.63 km3, (Þórarinsson, 1967), and the dense rock equivalent of the tephra from the opening explosive phase is 0.03 km3 (Gudnason et al., 2018).
The eruption began on 2 September 1845 with an initial sustained explosive phase estimated to have lasted about one hour (Gudnason et al., 2018). The tephra fallout reached the Orkney and Shetland Islands in less than 24 hours (Larsen et al., 1999). The eruption ended in April 1846. This eruption was followed by 101 years of quiescence leading up the 1947–1948 eruption. Some lava from the 1947–1948 eruption followed the same path as this lobe of the 1845–1846 lava, and its terminus is ~1.7 km up slope (Fig. 47).
Continue southward on Route 268 ~7.4 km to left turn (east) on a road signed for Selsund. Take this road ~1.4 km to a right turn (south); after 0.4 km make a left turn and continue ~0.3 km and park at a wide point in the road as is convenient, Stop 3.6.
Stop 3.6. Selsund Fault
This stop is to view the subtle topographic features produced by strike-slip faulting in a Holocene lava flow. Einarsson (2010) summarizes mapping work on surface expressions of earthquakes in the South Iceland Seismic Zone, and Bergerat et al. (2003) provides an excellent explanation of the nature of the surface structures formed and their interpretation. Common features observed include: fractures (with or without vertical displacement), open fractures, and push-ups. The push-ups occur at compressive step-overs between fracture segments.
This site is near the southern end of the Selsund fault and the structures are interpreted to have formed during the Selsund earthquake on 6 May 1912. This earthquake was the first instrumentally recorded in Iceland and had an estimated magnitude of 7 (Bellou et al., 2005). This site has been mapped in detail by Angelier et al. (2004) and Bellou et al. (2005), and what is described below should be understood to be based on their mapping and interpretations.
From the road walk south or south-southwest (depending on where you parked) toward several low hillocks (Fig. 49). You will encounter the first hillock before you reach heavily moss-covered lava. This first hillock is interpreted as a push-up, but it is not connected to obvious fractures. Continuing south into the moss-covered lava, a fracture (accentuated by sheep traffic) is encountered which leads to another push-up centered on 63.9356°N, 19.97080°W. Immediately south of this push-up is a fracture with vertical displacement, down to the east (Fig. 50). A larger push-up occurs starting ~100 m farther south (centered on 63.9344°N, 19.9711°W). South of this push-up is an open fracture.
Continuing south, through the open fracture and across a grassy area and back into moss at ~180 m the fault is represented by two sets of fractures and at this point two ENE-trending sheep trails cut across the fractures. Notably, both Bjarnason et al. (1993) and Angelier et al. (2004) mapped these sheep trails carefully and interpreted 2–3 m of right-lateral strike-slip displacement. The greatest and most distinct offset is at the northern track at the western fracture group (63.9328°N, 19.9717°W).
Retrace the route back to Route 26, and turn left (south). About 14 km after rejoining Route 26 the road passes between fields that have been managed differently with regard to grazing, and the difference in soil loss and vegetation that is developed on the same underlying geology is spectacular. Note the high soil loss on southeast side. About 22 km after rejoining Route 26 the road rises over hills of the inner Búði moraine (Preboreal) and at ~27 km the outer Búði moraine (Younger Dryas). These moraines were deposited by glaciers grounded in 20–40 m of seawater (Ingólfsson et al., 2010). Continue on Route 26 to the junction with Route 1, the Ring Road.
Take Route 1 east (left) through Hella and Hvolsvöllur. About 18 km outside of Hvolsvöllur turn right (south) on Route 254, signed for Vestmannaeyjar (with a ferry symbol). Continue until you reach the Landeyjahöfn ferry terminal. This harbor and terminal went into service in 2010, greatly shortening the trip to the islands, from nearly three hours from Þorlákshöfn to just 35 minutes.
On a clear day the trip on the ferry Herjólfur offers excellent views of the volcanoes Eyjafjallajökull and Katla, in addition to views of the islands. On the way into the harbor at Heimaey the ferry passes first Elliðaey then Bjarnarey (Fig. 51). Both are excellent examples of Surtseyan volcanism. Steep sea cliffs reflect erosion to the palagonitized hyaloclastite core of the volcanoes, and both have subaerial features visible from the sea (subaerial lavas within the tuff cone crater at Elliðaey and a cinder cone atop Bjarnarey). The route into the harbor at Heimaey passes through narrows between the palagonite cliffs of Norðurklettar to the north and 1973 lavas to the south.
DAY 4. VESTMANNAEYJAR AND KATLA
Our day begins in the Vestmannaeyjar (Westman Islands) with several stops on the island of Heimaey, including a museum visit. We will take the ferry back to the mainland and resume our loop around Iceland with two stops on the slopes of the volcano Katla (Fig. 52), below the Mýrdalsjökull ice cap. Several optional stops are provided that would be particularly suitable for someone following the itinerary, but not visiting the islands. The day ends in the south-coast village of Vík.
Introduction to the Vestmannaeyjar
Always windy, never disappointing: the Vestmannaeyjar (Westman Islands) are an excellent place to experience Iceland’s geologic past—and future. The Vestmannaeyjar archipelago comprises ~18 islands and skerries; the largest, Heimaey, is the only populated island. The Vestamannaeyjar comprise a single volcanic system that lies at the propagating tip of the Eastern Volcanic Zone and are products of volcanism along a (mostly) submarine fissure system that has been active <100 k.y. The islands are manifestations of submarine-turned-subaerial Surtseyan volcanism and are <20 k.y. in age. The youngest island is Surtsey itself, born in an eruption from 1963 to 1967. Over the next millennia, Iceland will continue to add bedrock as eruptions continuing along this segment of the Eastern Volcanic Zone result in pillow lavas capped by hyaloclastite and finally subaerial lavas and tephra, joining the existing Vestmannaeyjar to the mainland.
Heimaey was originally several separate islands that became connected over time. Norðurklettar, the cliffs along the northern part of the island, are the remnants of the oldest eruptive centers on Heimaey. They date roughly to the end of the last glaciation (13–11 ka) (Thordarson and Höskuldsson, 2014). Stórhöfði (Stop 4.3) formed ~6500 years ago and formed its own island, which increased in size when Sæfell erupted ~6200 years ago, producing a tuff cone structure (the visible remnant of which are Stóri-Stakkar and Litli-Stakkar, the sea stacks). The eruption of Helgafell ~5900 years ago produced extensive lava flows that joined Stórhöfði-Sæfell with the islands to the north to form the island of Heimaey. The effusive to Strombolian Helgafell eruption also produced the spatter cone still visible to the southwest of Eldfell (Stop 4.1), the most recent addition to Heimaey (Fig. 53).
Eruptive products in the Vestmannaeyjar follow the alkaline compositional trend and are dominantly basalt. Minor amounts of intermediate composition material are present only on Heimaey; their presence indicates more extensive differentiation processes that occur only at central volcanoes and suggest that Heimaey is a nascent central volcano (Mattsson and Höskuldsson, 2003). Eruptive materials indicative of submarine phreatomagmatic eruptions (e.g., tephra, hyaloclastite, tuff cones) and subaerial eruptions (lava flows, scoria cones) are common throughout the Vestmannaeyjar.
Eldfell is the cone that rises above the lava fields along the eastern edge of Heimaey on the outskirts of town (Fig. 53). There are several paths to the summit at time of writing: (1) the footpath north from Fellavegur, the road that runs along the southwestern side of Eldfell between Eldfell and Helgafell; (2) the path from Eldfellsvegur along the western side of Eldfell; or (3) the more substantial road that stems south from Eldfellsvegur on the northern side of Eldfell.
Stop 4.1. Eldfell
Eldfell and its lava flows resulted from eruptive activity that began on 23 January 1973 and lasted until 3 July 1973. At 1:55 a.m. on 23 January, a fissure opened along the eastern side of Heimaey and began erupting fire fountains. Amid fears that the fissure might lengthen, activity would increase, or the harbor or airport would be closed, the entire population of the island was evacuated to the mainland within hours. After several days, activity coalesced around a single vent and began to build the cinder cone Eldfell. Initially, lava flows moved downhill to the east and northeast, away from the town and into the sea. In a desperate effort to keep the lava from closing the entrance to the harbor, sea water was pumped onto the advancing flows to chill them. The effort worked—the harbor is actually more protected from wind and waves now—but also forced lava to inundate portions of town more extensively than the flow was proceeding previously. A combination of lava and tephra completely buried over 400 buildings and covered the majority of town with several meters of tephra (Fig. 54). Much of this tephra was later used to repave roads on the island, and the geothermal heat emitted from the lava flows heated the town for 25 years. Overall, Heimaey increased 20% in size due to addition of the new lavas (Thordarson and Höskuldsson, 2014).
Compositionally, the Eldfell magmas are relatively uniform mugearite that can be modeled by 30% closed-system fractional crystallization of olivine + plagioclase + clinopyroxene + Fe-Ti oxides from parental hawaiite (Furman et al., 1991; Mattsson and Óskarsson, 2005). Crystallization of this assemblage at ~8 kb pressure produced small magma batches (Furman et al., 1991), and likely occurred in less than 10 years based on U-series dating (Sigmarsson, 1996) and lack of equilibrium phenocryst assemblages in the eruptive products (Mattsson and Óskarsson, 2005).
On clear days, Eyjafjallajökull and Hekla (on the mainland) are visible from the crater rim.
Eldheimar Museum is a short 10 minute walk from the base of Eldfell and is visible from the western edge of the crater. The easiest route there is to descend Eldfell to the west, walk south (turn left) on Eldfellsvegur, and turn right (west) onto Fellavegur, then right (north) after 72 m onto Helgafellsbraut. The museum is ~150 m north at the corner of Helgafellsbraut and Gerðisbraut (a.k.a. Austurgerði).
Stop 4.2. Eldheimar Museum
Eldheimar—the “Pompeii of the North”—opened in 2014 and quickly became an award-winner. The exhibition comprises two parts: the lower floor of the museum is centered around the excavation of a home buried in the 1973 Eldfell eruption, a discussion of the eruption, and its impact on the island’s inhabitants; the upper floor is dedicated to the 1963–1967 eruption of Surtsey, the youngest of the Vestmannaeyjar. This excellent exhibit used to be in a separate Surtsey Museum.
From Eldheimar, head south on Helgafellsbraut for 115 m. Turn right (west) onto Fellavegur. After 400 m, take a sharp right onto Strembugata. Continue for 500 m, then turn left (south) onto Höfðavegur. Stay on Höfðavegur for 5 km (the road changes names to Stórhöfðavegur after ~1.5 km when it merges with another road) until the road ends at Stórhöfði. Near the south side of the airport, the remnants of the tuff cone Sæfell are visible along the east side of the road.
Stop 4.3. Stórhöfði
Stórhöfði is the southernmost of the original small islands that coalesced to form Heimaey. From the parking area, we will take a short hike through the pāhoehoe flow that caps Stórhöfði to view the southern islands and sea stacks in the Vestmannaeyjar. They form directly in line with the propagating rift, which ends at Surtsey (Fig. 55).
Taking the return ferry back to the mainland, you will have a different light to examine the islands. Once back on the mainland, return to Route 1 and turn right, resuming eastward travel. You will very quickly cross the bridge over the river Markarfljót. The Markarfljót is an outwash stream for outlet glaciers on the north side of Eyjafjallajökull and the west side of Mýrdalsjökull. It is here that jökulhlaups during the 2010 eruption of Eyjafjallajökull threatened the bridge. Authorities destroyed sections of the embankments on which the road runs in order to preserve the bridge.
On the east side of the Markarfljót is the stunning, but often crowded, Seljalandsfoss. Our itinerary stays on Route 1, but it is worth mentioning that the beautiful Þórsmörk area lies up Route 249, the junction here. Unfortunately the route into Þórsmörk involves significant river crossings. Continuing on Route 1, the road passes south of high cliffs, often featuring wispy waterfalls (which get blown upward on particularly windy days) and springs emerging from the cliffs. These cliffs were cut by ocean wave erosion when relative sea level was high in early post-glacial time, after eustatic sea level rise but before isostatic rebound. Post-glacial fluvial sedimentary aggradation pushed the shoreline 2–7 km farther south.
