The concept of predominantly thermal mantle plumes
has evolved to thermochemical plumes that incorporate
both thermal and chemical heterogeneities in the mantle
(e.g., Davaille et al., 2002, 2003, 2005; Farnetani and
Samuel, 2005; Lin and van Keken, 2006a, b). Models of
thermochemical plumes suggest that upwelling characteristics depend on the buoyancy ratio, which is the ratio of
the stabilizing chemical density anomaly to the
destabilizing thermal anomaly. Low ratios suggest the
absence of a plume tail, and high ratios suggest long-lived
hot spot tracks.
In contrast to solely deep Earth processes accounting
for intraplate magmatism, an exogenous cause, namely,
impacts of large bolides, has also been postulated to instigate such magmatism (e.g., Rogers, 1982; Glikson, 1999;
Abbott and Isley, 2002; Jones et al., 2002, 2005; Muller,
2002; Ingle and Coffin, 2004). Large impacts in the ocean
basins can excavate some or all of the lithosphere and
potentially the uppermost portions of the asthenosphere,
thereby inducing voluminous decompressional melting
of mantle material. Oblique impacts may impart high
shear to the core-mantle boundary, triggering one or more
mantle avalanches and mantle plumes.
Thus, whether caused by endogenous or exogenous processes, and whether originating at the core-mantle boundary, the 670-km discontinuity, or within the upper mantle,
intraplate magmatism results from decompressional melting of mantle material as it ascends toward the Earth’s crust
through the asthenosphere and lithosphere.
Nature and source of intraplate magmatism type
examples
The three highly active intraplate hot spots described
earlier – Hawaii, Réunion, and Heard islands
(Figure 1) – are classified by Courtillot et al. (2003) either
as originating (Hawaii, Réunion) or possibly originating,
with more data (Heard), at the core-mantle boundary
(Figure 2). By implication, this implies that the entire
Hawaiian-Emperor chain (Figure 1) was created by a hot
spot with its origin at the core-mantle boundary. Louisville
Ridge (Figure 1) is classified similarly.
In stark contrast, petit spots appear to be sourced
entirely within the upper mantle (Figure 2). They form
where the outer rise of a subducting plate flexes and
develops normal faults, providing pathways for mantle
to ascend and melt via decompression (e.g., Hirano
et al., 2008). Melt volumes are orders of magnitude less
than hot spots such as Hawaii, Réunion, and Heard.
The upper mantle is also likely to be the source for the
vast majority of the tens of millions of intraplate seamounts rising >100 m above surrounding seafloor that
aren’t petit spots (Figure 2). Fissures and faults allow
anomalously warm and/or wet mantle to ascend and melt
via decompression, producing these features (e.g.,
Natland and Winterer, 2005). Faults may also influence
where seamounts form that possibly originate from deeper
in the mantle (e.g., McNutt et al., 1989).
The sheer volumes of igneous rock comprising the
Ontong Java, Manihiki, Hikurangi, and Caribbean flood
basalt provinces (Figure 1) would suggest an origin in
the lower mantle (Figure 2; e.g., Coffin and Eldholm,
1994; Eldholm and Coffin, 2000). Evidence is equivocal
for the Louisville hot spot being the source of Ontong
Java, Manihiki, and Hikurangi flood basalts (e.g.,
Mahoney et al., 1993; Neal et al., 1997; Richards et al.,
1989; Riisager et al., 2003; Antretter et al., 2004; Chandler
et al., 2012). The Caribbean flood basalts have been linked
to the Galapagos hot spot (e.g., Duncan and Hargraves,
1984; Kerr et al., 1997; Sinton et al., 1998), which has
been classified as possibly originating, with more data,
at the core-mantle boundary (Figure 2; Courtillot et al.,
2003).
