related to a unique mantle plume that has existed for more than 60 My (Figure 1.6). Iceland could thus be interpreted as a residual plume axis (tail) whose “cap” is cooled.
Figure 1.6. Seismic imaging of the upper mantle at the Icelandic hot spot
COMMENT ON FIGURE 1.6.– Imagery illustrating the narrow cylindrical shape of the mantle plume (A), with a radius of about 150 km (low velocity anomaly to a depth of at least 400 km) (from Wolfe et al. 1997). Vertical sections in the ICEMAN model by Allen et al. (2002) perpendicular (B) and parallel (C) to the rift. These sections illustrate the vertical conduit between 400 and 200 km deep and the horizontal plume above.
However, in spite of the numerous studies carried out over a period of 50 years, knowledge about hot spots and the internal phenomena that cause them is still very incomplete. Numerous controversies remain (Courtillot et al. 2003; Campbell 2007). The spatial stability of hot spots is also questioned (Tarduno 2010) and the very reality of deep plumes is still debated (Anderson 2001; Montagner 2010).
As regards the Icelandic hot spot, no satisfactory model is currently available to account for all the very large amount of structural, tectonic, geochronological and petro-geochemical data available. The basic problem concerns the relationships between (i) a geochemical signature, suggesting an origin from the underlying deep mantle, (ii) the extensive deformation related to the ridges adjacent to and across the island, inducing adiabatic fusion at greater or lesser depths in the mantle, and finally (iii) the inheritance and recycling of ancient geological structures affecting the crust, such as the ancient oceanic suture and mantle in this particular area of the Atlantic, which may pollute the observed geochemical signatures.
The first oppositions to the deep plume model appeared in the early 2000s with the work of Foulger and her collaborators (Foulger et al. 2001, 2005; Foulger and Anderson 2005). For these authors, the large amount of magma produced in Iceland is thought to originate from the reworking of the subducted and eclogitized oceanic crust of the ancient Iapetus Ocean.
Several conferences, organized by pro- and/or anti-plume groups, were devoted to these issues during the first decade of the 21st century (in particular, a Penrose Conference held in Iceland in 2003, and the Great Plume Debate in Scotland in 2005), but the debate is far from over (Campbell and Kerr 2007; Meyer et al. 2007; Foulger 2012). Recent studies are still rehashing cards and, even among “pro-plume” authors, the spatial and temporal evolution of the latter is still the subject of work (Barnett-Moore et al. 2017; Martos et al. 2018). We will review the latest developments concerning the Icelandic hot spot in Chapter 3 and Chapter 1 of Volume 2.
1.2.4. The Greenland–Iceland–Faroe Ridge
The Greenland–Iceland–Faroe Ridge (GIFR) consists of a shoal interspersed with three fairly deep channels (about 2,000 m), which steers the North Atlantic from Greenland to the Faroe Islands, leaning against Iceland. It forms a relief of crucial importance not only as a recording of the last phases of the opening of the North Atlantic, but also for understanding the formation of the current thermohaline circulation, one of the major drivers of the evolution of our climate.
The GIFR crust has an anomalous thickness, ranging from 25 to 40 km, equivalent to that of a continental crust. It is not associated with any interpretable magnetic anomalies except locally between Iceland and the Faroes (Nuuns et al. 1983; Hjartarson et al. 2017).
1.3. Geodynamic characteristics of Iceland
One of the characteristics of Iceland is its sustained volcanic activity, accompanied by hydrothermal activity and almost permanent seismicity, although most of the time it is of low intensity and localized. Another major feature of the island lies in the presence of large glaciers and all that it implies from a geodynamic, climatic, touristic and industrial point of view.
1.3.1. Seismicity
While seismic activity in Iceland underlines all plate boundaries (Figure 1.7), it is less continuous and less strong in the rift and is mainly concentrated in the transform zones that connect the active rift to the offshore segments of the ridge to the north and south of the island.
Figure 1.7. Seismicity in Iceland recorded by the SIL network between 1994 and 2007 (modified from Jakobsdóttir 2008). Volcanic systems according to Einarsson and Sæmundsson (1987). EVZ: East Volcanic Zone; NVZ: North Volcanic Zone; WVZ: West Volcanic Zone; ZFT: Tjörnes Fracture Zone; SISZ: South Iceland Seismic Zone
The Icelandic Rift corresponds to the onshore extensions of two segments of the North-MAR, the Reykjanes Ridge to the south and the Kolbeinsey Ridge to the north (Figures 1.2 and 1.5). It has three branches (Figure 1.7), all offset eastward from the axis of the North-MAR. These branches of the rift consist of fissure swarms associated with active central volcanoes, a significant portion of which are located beneath the major glaciers of the island; they can be guessed from the ice-free topographic map (Figure 1.2(b)). In the rift, seismicity is strongly linked to the functioning of magmatic systems (Figure 1.7). This seismicity of volcanic origin and the volcanism itself (for example, the recent eruption of the Bárðarbunga volcano and its large flow of basaltic lava) testify to an important active magmatism supposed to be fed by the Icelandic hot spot.
The shift of the Icelandic rift with respect to the axis of the North-MAR is accommodated by two transform zones, the South Iceland Seismic Zone (SISZ) to the south and the Tjörnes Fracture Zone (TFZ) to the north. Both are inherited from an older history, that of the opening of the Atlantic (Chapter 3) and linked to the “rift jump” phenomenon (Chapter 2). It is in these two transform zones that seismicity is the most sustained (Figure 1.7) with both quasi-permanent microseismicity and major recurrent seismicity whose magnitude can exceed 7 (section 2.2.1).
The disposition of the two rift branches in the south of the island (WVZ and EVZ) resembles that of the overlapping spreading centers (OSC) described on fast-spreading ridges, rather than that of a true transform zone. Nevertheless, the associated seismicity characterizes well the left-lateral transform motion of the SISZ, connecting the EVZ to the Reykjanes Ridge. At present, the two branches of the rift do not have the same level of activity, and the WVZ accommodates only a small part of the extension; the question of whether this present low activity is representative of the last thousands or millions of years will be discussed in Chapter