At 20.8 km there is a pullout on the left (63.5429°N, 19.6627°W) that provides the best opportunity, on clear day, to get a relatively close view of Eyjafjallajökull. The 2010 eruption of Eyjafjallajökull forced the evacuation of the farms in the area, including Þorvaldseyri between here and the volcano. The junction to the road to Skógafoss, another spectacular waterfall, takes off to the left (north) at 29.1 km. This is also the starting point for the hike to the Fimmvörðuháls pass area where the 2010 eruption sequence began with effusive basaltic eruptions between Eyjafjallajökull and Mýrdalsjökull. The hike is a vigorous full-day round-trip (~25 km).
Several kilometers after Skógafoss, the road passes through lupine fields that were planted to stabilize windblown ash from the 2010 eruption. At 35.2 km, immediately after crossing the Jökulsá river turn left on Route 221, signed for Sólheimajökull. Park in a pullout on the right side of the road at 3.8 km, as you approach a notable bedrock ridge. This parking area is ~0.5 km before the main parking area for tourist traffic to the glacier Sólheimajökull. If you miss the intended pullout you can easily continue to the main parking lot, turn around and return. Stop 4.4
Stop 4.4. Sólheimajökull
The purpose of this stop is to examine Sólheimajökull, an outlet glacier of the Mýrdalsjökull ice cap which covers the summit area of the volcano Katla.
From the parking area, walk a short distance along the road looking for a trail that takes off to the east; once found, follow the trail (Fig. 56). For a period of years this trail provided the principal access for glacier walks on Sólheimajökull. The trail heads northeast up a small valley and is ~1.5 km in length leading to an overlook of Sólheimajökull a bit above its terminus. The mountain ridge that separates this valley from the main glacial valley of Sólheimajökull is called Jökulhaus. The trail begins by ascending over low moraines. Several recessional end moraines are passed along the route. Of note is a small but well-defined moraine at 63.5299°N, 19.3490°W. This point marks the position of a lobe of Sólheimajökull that flowed on the east side of Jökulhaus during its maximum recent advance at ca. 1904 when the glacier was thick enough to cover the northern end of Jökulhaus (see 1904 map, Fig. 56). The glacier extended farther down valley and covered much of Jökulhaus at ca. 1860 (Friis, 2011).
The fluvial sediment and channel pattern along the floor of the valley clearly reflect stream flow, but upon arriving at the end of the hike, one finds that there is no current source. The trail ends at a steep drop off overlooking the surface of Sólheimajökull. As recently as 2007 the valley we ascended carried outwash because the glacier was significantly thicker at this point (aerial photographs in Friis, 2011). Immediately south of the glacier overlook point is a small (~3.5 m) round hill interpreted as a kame (Fig. 57).
Sólheimajökull is widely known for its rapid retreat, which was featured, via time-lapse photography, by James Balog of the Extreme Ice Survey, in the 2013 documentary film “Chasing Ice.” The glacier has retreated ~625 m from 2007 to 2015 (Burkhart et al., 2017).
Focusing on the glacier, a conspicuous tephra layer can be traced along the near southeast margin of the glacier; its folded trace can be followed back across the glacier (Figs. 56 and 58). This is the Katla 1918 tephra, ~20 cm thick within the glacier. The 1918 eruption of Katla is, at the time of this writing, the last confirmed Katla eruption (several subsequent jökulhlaups may have reflected small subglacial eruptions). Katla volcano is discussed in the next stop description (Stop 4.5).
Return along the trail back to the parking area. The main parking area 0.5 km farther up the road is an Optional Stop in this itinerary (63.5303°N, 19.3708°W). From this parking area, a short hike leads to the terminus of the glacier. About 400 m from the parking area is a sign showing the annual glacier retreat observations made by Hvölsöllur School. Due to the rapid retreat of Sólheimajökull at this time, the length of the hike and character of the terminus change from year to year. Guide services provide glacier walks from the main parking area.
Return to Route 1 and head east. At ~3.4 km turn left (north) at the road signed for Mýrdalsjökull and Sólheimahjáleiga. At ~1.2 km is the parking area for a glacier tour operator. It may be necessary to park in their parking lot; if you do so, ask for permission. The road continues but has become very poor in recent years and may soon be impassable to all but genuine high-clearance 4-wheel drive vehicles. Whether on foot or by vehicle, continue up the road until another road branches off on the left at a little more than 1 km past the parking lot. Turn into this and park if you have a vehicle, Stop 4.5.
Stop 4.5. Sólheimar Ignimbrite
The objective of this stop is to visit the Sólheimar ignimbrite, the product of a large silicic eruption of Katla volcano. A general discussion of Katla precedes the stop description.
Katla is a large volcano with a 600–750-m-deep, 14-km-wide, summit caldera, filled by the Mýrdalsjökull ice cap (Björnsson et al., 2000). It is the central volcano of the Katla volcanic system. Katla lies within the Eastern Volcanic Zone (in the southern portion sometimes identified as the South Iceland Flank Zone). With at least 21 known historic eruptions (Larsen, 2010), Katla is generally considered the second most active volcano in Iceland during historic time (after Grímsvötn and just ahead of Hekla). The last confirmed eruption was in 1918. Katla is noteworthy from a hazards perspective for several reasons beyond its frequency of eruption. Eruptions of Katla are subglacial and thus have a high explosivity and fragmentation producing a threat to air traffic across the North Atlantic. These eruptions also produce jökulhlaups, threatening communities and infrastructure below. While most eruptions from Katla are basaltic, it is also capable of producing large explosive rhyolitic eruptions (e.g., Larsen, 2010) creating a greater regional impact (e.g., Elíasson, 2014).
Katla is a bimodal volcano, erupting predominantly transitional alkali basalts and mildly alkali rhyolites, with subordinate intermediates of basaltic andesite to dacite composition (Lacasse et al., 2007). Based on texture and geochemistry, the sparse intermediate magmas are interpreted as the products of magma mingling and hybridization (Lacasse et al., 2007).
The fissure swarm of the Katla volcanic system extends northwest from Katla as the Eldgjá fissure system. Eldgjá is notable for the 934–940 eruption which produced not only a large volume of tephra, but also the largest effusive eruption on Earth in the last 1100 years (Thordarson and Höskuldsson, 2014). This eruption will be discussed further at Stop 5.2.
There are two options for the stop. First, for those seeking to avoid steep grassy slopes and short stream-leaps, there are limited partial exposures of the ignimbrite available in ledges west and southwest of the stop location (coordinates above), staying at plateau level. If staying at this location, one can seek features described below for the main site.
For those who are comfortable with descending steep grass slopes and leaping across small streams, an excellent outcrop of the Sólheimar ignimbrite (Fig. 59) can be reached as follows. The destination is a set of tan/buff rounded ledges north of the stop coordinates, visible across several gullies at (63.5051°N, 19.3207°W). From the recommended parking area, continue up the side road until it rejoins the main road and stay close to the edge of the drop off to the small stream below. Just ahead, the stream goes over a small falls. The easiest place to descend the slope and step over the stream is just upstream of the falls. Ascend back up the other side, and angle northwest across the slope until you see a good place to descend to the next small stream. Cross the stream and ascend up the steeper lower slopes, then side-hill southwest until you reach the ledge exposures.
This site provides a complete exposure through the Sólheimar ignimbrite, and it has been described and interpreted by Tomlinson et al. (2012) as follows. The main ignimbrite is 4.7 m thick and consists of an ash matrix with pumice clasts up to 20 cm. In addition to simple pumice clasts there are also mingled pumices and finer scoria, obsidian, and lithic clasts. Below the ignimbrite is a thin (<0.2 m) fine surge deposit, two reversely graded airfall deposits (1.25 m), and a pyroclastic flow deposit (0.9 m). Plagioclase and clinopyroxene are the only phenocrysts. Overlying the ignimbrite is a series of coarse (lapilli dominated) surge deposits with abundant pumice, including coarse tail grading (2.75 m, with top not exposed). This section described above is interpreted to represent different phases of a single eruptive event. Though rhyolite dominated, the juvenile clasts include basaltic andesite and andesite (icelandite). Tomlinson et al. (2012) interpret the silicic composition to have originated by fractional crystallization of a basaltic parent magma, while Lacasse et al. (2007) interpret Katla rhyolites to originate by crustal melting, though both interpret intermediates as the products of mixing.
The Sólheimar ignimbrite has not been directly dated. It is often correlated with the Vedde ash (e.g., Lacasse et al., 2007), a 12.1 ka tephra found in marine and terrestrial deposits across Europe and the North Atlantic and in Greenland ice cores (Lane et al., 2012). However, Tomlinson et al. (2012) note that the ignimbrite was not deposited on ice as would be expected at ca. 12 ka and lacks the bimodal component of the Vedde ash so this correlation is not certain. There are other candidate Katla tephra correlations for the Vedde ash. The estimate of the volume of the Katla eruption associated with the Vedde ash is ~3.3 km3, making it the largest known explosive eruption of a modern Icelandic volcano.
Return to Route 1 and turn east (left). For the first several kilometers, the steep flat-topped hill Pétursey is ahead on the right (south). Pétursey is a Surtseyan volcano that formed as an island before fluvial aggradation pushed the shoreline south of it. A similar feature, Hjörleifshöfði, will be visited at Stop 5.1.
About 10.3 km after rejoining Route 1, a right turn (south) on Route 218, signed for Dyrhólaey, leads to an Optional Stop (63.4037°N, 19.1038°W). The cape at Dyrhólaey (west of main parking area) is commonly described as the southernmost point on the mainland of Iceland. While this was true prior to 1918, the southernmost point of Mýrdalssandur to the east now extends farther south (see Stop 5.1). Dyrhólaey is lower than Pétursey but also constitutes the erosional remnants of a Surtseyan volcano. Both the explosive Surtseyan facies, dominated by palagonite tuff, and the later subaerial facies basalt lavas are accessible in outcrops.
Another turnoff for an Optional Stop is 17.9 km from rejoining Route 1 after Stop 4.5. To visit this stop turn right (south) on Route 215 signed for Reynishverfi. At the end of the road is a large parking area (63.4023°N, 19.0445°W) for a beach called Reynisfjara which forms a spit extending westward from the cliffs of Reynisfjall, in front of the lagoon Dyrhólaós, and nearly to Dyrhólaey. The attraction here is a spectacular array of columns exposed at the base of Reynisfjall (Fig. 60), representing multiple generations of igneous intrusion. Large (up to ~0.8 m diameter) vertical columns rising from the beach are interpreted as having formed in a sill and smaller columns above and north formed in an inclined sheet (Gudmundsson, 2017). When the tide is out columns can be viewed from below in a cave to the east of the main column exposure. The main mass of the sea cliffs are composed of palagonite tuffs and breccias intruded by dikes and overlain by lavas, again reflecting emergence from a submarine environment. The Reynisdrangar pinnacles just off shore will also be visible from Vík. Note: the current is deceptively strong at Reynisfjara—stay well back from the sea.
Continuing eastward from the Optional Stop areas, Route 1 descends to Vík, the stopping point for this day.
DAY 5. SOUTHEAST COAST AND VATNAJÖKULL
The drive from Vík to Höfn along Route 1 offers a microcosm of arguably the most dynamic landscapes in Iceland—unforgiving sandurs; vast glaciers and their moraines and drainages; roadside lava fields from Iceland’s most historically impactful lava flows; eroded cliff faces; and towering volcanic edifices (Fig. 61).