Models for the formation of the Ontong Java Plateau
include various plume incarnations (e.g., Bercovici and
Mahoney, 1994; Farnetani and Richards, 1994; Larson,
1997; Neal et al., 1997; Ito and Clift, 1998; Tejada et al.,
2004; Hoernle et al., 2010), bolide impact (e.g., Rogers,
1982; Jones et al., 2002, 2005; Ingle and Coffin, 2004),
and upwelling eclogite (Korenaga, 2005). None of the
models to date satisfy all of the observations (e.g., Kerr
and Mahoney, 2007). Plume models have also been
invoked for the Manihiki and Hikurangi plateaus (e.g.,
Larson, 1997; Ito and Clift, 1998; Hoernle et al., 2010).
In contrast to Ontong Java, plume models seem to satisfy
most observations from these two plateaus. The dismembered nature of the Caribbean flood basalt province
means that it doesn’t lend itself readily to modeling; however, to date only the plume model has been invoked for its
formation (e.g., Duncan and Richards, 1991; Kerr et al.,
1997).
Summary
Intraplate magmatism in the ocean occurs over many
orders of magnitude in scale, and the predominant volcanic rock type is basalt. Type examples of robust active
intraplate magmatism – the islands of Hawaii, Réunion,
and Heard, all probably sourced from the core-mantle
boundary – show a general pattern of alkaline
pre-shield, tholeiitic shield, and alkaline post-shield volcanism. Classic seamount chains such as the HawaiianEmperor and Louisville, also probably sourced from the
core-mantle boundary, display differing degrees of elemental and isotopic compositional variations, with the
Hawaiian-Emperor chain quite heterogeneous and Louisville rather homogeneous. The massive oceanic plateaus (Ontong Java, Manihiki, Hikurangi) and ocean
basin flood basalts (Caribbean, East Mariana, Nauru,
Pigafetta), all likely sourced from the core-mantle
boundary, although perhaps instigated by exogenous
mechanisms, range from homogeneous (e.g., Ontong
Java) to heterogeneous (e.g., Caribbean) in their elemental and isotopic compositions. Finally, the tens of millions of seamounts, including petit spots, that rise
>100 m above surrounding seafloor probably originate
376
INTRAPLATE MAGMATISM
has evolved to thermochemical plumes that incorporate
both thermal and chemical heterogeneities in the mantle
(e.g., Davaille et al., 2002, 2003, 2005; Farnetani and
Samuel, 2005; Lin and van Keken, 2006a, b). Models of
thermochemical plumes suggest that upwelling characteristics depend on the buoyancy ratio, which is the ratio of
the stabilizing chemical density anomaly to the
destabilizing thermal anomaly. Low ratios suggest the
absence of a plume tail, and high ratios suggest long-lived
hot spot tracks.
In contrast to solely deep Earth processes accounting
for intraplate magmatism, an exogenous cause, namely,
impacts of large bolides, has also been postulated to instigate such magmatism (e.g., Rogers, 1982; Glikson, 1999;
Abbott and Isley, 2002; Jones et al., 2002, 2005; Muller,
2002; Ingle and Coffin, 2004). Large impacts in the ocean
basins can excavate some or all of the lithosphere and
potentially the uppermost portions of the asthenosphere,
thereby inducing voluminous decompressional melting
of mantle material. Oblique impacts may impart high
shear to the core-mantle boundary, triggering one or more
mantle avalanches and mantle plumes.
Thus, whether caused by endogenous or exogenous processes, and whether originating at the core-mantle boundary, the 670-km discontinuity, or within the upper mantle,
intraplate magmatism results from decompressional melting of mantle material as it ascends toward the Earth’s crust
through the asthenosphere and lithosphere.
Nature and source of intraplate magmatism type
examples
The three highly active intraplate hot spots described
earlier – Hawaii, Réunion, and Heard islands
(Figure 1) – are classified by Courtillot et al. (2003) either
as originating (Hawaii, Réunion) or possibly originating,
with more data (Heard), at the core-mantle boundary
(Figure 2). By implication, this implies that the entire
Hawaiian-Emperor chain (Figure 1) was created by a hot
spot with its origin at the core-mantle boundary. Louisville
Ridge (Figure 1) is classified similarly.