Leave Vík heading east on Route 1. At ~5.5 km note the caves at the base of the cliff, Skipahellir. Boats were launched from the area of these caves, now ~2 km from the sea, until ca. 1660, demonstrating the subsequent fluvial aggradation (Björnsson, 2017). The edge of the Mýrdalssandur is encountered ~8.2 km east of Vík where a bridge crosses the river Múlakvísl. Mýrdalssandur is an outwash plain (sandur) from glacial outburst floods (jökulhlaups) originating under Mýrdalsjökull. Some jökulhlaups are associated with eruptions, and others are not. The Múlakvísl originates as outwash from the large piedmont outlet glacier Kötlujökull on the southeast side of Mýrdalsjökull. Kötlujökull causes the greatest proportion of jökulhlaups from the Katla caldera. The 1918 Katla eruption caused a major jökulhlaup across the sandur which peaked at 500,000 m3/s discharge (Tómasson, 1996). A 2011 jökulhlaup, apparently non-volcanic, washed out the bridge over the Múkavísl.
Approximately 12.4 km east of Vík, turn south (right) on a dirt road with a small sign to Hjörleifshöfði. The road is constructed on outwash of the sandur plain on the west side of the steep sided hill Hjörleifshöfði. At ~1.5 km, pause to appreciate exposures of columnar jointed basalt lava in the west-facing cliffs. Continue southward, wrapping around the southwest corner of Hjörleifshöfði to a parking area near a prominent cave ~2.6 km from Route 1. Stop 5.1.
Stop 5.1. Hjörleifshöfði
Hjörleifshöfði seems like an island in Mýrdalssandur, and that is just what it is. The steep-sided morphology and overall geology of Hjörleifshöfði, like Pétursey which was noted in the “On Route” section between Stop 4.4 and Vík, bear a distinct similarity to the islands of the Vestmannaeyjar from Day 4.
Hjörleifshöfði has been studied in detail by Watton et al. (2012, 2013). Hjörleifshöfði formed by eruptions when this area was south of the shoreline. The eruptive process will have been like Surtsey in 1963–1967, with submarine construction then phreatomagmatic subaerial eruptions, and some subaerial eruptions giving rise to ‘a‘ā lava flows like the one highlighted on the drive in as well as an intact cinder cone on the upper surface. Areas of silicic accretionary lapilli are found on the upper surface, and chemically match the Solheimar ignimbrite (Watton et al., 2012).
The cliff at this stop (Fig. 62) exposes three lithofacies of Watton et al. (2013): massive volcanic breccia (VB) low in the cliffs, overlain by imbricated planar cross-bedded hyaloclastite breccia (GHip), unconformably overlain by trough cross-bedded volcaniclastic sandstone/breccia (VStcb). None of these facies here is interpreted as a primary volcanic deposit. In the interpretation of Watton et al. (2013) these facies represent: VB = eroded remnant of an older edifice (distally reworked deposit); GHip = delta deposits from material leaving a submarine channel; and VStcb = material reworked by tidal and/or wave processes.
As in the Vestmannaeyjar, the steep margins of Hjörleifshöfði reflect wave erosion reaching the more palagonitized core of the volcano. Since the formation of Hjörleifshöfði and the erosional oversteepening of its margins, outwash of Mýrdalssandur has aggraded, pushing the shoreline southward leaving this former island >2 km from the sea. The shoreline may have been at Hjörleifshöfði at the time of settlement (Björnsson, 2017). Deposition during the jökulhlaups of the 1918 Katla eruption pushed the shoreline >600 m south (estimated from Tómasson, 1996) making this area (Kötlutangi) the new southernmost point of the mainland of Iceland.
Return to Route 1 and head east (right). The road continues through Mýrdalssandur, with strips of vegetation planted to control windblown sand along both sides of the road. Closures along this stretch of road are common; sandstorms with winds >40 m/s occur in all seasons and can easily sandblast vehicles and threaten human safety. Soon after more rugged hillocks begin appearing by the road a parking area on the left at ~20.2 km is Stop 5.2.
Stop 5.2. Eldgjá Eruption and Álftavershólar Cones
The small cones at this stop are a manifestation of the largest effusive eruption on Earth in the past 1100 years, the Eldgjá eruption of 934–940. This basaltic eruption produced 20 km3 of lava (when the portion buried under the Mýrdalssandur is taken into account, Sigurðardóttir et al. 2015) and 6 km3 of tephra (Thordarson et al., 2001).
The name Eldgjá (fire fissure) refers, geographically, to an 8-km-long canyon that is part of the 75-km-long Eldgjá fissure in the fissure swarm of the Katla volcanic system. The canyon is not literally a fissure formed during the 934–940 eruption, but rather is a graben that was reactivated during the event (Thordarson and Höskuldsson, 2014). The Eldgjá fissure also erupted along a 15 km length under Mýrdalsjökull producing at least 30 tephra units (Larsen, 2010). Eldgjá lavas flowed up to 45 km from the fissure to sea. However, good exposures of the lava are uncommon because it is covered in many places by the younger Laki lavas or outwash of Mýrdalssandur. Here we see the Eldgjá lava as the Álftavershólar rootless cone field (Álftaversgígar). As described for Stop 3.1, rootless cone fields form when a lava flows into a lake or marsh with a saturated substrate and underlying water flashes to steam, typically erupting a combination of scoria from the lava flow and lithic clasts from the underlying material. This is one of several rootless cone fields formed by the Eldgjá lava where it flowed into the coastal lowlands.
Get back on Route 1 heading east. For the next ~38 km until the town of Kirkjubæjarklaustur, the road cuts through lava flows: first from the 934 Eldgjá eruption, then flows from the 1783–1784 Laki eruption that in many places cover Eldgjá flows. Laki and Eldgjá lavas are often distinguishable by their moss cover—Laki flows tend to have a light green moss (Fig. 63) (Racomitrium, for the bryophyte lovers amongst us; Einarsson, 2005) that generally enhances the distinct flow morphologies, while Eldgjá flows often have a darker, less coherent moss cover and less distinct flow outlines and morphologies. Lavas from the Laki eruption flowed around the topographic high of the Síða highlands and coalesced in two locations. In both areas, the lava flow exploited a river gorge to channelize the flow and propagate the lava further. A larger flow field called Eldhraun formed near the gorge of the river Skaftá to the southeast, and a smaller field referred to as Brunahraun formed from flow through the Hverfisfljót gorge to the northeast. We will visit the southeastern flow (Eldhraun) at a Katla Geopark site, ~12 km from the previous stop.
Stop 5.3. Laki Lava
The 1783–1784 Laki eruption (the effects of which are mentioned in the Historical Icelandic Eruptions section of the Introduction) produced both lava and tephra that significantly impacted Icelanders in the Fire District. At the time of this writing a short film about the eruption can be viewed at the Tourist Information Office (Skaftárstofa) in Kirkjubæjarklaustur. The eruption occurred along a 27-km-long northeast-trending fissure with >140 vents and craters, including the 200-m-tall cone Laki (Thordarson and Self, 1993; Thordarson et al., 1996, 2003). Eruptive activity was characterized by intermittent episodes of several days of tephra production followed by longer periods of lava effusion. In total, the Laki lava field covers ~600 km2, composed largely of individual flow lobes 1–5 m thick (Guilbaud et al., 2005). The eruptive products of the 1783 eruption are remarkably homogeneous quartz-normative tholeiitic basalt (Thordarson et al., 1996). Mineral phases are typical of Icelandic tholeiites: plagioclase, clinopyroxene, olivine, and Fe-Ti oxides (e.g., Grönvold, 1984). Vesicularity and glass abundance are highly variable throughout the flow and tephra.
Lava first flowed from vents along the southwestern part of the Laki fissure through the Skaftá gorge on 12 June 1783 and steadily advanced southward and laterally. Here, near the modern channel of the Skaftá, the lava from the southeastern flow field advanced until about 13 July. Eldhraun itself is dominated by pāhoehoe of varying textures, including rubbly, slabby, and spiny varieties. The reader is referred to Guilbaud et al. (2005, 2007) for an extensive summary of the Laki flow field morphologies, emplacement mechanisms, and flow progression.
As the eruption progressed, activity along the fissure propagated to the northeastern part of the fissure. A second flow field—Brunahraun—formed ~20 km to the northeast of our present location after lava inundated the gorge of the river Hverfisfljót on 3 August. That flow remained active until the end of October 1783, and the remainder of Laki’s activity was confined to the fissure area until eruption cessation in February 1784. We will drive through Brunahraun between Stops 5.4a and 5.4b.
Return to Route 1 and turn east toward Kirkjubæjarklaustur. The tan-colored cliffs exposed along the road are part of the Síða Formation, a series of volcaniclastic and lava deposits of Plio-Pleistocene age that stretch for ~60 km near Route 1, ending at the Lómagnúpur promontory. Stay on Route 1 through the roundabout, heading toward Höfn. After 4.5 km turn left onto the road to Keldunúpur and park along the road.
Stop 5.4a. The Síða Formation at Keldunúpur
Alternating lava flow and hyaloclastite sequences of Plio-Pleistocene age exposed in paleo-seacliffs are prominent features in the Síða Formation. These sequences have received research attention due to the combination of their accessibility, magnitude, and the presence of large flame-like apophyses of lava that intrude overlying hyaloclastite (Banik et al., 2014). The Síða Formation is composed of repeating “standard depositional unit” (SDU) (after Bergh, 1985; Bergh and Sigvaldason, 1991) sequences comprising three cogenetic lithofacies: basaltic lava, a zone composed of basalt-clast-bearing hyaloclastite breccia containing localized cross-bedding structures, and overlying bedded hyaloclastite (Fig. 64). The Síða Formation unconformably overlies the Fljótshverfi Formation, an eroded sequence of flat-lying lavas and sedimentary units formed prior to the Ice Age (Thordarson and Höskuldsson, 2014). Any of the SDU facies may be absent at a given location and/or reappear farther along the outcrop. In addition, sedimentary diamictite or mudstone, although not formally listed as part of the SDU, is commonly associated with the sequences. Bergh (1985) and Bergh and Sigvaldason (1991) further divide the basaltic lavas into three sub-units: (1) regular columnar-jointed basalt and (2) cube-jointed basalt with an irregular thickness, with the latter as the source for apophyses of lava into the overlying hyaloclastite unit; and (3) pillow and isolated pillow basalts that are only locally present (Bergh, 1985). Typically, basalt in the Síða sequences consists of a colonnade of columnar-jointed basalt ~5–30 m thick overlain by an entablature of kubbaberg (cube-jointed basalt) ~5–20 m thick. This succession is a common feature in Icelandic lava/hyaloclastite sequences (e.g., Sigvaldason, 1968; Sæmundsson, 1970). The columnar-jointed basalt—where present—is ubiquitously overlain by cube-jointed basalt in the Síða Formation.
Several interpretations for eruptive environments and emplacement mechanisms and conditions have been proposed for the Síða deposits. Notable among these is the work of Noe-Nygaard (1940) first attributing the production of the Síða sequences to a subglacial eruptive environment. Other studies in the Síða area have produced two competing hypotheses regarding the eruptive environment and the conditions of emplacement of the Síða Formation: Bergh (1985) and Bergh and Sigvaldason (1991) contend that the units were sourced by eruptions from subaquatic basaltic fissures in the Eastern Volcanic Zone and emplaced on the adjacent shelf, while more recently Smellie (2008) proposed a deep subglacial setting for eruption, formation, and emplacement of both the lava and hyaloclastite units. Banik et al. (2014) also favor a subglacial eruptive environment, but their results indicate that the eruptive vent site was essentially subaerial during formation of at least some of the hyaloclastite and during eruption of the lava—results compatible with a relatively thin glacier that was melted through above the vent site in the early stages of eruption. These authors further suggest that the lava flow at Keldunúpur is in fact a sill formed where subaerially erupted lava invaded down into the unconsolidated hyaloclastite and that the flame-like apophyses represent breakouts of magma from the top of the sill intruding into the overlying hyaloclastite (Banik et al., 2014).
Return to Route 1 and turn left to continue driving east toward Höfn. Approximately 10.5 km after Keldunúpur, the road enters Brunahraun, the eastern arm of the 1783–1784 Laki lava. After another ~21.7 km, we will turn off Route 1 for a glimpse of Lómagnúpur—the tallest cliff face in Iceland with a vertical extent of >600 m. If the weather is clear, Skeiðarársandur and Skeiðarárjökull, an outlet glacier of Vatnajökull, will also be in view (Fig. 65).
Stop 5.4b. The Síða Formation at Lómagnúpur
The exposure at Lómagnúpur offers a spectacular glimpse of the true magnitude of the Síða Formation. At least 5 SDUs are visible in the southern and southeastern faces of the cliff, including some of the most distinct apophyses of lava that invade the overlying hyaloclastite in the Formation (Fig. 66).
Returning to Route 1, we will continue to head east across Skeiðarársandur toward Vatnajökull, Skaftafell, and Höfn. At ~6.7 km after rejoining Route 1 from Stop 5.4b, a road takes off to the left (north) which leads to the moraine complex discussed at Stop 5.5. At ~15.6 km there is a parking area is on the left (north) where we will stop to get an overview of the Skeiðarársandur, Stop 5.5.
Stop 5.5. Skeiðarársandur
This site is approximately halfway across Skeiðarársandur, an outwash plain (sandur) between the glacier Skeiðarárjökull and the sea (Fig. 67). It is the largest active glacial outwash plain on Earth (Gomez et al., 2000). Skeiðarárjökull is a major outlet glacier of the Vatnajökull ice cap, the largest glacier in Iceland. Spreading out on the lowlands in front of the mountains Skeiðarárjökull is classified as a piedmont glacier. Near the shore Skeiðarársandur is ~50 km wide. This great sandur is swept by glacial outburst floods (jökulhlaups) every 1–7 years (Gudmundsson et al., 1995). The dynamic nature of Skeiðarársandur is why it was the last part of Route 1, the Ring Road, to be completed, with the construction of the first bridge over the river Skeiðará in 1974 (Björnsson, 2017).
As with Mýrdalsjökull, some jökulhlaups are associated with volcanic eruptions and others are not (e.g., the frequent small non-eruptive jökulhlaups from the Skaftá cauldron, Einarsson, 2010). The most frequent source of jökulhlaups from Vatnajökull is the Grímsvötn volcanic system. The most recent major eruptive jökulhlaups were in 2011, 2004, and 1996. The 1996 jökulhlaup had a peak discharge of 45,000 m3/s. The parking area is on the east side of a dry channel called Háöldukvísl that carries flow during some jökulhlaups.
Between this site and the terminus of Skeiðarárjökull is a prominent moraine complex that marks the extent of the glacier around 1890 at the end of the Little Ice Age. The moraines can be reached by road a bit west of this stop. A smaller, well-defined, moraine present on the western edge of Skeiðarársandur is from ca. 1749 (Gomez et al., 2000). The glacier has retreated up to 4 km from its 1890 position. Skeiðarárjökull is a surge-type glacier and underwent a surge 1991 that advanced the front up to 1 km at a rate of 9.4 m/day (Waller et al., 2008).
Continue eastward on Route 1. An Optional Stop is at a parking area on the right (63.9846°N, 16.9603°W) after the road bends north toward the nearby glaciers. On display are the twisted remains of an earlier bridge that was destroyed by the 1996 jökulhlaup. There are dramatic views of the outlet glaciers Skaftafellsjökull and Svínafellsjökull, and to the east of the latter the volcano Öræfajökull with the highest peak in Iceland, Hvannadalshnjúkur (2110 m). Öræfajökull is a stratovolcano, the southern of two volcanic systems that define the Öræfi Volcanic Belt, the easternmost flank zone in Iceland. Öræfajökull has had two major historic eruptions, both of which produced substantial jökulhlaups, in 1362 and 1727 (e.g., Roberts and Gudmundsson, 2015). The 1362 event was one of the largest explosive historic eruptions in Iceland, a Plinian eruption that produced 2 km3 (dense rock equivalent) of homogeneous rhyolitic tephra (Selbekk and Trønnes, 2007). Öræfajökull’s eruptive products are bimodal and slightly sodic alkaline (Prestvik et al., 2001).
Do not take the turnoff for Skaftafell. Stay on Route 1 for another ~0.8 km, turning left at a road signed for Svínafellsjökull. Take this road to a parking area near the terminus of the glacier Svínafellsjökull, Stop 5.6.
Stop 5.6. Svínafellsjökull
The parking area is at the end of Hafrafell, the bedrock ridge separating the glaciers Skaftafellsjökull and Svínafellsjökull, near the proglacial lake in front of the latter. The strata in the cliff above the parking area (Fig. 68) consist of lava flows intercalated with subglacial volcanics, tillites, and other sedimentary rocks (Helgason, 2007). These rocks are part of a sequence of volcanic and sedimentary rocks that record the glacial and volcanic history of the area from 3.94 Ma (the age of the lowest lava here) to 0.21 Ma (the age of the lavas in a paleovalley carved into the sequence seen as a massive outcrop above the well stratified lavas, visible from the moraines as in Fig. 68) (Helgason, 2007; Helgason and Duncan, 2014).
A trail skirts the base of the cliff along the proglacial lake with excellent views of the terminus of Svínafellsjökull. Dikes and sills cut the strata, and dikes with glassy margins are encountered along the trail. Glacial striae are well developed on some bedrock surfaces.
Svínafellsjökull is an overdeepened glacier, with a subglacial surface below sea level in its lower reaches (Swift et al., 2018). The proglacial lake is confined by an end moraine complex. Time permitting, one can hike into the moraines to the southwest and south of the parking area. Lee (2016) summarizes the features of the moraine complex as consisting of a large composite moraine that first formed ~2500 years ago (Neoglacial) with superimposed Little Ice Age moraines and lastly recessional moraines deposited since 1935 including some well-defined push moraines (Fig. 68). During the Little Ice Age, and in earlier times in the Neoglacial, the termini of Svínafellsjökull and Skaftafellsjökull met below Hafrafell, as marked by medial moraines in the area (e.g., Hannesdóttir et al., 2015).
Note: at the time of writing, a warning has been in place since June 2018 for potential catastrophic slope failure of the mountainside to the east of Svínafellsjökull due to the appearance of extensive ground cracks. Unstable slope volume is estimated at 60,000,000 m3 and slope failure may occur all at once or in smaller events. Travelers are advised not to go on the glacier and to limit their time at viewpoints.
Return to Route 1. For those with more time to spend in the area, a rewarding Optional Stop is available by backtracking to the Skaftafell turnoff (800 m west), and proceeding down the road to the Skaftafell Visitor Centre for Vatnajökull National Park (64.0165°N, 16.9666°W). An excellent hike, which can be done as a loop or out-and-back, is the trail to the waterfall Svartifoss (spectacular columns), backtracking to the trail to Sjónarnípa (spectacular overlook on to the glacier Skaftafellsjökull).
The itinerary continues east on Route 1. The road passes several more outlet glaciers of Öræfajökull. About 12 km after resuming travel on Route 1, another Optional Stop is accessed from a parking area on the left, signed Háalda (63.9280°N, 16.7860°W). About 30 m ENE from the parking area is Háalda, a 50 × 70 m oval kettle (Fig. 69) formed by deposition of an ice block during the 1727 jökulhlaup from Öræfajökull (Roberts and Gudmundsson, 2015). Outwash was deposited around the block, which subsequently melted, leaving the pit.
Near the farmstead Hof, Route 1 approaches an erosional escarpment exposing 0.78–2.7 Ma volcanic strata extending south in front of Öræfajökull. These strata are also exposed in the conspicuous bedrock promontory on the shoreline to the south, Ingólfshöfði. The road wraps around the south end of Öræfajökull and approaches the high ridges of a moraine complex. Turn left at the southern end of the moraine ridges on a road signed for Kvíármýrarkambur and proceed to a parking area, Stop 5.7.
Stop 5.7. Kvíárjökull
The parking area is just within the largest latero-terminal moraine complex in Iceland, marking the maximum late Holocene extent of the glacier Kvíárjökull (e.g., Spedding and Evans, 2002). The latero-terminal moraine system is defined by a set of high ridges, Kvíármýrarkambur on the south and Kambsmýrarkambur on the north, rising to a height of >100 m above the surrounding coastal plains (Fig. 70). The moraine system reaches to within 1.5 km of the current shoreline. The proximal slopes of the moraines are considerably steeper than the distal slopes. The latero-terminal moraine complex is breached by outwash streams at the eastern end (parking lot area) and northeast, and where the southern moraine meets the mountain front and a small lobe of the glacier cut through the moraine.
Kvíárjökull is an outlet glacier of Öræfajökull, the small ice cap covering a stratovolcano of the same name introduced in the On Route section between Stops 5.5 and 5.6. Ice flowing off of the north side of the Öræfajökull massif merges with Vatnajökull. On all other sides, the Öræfajökull ice cap feeds outlet glaciers like Kvíárjökull. Because it extends beyond the mountain front, the lower portion of Kvíárjökull is classified as a piedmont lobe, though lateral spread is confined by the large latero-terminal moraines (Bennett et al., 2010). In front of the terminus are a proglacial lake and an outwash plain with scattered kettle lakes.
The high latero-terminal moraines are interpreted to have initially formed during Neoglacial cooling ~3200 years ago and have been progressively built by repeated occupations (e.g., Spedding and Evans, 2002). During Little Ice Age (ca. 1550–1900) advance Kvíárjökull again reached the extent marked by the latero-terminal moraine complex, as documented in the Danish maps from 1903 and 1904. Surveyed points on the glacier in these maps show that it would have been possible to see over the moraines from parts of the piedmont lobe, which is certainly not true today. The latero-terminal moraine system is breached at several points, including its end where the river Kvíá flows from the proglacial lake to the sea. A notable breach occurs at Kambsskarð where the southern moraine ridge meets the mountain front. A lobe of ice has spilled through this breach in the past and a beautiful set of accreted lateral moraine ridges on the mountain front (Iturrizaga, 2008) marks previous glacial occupations of this breach (Fig. 71).
The area immediately west of the parking lot consists of hummocky recessional end moraine deposits. Well-defined low (<3 m) inset lateral push moraines formed during retreat from the Little Ice Age maximum (Bennett et al., 2010) diagonal up the proximal side of the southern moraine ridge (Fig. 72). Though generally retreating since the end of the Little Ice Age, Kvíárjökull underwent periodic advances during the 1980s and 1990s (Bennett and Evans, 2012) and more recently advanced ~200 m along its northern terminus in the winter of 2013–2014 producing a push moraine (Phillips et al., 2017). Time permitting, one gains a better perspective hiking up on to Kvíármýrarkambur (the southern moraine ridge), and the longer walk along the ridge to the Kambsskarð breach (Fig. 73), encountering spillover channels along the way, is well worthwhile (~3 km each way).
The explanation for the large latero-terminal moraines of Kvíárjökull is that it carries an unusually high sediment load. Kvíárjökull is characterized as debris-charged (e.g., Spedding and Evans, 2002; Bennett et al., 2010). One of the most important sources of this debris is rockfall (Spedding and Evans, 2002). The glacial valley through which Kvíárjökull flows is unusual in Iceland for its steep continuous valley walls carved into thick rhyolite lavas (Fig. 71).
Returning to Route 1, continue northeast for ~19.5 km until Jökulsárlón, the main glacial lagoon of Breiðarmerkurjökull. North of the road are several other glacial lagoons (Fjallsárlón, Breiðárlón) and moraine complexes.
Stop 5.8. Jökulsárlón
Jökulsárlón (Fig. 74) is one of Iceland’s most popular tourist stops for a reason. Floating bergs calved from the retreating Breiðarmerkurjökull occupy a 110-m-deep lake with classic “glacial blue” water; wildlife (seals and eiders are common) is only meters away; and—if you’re lucky—the beach is awash with otherworldly ice blocks. Jökulsárlón represents an environment in flux. The lagoon formed in the 1930s; as Breiðamerkurjökull retreated >4 km from its Little Ice Age moraines during the twentieth century, the lagoon grew to its present (as of 2009; estimates at the time of writing suggest ~23 km2) surface area of 21 km2 (Schomacker, 2010)—an increase of almost 50% over the 2001 area of 15 km2 (Björnsson et al., 2001). As the lagoon grows, the heat capacity of the ice-marginal lake and the change into calving glacier fronts accelerates ice loss from Vatnajökull, creating a positive feedback of ice loss (Schomacker, 2010). Both the ice flux toward the lake and the calving rate into the lake have increased rapidly during the past two decades. Model calculations suggest a similar retreat rate of the calving front position for the next 70 years which would imply an average lake growth rate of 0.5 km/yr. At that rate, Breiðamerkurjökull would retreat from the lake depression after 200 years and almost vanish in 400 years (Björnsson et al., 2001). The clear blue water contrasts with the proglacial lakes in Iceland, reflecting the influence of seawater which enters the lagoon on some tides. We will likely make two stops here—one before the bridge to examine the lagoon and the Little Ice Age moraine that bounds it, away from most of the tourist activity, and another stop to appreciate the beach. In addition to icebergs, the beach provides the opportunity for a distant view of the ramparts of the Kvíárjökull latero-terminal moraines.