In stark contrast, petit spots appear to be sourced
entirely within the upper mantle (Figure 2). They form
where the outer rise of a subducting plate flexes and
develops normal faults, providing pathways for mantle
to ascend and melt via decompression (e.g., Hirano
et al., 2008). Melt volumes are orders of magnitude less
than hot spots such as Hawaii, Réunion, and Heard.
The upper mantle is also likely to be the source for the
vast majority of the tens of millions of intraplate seamounts rising >100 m above surrounding seafloor that
aren’t petit spots (Figure 2). Fissures and faults allow
anomalously warm and/or wet mantle to ascend and melt
via decompression, producing these features (e.g.,
Natland and Winterer, 2005). Faults may also influence
where seamounts form that possibly originate from deeper
in the mantle (e.g., McNutt et al., 1989).
The sheer volumes of igneous rock comprising the
Ontong Java, Manihiki, Hikurangi, and Caribbean flood
basalt provinces (Figure 1) would suggest an origin in
the lower mantle (Figure 2; e.g., Coffin and Eldholm,
1994; Eldholm and Coffin, 2000). Evidence is equivocal
for the Louisville hot spot being the source of Ontong
Java, Manihiki, and Hikurangi flood basalts (e.g.,
Mahoney et al., 1993; Neal et al., 1997; Richards et al.,
1989; Riisager et al., 2003; Antretter et al., 2004; Chandler
et al., 2012). The Caribbean flood basalts have been linked
to the Galapagos hot spot (e.g., Duncan and Hargraves,
1984; Kerr et al., 1997; Sinton et al., 1998), which has
been classified as possibly originating, with more data,
at the core-mantle boundary (Figure 2; Courtillot et al.,
2003).
Models for the formation of the Ontong Java Plateau
include various plume incarnations (e.g., Bercovici and
Mahoney, 1994; Farnetani and Richards, 1994; Larson,
1997; Neal et al., 1997; Ito and Clift, 1998; Tejada et al.,
2004; Hoernle et al., 2010), bolide impact (e.g., Rogers,
1982; Jones et al., 2002, 2005; Ingle and Coffin, 2004),
and upwelling eclogite (Korenaga, 2005). None of the
models to date satisfy all of the observations (e.g., Kerr
and Mahoney, 2007). Plume models have also been
invoked for the Manihiki and Hikurangi plateaus (e.g.,
Larson, 1997; Ito and Clift, 1998; Hoernle et al., 2010).
In contrast to Ontong Java, plume models seem to satisfy
most observations from these two plateaus. The dismembered nature of the Caribbean flood basalt province
means that it doesn’t lend itself readily to modeling; however, to date only the plume model has been invoked for its
formation (e.g., Duncan and Richards, 1991; Kerr et al.,
1997).
Summary
Intraplate magmatism in the ocean occurs over many
orders of magnitude in scale, and the predominant volcanic rock type is basalt. Type examples of robust active
intraplate magmatism – the islands of Hawaii, Réunion,
and Heard, all probably sourced from the core-mantle
boundary – show a general pattern of alkaline
pre-shield, tholeiitic shield, and alkaline post-shield volcanism. Classic seamount chains such as the HawaiianEmperor and Louisville, also probably sourced from the
core-mantle boundary, display differing degrees of elemental and isotopic compositional variations, with the
Hawaiian-Emperor chain quite heterogeneous and Louisville rather homogeneous. The massive oceanic plateaus (Ontong Java, Manihiki, Hikurangi) and ocean
basin flood basalts (Caribbean, East Mariana, Nauru,
Pigafetta), all likely sourced from the core-mantle
boundary, although perhaps instigated by exogenous
mechanisms, range from homogeneous (e.g., Ontong
Java) to heterogeneous (e.g., Caribbean) in their elemental and isotopic compositions. Finally, the tens of millions of seamounts, including petit spots, that rise
>100 m above surrounding seafloor probably originate
376
INTRAPLATE MAGMATISM