After leaving Jökulsárlón, continue northeast on Route 1 for ~75 km. The landscape is dominated by sandur, while Vatnajökull looms to the north. When Route 1 turns left near the town of Höfn, veer right onto Road 99 (Hafnarvegur) to enter Höfn.
DAY 6. THE EASTFJORDS
Today we will transition from the southern coast (Höfn) to the northern interior (Reykjahlíð), experiencing a great diversity of Icelandic geology along the way (Fig. 75). The day will start with a farewell overlook of the historically active Öræfajökull volcano and the southern outlet glaciers of Vatnajökull ice cap. As we descend from our overlook we will travel back in time (ca. 4–7 Ma) and find ourselves in a rare exposure of the plutonic roots of ancient central volcanoes (Vesturhorn, Slaufrudalur, Austurhorn). These intrusions, exposed by glaciation and visible along the coast, reveal the complex interactions of mafic, intermediate, and felsic magmas that likely occur beneath active central volcanoes in modern Iceland. After exploring this intrusive coastline, we will drive through altered mafic and felsic hills thought to be the eruptive remnants of the plutons we’ve left behind. We will then travel for several hours through the layered, tilted, basalt stacks—cross-cut by dikes and glacially carved into fjords—that typify the Icelandic periphery. Our day will end in the active rift valley of the Northern Volcanic Zone, where we will encounter geothermal fields, central volcanoes, rootless cone fields, maars, subglacial ridges, and table mountains.
Leave Höfn by heading north on Route 99. You will travel a short distance (<5 km) before reconnecting with Route 1. Turn right to head east. After traveling ~7 km, you will enter the tunnel Almannaskarðsgöng, which will transport you beneath the steep mountainous terrain of Almannaskarð. If the weather is fair and the views are clear, be prepared to make a sharp turn off from Route 1 (turning left/west) ~200 m after exiting the tunnel. You will ascend a steep road, reaching the crest of the mountain after driving ~1 km. Before the construction of the tunnel, this road served as the main passageway between southern Iceland and eastern Iceland. The through-road is now closed to the public, but (on a clear day!) you can enjoy sweeping views of the southern coast and its glacial features from a small but established tourist spot (parking lot, informative signs, etc.). If the weather is poor or the clouds or fog will obscure your view, skip this stop and continue with the driving directions below.
Stop 6.1. Southeastern Coast Overlook
Upon arrival at the top of the pass Almannaskarð (on a clear day!), you will get a panoramic overview of the landscape you experienced yesterday (Fig. 76). To quote Icelandic naturalist Stefán Stefánsson, “He who has not looked out across East Skaftafellssýsla from Almannaskarð on a fine summer’s day has not seen Iceland.” A touristic sign (from which this quote was taken) identifies major features in the landscape, including Öræfajökull and Hvannadalshnjúkur (Iceland’s highest peak located atop Öræfajökull), as well as dozens of other peaks, ridges, outlet glaciers, and glacial valleys emerging from Vatnajökull. You will get a perspective of the Höfn area (from whence you came), including several low islands scattered across the lagoon. The touristic sign at the viewpoint also provides information about the establishment of Höfn as a population center in southeastern Iceland, summarizing a history of productive agriculture and fishing coupled with the challenges of isolation in a dangerous and dynamic landscape.
After descending from the top of Almannaskarð (or after emerging from the tunnel if the weather prohibits a trip to the top), continue east along Route 1 for 39 km to reach Stop 6.2a. As you drive you will leave Vesturhorn (a peak composed of ca. 4.2 Ma intrusive rocks) in your rearview mirror and approach Austurhorn (a peak composed of ca. 6.5 Ma intrusive rocks) (zircon ages from Padilla et al., 2018). As you drive along the lagoon between these western and eastern “horns,” you will pass through a small valley cutting through the ca. 6.4 Ma Slaufrudalur pluton (64.3057°N, 14.9947°W). You will also cross the Jokulsá í Loni, a large braided river draining altered, Pleistocene and Pliocene rhyolitic hills (64.4211°N, 14.8828°W). You will reach your destination as you round the “horn” of Austurhorn (Fig. 77). You will encounter a valley bisected by a small river at 64.4211°N, 14.5378°W. Park near the bridge and walk along the road until you are close to a small but prominent outcrop, Stop 6.2a, ~30 m west of the road (Fig. 78).
Stop 6.2a. Austurhorn Roadside Outcrop
Former magma conduits—dikes and sills—are relatively common features that you will see as you travel around the periphery of Iceland. Former magma reservoirs—plutons and intrusions—are far rarer, largely due to Iceland’s shallow level of erosion (no more than a few km). This morning you will explore one of the few—and one of the largest—exposures of intrusive silicic rock in Iceland.
Approximately 6.5 Ma, along the SE coast of Iceland, mafic and felsic magmas were intruded into lavas and tuffs erupted from now-extinct Miocene central volcanoes (Álftafjörður and Lón; Walker, 1964; Blake, 1966, 1970; Ross and Mussett, 1976). These magmas, which we now refer to as the Austurhorn Intrusive Complex, were emplaced at ~2 km depth (Blake, 1966; Walker, 1960, 1964) and now occupy a footprint of ~15 km2 as delineated by a 1-km-wide contact metamorphism aureole (Walker, 1964; Blake, 1966, 1970; Ross and Mussett, 1976). These depths and dimensions match expectations for active magma bodies beneath modern central volcanoes (e.g., Furman et al., 1992b). This suggests that Austurhorn can provide rare insight into modern sub-volcanic magmatism at Icelandic central volcanoes. The compositional complexity observed at Austurhorn helps to explain the origin of mafic-felsic features observed in young eruptive deposits (e.g., magma mixing preserved in pumice that we will observe at Askja, Day 7).
The intrusive rocks at Austurhorn include gabbro (mafic), granophyre (felsic), and intermediate rocks. The intermediate rocks at Austurhorn are the result of complex mixing, mingling, and remelting of mafic and felsic magmatic endmembers (Walker, 1966; Furman et al., 1992a, 1992b; Padilla et al., 2016). The construction of Austurhorn was a dynamic process which involved multiple pulses of magma interacting with previously emplaced melts and mushes and rock. Zircon ages suggest that the construction of Austurhorn likely spanned ~320 k.y., beginning ca. 6.67 Ma and continuing until ca. 6.35 Ma (Padilla et al., 2016). The compositional end members (i.e., the most mafic and the most felsic) are somewhat older than the intermediate rocks exposed in the intrusive complex (Padilla et al., 2016).
The “horn” at Austurhorn (East Horn) is a ridge in the middle of the intrusive complex that contains the peaks Hvalnesfjall and Þúfuhraunstindur (Fig. 79). The majority of the gabbro at Austurhorn is confined to these peaks, and is mostly surrounded by granophyre (Padilla et al., 2016). The intermediate mixed, mingled, and melted portions of Austurhorn account for ~30–40% of the intrusive complex, exposed in a somewhat-continuous zone that extends from the base of Þúfuhraunstindur, south to the peninsula Hvalnes, and north to the point Krossanes (Fig. 79). This zone, bounded by granophyre and country rock, is referred to as the “net-veined complex” (e.g., Blake 1966, Mattson et al., 1986, Furman et al., 1992a, Weidendorfer et al., 2014) and the “mafic-silicic composite zone” or MSCZ (Padilla et al., 2016). In this guide, we refer simply to the “composite zone.”
We will spend our morning in the Austurhorn composite zone, exploring mafic-felsic interactions preserved in the rock record. For evidence of molten interaction between mafic and felsic magmas, look for pillow-like mafic enclaves with crenulate, cuspate, and/or chilled (glassy) margins (e.g., Fig. 80A). These indicate magma mingling, with differences in temperature and viscosity inhibiting compositional exchange between the two end-member magmas. In your quest for evidence of molten interaction between mafic and felsic magmas, you may also observe diffuse boundaries with a visible gradation from mafic to intermediate to felsic composition (e.g., Fig. 80B). These features can be interpreted as magma mingling between mafic and felsic endmembers, or disruption and dispersion of crystal mushes. In addition to evidence of molten-molten interaction, you will discover evidence of mobile magma that interacted with chilled rocks during construction of the pluton. As you consider molten-solid interactions, look for angular clasts of mafic rock (seemingly shattered or brecciated enclaves) suspended in felsic matrixes (e.g., Fig. 80C). You will also encounter tabular bodies of varying compositions (mafic, intermediate, felsic) recording histories of ongoing, repeated, cross-cutting intrusions expressed at multiple scales, such as dikes and sills that are meters in width; and veins that are millimeters in width (e.g., Fig. 80D).
Myriad examples of mafic-felsic interactions can be observed at a small outcrop ~30 m from Route 1 at Stop 6.2a. This is a great place to train your eye to recognize different features, and to discuss the stories recorded in the rock record. This outcrop is best followed by a walk along the beach to further examine the diversity of mafic-felsic interactions exposed across a larger, sea-polished, area.
Stop 6.2b. Austurhorn Lighthouse, Hvalnes
This stop is a continuation of Stop 6.2a: Austurhorn Roadside Outcrop. If weather, tides, and time permit, cross the road and walk back (SSW) along the coastline and reconvene at the lighthouse on Hvalnes (~2.5 km, ~1.5 miles, on easy terrain). Alternatively, you may choose to drive back to the lighthouse, park, and then examine outcrops exposed along the cliffs and shore. From the lighthouse, you will have a beautiful view of Hvalnesfjall and Þúfuhraunstindur (the gabbroic peaks rising to the north; Fig. 79), the lagoon Lón protected by an extensive baymouth bar, and the Vesturhorn intrusive peak on the western horizon (Fig. 81).
From the Hvalnes lighthouse, continue traveling northeast on Route 1 through the Eastfjords toward the small village of Djúpivogur (~54 km). Early in our drive from Austurhorn to the Eastfjords (within ~15 km), we will pass altered rhyolite in the hills and cliffs that rise above Route 1 (e.g., at 64.4713°N, 14.4980°W). As you look out the window, train your eye to recognize evidence of rhyolite in the landscape by distinguishing the different colors of alteration products (e.g., rhyolite altering to golden, warm hues, basalt altering to greens and grays). As we continue on our journey toward Djúpivogur we will (largely) leave rhyolite behind and begin weaving our way along the coastlines of fjords that were glacially carved into thick piles of basaltic lava flows erupted during the Tertiary. Djúpivogur is positioned below the prominent pyramidal peak of Búlandstindur and on the coast of the long, narrow, Berufjörður. Djúpivogur is home to the Langabúð cultural center and heritage museum, as well as an outdoor sculpture installation by Sigurður Guðmundsson called Eggin í Gleðivík, which represents the eggs of 34 local bird species. The main turnoff used to reach Djúpivogur from Route 1 is located at 64.6575°N, 14.3115°W. The outcrop that we will visit for Stop 6.3 is located ~50 m past the turn off for the Eggin í Gleðivík. Park on the side of the road at ~64.6610°N, 14.3064°W and walk ~35 m up the grassy hillside toward the linear dike exposure (Fig. 82).
Stop 6.3. Dike Observation and Overlook
The area around Djúpivogur has a high concentration of closely spaced dikes which cut through the thick pile of westward dipping Tertiary lavas that compose the bedrock of the Eastfjords (Fig. 83). The hundreds of dikes in this region tend to be 3–6 m in width, sub-vertical, striking NNE. They are part of a larger dike swarm that is ~9 km wide, extending ~30–40 km NNE-SSW. The highest concentration of dikes is found near the central axis of the swarm, with numbers declining near the margins (Thordarson and Höskuldsson, 2014). The number of dikes declines in frequency from the base to the surface of the Tertiary lava pile.
This dike swarm is likely associated with the 10–12 Ma Álftafjörður central volcano, located to the southwest of Djúpivogur (e.g., Moorbath et al., 1968; Walker, 1974; Gudmundsson, 1995; Paquet et al., 2007; Urbani et al., 2015). The dikes observed at this stop were likely sourced from a shallow magma body, and were emplaced at a maximum depth of 1.3–1.5 km below the original (pre-erosional) surface of the crust (Walker, 1974; Paquet et al., 2007). They are thought to be associated with a spreading rate of ~8 mm/yr, with magma-induced spreading outpacing tectonic-spreading (Urbani et al., 2015). The dikes here likely fed fissure flows, similar to what we observed from Laki on Day 5.
As you approach the dike outcrop on foot, notice the near-vertical orientation of the structure, which stands above the landscape due to its resistance to erosion relative to the jointed basalt flows through which it intruded. As you get closer to the dike, observe the near-horizontal orientation of its thermal contraction joints. Dikes, like lava flows, will lose heat to their cooler surroundings. The rocks contract perpendicularly to the direction of cooling, and form thermal contraction joints parallel to the direction of dominant heat loss. Because this dike lost its heat horizontally to the surrounding bedrock, instead of up to the sky and down to the ground (like a lava flow), it does not have vertically oriented columns. As you continue to explore the outcrop, look for the contact between the dike and the country rock. Those comfortable with a steep ascent may choose to visit the top of the dike exposure in pursuit of this quest.
Before walking downhill back to the vehicles, pause to look north across Berufjörður and gain an appreciation of the thickness of the Tertiary basalt pile and the number of lava flows involved in its construction. You will notice the gentle western dip of the volcanic strata, as well as a number of dikes (perhaps even the continuation of the dike on which you stand) cutting across the steep valley wall. Survey the features downslope of your current vantage point, and extend your view to the west. You will notice several exposures of neighboring dikes—smaller in size but vertically exposed, laterally extensive, and aligned in a NNE direction.
Resume travel northwest on Route 1 along the shores of Berufjörður. At ~10.7 km is a small pullout on the right before some small hay fields; pull in, Stop 6.4.
Stop 6.4. View of Flank Rhyolites of Breiðdalur Central Volcano
This location provides a view across Berufjörður at the tilted strata and rugged peaks on the other side (Fig. 84). The peaks are the erosional remnants of rhyolite lavas and intrusive plugs on the southeastern flanks of the ca. 9 Ma (Martin et al., 2011) Breiðdalur central volcano, the focus of Walker’s (1963) pioneering study. Walker (1963) termed these rhyolite lavas the Röndólfur group. Walker (1963) described the setting as follows, “Each mountain has a pedestal of basalt and andesite lavas cut by agglomerate-bearing vents, and each is capped by precipitous and sometimes pinnacled monolithic masses of rhyolite.”
The peak Stöng and ridge Slöttur (Fig. 83) expose a >200 m thick non-porphyritic rhyolite lava. Röndólfur is a porphyritic rhyolite lava intruded, on the north side of the ridge, by shallow plugs (Walker, 1963). These flank rhyolites lack the propylitic hydrothermal alteration that characterizes the core of the Breiðdalur central volcano to the northwest (Walker, 1963). The thin strata below the rhyolite peaks are a sequence of flank basalt and intermediate lavas whose dip reflects the combination of the regional tilt (~9°W) and the slope of the flanks for the Breiðdalur central volcano (3–7°S, with local variations) (Walker, 1963).
Following the ridge-top southeast of the rhyolite peaks basalt strata dominate the sequence all of the way to the ridge top. Also note lateral moraines on the lower slopes below these continuous sections.
Return to Route 1 and continue your journey north to Egilsstaðir. As you drive you will see road signs giving you the option of following three different routes to Egilsstaðir. You may choose to stay on Route 1 for your entire journey (~143 km) or turn off of Route 1 onto Route 939 at the head of Berufjörður (~9.5 km to turnoff, ~73 km total distance to Egilsstaðir), or turn off on Route 95 at Breiðdalsvík (~51 km to turnoff, ~133 km total distance to Egilsstaðir). Route 1 will take you along the low coast, weaving in and out of the fjords before turning inland at Reyðarfjörður (still on Route 1) for the final ~35 km to Egilsstaðir. The alternative routes (Route 939 and 95) provides beautiful views of waterfalls and dikes bisecting the Tertiary basalt stack, and will take you past Þingmúli, a Tertiary central volcano made famous by the early work of petrologist Ian Carmichael (e.g., Carmichael, 1964). In the pass area Route 95 also passes through the propylitized core of the Breiðdalur central volcano. The tradeoffs for the beauty of Routes 939 and 95 are the conditions you will encounter on your drive: very steep mountain passes with large stretches of unpaved roads and several hairpin turns. Driving can be quite treacherous in inclement weather and fog. For this guided tour, we recommend sticking to Route 1, but encourage experienced mountain drivers to make their own informed decisions. Gasser et al. (2018) provide an excellent guide to further explore the geology of east Iceland.
Continuing on Route 1, an Optional Stop is at approximately 49 km from stop 6.4 where the road rounds the point at Streitishvarf, by the lighthouse Streitisviti. Numerous dikes are exposed at this location including a 10–11 Ma composite dike studied most recently by Eriksson et al. (2011). The dike is 26 m wide and has a felsic core and mafic margins separated by a thin hybrid zone (Eriksson et al., 2011, and references therein).
Route 1 then passes through the town of Breiðdalsvík. The town hosts the Breiðdalssetur Research and Heritage Centre, a facility that includes geology displays and the collected papers of volcanologist/petrologist George P.L. Walker (open to groups and scientists only, by appointment). The route winds into Fáskrúðsfjörður with more exposures of west-dipping basalt-dominated stratigraphy, including the Hólmar, Grjótá, and Grænavatn basalt groups (Óskarsson and Riishuus, 2013; Óskarsson et al., 2017). At approximately 102 km from Stop 6.4 the road enters a 5.9-km-long tunnel which passes through the mountains that separate Fáskrúðsfjörður from Reyðarfjörður, reducing the trip by ~36 km. When you emerge from the tunnel you are looking out on Reyðarfjörður, and across the fjord to the ENE is the Alcoa Fjarðaál aluminum smelter. This smelter was completed in 2007 and is powered by the Fljótsdalur Power Station (hydroelectric), built primarily to serve this facility. At the head of the fjord Route 1 heads west then north through the U-shaped glacial valley Fagridalur. The low pass (342 m) and drainage divide in Fagridalur is extremely subtle; see if you can pick out where the northward drainage begins.
At Egilsstaðir—a bustling town that acts as central hub for resources and industry in Eastern Iceland—you will likely want to stop for gas, restroom facilities, and food at the intersection of Route 1 and Route 95. If you have time to explore the town, be sure to visit the East Iceland Information Centre (directly across Route 1 from the N1 gas station) and the East Iceland Heritage Museum, which is located ~0.6 km away.
Leaving Egilsstaðir, travel along the southeastern shore of the lake Lögurinn (also called Lagarfljót, a name that applies to both the lake and the river downstream of the lake) using Route 95 (retracing your route if you came this way). After traveling for 11 km you will reach a fork in the road. Veer to the right (west) and continue on Route 931 for an additional 20 km. Your drive will keep you close to the shore of Lögurinn as the road winds through a tall, dense, birch forest (a rare Icelandic experience!). For this guided tour, with a long journey ahead to Reykjahlíð, we will stop at an overlook of the bridge that crosses Lögurinn (Fig. 85) before turning back to Egilsstaðir and Route 1. To reach this (informal) viewpoint, look for a small turn-off road on the left (~65.0701°N, 14.8294°W) just before Route 931 takes a broad, sweeping, turn to connect with the bridge. Follow the small road uphill for a short distance (~100 m) before parking and leaving your vehicle to take in the view (~65.0693°N, 14.8299°W). If you are on a less compressed schedule, take the time to cross the bridge, turn left (southwest) on Route 933, and travel an additional ~6.5 km to the Snæfellsstofa Visitor Centre (65.0443°N, 14.9476°W). This charming establishment has an educational exhibit (and interpretive guides on staff) that will help you learn about local flora, fauna, geology, climate, and the eastern sectors of Vatnajökull National Park.
Stop 6.5. Hallormsstaðaskógur and Lögurinn Overlook
From this elevated vantage point, you can see several prominent features on the landscape, including Tertiary lavas exposed in the steep cliffs across the valley, birch and larch trees of Hallormsstaðaskógur National Forest, and Lögurinn fed by the braided river channel of the river Jökulsá (Fig. 85).
At the time of settlement (ca. 874) ~30% (>28,000 km2) of Iceland was covered by forests dominated by downy birch (e.g., Snorrason, 2004; Pesonen et al., 2009). As of 2008, forest coverage was only ~1% (121 km2) (Snorrason, 2004), due to centuries of deforestation in pursuit of grazing land for livestock, firewood, charcoal, and construction timber (Loftsson, 1993; Snorrason, 2004). While this 96% decline in forest cover is shocking, the modern forest coverage is as substantive as it is due to reforestation efforts that began in 1899 (Bragason, 1995). In modern times, Iceland has planted more than one million seedling trees per year beginning in 1985, with that number rising to four to five million seedlings per year at the turn of the twenty-first century (Snorrason, 2004). These reforestation efforts have been part of the Icelandic Forest Inventory project, which was initiated as a show of commitment to the Kyoto Protocol (Snorrason, 2004).
Hallormsstaðaskógur National Forest, where you find yourself at this stop, is managed by the Icelandic Forest Service. It is both Iceland’s first established (1905) and most expansive (740 km2) protected forest land. This forest, like many in Iceland, is dominated by birch trees (particularly downy birch, which is native to Iceland), accompanied by several introduced species, such as Siberian and Russian larch (e.g., Pesonen et al., 2009), and varieties of pine, spruce, and willow (Bragason, 1995). These trees have been successful in Icelandic reforestation projects due to their tolerance for infertile and eroded soils (Loftsson, 1993; Blöndal, 1993; Pesonen et al. 2009).
Lögurinn occupies a glacially carved U-shaped valley. At ~25–35 km in length (Thordarson and Höskuldsson, 2014), ~2.5 km wide (at its widest extent), and ~112 m deep (at its deepest), Lögurinn is Iceland’s third largest lake (Thordarson and Höskuldsson, 2014). This glacially carved valley reaches ~90 m below sea level, and is famous for being home to the fabled Lagarfljótsormur (Lagarfljót Worm), a water monster likened to the Loch Ness Monster in Scotland. While you might not spot the fabled Lagarfljót Worm at this overlook, you will see an interesting fluvial interaction where the sediment-laded braided channels of Jökulsá (near the confluence of the Jökulsá í Fljótsdal and Kölduá) encounter the relatively still, deep, water of Lögurinn (Fig. 86). A great deal of sediment falls out of suspension at this transition point, creating large sandbars and sediment lobes. Much of the sediment that stays in suspension and continues downstream in the calmer waters is likely glacial flour derived from the meltwater headwaters at Vatnajökull. A hydroelectric dam downstream on the Lagarfljót helps to power the region with a capacity of 7.5 MW.
After taking in the view return to your vehicle and follow your path back Egilsstaðir (20 km NNE on Route 931, 11 km NNE on Route 95 to the intersection with Route 1). You may wish to stop for fuel and refreshments again, as this will likely be your last opportunity to do so before arriving in Reykjahlíð (165 km). When you are ready, reconnect with Route 1 heading north and west toward Reykjahlíð (from Lagarfljót and Route 95 you would turn left; from the gas stations and shops at Egilsstaðir you will turn right). On the drive from Egilsstaðir you will continue to see the tilted strata and vegetated hilltops of the Tertiary basalt pile until, all of a sudden, you find yourself transported to an otherworldly, barren landscape. Pull into a parking lot on the right (north) side of the road ~96 km from Egilsstaðir. Stop 6.6.
Stop 6.6. First Impressions of the Rift
You have entered the Northern Volcanic Zone. Tomorrow, on your excursion to Askja, you will travel along the axis of the rift and see the young volcanic features of this inhospitable landscape up close and personal. Today (if views are clear), we take advantage of this vantage point for a preview of the day ahead (Fig. 87). You will observe barren, flat lava plains interrupted by steep, subglacially erupted ridges rising in parallel along the strike of the rift. In the middle distance you will see Herðubreið—a prominent table mountain that holds the regal title of “Queen of Iceland.” On a clear day, you will be able to see as far south as the flanks of Askja and the icy expanse of Vatnajökull. After stretching your legs and taking in the view, return to your vehicle and continue to observe these rift-related landforms on your trek across the Northern Volcanic Zone to Reykjahlíð (68 km to go).
As you travel through the Northern Volcanic Zone to Reykjahlíð you will become accustomed to the vast lava fields, steep subglacial ridges, and table mountains dotting the landscape to the south (Fig. 87). You will start to notice large, regularly spaced cairns on the north side of the road, which served as important navigation guideposts before the development of Route 1. As you approach Reykjahlíð you will see a subglacially emplaced rhyolite dome named Jörundur on north side of the road positioned just outside the caldera rim of the Krafla central volcano (Fig. 88; 65.6460°N, 16.6416°W; Jónasson 1994). As you crest Námafjall before descending into Reykjahlíð, you will get a view of Mývatn (Midge Lake) and some of the phreatomagmatic landforms (maars, rootless cones, etc.) that you will explore on Day 8 of this trip. Upon arrival in town, navigate to your accommodation for the night.
DAY 7. ASKJA (DYNGJUFJÖLL)
Today we leave the beaten path and venture into the Icelandic interior (Fig. 89), following the trend of the Northern Volcanic Zone to Askja (technically Askja is the caldera and Dyngjufjöll is the name of the mountain range/massif but both terms are used for the volcano in the literature). During the official 2019 GSA Northeastern Section Meeting field trip (for which this guide was originally designed), we will leave our vehicles in Reykjahlíð and travel into the interior in a chartered mountain bus. WARNING: Should you decide to venture into the interior on a self-guided tour, it is absolutely critical that you do so with a vehicle that is properly suited for an Icelandic “F-road” (a rugged off-road experience). Do not attempt the drive to Askja without four-wheel drive, high clearance, spare tires and tire changing tools, a full fuel tank, a charged mobile or satellite phone with pre-programmed emergency numbers, and extra food, water, and clothing. The road to Askja weaves through lava fields, cuts across shifting sands, and includes several river crossings (Fig. 90). Weather in the interior is more severe than on the periphery of Iceland. A sunny morning in Reykjahlíð might be followed by a windy, snow-covered, hike through the Askja caldera (Fig. 91). A sunny start to a hike at Askja might quickly, without warning, turn into a rainy, foggy, windy return. Katabatic winds from Vatnajökull can be cold and can carry abrasive sand. River levels can change rapidly with glacial meltwater on sunny afternoons. A trip to Askja is a rare treat, and one that will leave a lasting impression, but approach the experience with proper respect for the remote, extreme, conditions you will encounter.
As you leave Reykjahlíð, travel east on Route 1, retracing the path you took yesterday on your journey from Egilsstaðir. From the intersection of Route 848 and Route 1 on the northeastern shore of Mývatn, you will travel 33 km to the turnoff to Route F88 on the right (south) side of the road (65.6260°N, 16.2505°W). Pay careful attention to signs that may warn of hazards or road closures on your route to Askja. Continue south on Route F88. Please note that all distances quoted in this section were estimated using Google Maps measurements and may vary in detail from the reading on your odometer.
On Route Observation 7a. Hrossaborg Tuff Cone
This is not an official stop, but rather an observation to make as you drive through the landscape of the Northern Volcanic Zone. Today will be a long day. If you decide to stop and leave your vehicle, be efficient with your time. This same cautionary note applies to other descriptions for this day accompanied by the subtitle “On Route Observation” (instead of the “Optional Stop” designation used elsewhere in this guide). These observations can (mostly) be made from the window of your vehicle, or with impromptu pauses while you journey through the interior.
At the junction of Route 1 and Route F88 you will see a dissected tuff cone called Hrossaborg—or Horse Pen—which formed ~7000 years ago (Thordarson and Höskuldsson, 2014; Fig. 92). This small-scale eruptive feature has a volume of ~0.1 km3 and a height of ~50 m. The construction of this tuff cone began with a fire-fountaining eruptive event, but it transitioned to a violent phreatomagmatic constructive phase when the river Jökulsá á Fjöllum flooded the eruptive vent. From your vehicle you will be able to see tilted volcanic stratigraphy, which consists of basaltic pumice, lapilli, pyroclastic surge and flow deposits, and blocks and bombs. Thordarson and Höskuldsson (2014) provide a detailed explanation of the evolution and stratigraphy of Hrossaborg and the interested reader is directed to that reference for further information.
You will travel alongside a mighty glacial river, Jökulsá á Fjöllum, for the first ~60 km of your journey. You will cross smaller tributary rivers that feed into Jökulsá á Fjöllum several times on your journey into Askja (Fig. 91). The first occurs ~39 km from the Route 1 turnoff (65.3370°N, 16.0597°W); the second occurs 54 km from the Route 1 turnoff (65.2291°N, 16.1895°W); and the third occurs 59 km from the Route 1 turnoff (65.1921°N, 16.2166°W). This third river crossing is located near the Thorsteinsskáli Hut, which is positioned in an oasis-like patch of green vegetation in otherwise barren landscape, very close (~4 km) to the base of the steep-sided Herðubreið. You will continue driving for ~20 km before F88 terminates at F910 (65.0724°N, 16.3806°W).
On Route Observation 7b. Pāhoehoe, Desert Pavement, and Ventifacts
Between 65.6260°N, 16.2505°W and 65.1921°N, 16.2166°W
During the early stages of your journey south into the Icelandic interior you will observe the Jökulsá á Fjöllum on your left (east) as it flows from its headwaters at Vatnajökull northward to the Atlantic Ocean. A subglacially erupted ridge can be observed across the river on the eastern horizon (Fig. 93A; Fig. 87). You will also see outcrops of pāhoehoe lava emerging from sandur deposits associated with jökulhlaups from Vatnajökull that raged beyond the confines of the valley of the Jökulsá á Fjöllum (Fig. 93B). As you penetrate farther south, fine-grained sediment (clay, silt, sand, fine gravel) are winnowed away by the strong katabatic winds blowing from Vatnajökull. The result of this is vast expanses of desert pavement (tightly packed gravel surfaces) pockmarked by boulder-sized ventifacts that have been polished into aerodynamic shapes by windblown sediment (Fig. 91A). If you leave your vehicle, please tread lightly and stick to the road as much as is possible. Desert pavements are fragile surfaces and even small disruptions in the microtopography can initiate significant erosion as the wind continues to blow.
On Route Observation 7c. Herðubreið and other Prominent Landforms
When skies are clear, the table mountain Herðubreið, the Queen of Iceland, is a towering (1682 m) and distinctive landmark that catches the eye for miles around (e.g., Fig. 87; Fig. 91A; Fig. 94; Fig. 104; Fig. 108). Herðubreið is a classic example of a tuya (also known as a table mountain). Its lower flanks were erupted under thick glacial ice, creating a confined, conical, deposit of hyaloclastite (also known as móberg). Its upper 300 m are composed of subaerially erupted basaltic lava which armor the steep-sided landform. The elevation at which Herðubreið transitions from subglacial móberg to subaerial lava (~1350 m) has been used to determine a paleo-ice thickness of ≥800 m (Thordarson and Höskuldsson, 2014). Refer back to the Introduction section of this guide for more details about tuya formation.
If visibility is clear on this stretch of the drive, you should also be able to spot Vatnajökull on the southern horizon, Snæfell (a stratovolcano on the northeast margin of Vatnajökull) off to the southeast, Dyngjufjöll (Askja) to the southwest, and the Kollóttadyngja shield volcano to the west (Fig. 94).
On Route Observation 7d. 1875 Pumice
As you approach the junction of F88 and F910, and as you progress closer and closer to Askja, you will start to notice white and golden pumice scattered on the landscape. These pumice deposits emphasize the ripple marks in gravel deposits shaped by strong winds blowing north from Vatnajökull. This pumice was expelled from the Askja central volcano in the 1875 eruption which formed the Öskjuvatn caldera (Stop 7.1). Pumice will become more and more concentrated on the landscape as you approach Askja until it eventually starts to dominate the landscape (Fig. 95).
At the junction of F88 and F910, take the right (southwest) fork of F910 and continue for ~12 km before reaching your fourth river crossing (65.0466°N, 16.5851°W). At this fourth river crossing, you will be ~0.5 km and one additional stream crossing away from your arrival at the Dreki Hut (65.0421°N, 16.5946°W). This hut has restroom facilities that can be accessed for a fee. You may choose to stop and take advantage of this as there are no other facilities as you continue up in elevation to the interior of the nested Askja calderas.
When you are back in your vehicle after stretching your legs and using the facilities, continue uphill (west-northwest) on Route F894. This road will run adjacent to, then cut through, lava flows from an effusive caldera-fill eruption that occurred in 1961 (Fig. 96). Route F894 will terminate after ~8 km in the Vikraborgir car park (65.0670°N, 16.7251°W; Fig. 97). This parking spot serves as a trailhead for both the Öskjuvatn hike (Stop 7.1) and the 1961 lava flow exploration (Stop 7.2).
Stop 7.1. 1875 Öskjuvatn Caldera
Begin your hike by walking through cinder cones associated with the fissure from which the 1961 lava erupted (Fig. 97). Early in your hike, traveling south, the landscape will be dominated by a broad and flat caldera floor, surrounded by the high rim of a 10 ka caldera. The mountainous terrain forming the eastern caldera rim is called Thorvaldsfjall. This hyaloclastite formation (erupted subglacially) reaches 1516 m above sea level—the highest elevation of the Dyngjufjöll volcanic center (Sigvaldason, 2002). The western caldera rim is slightly lower, with a maximum elevation of 1380 m above sea level. This block (called Vesturfjöll) is composed of interglacial basaltic lavas, as well as hyaloclastites that were emplaced subglacially (Sigvaldason, 2002). The northern perimeter of the caldera is defined by a block called Norðurfjöll (Stop 7.2). In total, the 10 ka caldera is ~8 km in diameter and has a footprint of ~45 km2.
As you walk south to Öskjuvatn, the bedrock beneath your feet is caldera-fill basaltic lava that predates the 1875 eruption (Sigurdsson and Sparks, 1981). The tephra blanketing the bedrock will gradually transition from being dominantly mafic in the north (from the 1961 basaltic eruption) to dominantly felsic in the south (from the 1875 rhyolitic eruption). This transition will become stark after ~1.5 km of hiking on the established trail (65.0529°N, 16.7283°W). The 1875 pumice spans a range of light colors, from golden to white (Fig. 98A). The careful observer will notice abundant pumice with evidence of pre-eruptive (potentially eruption-triggering), molten, mafic-felsic magma mingling (Fig. 98B), as well as leucocratic xenoliths that are trondhjemitic in composition (Fig. 98C; Sigurdsson and Sparks, 1981). In cold summers, these eruptive products may be obscured from view by snow (Fig. 91).
After ~2.5 km of hiking south on an established path, if clouds and fog permit, you will encounter a stunning view of Öskjuvatn (Askja Lake) and Víti (Fig. 99). Hike to the northern rim of Víti (Hell; 65.0478°N, 16.7245°W), then take the perimeter trail to visit its southern rim to get a closer (~0.3 km one way, ~0.6 return). You may also walk down to the shore of Öskjuvatn to visit an exposed outcrop of 1875 tephra on the small peninsula positioned downhill and to the south-southeast of Víti, ~0.6 km from the Víti-Öskjuvatn overlook (65.0438°N, 16.7247°W; Fig. 100). On your walk down to the lake, if seasonal conditions permit, you may encounter swathes of small, closely spaced, gravel pyramids (Fig. 101). This gravel was initially concentrated in “suncups” in melting snow and ice, inverting their geometry (from cups of gravel to piles of gravel) as snow melt intensified.
Öskjuvatn is a caldera positioned within a larger system of nested Askja calderas. The Öskjuvatn caldera has dimensions of 3 × 4 km and a depth of 267 m (Sigurdsson and Sparks, 1981). The caldera-fill lake has an area of 11 km2, a maximum depth of 220 m, and an average depth of 114 m (Gylfadóttir et al., 2017).
The main event associated with the formation of the Öskjuvatn caldera is a Plinian eruption of rhyolite that occurred in 1875. Historical accounts (e.g., Thoroddsen, 1913, summarized by Carey et al., 2009) explain that a “dry” (non-phreatomagmatic) subplinian ashfall event began at ~9:00 p.m. local time on 28 March 1875. The next morning, at ~5:30 a.m., a “wet” phreatoplinian eruption began and lasted for approximately one hour. After a brief respite of ~30 minutes, a Plinian phase initiated, as determined by renewed pumice fall in eastern Iceland. Víti—the small, ~250 m maximum dimension, crater—was created by a phreatomagmatic eruption shortly after the Plinian stage waned (Thordarson and Höskuldsson, 2014). This eruption event was unique for its abrupt transitions between “wet” (phreatomagmatic) and “dry” (non-phreatomagmatic) explosive styles. These eruptive-style transitions were accompanied by fall deposits and surge deposits, which can be distinguished in the rock record (Carey et al., 2009). Within ~8.5 hours the majority of the 0.8 km3 tephra (0.2 km3 dense rock equivalent) had been explosively expelled from Askja (Sigurdsson and Sparks, 1981). In total, the explosive eruptive event concluded after 17 hours. Tephra was dispersed to the east due to prevailing wind direction, and proximal tephra deposits in the caldera itself were deposited on seasonal snow and ice (Carey et al., 2009).
The initiation of the volcanic unrest preceding the caldera-forming eruption of 1875 can be traced to February 18 of that same year, when a basaltic eruption began at the Sveinagjá fissure swarm ~70 km north of Askja. This fissure eruption continued for six months, ultimately erupting 0.3 km3 of basaltic lava (Sigurdsson and Sparks, 1981). The concurrent eruption at Askja and Sveinagjá is one of many lines of evidence of the connectivity between central volcanoes and fissure swarms in Iceland (also observed at Torfajökull-Veiðivötn, Grímsvötn-Laki, Krafla caldera and fissures, and many other volcanic systems).
Following the 1875 eruption (and subsequent activity associated with the Askja Fires) the volume of caldera subsidence at Öskjuvatn outpaced the volume of magma ejected by a factor of 10 (Thordarson and Höskuldsson, 2014). Öskjuvatn is down-dropped from the Þorvaldsfjall caldera rim block, and is bounded on its southern and eastern margin by a fault scarp up to 5 km long and 400 m high (Sigvaldason, 2002). It took 44 years after the 1875 Plinian event for Öskjuvatn to reach its current dimensions, progressively collapsing in a stepwise fashion. It took ~30 years for the caldera depression to fill with water to the level we observe in the crater lake today (Carey et al., 2009; Hartley and Thordarson, 2012; Thordarson and Höskuldsson, 2014; Gylfadóttir et al., 2017).
Öskjuvatn remains dynamic in the modern era, and its hazards are not limited to eruptive activity. In July of 2014 a large rockslide originated on the southern caldera rim, and ~107 m3 of rock entered the lake from a height of 150–370 m above lake-level. The result was a tsunami that traveled ~3 km across the lake in 1–2 minutes and had a 60–80 m runup on the northern shore (where tourists hike and appreciate the view). Following this rockslide event, the overall elevation of the water level in Öskjuvatn has increased by ~0.5 m. The rockslide scar is visible on the caldera rim on the southeastern margin of the lake (Gylfadóttir et al., 2017).
Stop 7.2. 1961 Lavas and the Outer Caldera Rim
The Vikraborgir parking lot is established in the midst of the 1961 lava field, at the base of cinder cones marking the fissure from which the flows emanated (Fig. 97). Because this lava field was emplaced in 1961, we have the benefit of learning from trained geologists who observed the eruption and recorded its progression with great detail. Specifically, Sigurdur Thórarinsson and Gudmundur Sigvaldason released a report in 1962 that explained that the eruption started on 26 October 1961, after approximately two weeks of solfatara activity at the eruptive site. The eruption continued for five weeks, and ultimately covered an area of 11 km2. The eruption initiated with a 0.7-km-long fissure, oriented east-west. The opening phase of the eruption was dramatic, with lava fountains lofting spatter 500 m in the air. Within just 10 hours of the eruption starting, the lava flow had reached a length of 7.5 km and an area of 6 km2. The initial phase of lava flows, which continued for 10 days (until 5 November), were ‘a‘ā. This ‘a‘ā phase was then followed by more fluid pāhoehoe flows. The transition from ‘a‘ā to pāhoehoe was accompanied by mineralogical changes; notably, the early phase of the eruption was phenocryst poor with plagioclase and pyroxene but no olivine. The later phase of the eruption included olivine, and a general increase in phenocryst content (Thórarinsson and Sigvaldason, 1962). These olivine-bearing pāhoehoe flows are prevalent near the Vikraborgir parking lot.
To begin your exploration of the 1961 lava field, first look for a sign pointing north to Dyngjufell (Fig. 102). From this location, descend to explore the young, well-preserved, lava flow features. With time and attention to detail you will encounter ropy pāhoehoe, slabby pāhoehoe, and entrail pāhoehoe (Fig. 103). This lava field also contains several examples of drained lava tubes (Fig. 104) and ‘a‘ā (Fig. 105).
If time permits, you may choose to walk ~1 km across the lava (a slow and careful process) to visit the 40-m-high caldera wall (Fig. 106). Alternatively, the caldera rim can be appreciated from a distance. The caldera wall is the southern perimeter of a landform called Norðurfjöll. The rocks of the caldera margin are dominantly hyaloclastites (erupted subglacially), though there are also instances of subaerially emplaced basalts that erupted during interglacial periods (e.g., at Vesturfjöll on the western margin of the caldera; Sigvaldason, 2002). Rare pockets of rhyolitic tephra from an eruption that occurred at Askja 10,000 years ago have been reported at various sites around the Askja caldera, including two sites at Norðurfjöll (in crater-like depressions, interlayered with basaltic tephra, and paved over by lava flows; Sigvaldason, 2002). Caldera collapses are often associated with explosive rhyolitic eruptions (e.g., 1875 Askja and the Öskjuvatn caldera). An alternative explanation has been proposed, however, to explain caldera formation of the ca. 10 ka Askja caldera. A study conducted by Sigvaldason (2002) reveals tectonic evidence for uplift of Dyngjufjöll central volcano (e.g., steep faulting at the caldera margin, uplift of caldera margin blocks, etc.). He attributed this uplift to depressurization of the crust in response to deglaciation, as well as to overpressurization of pre-eruptive magma. This deglaciation, magmatic overpressurization, and subsequent crustal uplift may have contributed to volcanic productivity at Askja (e.g., caldera fill lavas). These lava flows would have simultaneously depleted the subterranean magma reservoir, and overloaded the overriding crust, promoting down sagging of the crust and the depressed caldera topography visible at Askja today.
The 2019 GSA Northeastern Section guided field trip planned for an out-and-back excursion to Askja. Upon completing your time at Öskjuvatn (Stop 7.1) and the 1961 lava field (Stop 7.2), trace your steps back to Reykjahlíð. This entails descending on F894 for ~8 km to the Dreki Hut, following F910 ENE for ~12.5 km to the junction with F88, then traveling north on F88 for ~80 km until intersecting with Route 1. If you are returning to Reykjahlíð, turn left (west) and travel the 33 km back to town.
If you are using this field guide to explore Iceland on your own schedule, you may wish to reserve (in advance) a spot at either the Dreki Hut or the Thorsteinsskáli Hut and campground in order to spend more time exploring the northern interior. Additional destinations to consider include hiking at Herðubreið or Kollóttadyngja (Fig. 94), or venturing farther south to Kverkfjöll (Fig. 107) or the 2014–2015 Holuhraun lava fields (Fig. 108).
DAY 8. MÝVATN, KRAFLA, AND TJÖRNES
Today we will travel a loop starting and ending at Reykjahlíð (Fig. 109). The circuit will start with an overview of the geologic features in the immediate Lake Mývatn area from the rim of the maar Hverfjall. We will then drive into the Krafla caldera, which is the site of one of the more recent effusive eruptions in Iceland (1975–1984 Krafla Fires), and also a high-temperature geothermal field and power plant. Our route will take us to two spectacular features along the river Jökulsá á Fjöllum: the great falls of Dettifoss, and the scoured shallow intrusions of Hljóðaklettar. The remaining stops will focus on tectonic features: a rift graben formed rapidly during the Krafla Fires, and the Húsavík faults, which are part of the Tjörnes Fracture Zone.
From Reykjahlíð Route 1 goes around Mývatn to the north and west, and Route 848 goes around the lake on the east and south. From the intersection with Route 1 in Reykjahlíð, take Route 848 ~2.1 km to a left (east) turn signed for Hverfjall (65.6106°N, 16.9173°W). Take the dirt road to Hverfjall, bending left (northeast) as it encounters the cone. Park in the established parking area, proceed to the trailhead, and make your way up the short but steep trail to the top of Hverfjall, Stop 8.1.
Stop 8.1. Hverfjall
Hverfjall, a maar with relatively low and gentle slopes and a wide-mouthed crater, is a prominent landform standing ~100 m above its surroundings near the eastern shore of Mývatn (Fig. 110). The construction of Hverfjall occurred ~2500 years ago in an event called the Hverfjall Fires (Sæmundsson, 1991; Mattsson and Höskuldsson, 2011). The eruptive activity was concentrated along a NE-oriented fissure associated with the Krafla central volcano, restricted to the ~1.5-km-wide Grjótagjá-Krummaskarð graben (Mattsson and Höskuldsson, 2011).
Until recently, it was thought that the formation of Hverfjall could be attributed to discrete activity at a single vent (e.g., Sæmundsson, 1991). However, a relatively new interpretation of the local landscape and volcanic stratigraphy identifies three discrete eruptive phases associated with multiple vents (Mattsson and Höskuldsson, 2011). First, there was sustained phreatomagmatic explosivity from the Hverfjall vent in proto-Mývatn, a shallow lacustrine environment. Diatoms thought to originate from a lake environment are distributed throughout the Hverfjall tephra, supporting the idea of eruption occurring through a paleolake. Tephra was deposited close to the Hverfjall vent, falling out from a low, sustained eruption plume. This fallout created the Hverfjall tephra ring. A second phase of eruption took place with subaerial eruptions of scoria and lava flows. This subaerial eruptive activity led to the construction of two overlapping scoria cones at Jarðbaðshólar, ~3.6 km north of Hverfjall, near the present-day location of the commercial Mývatn Nature Baths. The associated subaerial lava flows flowed to the west and reached the present-day shoreline of Mývatn. These subaerial eruptions likely depleted the magma available to feed the phreatomagmatic activity centered at the Hverfjall vent, and activity began to wane. No longer able to sustain an explosive phreatomagmatic column, the Hverfjall eruption entered its third and final stage. Base surges (some “wet” involving particle transport in the presence of liquid water, and others “dry” involving particulates in steam) dominate the final deposits emplaced at Hverfjall (Mattsson and Höskuldsson, 2011). As you explore Hverfjall, look for evidence of wet surge deposits, such as accretionary lapilli found in outcrops just inside the rim of the crater at the top of the trail and sporadically at the same level (Fig. 111).